Resonant and Non-resonant type Pre-ionization and Current Ramp-up Experiments on Tokamak Aditya in the Ion Cyclotron Frequency Range S.V. Kulkarni, Kishore Mishra, Sunil Kumar, Y.S.S. Srinivas, H.M. Jadav, Kirit Parmar, B.R. Kadia, Atul Varia, R. Joshi, Manoj Parihar, Manoj Kumar Gupta, Nilam Ramaiya, Joydeep Ghosh, P.K. Atrey, R. Jha, Y.S. Joisa, Rakesh Tanna, S.B. Bhatt, C.N. Gupta, P.K. Kaw, ICRH Group and Aditya Team Institute for Plasma Research Bhat, Gandhinagar, 382 428, Gujarat, India e-mail: kulkarni@ipr.res.in Abstract The steady state operation of the tokamak with superconducting magnets and central solenoid with cryostat reduces available loop voltage and hence plasma production becomes difficult. In order to overcome this problem, different pre-ionization techniques such as ECRH and ICRH type pre-ionization schemes are being investigated. The advantage of these schemes is that one can use same RF system for pre-ionization which is being developed for plasma heating, current drive and start-up experiments. Here we report the ICRF preionization experiments carried out in ICRF range using poloidal type fast wave antenna, and 200 kw RF system at 24.8 MHz frequency. The toroidal magnetic field is varied from 0.825 Tesla to 0.075 Tesla due to which the resonance layer moves from center towards high field side and finally goes out of plasma minor radius at ~0.5T. The RF power is varied from 20 kw to 120 kw and also pressure of hydrogen gas is varied. The diagnostics used are Langmuir probes, visible camera, spectroscopy, soft X-ray detection techniques, diamagnetic loop and microwave diagnostics like interferometer and reflectometer. After plasma production at different magnetic fields, the pre-ionization experiments are carried out at different loop voltages to ramp up the current and we could ramp up current at all available loop voltages starting from 22 volts to 8 volts to get normal plasma discharge of 90 ka and 90 ms duration. The parameters like pressure, magnetic field, loop voltage, RF power etc. are systematically varied to have full understanding of the pre-ionization and current ramp up in tokamak Aditya. It is observed that at lower loop voltages the current ramp-up is possible only in presence of RF preionization. 1. Introduction Pre-ionization is an essential part of the tokamak operation to have highly reproducible plasma as well as to have full control over plasma position, and plasma development as well as to save volts-seconds to have long duration plasma. Moreover in superconducting tokamaks, there is an additional factor of cryostat due to which the applied loop voltage from Ohmic transformer penetrates with attenuation as well as gets delayed and hence the available loop voltage requires some source of pre ionization. A large Ohmic voltage is required for producing plasma in large tokamaks and also there are certain limitations in the design of Ohmic transformer [1, 2]. Also a beam of runaway electrons is produced by the high initial loop voltage, increasing the energy losses from the Ohmic discharge at the quasi-steady stage. Another important point is that the discharge should always get initiated at the right position for reproducibility and consistency of the discharges. Also in a superconducting tokamak the normal glow discharge cleaning becomes very difficult and one needs to find alternatives. Different types of pre-ionization schemes like nuclear radiation, filamentary discharge, ultraviolet source, X ray and electron cyclotron resonance assisted microwave sources etc. have been used in different tokamaks [3]. Recently, new schemes like ion cyclotron resonance based RF systems are being tried in few tokamaks [3, 4]. Normally ion cyclotron resonance heating schemes are usually available in
most of the tokamaks. If one can produce pre-ionization using same fast wave antenna and RF system, then one can save additional port of the tokamak as well as decrease the burden of having additional pre-ionization system. Introduction of RF waves in plasma in ICRH range has many applications starting from preionization, fast wave and ion Bernstein wave heating, stabilization of some of the modes like ELMs in the plasma, vessel wall conditioning, current drive via momentum transfer from wave to the plasma etc. With the help of ICRH waves one can do either electron heating, ion heating or both [5] to increase the temperature and pressure of the plasma. ICRF plasma production and its assisted low voltage ohmic start up have been demonstrated in TEXTOR-94[6]. Interestingly without ICRF assistance no start up is achieved on application of low loop voltage. The plasma production through ICRF is believed to be mainly because of absorption of RF wave energy by electrons in presence of toroidal magnetic field. The RF electric field parallel to magnetic field may be