Docstoc

RF Heating technology

Document Sample
RF Heating technology Powered By Docstoc
					RF Heating and Current Drive Systems Wukitch, Parker, Porkolab, Temkin RF systems have been used extensively in tokamak plasmas for heating, driving current and controlling plasma instabilities. These systems integrate wave physics, coupler design, transmission network, and source characteristics to produce a desired physics application. Each component of RF systems is fairly well developed and their path to application to DEMO from ITER is fairly well defined. The US can capitalize on its historical focused source development, extensive experimental and computational assets to develop the RF systems needed for DEMO. Sources for ion cyclotron resonance heating and current drive in a DEMO would be CW tetrode based transmitters in the 50-100 MHz range. High power (2 MW) sources are readily available but the transmitters envisioned would benefit from incorporation of new solid state drivers to increase their reliability and robustness. In addition to transmitting power, transmission networks have evolved to isolate the transmitter from the variable load presented by the plasma. ITER plans to utilize passive elements for isolation but are inadequate for maximizing coupled power to the plasma. Another approach is to utilize active elements like ferrite tuners that can be utilized in a feedback loop to maximize input power. Such a system was recently successfully demonstrated but would require further development. The antenna proximity to the plasma edge presents significant challenges. The present ITER antenna utilizes an antenna in a cavity approach where the antenna elements are completely recessed in a cavity behind the first wall. This reduces the thermomechanical loads but lowers antenna loading and narrows the antenna bandwidth. Thus, the antenna is more sensitive to load variations that can only be partially mitigated by the matching network. An alternative approach is to utilize wider bandwidth antennas that are placed 5-10 cm in front of the first wall. These antennas have good coupling and are obvious candidates for active matching but have higher thermo-mechanical loads. Power handling reliability issues and impurity generation are issues in both cases. The sensitivity of impurity generation and plasma scrape-off modifications to specific geometries, plasma edge conditions and first wall materials require integrated test for validation of amelioration techniques. Antenna coupling over long distances and heating efficiency in the presence of competing absorption mechanisms should also be explored. Lower hybrid current and heating sources for DEMO would be ~5 GHz CW klystrons with a unit size of ~ 1 MW. Source development does not represent a huge technological challenge and are likely to be developed by Japan and EU for ITER if LHCD is used. The same may be said for waveguide transmission systems although special overmoded waveguides that preserve phase may be required to reduce transmission losses to acceptable levels. At the present time, the US program lacks a plan and funding to develop such tubes or transmission lines and this should be remedied in the future. The slow wave launching structure required to launch current-driving waves represents a significant challenge. Long distance coupling is the biggest issue for LHCD. Good

coupling requires that the density at the mouth of the launching grill exceed the cutoff density and remain below a critical accessibility density. These conditions must be maintained for the waves to penetrate through the pedestal at the separatrix without strong reflection until the waves reach the region of the plasma where deposition is desired. While JET has made proof of principle experiments utilizing gas injection, an integrated test with an RF system having a DEMO-relevant power density, configuration, and plasma conditions is required. Another key question is the degree of spectrum control and directivity required for physics applications. This impacts the launcher design where the classical module has better spectral control and high directivity but the passive active module, favored for ITER, has better thermo-mechanical characteristics. ECH/ECCD will play a major role on ITER, but significant advances will be needed for application of ECH/ECCD on DEMO. The sources for ECRF are a major technological challenge for DEMO and the specific frequency required will depend on DEMO’s magnetic field and density. The worldwide progress in gyrotron development suggests that higher frequency gyrotrons needed for Aries RS, at 220-250 GHz, should be realizable in a 5-10 year time period. For a lower field AT plasma, a 324 GHz gyrotron is envisioned. Short pulse gyrotrons at MIT have shown that 1 MW power levels can be obtained at frequencies as high as at least 300 GHz but industrial development is necessary to demonstrate the required gyrotron stability and an efficiency exceeding 50%. Development of 2 MW gyrotrons is also important to reduce cost and system complexity. Fast gyrotron tuning should also be pursued because of significant implications for control of instabilities. The US should resume its gyrotron research program, particularly for the higher frequencies where a more aggressive development program would be required. All tubes developed should also have significant effort expended to determine the tube lifetime and production issues. The required waveguide transmission system is a more mature technology. However, transmission efficiency at the high power levels, at least 2 MW, and the higher frequencies, up to 300 GHz or higher, will require development. Mode conversion losses (scale with frequency to the 3/2 power) may lead to unacceptable losses. Thermomechanical issues at the final mirror must be successfully addressed for ITER and the designs for DEMO will benefit from the ITER experience. However, the increased reliability and power requirements of DEMO may prove to be a significant challenge. Although the wave propagation and absorption are theoretically well developed in all frequency regimes, the intention is to develop RF into a tool. Optimizing an RF scenario to deliver power or driven current in a prescribed manner needs to account for RF system and the plasma response. The computational requirements, both software and hardware, should not be underestimated or undervalued. Furthermore, there will also need to be an integrated effort to verify and validate these codes against one another in standardized cases and experiments. The return for such an investment can be quite substantial. For example an integrated antenna plasma model could greatly accelerate antenna optimization in ICRF and LHRF. These models will also be important in developing simulations for evaluation of feedback systems designed to control current profiles and plasma instabilities and some will likely be tested in ITER.