Study of the double null divertor configuration in DTT

On the way to the development of a fusion reactor based on the Tokamak configuration, the Divertor Test Tokamak facility (DTT) [1] in construction in Italy should provide useful information for the DEMO [2] reactor in the field of the power and particle exhaust. DTT is designed to accept the Single Null divertor (SND) and also divertors optimized for all the present more promising configurations like the Snowflake divertor (SFD), the X divertor (XD), the Super-X (SXD), the X-point target (XPD) and the double null (DND). The DND in particular has gained a new attention as a DEMO candidate considering its ability to reduce the peak heat flux at the divertor targets splitting the power on twice the surface, but this geometrical advantage can in principle be overcome on the physical side by the shorter connection length and the additional engineering complications and costs associated to the need of a double divertor with its pair pumping systems. In this paper we present the analysis carried out for the DND configuration in DTT to evaluate its advantages/disadvantage with respect to the SND one in terms of pumping system. To study the engineering requirements of DND its power exhaust handling capability has been analysed both in the optimal case of two exactly symmetric divertors (in terms of pumping and main specie/seeding gas puffing locations) than in the simpler case of a secondary divertor without pumping. In all cases full tungsten divertors and wall have been considered and neon gas has been used as seeding impurity, the analysis has been done at the maximum DTT heating power of PTOT = 45 MW which corresponds to a PSOL?32 MW and at separatrix density between nsep = 6?1019 and nsep = 10?1019 m-3. In addition, the transport coefficients have been set up at the separatrix to provide an outer mid-plane heat flux decay length of 1.0 mm in SND, in agreement with the present Eich scaling [3] prediction at the previous DTT parameters. The SOLEDGE2D-EIRENE [4,5] edge code has been used for the analysis for its ability to deal with all configurations and to extend the fluid domain up to the first wall.