The effect of lithium wall conditioning in TFTR on plasma–surface interactions

doi:10.1016/S0022-3115(98)00863-0

Journal of Nuclear Materials

Volumes 266–269, 2 March 1999, Pages 1303-1308

https://doi.org/10.1016/S0022-3115(98)00863-0 Get rights and content

Abstract

Five TFTR deuterium supershots with increasing Li pellet injection are analyzed in detail. Five chords of experimental H-α measurements are compared to predictions from a series of computational models. First, experimental data from the discharge is used in the TRANSP plasma transport code to predict the ion flux to the wall. Then a modified version of the DEGAS neutral transport code which includes both reflection, desorption and sputtering of hydrogenic species from the wall is used to determine the neutral density profile across the machine. This data combined with the known density and temperature contours predicts values for the magnitude of H-α light observed for 16 viewing angles of the diagnostic. To match the experimental data, the wall reflection, desorption and sputtering coefficients were altered using data from VFTRIM-3D to include the effect of the added Li. In addition, the first 10 cm of the stainless steel wall adjoining the C inner bumper limiter was treated as C-covered; the highest flux area of the inner wall was treated as a sink; and the lower reflection coefficients for a Li-wall rather than a C-wall were used over an increasingly larger area of the inner wall as the Li concentration in the discharges increased.

Introduction

The performance of TFTR is greatly enhanced when the carbon inner bumper limiter has been scoured of its imbedded D by the repeated production of high power discharges fueled only by the desorbed gas. When Z_eff reaches a value of approximately 6 all of the easily desorbed surface D has been removed. The wall is then fully conditioned and a `supershot' plasma is formed [1].

In May 1994, a series of supershots were performed keeping all parameters constant except for the addition of Li pellets. The first shot of the series had no Li injection and served as a baseline. The subsequent four shots had the same amount of Li injected into each, thus increasing the total Li content on the walls during the series. The performance of the plasma improved as the total Li content increased. This paper endeavors to explain why in terms of the plasma-surface interactions.

The primary diagnostic used in this work is a spectroscopic measurement of the H-α light emanating from differing chords across the minor radius of the plasma. This data is compared to modeling results for the discharge. In previous work a variety of plasma parameters in the models were altered 2, 3to produce a fit to the data. Changes of that type were inadequate to match the data from these experiments. In this work the only variables changed in the model are the plasma–surface interaction of the ions and neutrals on the walls. These variables include the absorption/re-emission characteristics, the sputtering coefficient of trapped D, and the reflection coefficients.

Section snippets

Simulation

The TRANSP plasma analysis code is used to model the time evolution of the plasma parameters 3, 4. The energy, particle, and magnetic field dynamics are computed based on the measured plasma profiles and location of the last closed flux surface. The measured inputs include time-dependent profiles of the electron density, electron and carbon temperatures, and carbon toroidal velocity. The plasma scrape-off length used in TRANSP was held constant in all cases at 1.60 cm.

These measured parameters

Experiment

The five TFTR discharges examined in this study are # 76649, 76650, 76651, 76653 and 76654 from May 23, 1994. These TFTR supershots were performed after a long series of wall conditioning which desorbed the inner bumper limiter of deuterium. Supershots are circular cross-section plasmas which ride on the inner bumper limiter and are only fueled by the neutral beams. Therefore the interaction with the wall dominates recycling. There was no Li injection in the first shot and then two identical

Results

The presence of increasing accumulated Li on the walls had several effects. First, the Z_eff at 4.2 s increased from 2.3 to 3.0. This increase is inevitable with the addition of an impurity. However, the plasma performance improved dramatically. The total stored energy increased from 3 to 4.5 MJ. The central electron line density at that time also increased from 2.5 to 3.9 × 10¹⁹ m⁻³ due to a peaking of the density profile. The peak electron density (at r/a=0) rose from 5.1 to 6.5 × 10¹⁹ m⁻³ while

Conclusion

The wall model chosen has a significant impact on both “I” and being able to match the experimental HAIFA data. To properly include the effect of conditioning, absorption only occurs on the segments that have the highest ion flux. This conditioning does not remove all of the embedded D, since the sputtering of D from those areas is essential to produce a good fit to the data. Once Li is added absorption on the high-particle-flux-receiving segments remains strong since Li also can trap

Acknowledgements

This work was supported under the TFTR collaborators program, subcontract DOE PPPL S-03991-G.

References (9)

R.V. Budny et al.
J. Nucl. Mater.
(1992)
D. Heifetz et al.
J. Comput. Phys.
(1982)
D.N. Ruzic
Nucl. Instrum. Methods B
(1990)
J.D. Strachan, et al., Phys. Rev. Lett. 72 (1994)...

