Description
The dimensionally matched deuterium-tritium pulse pair was executed in the JET DT campaign under JET L-mode conditions to complement the earlier work on the dimensionless scaling between D and T [1]. Depending on the NBI power level, the experiment showed 13-16% improvement in the energy confinement time in favour of the tritium pulse. This favourable isotope scaling can also be seen in the effective heat diffusion coefficients at all plasma radii. The phase and amplitude profiles based on the density profiles from the profile reflectometry data in response to the gas puff modulation were virtually identical between the matched deuterium tritium plasma pair. Consequently, there is no room for any isotope scaling in the particle transport channel in the core plasma, at least at $ρ_{tor}$ ≤ 0.95. This can also be seen in the derived particle transport coefficients which are identical between the deuterium and tritium pulses [2]. To complement the absence of isotope scaling in the particle transport channel, the particle transport EDGE2D-EIRENE simulations found that the radial ionisation profiles at the plasma edge are very similar between the dimensionally matched deuterium and tritium identity pulses.
A similar deviation from the gyro-Bohm scaling, i.e. strong isotope scaling favouring tritium in the heat transport channel was found with gyrokinetic simulations in the edge at $ρ_{tor}$ = 0.95 under these same experimental JET L-mode conditions. According to GENE simulations, the edge isotope effect at $ρ_{tor}$ = 0.95 is much larger than that found in the core plasma $ρ_{tor}$ = 0.6. Supported both by the experiment and simulations, the isotope scaling originates dominantly from the electron heat transport channel and from the edge part of the plasma. Furthermore, both the experimental results and the GENE simulations suggest that the isotope mass scaling is a nonlinear function of the isotope mass itself, being significantly stronger between deuterium and tritium than between hydrogen and deuterium, at least in JET L-mode conditions. The study will be expanded to cover the mixed H-D-T plasmas.
[1] T. Tala et al., Nucl. Fusion 63, 112012 (2023).
[2] A. Salmi et al., Plasma Phys. Control. Fusion 65, 055025 (2023).