1. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 PWI aspects of the FAST (Fusion Advanced Studies Torus) project Presented by G. Maddaluno Outline Main objectives and parameters of the FAST project Modelling of the FAST core/SOL plasma and evaluation of the divertor heat loads. Assessment of the divertor heat loads by ELMs Conclusions Univ. of Rome Tor Vergata Univ. of Catania

2. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 FAST objectives FAST (Fusion Advanced Studies Torus) is the proposal of the Italian Association on Fusion for a satellite facility in the frame of the EU Accompanying Programme. It is conceived to meet EFDA programmatic missions 1 to 5 (Burning Plasmas, Reliable Tokamak Operation, First Wall Materials & compatibility with ITER/DEMO relevant Plasmas, Technology and Physics of Long Pulse & Steady State, Predicting Fusion Performance) in support of ITER towards DEMO by integrating a set of conditions that must be as close as possible to those expected on ITER, in terms of physics parameters as well as of technical terms. FAST parameters have been chosen to satisfy the following conditions: ITER relevant geometry; production and confinement of energetic ions in the half-MeV range in order to obtain the presence of dominant electron heating; large ratio between the heating power and the device dimensions to investigate the physics of large heat loads; pulse duration (normalized to the plasma current diffusion time) similar to that of ITER to study AT plasma scenarios.

3. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 FAST parameters Plasma Current (MA)6.5 B T (T)7.5 Major Radius (m)1.82 Minor Radius (m)0.64 Elongation k 95 1.7 Triangularity δ 95 0.4 Safety Factor q 95 3 (m -3 ) 2x10 20 Flat-top (s)13 H&CD power (MW)30 ICRH (60-80 MHz) 30 ECRH (170 GHz)4 LH (3.6 or 5 GHz)6 Port compatible with (45° inclined) NNBI 10 P/R (MW/m)22 Reference scenario (scaled)

4. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 FAST plasma wall interaction issues ITER relevant values of P/R (up to 22 MW/m, P = 40 MW, R = 1.82) All tungsten machine (Li divertor also considered) Impurity seeding (Ar, Ne) to mitigate divertor heat loads All actively cooled PFCs Design maximum heat load assumed = 18 MWm -2 Outer midplane power flux e-folding length p omp assumed = 0.005 m Closed divertor geometry (flux expansion factor at the target = 5)

5. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 Evaluation of the outer midplane power flux e-folding length From regression analysis on experimental power deposition profiles measured in JET H-mode discharges: λ TC q H-mode A(Z) 1.1 B φ0.9 q 95 0.4 P t0.5 n e,u 0.15 with A and Z the ion mass and charge, B φ the toroidal field, q 95 the safety factor, P t the outer target power and n e,u the upstream density on the separatrix λ p omp 1-2 10 -3 m From multi-machine scaling the application at the FAST H-mode scenario of a multi-machine scaling provides a value λ p omp 15 10 -3 m or 6.5 10 -3 m depending on the scaling being calculated with the measured power flux to the outer divertor or with the total input power. The average heat flux on the divertor has been calculated as q target [MW m -2 ] = f out P div cos p / (2 R out λ p target ), where f out is the fraction of P div flowing to the outer target (= 2/3), p is the tilt angle of the target in the poloidal cross section, assumed = 70° and R out = 1.6 m is the major radius of outer target

6. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 Modelling of the FAST core/SOL plasma To have reliable predictions of the thermal loads on the divertor plates and of the core plasma purity a number of numerical self- consistent simulations have been made for the H-mode and steady-state scenario by using the code COREDIV. The COREDIV code treats the coupled SOL-bulk system by imposing the continuity of energy and particle fluxes and of particle densities and temperatures at the separatrix. The code solves self-consistently radial 1D energy and particle transport of plasma and impurities in the core region and 2D multi-fluid transport in the SOL A simple slab geometry (poloidal and radial directions) with classical parallel transport and anomalous radial transport is used for the SOL and the impurity fluxes and radiation losses caused by intrinsic and seeded impurity ions are calculated fully self consistently.

7. PFC Material W W + 0.02% ArLi + 0.7% Ne Scenario H-mode reference Full NICD H-mode reference H-mode extreme H-mode reference I p (MA) 6.522 8 B T (T) 7.53.5 7.58.57.5 (10 20 m -3 ) 211.3252 P ADD (MW) 3040 304030 n sep ( 10 20 m -3 ) 0.730.290.400.761.750.92 Z eff 1.12.41.71.41.11.8 f RAD (%) 197264574631 T eplate (eV) 57867617634 P DIV (MW)22.79.212.211.717.817.1 q t (MWm -2 ) 20.68.311.110.616.115.5

8. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 Main results of COREDIV modelling In the H-mode reference scenario (I p = 6.5 MA, B T =7.5 T, = 2.0 10 20 m 3, P ADD = 30 MW) impurity seeding could reveal not essential, with a beneficial effect on the core Z eff ( 1), the outer divertor heat load exceeding only marginally the design value of 18 MW m -2. In the full NICD scenario (I p = 2.0 MA, B T = 3.5 T, = 1.0 10 20 m -3, P ADD = 40 MW), without impurity seeding, a slight increase (to 1.3 10 20 m -3 ) of the foreseen density is needed for reducing core Z eff to acceptable values (that is not possible with impurity seeding). In the extreme H-mode scenario (I p = 8.0 MA, B T = 8.5 T, = 5.0 10 20, P ADD = 40 MW), the impurity seeding is needed for decreasing the power flowing to the divertor, the core Z eff value staying always below 1.2.

9. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 Tungsten sputtering yields

10. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 Preliminary ELMs heat load assessment Assumptions: 1.ELM energy W ELM ~ 0.15 W PED [1] ~ 0.15 0.4 W TOT ; 2.for H-mode reference scenario, with n e /n eGW 0.3, W ELM 1.5 MJ; 3.all the ELM energy W ELM reaches the divertor; 4.the fraction of the ELM energy mostly contributing to material damage, i.e. the one deposited in short timescales, is about 40% for low collisionality [2]; 5.similar spatial deposition profile as inter-ELM and a factor 2 asymmetry in the in- out ELMs energy deposition 6.the heat deposition time depends on the parallel ion loss time, scaling according R/ T ped the energy density on the inner divertor is expected to be about 1.0 MJ m -2, to be compared with the threshold for damage (0.3 MJ m -2 ), scaled from the one adopted by ITER for avoiding too strong W erosion. [1] Loarte A. et al 2003 Plasma Phys. Control. Fusion 45 1549 [2] Eich et al 2005 J. Nucl. Mater. 337–339 669

11. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 W ELM /W ped vs. ped *

12. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 At low * the fraction of W ELM contributing to target damage is 40%

13. EU PWI TF- 7th General meeting –Frascati 27- 29/10/2008 Conclusions ITER & DEMO relevant plasma wall interactions regimes are achievable in FAST (P/R; ELMs); COREDIV simulations show that in all the foreseen scenarios steady state divertor heat loads can be kept under the design value while preserving plasma purity, allowing for impurity seeding when a larger fraction of radiated power is necessary. The preliminary assessment of ELMs power flux results in a divertor heat flux 3-4 times larger than the safe limit, a factor that can be recovered by the present mitigation tools. The involvement of the largest possible number of Associations is mandatory to realize FAST. If the FAST project were to go on, the skill and the expertise existing inside the EU PWI TF can play a key role in withstanding and solving the related plasma wall interaction problems.