Annual Report (MS-Word)
Effective operation of the new divertor has been obtained in all its design configurations i.e., the plasma current and the combined heating power respectively reached 6 MA and 32 MW. The experimental program proceeded as planned and addressed issues directly relevant to ITER in the correct geometry, with the right plasma parameters and in the relevant size range. The divertor has proved effective in handling the plasma exhaust heat load and so preventing the rapid impurity influx which previously terminated high performance pulses. Thus, with the new divertor configuration, operation with detached plasma and radiative power transfer in the divertor has been obtained. Here, it was found that nitrogen radiates more in the divertor than in the main plasma but leads to large values of Zeff at high power levels. At low powers argon can radiate 80% of the input power with Zeff = 2, but significant radiation occurs in the main plasma. Confinement is reduced with rapid ELMs to give H = 1.5, and this is true even with the heating power far above the H-mode threshold. Steady state operation has been obtained with ne, Zeff and Prad constant for 20 s at 2 MA and 9 s at 3 MA in ELMy H-mode discharges. Here, H-factor was two. In the high fusion performance campaign, nT within 10% of the JET best of 9 x 10 20 m- 3 keVsm-3 was obtained together with (1) the equivalent QDT around unity transiently (0.7 for 1 s and steady), (2) D-D neutrons equal to 4.7 x 1016 s -1 and the stored energy of 12 MJ. Here, a power step down technique was utilized, in order to delay the ELM and MHD activity as well as to obtain higher H-factor and reduce dWp / dt fraction. It was also found that high shear (which is accomplished by increased triangularity) and high flux expansion are effective ways to extend the ELM-free period. These high values are obtained despite the smaller plasma volume with the 'new' JET with divertor compared to the 'old' JET. Moreover, they are sustainable towards a steady state. Previously, this type of discharge ended with a roll-over of performance followed by a massive impurity influx. With the new divertor the impurity influx is prevented, but the performance roll-over still occurs. In addition, high values of > 3% have been maintained for more than 3 s in steady state, utilizing similar approaches as JT-60. Here, > 3, > 2 and 2.0 < H-factor < 2.2 at 1 MA / 1.7 T.
Nondimensional scaling studies of thermal diffusivity, carried out for steady state ELMy H-mode discharges, were closer to gyro-Bohm scaling. However, Bohm-like feature was observed at the heating power close the H-mode power threshold. Physics interpretation of this phenomena has been under discussion between JET and JT-60. Studies of TAE mode were also carried out by utilizing the saddle coil as exciter, and KAE was first observed at JET. Experiments to assess the effect of controlled toroidal field ripple have been carried out by powering the 32 TF coils in two sets of 16. It was found that a slightly increased ripple fraction can reduce the H-mode threshold power. For larger ripple rate, however, degradation of the H-mode quality was observed. The CFC target plate was replaced by Beryllium target of similar design in April '95, and it was found that (1) Beryllium tiles do not limit the power handling capabilities, (2) general plasma behaviors with beryllium was very similar to that with CFC target and (3) at high energy deposition the plate shows melt damage, and the protection by a vapor shield is not apparent.
Substantial portion of the current JT-60 program is devoted to the contribution to the broad range of ITER Physics R&D. Major efforts are focused on the improvement of confinement and exploration of the steady-state discharge, where (1) achievement non-inductive full current drive with a high bootstrap-current fraction and (2) optimization of the divertor function to control the particle and heat flux are the principal issues of investigation. As to the former, installation of the negative-ion based 500 keV NBI system has progressed successfully, and preparatory work for the joint experiment with TFTR is in progress. The latter is represented by the pumped divertor modification program, a part of which engineering design has been completed and the manufacturing procedure has started in 1995.
The fusion performances of H-mode and high- H-mode discharges in JT-60 are in principle constrained by the edge density limit, which is determined by the appearance of ELMs. Optimization of the plasma configuration to make the ballooning instability benign, in particular the triangularity, was attempted to improve the edge MHD stability. The increase of edge density limit for the ELM-free discharges as well as improvement of beta limit have been accomplished with increased triangularity of < 0.45. Sustainment of 1 MA full current drive condition in high plasmas for 2 s was thereby achieved in high steady-state campaign in August '95. Here, albeit not optimized, = 2.3 - 2.8, = 2.4 - 3.1 and H-factor was 1.9 - 2.4. Further enhancement of the integrated fusion performances of quasi-steady-state high plasmas, as an alternative candidate to the low q long pulse scheme, are also discussed with JET, which produces similar discharges. In regard to the remote radiative cooling of divertor plasmas and the investigation of detached plasma characteristics, various species of working gasses were puffed in to the divertor region. It was found that neon puffing into the divertor was effective to reduce the power flow to the divertor plates without the deterioration of confinement properties. Chemical sputtering was found to be the dominant mechanism of carbon impurity generation at a high wall temperature of around 300 C. Water-cooled divertor operation has been successfully carried out, which apparently reduced the carbon impurity concentration in a plasma.
Exploitation of the high current regime was also undertaken in hot-ion H-mode discharges, where Ip was raised up to 4.5 MA. Here, 5 MW of ICRH power was coupled to the H-mode plasmas to successfully suppress sawteeth, and the stored energy of 8 MJ was obtained. However, the volume averaged fusion product was about half of the record value achieved in high H-mode plasmas, largely due to exceedingly broadened beam deposition profile and the effect of ripple induced fast ion losses. Instead, the capabilities of shear reversal discharges were examined later in the year, in which a JET scientist participated. Transition to the enhanced shear reversal mode similar to that of TFTR was observed. However in JT-60, reduction of thermal diffusivities both in ion and electron channels was observed at the region of qmin. Comparative studies of transport barriers observed in high discharges and shear reversal plasmas are in progress, with an emphasis on the current density profile, of which measurement was made available as a result of the collaborative work with TFTR.
