M. Sakamoto

Recycling and wall pumping in long duration discharges on TRIAM-1M

Recycling and wall pumping have been studied comparing low (?1018?m-3) and high (?1019?m-3) density long duration plasmas in TRIAM-1M. The recycling coefficient of each plasma increases with time. There exist two time constants in the temporal evolution of the recycling coefficient. One is a few seconds and the other is about 30?s. The wall pumping rates of low and high density plasmas are evaluated to be ?1.5?1016?atoms m-2?s-1 and ?4?1017?atoms m-2?s-1, respectively. The difference may be caused by the total number of diffused ions and the charge exchange neutral flux with an energy of less than 0.7?keV. In an ultra-long discharge (?70?min), the recycling coefficient becomes 1 or more before again decreasing below 1, i.e. the wall repeats the process of being saturated and refreshed. In high power and high density experiments, wall saturation phenomena have been observed. The discharge duration limited by wall saturation decreases with an increase in density.

Overview of steady state tokamak plasma experiments in TRIAM-1M

An overview of steady state tokamak studies in TRIAM-1M (R0 = 0.8 m, a × b = 0.12 m × 0.18 m and B = 8 T) is presented. The current ramp-up scenario without using centre solenoid coils is reinvestigated with respect to controllability of the current ramp-up rate at the medium density region of (1–2) × 1019 m−3. The plasma is initiated by ECH (fundamental o-mode at 170 GHz with 200 kW) at B = 6.7 T, and the ramp-up rate below the technical limit of 150 kA s−1 for ITER can be achieved by keeping the LH power less than 100 kW during the current ramp-up phase. The physics understanding of the enhanced current drive (ECD) mode around the threshold power level has progressed from a viewpoint of transition probability. A transition frequency, ftrans, for the ECD transition is determined as a function of PCD. At ~70 kW no transition occurs for an ftrans value of ~0.017 Hz, meaning almost zero transition probability. With increasing PCD > Pth, ftrans increases up to 10 Hz, and the transition tends to occur with high probability. The record value of the discharge duration is updated to 3 h 10 min in a low and low power (<10 kW) discharge. The global particle balance in long duration discharges is investigated, and the temporal change in wall pumping rate is determined. Although the density was low, the gas supply had to be stopped at 30 min after the plasma initiation to maintain the density constant. After that the density was sustained by the recycling flux alone until the end of the discharges. In addition to the recycling problem, in the high power and high density experiments, the localized PWI affects the SSO of the tokamak plasma. The effects of enhanced influx of metal impurities (Fe, Cr, Ni, Mo) on sustainment of the high performance ECD plasma are investigated. In order to evaluate the helium bombarding effects on the plasma facing component and hydrogen recycling in the future burning plasma, microscopic damage of metals exposed to long duration helium discharges was studied. The total exposure time was 128 s. From thermal desorption experiments for the specimens the amount of retained helium was evaluated as 3.9 × 1020 He m−2 and the scale length to be ~1 mm in the SOL.

Global particle balance and wall recycling properties of long duration discharges on TRIAM-1M

The longest tokamak discharge, with a duration of 11u2009406u2009s (3u2009h 10u2009min), has been achieved. The global particle balance has been investigated. In the longest discharge, the global balance between the particle absorption and release of the wall was achieved at t ~ 30u2009min. After that, the plasma density was maintained by the recycling flux alone until the end of the discharge. The maximum wall inventory is about 3.6 × 1020u2009H at t ~ 30u2009min, but it is finally released from the wall at the end of the discharge. The hydrogen release seems to be caused by the temperature increase in the whole toroidal area of the main chamber. Moreover, it has been observed that there is a large difference between the properties of wall recycling in the continuous gas feed case (i.e. static condition) and in the additional gas puff case (i.e. dynamic condition). In the static condition, the effective particle confinement time increases to ~10u2009s during the 1u2009min discharge and it increases to ~100u2009s before the global balance in the longest discharge. In the dynamic condition, the decay time of the electron density just after the gas puff, i.e. the effective particle confinement time, is constant at 0.2–0.3u2009s during the discharge. The large difference in the effective particle confinement time between the static and dynamic conditions seems to be caused by the reduction in the recycling coefficient due to the enhanced wall pumping resulting from the additional gas puff.

Steady state experiments on current profile control and long sustainment of high performance LHCD plasmas on the superconducting tokamak TRIAM-1M

The main purpose of TRIAM-1M (R0 = 0.8?m, a ? b = 0.12?m ? 0.18?m, B = 8?T) is to study the route towards a high field compact steady state fusion reactor. In the advanced steady state operation programme, a heating mechanism for the high ion temperature mode with an internal transport barrier has been studied, an enhanced current drive mode in an extended (higher power and higher density) operation regime has been obtained, current density profile control experiments using multicurrent drive systems have been performed and the effects of wall recycling, wall pumping and wall saturation on particle control have been investigated.

Enhanced current drive efficiency in a long discharge on TRIAM-1M

The enhanced current drive (ECD) efficiency mode, which is characterized by a spontaneous increase of current drive efficiency ηCD from (0.3-0.4) × 1019A/W m-2 to (0.6-1.0) × 1019 A/W m-2, is observed in long pure LHCD plasmas on TRIAM-1M. The energy confinement time is also improved due to the increase of line averaged electron density, and of the ion and electron temperatures. The current drive efficiency is proportional to the electron density. The transition to ECD mode occurs at a critical density, which depends slightly on the refractive index in the toroidal direction N|| of the injected wave.