There are more references available in the full text version of this article.

Cited by (8)

Elastic backscattering as a method for the measurement of the integral lithium content in thin films on fusion-relevant substrates
2019, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms
Citation Excerpt :
Plasma Material Interactions in fusion devices have a crucial impact on plasma performance. It has been demonstrated, that the deposition of thin lithium films on plasma-facing components (PFCs) can improve the plasma performance by reducing the power threshold for access to high confinement modes (H-modes) [1–4]. Lithium has been also successfully used as plasma-facing material (PFM) in liquid limiters [5–8] and is considered to be used as liquid divertor material [9–11].
Different ion beam analysis techniques for the study of thin lithium-containing layers on top of fusion relevant materials are discussed and compared to each other. Elastic backscattering analysis (EBS) with protons is determined to be one of the most promising techniques and allows measurements of Li layers with thicknesses from ∼100 nm up to ∼600 μm, as shown by SIMNRA simulations. The best sensitivity for thin films (∼100 nm) can be achieved using 4 MeV protons with 170° scattering detection angle for layers on Mo and W substrates, and 2 MeV for C substrates. Experimentally EBS measurements were successfully tested for Li films with thicknesses from ∼50 nm up to ∼400 nm after air exposure. The Li films become strongly inhomogeneous and require averaging over multiple measurements in nearby areas. This necessitates averaging over multiple nearby measurement points, and limits the overall precision of the measurement.
Vertical flow in the Thermoelectric Liquid Metal Plasma Facing Structures (TELS) facility at Illinois
2015, Journal of Nuclear Materials
Citation Excerpt :
Liquid lithium suffers from none of these deficiencies, and as a result may provide better power handling than tungsten. Lithium PFCs also offer the ability to reduce edge recycling [7,8], resulting in a hotter edge region, increasing the volume of the fusion device at fusion temperatures [9]. Lithium has been shown to aid in suppressing ELMs [10], and to enhance plasma performance [11–13].
Flowing liquid metal PFCs may offer a solution to the issues faced by solid divertor materials in tokamak plasmas. The Liquid–Metal Infused Trenches (LiMIT) concept of Illinois Ruzic et al. (2011) is a liquid metal plasma facing structure which employs thermoelectric magnetohydrodynamic (TEMHD) effects to self-propel lithium through a series of trenches. The combination of an incident heat flux and a magnetic field provide the driving mechanism. Tests have yielded experimental lithium velocities under different magnetic fields, which agree well with theoretical predictions Xu et al. (2013). The thermoelectric force is expected to overcome gravity and be able to drive lithium flow along an arbitrary direction and the strong surface tension of liquid lithium is believed to maintain the surface when Li flows in open trenches. This paper discusses the behavior of the LiMIT structure when inclined to an arbitrary angle with respect to the horizontal.
Liquid lithium divertor system for fusion reactor
2009, Fusion Engineering and Design
One of the most critical issues for the steady state fusion reactor is the heat flux in the divertor target. This paper proposes a liquid lithium divertor system to solve this problem. The proposed divertor system consists of a liquid lithium target, an evaporation chamber and a differential evacuation chamber. The heat coming from the fusion plasma along the divertor leg is removed by evaporation of lithium. The lithium vapor is condensed on the wall and is circulated with a pump. The coolant temperature for the wall is high enough to drive a power generator. Narrow slits along the divertor leg and the differential evacuation chamber reduce leakage of lithium vapor to the plasma chamber. A preliminary estimation predicts that the lithium ion density in the core plasma is lower than the plasma density.
Characteristics of lithium thin films under deuterium ion implantation
2003, Journal of Nuclear Materials
To clarify characteristics of thin Li films as a plasma facing material, effect of modification of chemical composition of μm-thick deposited Li films introduced during the sample preparation and implantation with 1-keV deuterons on deuterium retention and thermal desorption properties is studied. Accelerator analysis has been employed to measure the thickness, the retention of D, H, O and C in the Li film, while thermal desorption spectroscopy has been utilized to investigate the chemical composition. The Li films exposed to moist atmosphere and subjected to D implantation show almost perfect isotope exchange from LiOH to LiOD. Carbon deposition during ion implantation is found to form a layer mixed with Li. With increasing the deposited amount of C, D₂ desorption behavior approaches that of pure C.
Measurements and modelling of solid phase lithium sputtering
2002, Nuclear Fusion
Liquid lithium divertor system in a spherical tokamak reactor
2009, IEEJ Transactions on Fundamentals and Materials

View all citing articles on Scopus

View full text