Edge-localized mode avoidance and pedestal structure in I-mode plasmas

Abstract

I-mode is a high-performance tokamak regime characterized by the formation of a temperature pedestal and enhanced energy confinement, without an accompanying density pedestal or drop in particle and impurity transport. I-mode operation appears to have naturally-occurring suppression of large ELMs in addition to its highly favorable scalings of pedestal structure and overall performance. Extensive study of the ELMy H-mode has led to the development of the EPED model, which utilizes calculations of coupled peeling-ballooning MHD modes and kinetic-ballooning mode (KBM) stability limits to predict the pedestal structure preceding an ELM crash.We apply similar tools to the structure and ELM stability of I-mode pedestals. Analysis of I-mode discharges prepared with high-resolution pedestal data from the most recent C-Mod campaign reveals favorable pedestal scalings for extrapolation to large machines -pedestal temperature scales strongly with power per particle P net /n e , and likewise pedestal pressure scales as the net heating power (consistent with weak degradation of confinement with heating power). Matched discharges in current, field, and shaping demonstrate the decoupling of energy and particle transport in I-mode, increasing fueling to span nearly a factor of two in density while maintaining matched temperature pedestals with consistent levels of P net /n e . This is consistent 2 with targets for increased performance in I-mode, elevating pedestal β p and global performance with matched increases in density and heating power. MHD calculations using the ELITE code indicate that I-mode pedestals are strongly stable to edge peeling-ballooning instabilities. Likewise, numerical modeling of the KBM turbulence onset, as well as scalings of the pedestal width with poloidal beta, indicate that I-mode pedestals are not limited by KBM turbulence -both features identified with the trigger for large ELMs, consistent with the observed suppression of large ELMs in I-mode.

Multi-device studies of pedestal physics and confinement in the I-mode regime

“…The instability causing these fluctuations is unclear. Similar to the C-Mod results [3], the pedestal p in I-mode is low, more than a factor of 4 below the critical gradient for peelingballooning mode stability reached before the ELMs in the H-mode phase as computed with ELITE [24], and also roughly a factor of 2 below the ideal infinite-n ballooning mode limit.…”

“…As can be seen, however, there is a wide range from 0.6<H 98,y2 <1.3, and the DIII-D discharges so far tend to be on the low end of the range, H 98,y2 up to 0.8. In both C-Mod and AUG,  E in I-mode does not degrade significantly with power as in H-mode [9,3,6,11], thus the best normalized and absolute performance is achieved when maximum power can be applied and pedestal pressures achieved. This trend contrasts with the  E~P -0.7 scaling of H 98,y2 .…”

“…However, C-Mod has demonstrated that the density can be increased by fueling after the transition to I-mode [9]. With sufficient input power, T ped can be maintained [3]. The limits in density, as for pressure, are generally set by I-H transitions.…”

“…Particle transport will also be needed to remove helium 'ash' produced by fusion reactions. Without a particle barrier, plasma densities can also be more readily controlled by fueling or pumping, important for burn control [3]. Another critical advantage over H-mode is that the regime is naturally ELM-free, giving stationary conditions for many energy confinement times  E without the need for active means to suppress ELMs.…”

20 years of research on the Alcator C-Mod tokamak

“…Stability calculations with ELITE are consistent with operation near the high-n or ballooning side of the peeling-ballooning stability diagram. 151 This combination of conditions exist on C-Mod in a restricted window in shaping (d U < 0.3, d L > 0.7, and j < 1.6).…”