Thomson scattering on low and high temperature plasmas

Worldwide research is ongoing, to develop and build the tokamak ITER to generate energy based on controlled nuclear fusion. The principle design concept of ITER is a donut-shaped vessel wherein the fusion fuel, a hot plasma of hydrogen isotopes, is contained by high magnetic fields. The fusion power can be produced at a plasma temperature of ~100 million degrees C and density of ~1020 m-3. In order to realize turn key fusion energy plants, a number of issues need to be addressed. Firstly, the control of the bulk plasma to prevent outflow of heat and particles due to instabilities, needs to be improved. Subject of research on many present-day tokamaks is the formation of magnetic islands (so-called tearing modes) due to instabilities in the magnetic field that confines the plasma. Tools to monitor and prevent growth of magnetic islands are therefore very important. Additionally, the mechanisms underlying the occurrence of confinement-friendly internal transport barriers have to be studied. This research can lead to improved plasma performance. The second issue that has to be addressed, is the erosion due to the high power load (10 MW/m2 (continuous) and > 1 GW/m2 (transient) due to Edge Localized Modes (ELMs )) on plasma facing components of the ITER divertor. The linear plasma generators Pilot-PSI and Magnum-PSI have been built to study plasma-wall interaction during these power loads. This research includes the third issue; tritium retention build-up in wall material. This thesis describes the development of Thomson scattering systems to study fast plasma phenomena in tokamaks as well as to study the quasi-continuous plasma of linear plasma generators. Thomson scattering is the most accurate method for measuring the electron temperature (Te) and density (ne) of a plasma: the accuracies of the systems described in this thesis are better than 3 - 4% and 4 - 8% for ne and Te, respectively. Basically, Thomson scattering is the process of acceleration of electrons due to an electromagnetic wave and as a consequence emission of radiation with the same frequency as that of the incoming wave, i.e. the wave is scattered elastically. The re-radiated light is Doppler shifted due to the velocity of the electron. Scattering on an ensemble of electrons results in a spectrum that resembles the electron velocity distribution, from which Te and ne can be retrieved. If the size of the incident wave is larger than the Debye length, then the light is collectively scattered by the electrons bunched in the Debye cloud of an ion. This so-called collective Thomson scattering can be utilized to measure the ion temperature (Ti). The first research challenge was the development of a high repetition rate Thomson scattering system for the TEXTOR tokamak (Jülich, Germany). A so-called double-pass intracavity laser was developed that generates a burst of 30 laser pulses of ~15 J each (with a repetition rate of 5 kHz). The system operates like a laser oscillator with the plasma as part of an 18 m long cavity. A fast detector equipped with CMOS cameras coupled to an image intensifier stage was developed. At a repetition rate of 5 kHz and a density of ne = 2.5×1019 m-3, density and temperature (range: 50 eV – 5 keV) profiles could be measured along the full plasma diameter of 900 mm long, with a spatial resolution of 7.5 mm. Coping with the plasma light background turned out to be the biggest issue: due to the long cavity and the laser pulse is relatively long (~1 ¿s) and a large detector gate window is required, resulting in a much higher plasma light contribution compared to single-pulse Thomson scattering systems. Nevertheless, a combination of high laser pulse energies (~12 J/pulse), careful plasma light monitoring and effective detector gating proved to be the solution to realize a reliable high repetition rate Thomson scattering system. The main step in laser development was the replacement of a high dope ruby rod by one with a low dope (0.03% Cr+), leading to a homogenous absorption of the pumping light from the flash lamps over the ruby rod cross section. The pumping-to-probing efficiency was improved by a factor of 1.5 and a significant minimization of the laser beam divergence, resulting in a better imaging efficiency of the viewing system. The diagnostic system enabled to record rotating magnetic islands during 2.2 ms with a repetition rate of 5 kHz, revealing the detailed density profile evolution inside the islands. Confinement properties of transport barriers were studied by measuring the time evolution of the ne and Te profiles. A second challenge was to develop a Thomson scattering system for the Pilot-PSI linear plasma generator, based on a frequency-doubled Nd:YAG laser. The stray light contribution of the system already existing at Pilot-PSI could be significantly reduced by application of a special carbon aperture system in the vacuum laser beam line, which enabled Thomson scattering measurements at a distance of 17 mm from a target surface exposed to a high power plasma beam. The sensitivity of the detector system was improved by more than a factor of 5 by application of a Generation III image intensifier at the front of the existing ICCD detector. The lower density and temperature limit of the system is 4×1019 m-3 and 0.2 eV, respectively. To achieve these values, the signal from 30 laser pulses (0.35 J/pulse, 10 Hz) needs to be accumulated. Instead of multiple Pilot-PSI discharges, now only one discharge is required to obtain accurate ne and Te profiles. This diagnostic has become a working horse for Pilot-PSI research and revealed different properties of the hydrogen plasma jet such as plasma confinement and indications for ion viscous heating. During ELM simulation experiments, single pulse TS measurements were successfully performed; using only 0.35 J scattering energy the time evolution of the plasma could be measured on shot to shot base. Subsequently, an advanced Thomson scattering system was designed and constructed for Magnum-PSI. This system features a frequency-doubled Nd:YAG laser, equipped with a 35 m long remotely-controllable laser beam line and a high etendue spectrometer based on a transmission grating. The system is designed to measure electron density and temperature profiles of a plasma column of 100 mm in diameter with a spatial resolution of 1.5 mm and features a lower density limit of 9×1018 m-3 (using 30 laser pulses of 0.55 J each, 10 Hz). First measurements at Magnum-PSI show that the design specifications are met and that on virtue of the high light collection power of the detection system even ne and Te profiles of the argon plasma expansion could be measured accurately at densities of 5×1018 m-3 and temperatures below 0.15 eV. In recent years the need arose for an accurate method to determine the ion properties in the plasma jet of the linear plasma generators. Therefore, the author initiated a feasibility study to find out whether CTS can be performed on Magnum-PSI to measure Ti, and moreover the macroscopic velocity of the plasma. It was demonstrated that Ti and the macroscopic velocity can be measured with an accuracy of 10% at ne = 5.0x1020 m-3 (test case: Ti = 2.5 eV, resolution 2.4 mm) and 15%, respectively. This can be achieved by accumulating 10 laser pulses of 1.2 J each, using the fundamental wavelength of a Nd:YAG laser. The proposed system may be used to prove that viscous heating of the ions in the plasma is the main cause for the ion temperature being much higher than the electron temperature in the magnetized plasma jet of Pilot-PSI and Magnum-PSI. Moreover, CTS experiments on Magnum-PSI can possibly prove that this technique is a viable ion temperature determination method for the ITER divertor; presently there are no good candidate techniques for ITER available