Improved accuracy and robustness of electron density profiles from JET's X-mode frequency-modulated continuous-wave reflectometers

JET's frequency-modulated continuous wave (FMCW) reflectometers have been operating well with the current design since 2005, and density profiles have been automatically calculated intershot since then. However, the calculated profiles had long suffered from several shortcomings: poor agreement with other diagnostics, sometimes inappropriately moving radially by several centimeters, elevated levels of radial jitter, and persistent wriggles (strong unphysical oscillations). In this research, several techniques are applied to the reflectometry data analysis, and the shortcomings are significantly improved. Starting with improving the equilibrium reconstruction that estimates the background magnetic field, adding a ripple correction in the reconstructed magnetic field profile, and adding new inner-wall reflection positions estimated through ray-tracing, these changes not only improve the agreement of reconstructed profiles to other diagnostics but also solve density profile wriggles that were present during band transitions. Other smaller but also persistent wriggles were also suppressed by applying a localized correction to the measured beat frequency where persistent oscillations are present. Finally, the burst analysis method, as introduced by Varela et al. [Nucl. Fusion 46 S693 (2006)], has been implemented to extract the beat frequency from stacked spectrograms. Due to the strong suppression of spurious reflections, the radial jitter that sometimes would span several centimeters has been strongly reduced. The stacking of spectrograms has also been shown to be very useful for stacking recurring events, like small gas puff modulations, and extracting transport coefficients that would otherwise be below the noise level.

IShTAR: A test facility to study the interaction between RF wave and edge plasmas

Existence of high electric fields near an RF antenna launcher causes a number of parasitic phenomena, such as arcing and impurity release, which seriously deteriorate the performance of an Ion Cyclotron Range of Frequencies (ICRF) heating scheme in fusion devices. Limited accessibility of the near antenna region in large-scale fusion experiments significantly complicates the associated experimental studies. The IShTAR test facility has been developed with the requirement to provide a better accessibility and diagnosability of plasmas in the direct vicinity of an ICRF antenna. The purpose of this work is to give a detailed description on the experimental setup and the available diagnostics. Furthermore, the paper will demonstrate the capability of the experiment to study phenomena near an ICRF antenna launcher which are relevant for large-scale fusion ion cyclotron resonance heating systems.

Theory of a cylindrical Langmuir probe parallel to the magnetic field and its calibration with interferometry

A theory for data interpretation is presented for a cylindrical Langmuir probe in plasma parallel to the magnetic field direction. The theory is tested in a linear low-temperature plasma device Aline, in a capacitive radio-frequency (RF) discharge. The probe is placed on a 3D manipulator, and a position scan is performed. To exclude strong RF perturbations, the probe is RF compensated. Using the theory, electron densities are obtained from the current at the plasma potential, where no sheath is present. Results are calibrated by line-integrated density measurements of a 26.5 GHz microwave interferometer. Reasonable agreement is observed for probe and interferometer measurements. Furthermore, preceding, more general probe theory is compared to the one developed in the current work and the application limits are discussed.

Turbulence level effects on conventional reflectometry using 2D full-wave simulations

Numerical simulations are critical in improving the capabilities of microwave diagnostics. In this work, the 2D finite-difference time-domain full-wave code REFMUL was applied to broadband turbulent plasmas using the conventional reflectometry setup. Simulations were performed with O-mode waves, fixed frequency probing, and I/Q detection. The plasma density, determining O-mode propagation, was modeled as the sum of a slab background plasma with a fluctuating component following a Kolmogorov-like amplitude k-spectrum. The density turbulence level δn _e/n _e was scanned over several orders of magnitude for simulated plasma flows of constant plasma velocity in either the radial or the poloidal direction. Simulations show trends, such as spectral broadening of the complex A(t)e^iφ(t) signals and increasing fluctuations in A(t) and φ(t) with increasing δn _e/n _e, that are similar for both plasma flow directions. These together with possibilities to reconstruct a poloidal wavenumber spectrum are discussed in view of extending the measuring capabilities. The onset of non-linear effects associated with phase runaway, as previously observed with other 1D and 2D codes, as well as radial Doppler effects is also observed and discussed.

