Y. Kamada

The addition of high power, low aspect ratio data from the NSTX and MAST experiments have motivated a new investigation of the effect of aspect ratio on confinement scaling. Various statistical methods, including those that incorporate estimates of measurement error, have been applied to datasets constrained by the standard set of criteria in addition to the range of ? and M(sub)eff appropriate to ITER operation. Development of scalings using engineering parameters as predictor variables results in ?-scaling coefficients that range from 0.38 to 1.29; the transformation of these scalings to physics variables results in an unfavorable dependence of ?? on ?, but a favorable dependence on ?. Because the low aspect ratio devices operate at low ?(sub)T and therefore high ?(sub)T, a strong correlation exists between ? and ?, and this makes scalings based on physics variables imprecise.

Our understanding of the physics of internal transport barriers (ITBs) is being advanced by analysis and comparisons of experimental data from many different tokamaks worldwide. An international database consisting of scalar and two-dimensional profile data for ITB plasmas is being developed to determine the requirements for the formation and sustainment of ITBs and to perform tests of theory-based transport models in an effort to improve the predictive capability of the models. Analysis using the database indicates that: (a) the power required to form ITBs decreases with increased negative magnetic shear of the target plasma, and: (b) the E×B flow shear rate is close to the linear growth rate of the ion temperature gradient (ITG) modes at the time of barrier formation when compared for several fusion devices. Tests of several transport models (JETTO, Weiland model) using the two-dimensional profile data indicate that there is only limited agreement between the model predictions and the experimental results for the range of plasma conditions examined for the different devices (DIII-D, JET, JT-60U). Gyrokinetic stability analysis (using the GKS code) of the ITB discharges from these devices indicates that the ITG/TEM growth rates decrease with increased negative magnetic shear and that the E×B shear rate is comparable to the linear growth rates at the location of the ITB.

The `stiffness' of thermal transport in ELMy H-modes is examined in a series of carefully chosen JT-60U plasmas, and measured temperatures are compared with the predictions of several transport models. A heating power scan with constant T_ped, a scan of pedestal temperature, T_ped, with constant heating power, and an on-axis/off-axis heating comparison are presented. In the power scan, a 45% increase in heating (and a 12% density rise) produces an approximately fixed core temperature profile in a group of five plasmas with the same pedestal temperature. With fixed heating power, we find that a 30–40% increase in T_ped is associated with similar increases in core temperature. Heating in the deep core is varied by employing different groups of neutral beams that deposit their power near the magnetic axis and farther from the axis. In these plasmas on-axis heating produces slightly more peaked temperature profiles although they have 60% more heating power inside r = a/2. Transport models are tested by solving the power balance equations to predict temperatures, which are then compared to the measurements. Predictions of the RLWB and IFS/PPPL models generally agree with the measured temperatures outside r~0.3a, but the Multimode model uniformly predicts temperatures that are too high except in the central region. Tests based on these discharges are not able to discriminate between the transport models of varying stiffness, so we conclude that larger changes are needed in the P_heat and T_ped scans.

The effects of the conductive heat flow on the energy transport concerning the temperature profile similarity have been analysed for (\ref{eq1}) high triangularity and (2) Ar seeded ELMy H-mode plasmas on JT-60U. The temperature profiles of the plasma core on the logarithmic plot are shape preserving and independent of the varying pedestal height. As the conductive heat flux increases, the effective heat conductivity, χ_eff, adjusts itself to sustain a constant ∇(ln T_i) in the plasma core. In high triangularity discharges, highly improved energy confinement quality accompanied by high core temperature is obtained when the boundary temperature rises with an increase in β_pol. On the other hand, the peaked density profiles in Ar seeded discharges allow the temperature at the pedestal shoulder to be higher relative to the cases without Ar gas injection, and the core temperature is also raised. As the conductive heat flux is reduced by increasing the radiation loss flux in Ar seeded H-mode plasmas, χ_eff decreases with a similar ∇(ln T_i) of the plasma core in the case without Ar gas injection. Improved particle confinement due to Ar gas puff raises the achievable density and contributes to the reduction in χ_eff.

With the high-field-side multiple pellet injection into high-triangularity high-β_p type I ELMy H-mode discharges, JT-60U has extended the density range with a favourable integrated performance. The pellet injected discharge has a high pedestal pressure compared with the gas fuelled discharges, and the pedestal temperature does not decrease even at high pedestal density. At high triangularity, the pedestal β_p and the edge α-parameter increase with increasing total β_p. Possible linkage among pedestal and core parameters has been proposed based on a variety of JT-60U experiments. Grassy ELMy discharges have been achieved at the low safety factor regime q₉₅<4 for the first time in JT-60U with high triangularity (~0.6). The full non-inductive operation has been demonstrated with grassy ELMs.

High-triangularity ELMy H-mode experiments (δ_x~0.45) have been carried out in JT-60U. In the low-density regime ( _e/n^GWx~0.20). In the high-density regime ( _e/n^GW>0.5), at fixed neutral beam heating power type-III ELMs are generated and the pedestal pressure reduces gradually with density whereas the ELM frequency at low triangularity is much lower at a given density. Compared to low-triangularity plasmas at low densities, higher pedestal temperature was obtained at high triangularity and the core temperature was also improved by almost the same factor. The profiles of temperature are stiff in the sense that there seems to exist a minimum scale length of temperature gradient. Since plasma shaping affects strongly the peripheral region of toroidal magnetic flux surfaces, the improvement of the edge stability by triangularity leads to higher pedestal temperature, which in turn raises the core temperature.

The 'stiffness' of thermal transport in ELMy H-modes [edge localized high-confinement modes] is examined in a series of carefully chosen JT-60U plasmas, and measured temperatures are compared with the predictions of several transport models. A heating power scan with constant T(subscript ''ped''), a scan of pedestal temperature, T(subscript ''ped''), with constant heating power, and an on-axis/off-axis heating comparison are presented. In the power scan a 45% increase in heating (and a 12% density rise) produces an approximately fixed core temperature profile in a group of five plasmas with the same pedestal temperature. With fixed heating power, we find that a 30-40% increase in T(subscript ''ped'') is associated with similar increases in core temperature. Heating in the deep core is varied by employing different groups of neutral beams that deposit their power near the magnetic axis and farther from the axis. In these plasmas, on-axis heating produces slightly more peaked temperature profiles, although they have 60% more heating power inside r = a/2. Transport models are tested by solving the power balance equations to predict temperatures, which are then compared to the measurements. Predictions of the RLWB and IFS/PPPL models generally agree with the measured temperatures outside r approximately 0.3a, butmore » the multimode model uniformly predicts temperatures that are too high except in the central region. Tests based on these discharges are not able to discriminate between the transport models of varying stiffness, so we conclude that larger changes are needed in the P(subscript ''heat'') and T(subscript ''ped'') scans.« less