Behaviour of carbon and boron impurities in the Madison Symmetric Torus

Temporally and spatially resolved measurements of carbon and boron impurity density are obtained in the reversed field pinch (RFP) for the first time. It is observed that, unlike in tokamaks and stellarators, the RFP does not exhibit a centrally peaked impurity profile in either standard plasmas where field lines have some degree of stochasticity, or improved confinement discharges where there exist well-nested flux surfaces for a substantial fraction of the plasma volume. Results from improved confinement discharges also indicate an outward convection of impurities from the core of the plasma. It is well known that plasma–wall interaction in fusion devices liberates impurities [1]. Deliberate injection of impurities at the edge of the plasma has also been applied in many fusion devices as a method to reduce heat load to the wall by forming a radiative edge [2]. Transport of these edge impurities in the plasma is governed by the electric field and thermal forces. Collisional (neoclassical) theory predicts that these edge impurities have a natural tendency to accumulate in the core of the plasma [3]. This has been experimentally observed in many experiments in high confinement regimes [4–7]. This poses a serious concern as these impurities can have deleterious effects on the performance of a fusion device such as fuel dilution and increased radiative loss. In the reversed field pinch (RFP), the global magnetic reconnection activities enhance plasma–wall interactions and liberate impurities into the plasma [8]. A good understanding of impurity generation and transport is crucial for RFP operation. Even though there exists a large number of theoretical and experimental studies on tokamaks and stellarators, there is a paucity of time-resolved and spatially localized information about impurities in RFP plasmas. Impurity studies in RFPs have been supported by line integrated emission measurements [9–12]. A large range in the transport parameters in different machines is found in these works. 3 Currently at the Laboratory for Laser Energetics, University of Rochester, NY, USA. 0741-3335/11/032001+06$33.00 © 2011 IOP Publishing Ltd Printed in the UK & the USA 1 Plasma Phys. Control. Fusion 53 (2011) 032001 Brief Communication In this paper, we present the first temporally and spatially resolved impurity density measurements in the Madison Symmetric Torus (MST) RFP (major radius R = 1.5 m, minor radius a = 0.52 m) [13]. Measurements of fully stripped boron and carbon impurities are obtained using charge-exchange recombination spectroscopy (CHERS). Experimental observations indicate a flat/weakly hollow impurity density profile for standard plasma discharges. Sawtooth (global magnetic reconnection) events in standard discharges result in a significant increase and the radial redistribution of impurities. (A description of sawtooth events in MST and their impact on main plasma parameters is available in previous publications. For example, see [14].) In improved confinement discharges, the hollowness of the radial profile increases, and the core impurity density slowly decays in time concurrent with an increase in impurity density in the outer regions of the plasma, indicating an outward impurity convection. The reason for this outward convection is not clear. Experiments are conducted in deuterium plasmas of toroidal plasma current∼400 kA, line averaged central electron density ne ∼ (0.8–1) × 1019 m−3 and central electron temperature ranging from 400 eV (in standard discharges) to above 1 keV (in improved confinement discharges). Improved confinement discharges are achieved by the ‘pulsed poloidal current drive’ (PPCD) scheme [15]. The main characteristics of PPCD discharges are low levels of magnetic fluctuations, higher core electron temperature and improved energy confinement [16–18]. Major sources of carbon impurity in MST are toroidal and poloidal graphite limiters. Recent boronization efforts in MST resulted in increased boron levels in the plasma. Measurements of carbon and boron impurities are done by collecting C VI emission at λ = 343.4 nm (n = 7 to n = 6 transition) and B V emission at λ = 298.1 nm (n = 6 to n = 5 transition), respectively. These emissions are stimulated by charge exchange between fully stripped ions and 50 keV hydrogen atoms injected radially using neutral beam injector. Distinct features of CHERS on MST are good spatial (∼2 cm) resolution and a dynamic background subtraction for good temporal resolution (up to 100 kHz) [19–21]. The spectrometer has recently been calibrated for radiant sensitivity using a tungsten–halogen lamp for absolute impurity density calculations from charge-exchange emission brightness. (A 10% uncertainty is estimated in the impurity density, mainly due to uncertainties in the beam attenuation calculations and transmission efficiencies of the viewing lenses.) Radial and temporal profiles of C6+ density are shown in figures 1(a) and (b), respectively. Data are averaged over a large number of sawtooth events. Profiles away (∼1.5 ms before) from the sawtooth event represent equilibrium profiles. The C6+ density has a flat/weakly hollow profile both at and away from the event. There is a global increase in the C6+ density at sawtooth events. The ‘burst’ of impurity density at a sawtooth event is more apparent from the sawtooth averaged temporal evolution shown in figure 1(b). The chord at ρ ∼ 0.75, where ρ is the radial distance from the magnetic axis normalized to the plasma minor radius, shows the maximum increase, which is consistent with the fact that sawtooth activity leads to the liberation of impurities from the wall. Impurities injected at a sawtooth event are expelled before the subsequent event. The radial profile of C6+ density for PPCD discharges shown in figure 2(a) indicates a clearly hollow profile and a steep gradient at ρ ∼ 0.6. Temporal behaviour of the C6+ density at three radial locations (ρ ∼ 0.02, 0.55 and 0.76) is plotted in figure 2(b). A slow decay of the core impurity density in time after the transition to improved confinement is apparent. The impurity density at the outer regions of the plasma, on the other hand, is slowly increasing. Measurements of B5+ density also show similar behaviour for both standard and PPCD discharges, as shown in figure 3. Standard discharges in MST have multiple, coupled tearing modes and the magnetic field lines have some degree of stochasticity for most of the plasma volume. Therefore, the flat