Parametrization of reflectometry fluctuation frequency spectra for systematic study of fusion plasma turbulence

We describe a way to parameterize power spectra extracted from fixed-frequency reflectometry data, with a view to systematic studies of turbulence properties in tokamak plasmas. Analysis of typical frequency spectra obtained from a new database suggests decomposition in a set of four key components: the direct current component, low-frequency fluctuations, broadband (BB) turbulence, and the noise level. For the decomposition in the identified components, different kinds of functions are tested and their fitting performance is analyzed to determine the optimal spectrum parametrization. In particular, for the BB turbulence, three models are compared qualitatively based on a number of representative spectrum test cases, notably the generalized Gaussian, the Voigt, and the Taylor model. In addition, quantitative performance testing is accomplished using the weighted residual sum of squares and the Bayesian information criterion in a large database including 350 000 spectra obtained in Tore Supra. Next, parametrization by the Taylor model is applied to Ohmically heated plasmas, and a BB energy basin is systematically observed in the core plasma region, which shrinks with decreasing radial position of the q = 1 surface. This basin might be explained by a drop of the density fluctuation level inside the q = 1 surface.

New density profile reconstruction methods in X-mode reflectometry

The reconstruction method published by Bottollier-Curtet and Ichtchenko in 1987 has been the standard method of density profile reconstruction for X-mode reflectometry ever since, with only minor revision. Envisaging improved accuracy and stability of the reconstruction method, functions more complex than the linear are evaluated here to describe the refractive index shape in each integration step. The stability and accuracy obtained when using parabolic and fixed or adaptative fractional power functions are compared to the previous method and tested against spurious events and phase noise. The developed relation from the plasma parameters to the best integration shapes allows for the optimization of the reconstruction for any profile shape. In addition, the density profiles can be reconstructed using less probing frequencies without accuracy loss, which speed up the reconstruction algorithm and enable real-time monitoring of faster density profile evolution.

Assessment of the measurement performance of the in-vessel system of gap 6 of the ITER plasma position reflectometer using a finite-difference time-domain Maxwell full-wave code

We conducted a first assessment of the measurement performance of the in-vessel components at gap 6 of the ITER plasma position reflectometry with the aid of a synthetic Ordinary Mode (O-mode) broadband frequency-modulated continuous-wave reflectometer implemented with REFMUL, a 2D finite-difference time-domain full-wave Maxwell code. These simulations take into account the system location within the vacuum vessel as well as its access to the plasma. The plasma case considered is a baseline scenario from Fusion for Energy. We concluded that for the analyzed scenario, (i) the plasma curvature and non-equatorial position of the antenna have neglectable impact on the measurements; (ii) the cavity-like space surrounding the antenna can cause deflection and splitting of the probing beam; and (iii) multi-reflections on the blanket wall cause a substantial error preventing the system from operating within the required error margin.

A linear radio frequency plasma reactor for potential and current mapping in a magnetized plasma

Langmuir probe measurements in front of high power ion cyclotron resonant frequency antennas are not possible or simply too noisy to be analyzed properly. A linear experiment is a radio frequency (RF) magnetized plasma discharge reactor designed to probe the rectified potential in front of such antennas but at low power level (1 kW) to next improve antenna design and mitigate sheath effects. The maximum magnetic field is 0.1 T, and the RF amplifier can work between 10 kHz and 250 MHz allowing ion cyclotron resonances for argon or helium. The first measurements with no magnetic field are presented here, especially 2D potential maps extracted from the RF compensated probe measurements yield ni ≈ 10(15) m(-3) and Te ≈ 2 eV for RF power lower than 100 W. Series resonances in the chamber are highlighted and allow to deduce the plasma parameters from a simple equivalent impedance model of the plasma in helium gas. Next studies will be focused on magnetized plasmas and especially magnetized RF sheaths.

A numerical study of fixed frequency reflectometry measurements of plasma filaments with radial and poloidal velocity components

A 2D finite-differences time-domain full-wave code is used to simulate the measurements of plasma filaments with fixed frequency O-mode reflectometry. The plasma is modeled by a linear slab plasma plus a Gaussian perturbation propagating in a direction that can vary from poloidal to radial. The plasma background density gradient is chosen in agreement with the steep edge transport barrier of H-modes in the ASDEX Upgrade (AUG) tokamak. Illustrative results are presented and different types of reflectometry responses are observed depending on filament sizes and propagation directions. The reflectometry signatures obtained here with numerical simulations support previous experimental findings on filament measurements.

Note: On the measurement of plasma potential fluctuations using emissive probes

In this Note, it is pointed out that emissive probes cannot be used to directly and reliably measure plasma potential fluctuations. An experimentally validated model demonstrates indeed that the floating potential fluctuations of an emissive probe which floats at the mean plasma potential depend not only on the plasma potential fluctuations but also on electron density and temperature fluctuations.