responsible for this neutral gas breakdown and initial plasma build up. The parallel electric field may be generated between antenna central strap and sidewall of the antenna. In TORE SUPRA [7,8], two ICRF antennas are used to produce plasma and study its wall conditioning efficiency in presence of permanent toroidal magnetic field of 3.8T [267]. ICRF power is applied in the range of 20-340 kw and electron density obtained is in the range of 1-5 x 10 11 cm -3. In JFT-2M tokamak successful startup and current ramp up is also achieved when ICRF plasma is applied to low loop voltage. Here we report the experiments on pre-ionization and current build-up at different loop voltages, pressures, magnetic fields and RF powers using the developed ICRH system of 200 kw for ICRH heating on tokamak Aditya [5]. The section 2 describes experimental set up. The section 3 explains the experimental procedure and section 4 explains the measurements and results of pre-ionization experiments and last section describes the conclusions and future work. 2. Experimental Set up A system of plasma heating by poloidal type fast wave antenna is currently being employed on Aditya tokamak [9, 10]. Aditya is a medium sized tokamak (R=0.75m, a=0.25m, B t =0.75T) with hydrogen plasma in circular channel cross-section. The ICRF system on Aditya is of 20-40 MHz frequency range and 200kW of RF power. A simplified block diagram of complete ICRF system on Aditya is shown in Figure 1. A 100-meter long 9 coaxial copper transmission line carries RF power from generator to tokamak Aditya. A series of SPDT switches enables to divert the RF power either towards Aditya or towards 50-ohm water dummy load system housed in generator hall for testing purpose. A matching network consisting of two stubs and two phase-shifters is placed between the antenna and generator for matching antenna-plasma impedance to that of the 50-Ohm impedance of the generator and transmission line. A Vacuum Transmission Line (VTL) section, which has separate vacuum system, isolates the Aditya vacuum to atmospheric pressure transmission line. The VTL section also facilitates the radial movement of antenna up to 3 cm in the scrape of layer (SOL) of Aditya plasma.
(VTL Section) Figure 1. ICRF-Aditya system schematic. 1-a, 1-b: dual directional couplers, 2:SPDT switch, 3:mechanical stub, 4:liquid phase shifter, 5: liquid stub, 6: mechanical phase shifter, 7: vswr probe section, 8:Fast wave antenna. Details of VTL section are shown in the right side of schematic. The VTL section has also a provision of torque rings to bring back the antenna to its original position in case of antenna movement during plasma disruption. A shorted strip line type fast wave antenna is placed inside the Aditya torous poloidally to inject RF power into the plasma. The poloidal type fast wave antenna (lengh-30 cm, width 10 cm) is made up of the stainless steel (SS304L) material with Faraday shield and has graphite tiles on plasma facing sides to protect the antenna from damage due to high energy particles as well as to decrease impurity generation from antenna during plasma. During present experiments, the antenna is kept at approximately 1 cm behind the circular graphite limiter of tokamak. Heating and pre-ionization experiments are carried out after the recent up-gradation of its VTL section and complete system up to 80 kw RF power [12]. The up-gradation of VTL section includes replacing the vacuum feed-through with a new coaxial ceramic vacuum window [13], the modification of inner conductor at T-junction of feeder and modification of outer conductor with a disk type perforated shorting plunger. The improved VTL section is adequately conditioned with short RF pulses and tested for high power operation up to 80 kw RF power, with a 50-Ohm dummy load. In order to characterize the plasma, the diagnostics used are Langmuir probe near the antenna, perpendicular CCD cameras from the top and side ports, diamagnetic loop for the measurement of stored energy, soft X ray for electron temperature and all spectroscopy diagnostics at different ports to detect carbon and other impurity lines along with H alpha measurements. The microwave interferometer and reflectometer diagnostics for density measurements along with all machine diagnostics for loop voltage and plasma current are also used for characterization of the plasma. 3. Experimental Procedure The experiments are carried out in following phases. In first phase, only RF plasma is produced at 24.8 MHz using a fast wave antenna without toroidal magnetic field and loop voltage at different RF powers and hydrogen pressures. In second phase, only RF plasma is produced at 24.8 MHz using a fast wave antenna, toroidal magnetic field and RF system of 200 kw. In this phase the toroidal magnetic field is varied from 0.825 Tesla to 0.075 Tesla.