Recent progress on TRIAM-1M

A steady state plasma with high performance and high current drive efficiency is reported. In 2.45 GHz LHCD plasmas Ti is studied as a function of ne at the edge of the high ion temperature (HIT) window. Different characteristic timescales are found for Ti and ne to enter the HIT regime and the observed hysteresis behaviour of Ti with respect to ne is attributed to this difference. The electromagnetic emission (<3.5 GHz) is studied in order to understand ion heating mechanisms in the HIT regime. The spectrum shows several sidebands whose peak frequencies correspond to the ion plasma frequency. The spectral narrowing of the width of the sideband shows a clear correlation with ion heating. In 8.2 GHz LHCD plasmas an enhanced current drive (ECD) regime where both current drive efficiency ηCD( = eICDR0/PLH ~1 × 1019 A m-2/W) and energy confinement time τE (~8-10 ms) are simultaneously improved is obtained at an e of 4.3 × 1013 cm-3 and B = 7 T under full current drive conditions. There exists a certain threshold power above which the ECD transition occurs. A hysteresis of ηCD is found around the threshold power, which is explained by the different characteristic time for the ECD transition in power rampup and rampdown schemes. Current profile control experiments are performed by using two opposite travelling LHWs. Current compensation (ΔICD/ICD < -10%) is clearly seen when the backward (BW) travelling LHW (8.2 GHz) is added to a target plasma whose current is driven by a forward travelling LHW (8.2 GHz). As the BW wave power is increased, however, the current tends to flow in the forward direction. The mechanisms of this non-linear behaviour of the driven current with respect to the BW wave power are discussed.

Current ramp-up experiments in full current drive plasmas in TRIAM-1M

Four types of plasma current ramp-up experiments in full non-inductively lower hybrid current driven (LHCD) plasmas were executed in TRIAM-1M: (1) current start-up by a combination of electron cyclotron resonance heating (ECRH) and LHCD, (2) tail heating by additional LHCD, (3) bulk heating by ECRH and (4) spontaneous ramp-up by a transition to enhanced current drive (ECD) mode. The time evolutions of plasma current during four types of ramp-up phase were adjusted by a simple model with two different time constants, which are a time defined by the total current diffusion time and a time constant for improving the current drive efficiency. In the case of (1) and (4), the latter time constant is significant during the current ramp-up phase. The improvement in the current drive efficiency in the ECD mode is likely to be caused by the increase in the effective refractive index along the magnetic field of the lower hybrid wave.

Static and dynamic properties of wall recycling in TRIAM-1M

A large difference between properties of wall recycling in the continuous gas feed case (i.e. static condition) and the additional gas puff case (i.e. dynamic condition) has been observed. In the static condition, the effective particle confinement time, τ* p , increases almost linearly to about 10 s during the 1 min discharge. In the dynamic condition, τ* p is 0.2-0.3 s during the 1 min discharge. This difference of τ* p is also confirmed in the ultra-long discharge. τ* p in the static condition becomes ∼100 s before the global balance between particle absorption and release of the wall is achieved at t ∼ 30 min. τ* p in the dynamic condition is, however, still on the order of ∼0.3 s. The large difference between τ* p in the static and dynamic conditions is attributed to a reduction in the recycling coefficient due to enhanced wall pumping resulting from the gas puff.

Modeling of global particle balance in steady-state magnetic fusion devices – Analysis of the recent data from the TRIAM-1M tokamak

Abstract Global particle balance analysis has been conducted to interpret the data from TRIAM-1M experiments, using a zero-dimension four-reservoir model to calculate the particle inventories in the core plasma, SOL region, gas phase, and wall materials. Two cases have been examined: a relatively short pulse (∼30 s) but high density (∼1019 1/m3) LHCD discharge at 8.2 GHz; and a long pulse (∼4000 s) but low density (∼1018 1/m3) LHCD discharge at 2.45 GHz. Model calculations have reproduced well the core and SOL plasma densities. Also, the observed wall pumping effects have been analyzed by this model. Measured and model prediction on the wall pumping rate are 4×1017 and 4.4×1017 1/m2/s at 8.2 GHz, and 1.5×1016 and 1.8×1016 1/m2/s at 2.45 GHz, respectively, both relatively good agreements. Also, a parameter sensitivity check has been conducted, and modeling results clearly indicate that the steady state core plasma density decreases with increasing the codeposition probability.

Current profile control experiments in the LHCD plasma on TRIAM-1M

Controllability of current profiles in long duration discharges is studied by two kinds of combination of wave spectra. First, by superposition of a higher N? spectrum wave at 2.45?GHz on a target plasma driven by a lower N? one at 8.2?GHz, a hollow j(r) profile has been achieved and sustained for 20?s. The current profile can be well controlled below a threshold power of 14?kW by varying the power at 2.45?GHz. A spontaneous transition phenomenon to a peaked j(r) profile, however, has occurred above the threshold power even after a reasonable steady state condition had been obtained. Second, using two oppositely travelling waves at 8.2?GHz the total current is clearly reduced, and the j(r) profile becomes peaked when the backward travelling lower hybrid waves (BWs) are superposed on the forward lower hybrid wave at a low power of 20?kW. When the BW power is increased above 27?kW, further current reduction is suppressed, and above 34?kW the direction of the current driven by the BWs is completely reversed and the j(r) profile becomes broad. Thus, current modification by BWs shows a highly non-linear behaviour with respect to the BW power.