Design and validation of the ball-pen probe for measurements in a low-temperature magnetized plasma

Ball-pen probes have been used in fusion devices for direct measurements of the plasma potential. Their application in low-temperature magnetized plasma devices is still subject to studies. In this context, a ball-pen probe has been recently implemented on the linear plasma device Mirabelle. Produced by a thermionic discharge, the plasma is characterized by a low electron temperature and a low density. Plasma confinement is provided by an axial magnetic field that goes up to 100 mT. The principle of the ball-pen probe is to adjust the saturation current ratio to 1 by reducing the electron current contribution. In that case, the floating potential of the probe is close to the plasma potential. A thorough study of the ball-pen probe operation is performed for different designs of the probe over a large set of plasma conditions. Comparisons between ball-pen, Langmuir, and emissive probes are conducted in the same plasma conditions. The ball-pen probe is successfully measuring the plasma potential in these specific plasma conditions only if an adapted electronics and an adapted probe size to the plasma characteristic lengths (λ(D), ρ(ce)) are used.

New signal processing technique for density profile reconstruction using reflectometry

Reflectometry profile measurement requires an accurate determination of the plasma reflected signal. Along with a good resolution and a high signal to noise ratio of the phase measurement, adequate data analysis is required. A new data processing based on time-frequency tomographic representation is used. It provides a clearer separation between multiple components and improves isolation of the relevant signals. In this paper, this data processing technique is applied to two sets of signals coming from two different reflectometer devices used on the Tore Supra tokamak. For the standard density profile reflectometry, it improves the initialization process and its reliability, providing a more accurate profile determination in the far scrape-off layer with density measurements as low as 10(16) m(-1). For a second reflectometer, which provides measurements in front of a lower hybrid launcher, this method improves the separation of the relevant plasma signal from multi-reflection processes due to the proximity of the plasma.

Fast sweeping reflectometry upgrade on Tore Supra

In order to study the temporal dynamics of turbulence, the sweep time of our reflectometry has been shortened from 20 to 2 μs with 1 μs dead time. Detailed technical aspects of the upgrade are given, namely, about the stability of the ramp generation, the detection setup, and the fast acquisition module. A review of studies (velocity measurement of the turbulence, modifications of the wavenumber spectrum, radial mapping of correlation time, etc.) offered by such improvements is presented.

Simulation of reflectometry Bragg backscattering spectral responses in the absence of a cutoff layer

Experimental reflectometry signals obtained in the absence of a cutoff layer, with the possibility of interferometric operation excluded, show a coherent and recurrent frequency spectrum signature similar to an Alfvén cascade signature. A possible explanation resides in the modulation of a resonant Bragg backscattering response by an Alfvén mode structure located at the center of the plasma whose frequency of oscillation modulates the backscattered signal in a conformable way. This situation is modeled and simulated using an O-mode full-wave Maxwell finite-difference time-domain code and the resulting signatures are discussed.

Turbulence in the TORE SUPRA tokamak: measurements and validation of nonlinear simulations

Turbulence measurements in TORE SUPRA tokamak plasmas have been quantitatively compared to predictions by nonlinear gyrokinetic simulations. For the first time, numerical results simultaneously match within experimental uncertainty (a) the magnitude of effective heat diffusivity, (b) rms values of density fluctuations, and (c) wave-number spectra in both the directions perpendicular to the magnetic field. Moreover, the nonlinear simulations help to revise as an instrumental effect the apparent experimental evidence of strong turbulence anisotropy at spatial scales of the order of ion-sound Larmor radius.

Study of ITER plasma position reflectometer using a two-dimensional full-wave finite-difference time domain code

The EU will supply the plasma position reflectometer for ITER. The system will have channels located at different poloidal positions, some of them obliquely viewing a plasma which has a poloidal density divergence and curvature, both adverse conditions for profile measurements. To understand the impact of such topology in the reconstruction of density profiles a full-wave two-dimensional finite-difference time domain O-mode code with the capability for frequency sweep was used. Simulations show that the reconstructed density profiles still meet the ITER radial accuracy specifications for plasma position (1 cm), except for the highest densities. Other adverse effects such as multireflections induced by the blanket, density fluctuations, and MHD activity were considered and a first understanding on their impact obtained.