For 24.8 MHz RF frequency, the second harmonic resonance layer lies at the center of the plasma at 0.75 T and when the magnetic field is varied, it goes away from the antenna towards high field side and finally vanishes at ~0.5 T inside the vacuum vessel. Then the experiments are carried out at fixed 0.75 T and fixed power of 70-80 kw and the pressure is varied to find the pressure window of RF produced plasma. The variation of power is also done at 0.75 T and fixed tokamak pressure of 7x10-5 Torr. In third phase, the experiments are carried out with RF power and the full loop voltage of 22 volts. The duration of pre-ionization pulse is from -150 ms with reference to Ohmic loop voltage starting at 0 ms. In this experiment, the over-lapping time of RF power with loop voltage is varied and also the RF power is varied. The adjustment of the vertical field as well as the magnitude, duration and the frequency of the gas puff are varied to have current build up. In fourth phase, the loop voltage of Ohmic transformer is decreased by decreasing the current through the transformer due to which the available volts-sec also decreases. After that the resistors in the ohmic transformer are changed to keep available volt-seconds constant and the current ramp-up experiments are carried out. In fifth phase, the current build up experiments are carried out at different loop voltages with 3 msec slowdown in rise time at fixed RF power and gas pressure. 4. Results and Discussions The Langmuir probes situated just above the fast wave antenna at same toroidal location could not draw significant ion saturation current when negatively biased in presence of ICRF produced plasma. While a different probe situated toroidally ~120 degree away from the antenna but poloidally at outboard bottom quadrant could measure ion saturation current in presence of ICRF plasma near limiter position. The Figure-2 (a) shows time traces of plasma current, applied loop voltage, ICRF pulse and estimated electron density from the measured ion saturation current in a typical ICRF produced plasma discharge. The figure shows that, after ~37ms of the application of ICRF power, electron density started increasing suddenly. This delay significantly increases with decrease in fill pressure, which may be attributed to the reduced collisionality and increase in probability of ionization. In Figure 2(b) the plasma current is fully evolved at a reduced applied ohmic loop voltage (~10V). The ICRF pulse has started at ~97ms before the application of ohmic loop voltage. The measured density from the Langmuir probe shows two distinct density time evolutions during ICRF plasma phase and ICRF assisted ohmic plasma phase. While the density remains steady during former phase, it changes in ohmic phase. This variation is may be due to changes in loop voltage, plasma position and time. Figure 3 shows the plasma breakdown initiation with ICRH power in absence of loop voltage. Reflectometry (e) Dia-magnetic loop signals are shown in Figure 3. The toroidal field scan, pressure scan and ICRF power scan were performed during the experiment in absence of loop voltage. The threshold value for minimum toroidal field, pressure and ICRH power was established during the experiment. It was observed that plasma breakdown is obtained in wide range with RF power which is visible from the H alpha signal as well as diamagnetic loop signal. The plasma is produced in a wide toroidal magnetic field ranging from 0.825 T to 0.075 T and also in the pressure range of 6x10-5 Torr to 4.0x10-4 torr.
Fig. 2(a) Fig. 2(b) Figure-2: Time traces of plasma current, applied loop voltage, ICRF pulse and estimated electron density from the measured ion saturation current in a typical ICRF produced plasma discharge (Fig. 2(a) and in Ohmic start-up and current ramp-up (Fig. 2(b) (RF Power:~55 kw) Figure 3. Time series of (a) H (a.u) (b) Pressure (value x calibration factor 4) (c) ICRF power (d) Reflectometry (e) Dia-magnetic loop signal additional extension in ~ 20 ms loop voltage is Figure 4 (a) Loop voltage comparison in Vacuum shot for similar OT current, (b) successful current ramp up at 10V loop voltage and 80-90kW RF power.
