Fusion.ucla.edu

Liquid Lithium Wall Experiments in CDX-U
R. Kaita,a R. Majeski, a S. Luckhardt,b R. Doerner, b M. Finkenthal,c H. Ji, a H. Kugel, a D. Mansfield, a D. Stutman, c R. Woolley, a L. Zakharov, a and S. Zweben a aPrinceton Plasma Physics Laboratory, Princeton University, Princeton NJ 08543-0451 USA bUniversity of California at San Diego, San Diego CA, 92093-0417 USA cJohns Hopkins University, Baltimore MD, 21218 USA X-ray array from the Johns Hopkins University withmultilayer mirror detectors selected for lithium line The concept of a flowing lithium first wall for a fusion radiation.[4] These can be used to determine the lithium reactor may lead to a significant advance in reactor design, influx into CDX-U discharges and the lithium content and since it could virtually eliminate the concerns with power density and erosion, tritium retention, and coolingassociated with solid walls. Sputtering and erosion tests are There is a small but growing experimental database o n currently underway in the PISCES device at the University transport of liquid lithium and the behavior of lithium i n of California at San Diego (UCSD). To complement this contact with plasmas. Sputtering and erosion tests are effort, plasma interaction questions in a toroidal plasma currently underway at divertor simulation facilities such as geometry will be addressed by a proposed new ground PISCES at the University of California - San Diego breaking experiment in the Current Drive eXperiment – (UCSD).[5] Lithium limiter experiments have also been Upgrade (CDX-U) spherical torus (ST). The CDX-U plasma performed on the T-11M device,[6] where a Capillary is intensely heated and well diagnosed, and an extensive Porous System was used to form a “self-restoring” liquid liquid lithium plasma-facing surface will be used for the lithium limiter surface.[7] However, introduction of large first time with a toroidal plasma. Since CDX-U is a small area lithium limiter targets and walls into existing tokamak ST, only ≈1 liter or less of lithium is required to produce a facilities has not yet taken place. This proposal describes toroidal liquid lithium limiter target, leading to a quick and the first experiments of this type, which will be done i n II. OVERVIEW OF LIQUID LITHIUM WALL EXPERIMENT Key liquid lithium-plasma interaction questions will beaddressed by a new experiment in PPPL’s existing CDX-Udevice.[1] The primary goal will be to measure theinteractions between the plasma and the lithium in anauxiliary-heated discharge whose surface contact is solelywith a large-area liquid lithium limiter. The objectives ofthese investigations are: Demonstrate operation of a toroidal plasma with liquid lithium as the sole plasma-wall contact. Initialoperation will be with the liquid lithium sample probesupplied by UCSD, followed by discharges utilizing a largearea liquid lithium pool as the target.
Investigate the effects of the toroidal plasma o n the lithium, including thermal and magnetohydrodynamic(MHD) interactions.
A first study of lithium transport will be conducted in CDX- U in FY99. Lithium will be introduced into CDX-U using the “spark plug” technique used in previous impurity injection experiments.[2] This study will use the existingset of CDX-U profile diagnostics, including a new Fig.1. Schemes for liquid wall studies on CDX-U.
multipoint Thomson scattering system,[3] and an ultrasoft The first investigations on the interaction of a spherical compact ST geometry (see Section VI for toroidal limiter torus (ST) plasma with liquid lithium will be performed using a liquid lithium sample probe. This probe consists of aheated sample manipulator which can be inserted into the CDX-U plasma chamber using an existing drive mechanism, as shown in Figure 1. The PISCES group at UCSD will design and construct the manipulator, and no modifications to the vacuum vessel are necessary because of the relatively small amounts of lithium (≈100 gm) used.
These sample exposures in CDX-U will permit the evaluation of effects that could not be investigated in the PISCES device, such as the dependence of lithium sputtering on ion angle of incidence, the importance of the magneticsheath on redeposition, and the role of the ion energy distribution function on the loss rate of lithium from the sample.[7] This initial experience with liquid lithium will couple into the design and installation of the toroidal liquid Table 1. Parameters of existing CDX-U facility.
