Density fluctuation measurements via beam emission spectroscopy (invited)

Abstract

Previous studies of plasma microturbulence have indicated that the fluctuation power scales with radial wave number, k⊥ , like k⊥−2→k⊥−3.5 for k⊥ ≥2 cm−1. This implies that low k fluctuations may dominate the spectrum. Beam emission spectroscopy (BES) has been developed to provide spatially localized measurements of density fluctuations in this low k region of the spectrum (k⊥ ≤2 cm−1). A 20-channel system has been installed on TFTR which images one of the heating neutral beams (via fiber optics) onto a set of photoconductive photodiode detectors. Fluctuations in the fluorescent Dα emission from the beam can be related to the local plasma density fluctuations via a model of the atomic excitation processes. The analysis of BES data utilizes many of the standard statistical analysis techniques such as power spectra, coherency and cross phase, and correlation analysis which are also used in the analysis of, for example, Langmuir probe data. In the case of BES however, these techniques require some special modifications to account for systematic effects such as photon statistics and fluctuations in the neutral beam density induced by the strong fluctuations near the plasma edge.

The beam emission spectroscopy diagnostic on the DIII-D tokamak

“…BES diagnostic systems have been previously deployed on several tokamaks: PBX-M, 1 TFTR, 2,3 Phaedrus-T, 4 and TEXT-U. 5 To deploy the BES diagnostic on DIII-D, numerous advances and technological innovations were required.…”

Plasma Turbulence Imaging via Beam Emission Spectroscopy in the Core of the DIII-D Tokamak

“…1. The purpose of these channels is to measure and isolate any fluctuation components on the neutral beam itself that do not represent local plasma fluctuations and thus should be subtracted from the measured signals [14]. Such common-mode fluctuations arise predominantly from fluctuations in the neutral beam ion source [19], as well as large-amplitude edge fluctuations that become imprinted on the beam due to fluctuating beam attenuation.…”

“…The optical and detection systems allow for multi-channel measurements and good localization in the radial-poloidal plane. For the system discussed here, 32 channels are deployed in a 5 (radial) × 6 (poloidal) 2D grid, plus two common-mode rejection channels [14]. Each channel images an approximately 0.9 cm (radial) × 1.2 cm (poloidal) region with channels located immediately adjacent to each other in each direction, for a total sampling area of approximately 5 × 7 cm.…”

Experimental Signatures of Critically Balanced Turbulence in MAST

“…We fit C (∆x = 0, ∆Z, ∆t = ∆t peak (∆Z)) taken at the time delay ∆t peak (∆Z) when the correlation function is maximum at a given ∆Z [30], to the function f τ (∆Z) = exp [− |∆t peak (∆Z)| /τ c ]. The reliability of this method relies on the temporal decorrelation dominating over the parallel spatial decorrelation, viz., we require τ c ≪ ℓ cos α/U φ .…”

“…[29]), where v BES is calculated at each radial location using the cross-correlation time delay (CCTD) method [30]; (iv) the estimated error in the calculation of v BES is > 20%; (v) p Z or p x > 0.5. The last two exclusion criteria pick out the cases when MHD modes are too strong; they are known to degrade the reliability of the BES data [29].…”