Non-dimensional transport study has also been undertaken to respond to the ITER physics R&D requirements. The H-factor for the ELMy H-mode discharges was nearly independent of *, which indicates that the transport of these discharges is Bohm-type. Behavior of MeV ions produced by the ICRF acceleration and the suppression of TAE modes with the plasma rotation have also been intensively investigated. In addition, H-mode transition power threshold study, which is also one of the urgent ITER physics R&D issues, was pursued to demonstrate experimentally for the first time that edge neutral density influences the transition criteria.
TFTR has explored a broad range of physics issues involved in D-T discharges, in which number of scientists from JET and JT-60 participated, and a maximum fusion power of 10.7 MW was achieved using 39.5 MW of neutral-beam heating. Indications of alpha heating was also observed in April '95 as the increase of Te after NB is ramped down. The energy confinement increased in D-T, relative to D-D plasmas, by 20% and the nT product by 55%. Improvement in thermal confinement is ascribed primarily to a decrease in ion heat conductivity, and experimentally derived < A > -1.8 dependence is stronger than expected from the previous experiments and from the transport theory. Extensive lithium pellet conditioning of the graphite inner wall has resulted in highly peaked supershot profiles at higher Ip, and thereby increased the confinement time to 0.33 s with 17 MW of tritium NBI. As indicated in the high task-assignment work with JT-60, importance of the peaked deposition profile was herewith reconfirmed. Evaluated confinement time was approximately 2.4 times the prediction of ITER-89P scaling, based on an average ion mass of 2.7. In addition, central triple product nHyd * Ti was extended to the record value of 8.3 x 10 20 m -3 s keV in February '95, where * is defined as WTOT / PTOT. The central fusion power densities achieved in the high-performance TFTR supershots, which are 1.5 to 2.8 MWm- 3, are compa-rable to or greater than those expected in ITER.
Measurements of the confined alpha particles have been performed and found to be in good agreement with TRANSP simulations. Here, alphas in higher energy range of 0.5 to 2 MeV have been detected with ablating lithium pellet diagnostic, while lower energy alphas were measured by the charge-exchange recombination spectroscopy. Measurements of the alpha ash profile have been compared with simulations using particle transport coefficients from He gas puffing experiments. With the same transport coefficients, it was concluded that helium ash accumulation would not quench ignition in ITER provided the density of helium at the plasma edge can be controlled. In addition, loss of alpha particles to a detector at the bottom of the vessel was found to be well described by the first-orbit loss mechanism. Stochastic orbit losses in the toroidal field ripple have also been investigated. At major disruptions, losses of energetic alphas estimated to be up to 10% of the alpha population have been observed to occur in 2 ms during the thermal quench phase while the total current is still unperturbed, which could induce a serious impact on first-wall components in a reactor.
The initial DT experiments in TFTR showed no signs of instability in the TAE frequency range and the alpha-particle loss rate remained a constant fraction of the alpha production rate as the alpha pressure increased, suggesting that deleterious collective alpha instabilities were not being excited. Theory has since shown that the ion Landau damping of the mode in TFTR is generally stronger than the alpha-particle drive. Further experiments to assess the effect of NBI and ICRF induced TAE activity on the alpha population are in progress with collaboration with other large tokamaks. ICRF heating of D-T supershots at 2T has resulted in 60% absorption to ions, which is consistent with modeling. Electron heating as well as the current drive have been demonstrated using the mode-converted ion Bernstein wave. It is planned to use ICRF heating to increase the alpha-particle pressure and to investigate the possibilities for ICRF current drive to access "advanced" tokamak regimes with improved confinement and stability of the core plasma.
A sudden transition to a new regime of enhanced confinement has been observed in TFTR plasmas with reversed magnetic shear and balanced NB injection above 16 MW in February '95, and considerable number of discharges was thereafter devoted to the investigation of confinement properties and stability analysis of this heuristic and robust operating regime. These plasmas are characterized by a rapid increase of the central density in the shear-reversed region, rising to n e o > 1 x 1020 m-3 with Ip = 1.6 MA, BT = 4.8 T, Tio = 24 keV, Teo = 8 keV, low toroidal rotation velocity and a pressure peaking factor of 8. The calculated bootstrap fraction reaches two thirds of the total plasma current. As measured by the motional Stark diagnostic, qmin ranges from 1.8 to 3 at a / r = 0.35 and qo from 2 to 5. Transport analysis indicates that the electron particle diffusivity decreases by a factor of 50 after the transition, which is comparable with the neoclassical prediction including off-diagonal terms. The inferred ion thermal loss is substantially lower than predicted by neoclassical theory in reversed shear region. In addition, the ion pressure gradient scale length is comparable to banana width of the thermal ions, which violates the standard neoclassical ordering. However, the electron thermal diffusivity is not changed significantly, and it is much larger than i or De. In the high-performance reversed shear plasmas obtained so far, values of * = 3.8 have been achieved. Ideal stability analysis indicated that disruptions for the highest performance plasmas are near the stability boundary for n = 1 infernal modes. Code extrapolations indicate that 20 MW of fusion power may reasonably be obtained with the modification of qmin and the increase of rmin / a and extending the duration of the high confinement region in the plasma core. These calculations indicate that due to the highly peaked pressure profile, thermal runaway of the core pressure due to alpha heating may occur. Further, experiments are planned to explore this exciting new regime of operation.