In early CY00, we will reconfigure the CDX-U t oaccommodate a large area liquid lithium limiter target as Diagnosis of the effects of a low-recycling limiter target o n shown in Figure 1. The primary goal of these investigations the CDX-U plasma will utilize the extensive set of CDX-U will be to produce an ST discharge in which the plasma-wall diagnostics (spectroscopy, tangential bolometer array, interaction is dominated by a liquid lithium surface. The soft x-ray diode arrays, multipoint Thomson scattering, effects of operating with lithium walls will be quantified for the first time, greatly reducing the uncertainties ofincorporating liquid lithium walls in larger toroidal devices Effects of the plasma on the lithium will be diagnosed with spectroscopy (monitoring the neutral lithium lineemission at 670.8 nm), an infrared camera measuring the In summary, the steps in this effort will be as follows.
temperature distribution of the lithium surface, and a10,000 frame per second fast visible camera viewing waves Investigate the effects of high power density (8-10 and turbulence on the liquid lithium surface. The possibility of measuring 2-D profiles of the lithium density in the MW/m ) but short pulse plasma interactions with liquid limiter region, using a laser-induced fluorescence technique lithium, first with the liquid lithium sample probe and then a separately funded and under development by Fusion Physics and Technology, is also being explored.
Study thermal and MHD effects on the lithium under standard and “off-normal” conditions such asdisruptions.
The CDX-U research will involve the introduction of alarge-area, toroidal liquid lithium limiter target (Section The liquid lithium limiter on the bottom of the plasma 4VI). A liquid lithium wall should have a very low recycling chamber can also be compared directly with a solid limiter coefficient, and a direct comparison of a solid limiter on top by changing the vertical plasma position, so that versus liquid lithium target will be made by running the the discharge is limited on the upper or lower surface of the plasma on a solid limiter on the upper surface or a liquid limiter on the lower surface of the vacuum vessel. Inaddition, the effect of liquid lithium walls can be compared The CDX-U facility has recently undergone an extensive to boron pellet conditioning. The Boron Low Velocity program of upgrades which has resulted in an increase of Edge Micropellet Injector that was developed on CDX-U the toroidal field to 2.3 kG with a “flattop” of 100 msec.
The new power supplies for the vertical and shaping fieldspermit discharges with plasma current up to 150 kA for Auxiliary RF heating will permit investigation of the greater than 25 msec. All power supplies (with the effects of high power density plasma interactions with exception of the two capacitor banks) are preprogrammed liquid lithium, using existing edge and core diagnostics.
and controlled by digital to analog waveform generators.
Discharge start-up and plasma current penetration are key The plasma geometry remains substantially unchanged, issues that need to be resolved in the formation of plasmas with the basic discharge parameters summarized in Table 1.
in the presence of a lithium limiter target. Deuteriumfueling efficiency and lithium impurity accumulation will The Ohmic heating system is capable of providing 0.2 MW also be investigated with the spectroscopic diagnostics o n to CDX-U, and the facility also has a radio frequency (RF) CDX-U. Possible surface coatings such as lithium hydride heating system[8] that is rated at 0.3 MW. The resulting might be formed which could necessitate plasma parallel and normal heat fluxes will be 8-10 MW/m2 and 2- 3 MW/m2, respectively, over 25 to 50 msec because of the be an issue for liquid lithium targets.