The RF power is varied from 18 kw to 120 kw and it was observed that below 18 kw RF power we do not get breakdown. The density of the plasma produced with RF is in the range of 10 10 /cc which is measured with reflectometry as well as Langmuir probes. The Figure 4 shows that the peak loop was reduced from 18.5 V to 10 V by changing R 0 switching resistance from 0.96 Ω to 0.48 Ω in Aditya pulsed power supply. In absence of ICRH pre ionization, we could not get normal ohmic discharge at 10 V loop voltage. The ICRH power of 80-90 kw was introduced and superimposed with loop voltage up to 5-7 ms which generated successful start-up and longer duration discharge (80 ka, 100 ms) at 10 V loop voltages. The Hydrogen Gas pressure of the order of 1x 10-4 torr was in pre-filled mode and applied at -300 ms before loop voltage is established. The proper equilibrium fields provided by B V coil gives rise to successful as well as reliable start-up. The time series of (a) Loop voltage (V) (b) Plasma current (ka) (c) H (a.u) (d) ICRF power (e) Soft X-rays (a.u) (f) diamagnetic signals with and without ICRH pre-ionization assisted start-up are shown in figure 4b. Figure 4b clearly shows that with ICRF preionization, successful start-up and current build up was p[possible at 10V loop voltage. Shot No. 24111-4 Shot No. 24117-4 Fig. 5(a) Fig. 5(b) Fig. 5(c) Fig. 5(d) Fig. 5(e) Figure 5: (a) and (b) Photograph of the pre-ionization plasma from top port, RF power 70 kw and Tf 0.75 Tesla. Fig. 5(c) shows the variation of relative intensity with radial position at different pressures at different loop voltages (5(d) and at different toroidal magnetic fields (5(e))
Figure 5 shows the photographs of pre-ionization plasma produced under different conditions. Fig. 5(c) shows the variation of visible intensity at different pressures which indicates slight variation in the area of intensity with pressure and Fig. 5(d) shows that there is not much variation in visible intensity with change in loop voltage. However there is lot more variation in the high intensity area with change in toroidal magnetic field (Fig. 5(e)). Fig. 6(a) Fig. 6(b) Figure 6: shows the estimated density using reflectometer as a function of toroidal magnetic field (a) and as a function of loop voltage (b) As shown in Fig. 6 the chord averaged density increases with increase in toroidal magnetic field and also with increase in loop voltage. Although there is slight variation in the estimated densities between reflectometer and Langmuir probe, probably it is due to chord average nature of density by reflectometer measurement. Fig. 7(a) Fig. 7(b)
Fig. 7(c) Figure 7: The variation of H α signal and delay with magnetic field (a), variation of H α signal and delay with magnetic field (b), and the variation of delay and H α signal with RF power (-9.6 db m =50 kw and -8 db m = 80 kw). From Fig. 7 it is clear that the delay in production of preionization with reference to application of RF power is a function of magnetic field, pressure and the RF power and decreases as we increase power, pressure and magnetic field. 5. Conclusions The pre-ionization experiments are carried out in tokamak Aditya at different loop voltages, different magnetic fields to vary the position of the resonance layer and also in the wide range of pressure range for the tokamak operation in ion cyclotron frequency range. It is observed that in presence of resonance layer the pre-ionization plasma mainly forms between antenna and the resonance layer and with proper vertical field and gas puff the current ramp up is possible even at lower loop voltages. Although we have produced pre-ionization plasma under non-resonant conditions we are yet to produce current ramp-up which is in progress It is expected that these results will be useful for the steady state operation of the tokamak SST- 1 where the available loop voltage is less due to penetration depth of the cryostat. References [1] Thomas P R et. Al, Contr. Fusion and Pla. Phy. V9F, part1, Europian Physical Society 1985 p283 [2] Moiseenkoi V E, Ukr. Fizch. Zh. 35 (1990) 214 [3] Lysojvan A. I. et al, Nuclear Fusion, V32-8 (1992) p1361 [4] Shevts O. M. et al, Nuclear Fusion 26 (1986) 23 [5] Kishore Mishra, S V Kulkarni, et. Al Plasma Physics and Controlled Fusion, 53, (2011) 095011 [6] Esser H G, et al, JNM 241-243, (1997) 861 [7] E de la Cal and Guthier E, PPCF, 39 (1997) 1083 [8] Mantsinen M J et al, Plas. Phy. Contr. Fusion, 45, (2003), A445 [9] Bhatt S. B. et al, 1989, Indian J. Pure Appl. Phys. 27: 710 742, [10] Bora D et al, 2005, Ion Cyclotron Resonance Heating System on Aditya, Sadhana, 30, Part 1, PP. 21-46 [12] Kishore Mishra et al, 2008, Recent Advances and Up gradation System on Aditya, IPR-Technical Report, IPR/TR-420/2008 [13] D. Rathi, K. Mishra, Siju George, A. Varia, Raj Singh, S.V. Kulkarni and ICRH RF Group, Development of high Power Co-axial Vacuum Window for ICRF, IPR-Technical Report, IPR/TR-146/2008, October 2008