Among the effects of the plasma on the liquid lithium A toroidally-continuous shroud will also be mounted below surface to be investigated are jxB forces that result from the limiter target and on the center stack. The purpose of MHD activity and disruptions. They can cause toroidal the shroud is to protect vacuum vessel structures in the currents to flow within the toroidally-continuous lithium vicinity of the limiter target when it is heated up to 500 limiter target. The degree to which the liquid lithium serves degrees C. This structure will be cooled with silicone as a conducting “shell” that affects plasma current diffusion pump oil, because of its low viscosity and formation, position control, and MHD stability will also compatibility with high vacuum and the presence of The amount of lithium to be used in the proposed displacement events (VDE’s) toward the liquid lithium experiments can be estimated from the dimensions of the limiter target. Halo currents induced by VDE’s might cause limiter target. If its depth is 0.5 cm, the quantity will be the lithium to “splash,” but its large surface tension and ≈1,000 cubic centimeters. Because this is comparable to adhesion may prevent this in practice.
the amount used previously on TFTR,[10-12] thisexperience is relevant to the lithium quantities, handling, Surface impurities may interfere with lithium’s normal and safety analysis required for the work on CDX-U.
181°C melting, as has been found in some previous lithiumexperiments.[10-12] The lithium temperature can be varied The facilities at PPPL that were used to prepare the lithium by heaters on the toroidal limiter container up to 500°C t o samples for the TFTR lithium experiments are available for address this issue, and the influx of lithium as it evaporates the proposed work on CDX-U. They include a glove box, will be monitored spectroscopically as a function of target vacuum chamber, and heater for handling lithium samples and testing various liquid metal container concepts prior t oinstallation in CDX-U.
Experiments will also be done to discharge clean thelithium surface. A 2.45 GHz radio frequency source i s The transfer procedures will be similar to those used for available, and it can operate CW at about the 5 kW level.
TFTR. The UCSD lithium sample will be mounted on a The resonance can be located radially in the vicinity of the probe drive, and will be brought to CDX-U in a transfer limiter target and swept over its surface by oscillating the container filled with argon. The probe assembly will then be connected to a valve on the CDX-U plasma chamber andpumped out before insertion into the vacuum vessel.
The toroidal container for the lithium target will beconstructed in sections as described above, and installed The first experiments will be performed with a liquid while the CDX-U plasma chamber is vented. Pieces of lithium sample probe inserted into the CDX-U plasma lithium will be transferred to CDX-U in an argon-filled chamber, and no modifications will be made to the vacuum container. The plasma chamber will then be filled with vessel because of the relatively small amounts of lithium argon, after which the lithium pieces will be distributed (<1 gm) used. In early CY00, the CDX-U chamber and evenly around the target container. The chamber will then pumping system will be modified to accommodate the be pumped down, and the toroidal container will be heated larger quantities of lithium required by the toroidal liquid lithium limiter. It is sufficient for this purpose to replace afew aluminum port covers with stainless steel, and t o install a cold trap on the turbopump.
In summary, the tasks to be undertaken are: The design for a toroidal liquid lithium target is intended t obe inexpensive and to simplify lithium handling. Lithium Investigate the effects of plasma on a lithium will be introduced to the assembled limiter target inside target, first with a liquid lithium sample probe, then with a CDX-U as a solid, and then melted in place. Resolidified large area liquid lithium toroidal limiter target.
lithium will be removed from CDX-U via limiter targetsector removal.
Monitor lithium influx spectroscopically and investigate the effect of a very low recycling limiter on the The annular toroidal limiter target will extend radially plasma. Measure the dependence of the lithium content i n between its inner and outer sidewalls, which are located atR=29 cm (=R0-a/4) and R=39 cm (=R0+a/4). The limiter the plasma core on the temperature (up to 500oC) of the target will be constructed of stainless steel in four 9 0 degree sectors. “Knife-edge” straight interfaces between thelimiter target’s sectors are kept tightly pressed against each other to prevent lithium leakage, with the pressure between the plasma and the lithium, including maintained by pairs of clamps adjacent to each interface the results of a forced disruption on the lithium, and the which force the sectors together. The limiter target sectors effects of lithium on current penetration and discharge wil be mounted on insulators that provide thermal and electrical separation between the limiter target and thevacuum vessel.
A liquid metal jet/droplet injector could be deployed after experiments in a toroidal plasma device where the the static toroidal liquid lithium experiments. This will dominant plasma-wall interaction will be with a liquid provide a test of the performance of flowing liquid metals lithium surface. The research will begin with a lithium in contact with the plasma boundary, an essential feature of sample probe (in collaboration with UCSD), and will be followed by the introduction of a large-area, toroidal liquidlithium limiter.
The liquid metal jet/droplet injector is being developed b yUCSD. Preliminary work on liquid metal droplet formation Auxiliary heating in both stages will be provided by 0 . 3 in the PISCES-A plasma device at UCSD has demonstrated MW of RF heating, with power deposition in the electron what is required to control a high surface tension liquid channel. The local power densities are expected to be in the metal, and many of the basic fluid properties have been investigated. A fast CCD imaging system has been used t o range of 8-10 MW/m . The CDX-U facility has an study the trajectory and evolution of liquid metals as they extensive set of diagnostics already in place that are pass through the discharge, and it has produced valuable capable of evaluating lithium-plasma interactions.
insight into what should be expected as large quantities of Measurements with these systems should be able to provide liquid metal are introduced into the plasma-vacuum system.
benchmarking for modeling future liquid lithiumexperiments.
The first stage would be to construct and operate a dropletinjector for seeding the plasma with a liquid metal curtainin the PISCES-A machine at UCSD. This would be followed by its installation on CDX-U. The experience gained i nthis project and the static large area toroidal liquid lithium Experiments with divertor plasmas can be done with small experiments can be used to investigate moving liquid modifications to the CDX-U facility. Installation of lithium, including jxB flows, in CDX-U.
additional poloidal field coils to produce a single nullconfiguration could be performed after the completion of the limiter experiments with the toroidal liquid lithiumlimiter target. The coils will be relatively simple, external The authors acknowledge the assistance of N. Pomphrey, to the vessel, and utilize existing power supplies.
B. Jones, and T. Munsat in the equilibrium reconstructionsfor the proposed CDX-U divertor plasma configuration.
The Tokamak Simulation Code (TSC) has been used t o This work was supported by USDOE Contract No. DE- determine if it would be possible to create a single-null discharge in CDX-U with a modest modification to theexisting coil system. The plan would be to add a pair of 180 kA-turn coils which are simple to wind and install aboveand below the vacuum vessel for future divertor operation.
et al. , Proceedings of the 17th IAEA
The result of a TSC calculation with the addition of these Fusion Energy Conference, Yokohama, Japan,
coils is shown in Figure 2. By choosing an upper or lower October 19-24, 1998, IAEA-CN-69/CDP/12 (1998) null, divertors made of molybdenum (a) and liquid lithium [2] F. M. Levinton and D. D. Meyerhofer, Rev. Sci.
(b) can be compared directly in a fashion similar to limiter Instrum. 58, 1393-1400 (1987)
experiments. This allows experiments with H-mode [3] T. Munsat and B. LeBlanc, Rev. Sci. Instrum. 70, 755-
wall that are closer to those on NSTX or Alcator C-Mod.
et al. , Rev. Sci. Instrum. 70, 572-576
et al. , J. Vac. Sci. Technol. A8, 1790-1797
al. , 26th
EPS Conference o n
Controlled Fusion and Plasma P h y s i c s ,
Maastricht,
al. , J. Nucl. Mater. 2 4 1 - 2 4 3 , 1190-
et al. , Physics of Plasmas 6, 2002-2008
et al. , Rev. Sci Instrum. 70, 493-497 (1999)
al. , Proceedings of the 1 7 t h
IEEE/NPSS Symposium on Fusion E n g i n e e r i n g ,
San Diego, CA, October 6-10, 1997, 869-872 (1998) al. , Proceedings of the 1 7 t h
IEEE/NPSS Symposium on Fusion E n g i n e e r i n g ,
Fig. 2. TSC simulations of single-null divertor plasmas in CDX-U.
San Diego, CA, October 6-10, 1997, 873-876 (1998)[12] D. K. Mansfield al. , Phys. Plasmas 3 , 1892-1897

Source: http://www.fusion.ucla.edu/kaitapub.pdf