DIII-D RESEARCH OPPORTUNITIES FORUM FOR THE 2008 EXPERIMENTAL CAMPAIGN
Login | Review | Submit | Logout | Help
Questions about this website? Contact Andrew LeBlanc Questions about ROF? Contact Chuck Greenfield
| Title | 351: Recovery of D from thick exfoliated C-codeposits mechanically or using localized thermal oxidation | ||
| Name: | Peter C. Stangeby ( |
Affiliation: | GA ,LLNL and U of Toronto |
| Research Area: | Hydrogenic Retention | Presentation time: | Not requested | Co-Author(s): | Phil West |
| Description: | DIII-D studies would be used to establish the conditions required to cause exfoliation of carbon codeposits and to optimize the process, e.g. to minimize dust formation and to avoid UFO exfoliated flakes entering the confined plasma where they could cause disruptions. Water-cooled surfaces in plasma-hidden regions, as used in JET (louvers) may be advantageous but heated ones may also have benefits, e.g. reduction of the D/C (simulating T/C) in the co-deposits. In initial studies, the exfoliated material created in the hidden region (lower pump entrance) would simply be collected in a tray which would be removed at the time of a vacuum opening. In later studies a conveyor-belt system could be tested. If heat-able co-deposition surfaces are used, then localized oxygen recovery of D (T) from the co-deposits could be used as an alternative to mechanical removal. The rest of the vessel would not need to be heated to the temperatures at which O-baking is effective at removing carbon co-deposits (and releasing the D (T)), i.e. 250-350C, but just the codeposits themselves would be heated while the (cold) vessel was filled with O2. If the O2 was introduced before exfoliation of the co-deposits occurred, then the heating would be applied to the deposition surface, while if after exfoliation, the heating would be applied to the tray into which the exfoliated material fell. Although lab studies have not yet been carried out on carbon co-deposition flakes/dust, it seems likely that their reactivity to hot O2 will be greater than for still-adhering co-deposit films. Such localized O-baking to recover tritium would not involve the problems caused by having to raise the entire vessel and all its internal components to high temperature.
Disruptions caused by exfoliated flakes entering the plasma was initially a problem in the Tore Supra DITS Project, (Deuterium Inventory in Tore Supra) although it was not a problem in the JET DTE1 campaign, presumably because of the basic differences in configuration between limiters and divertors: a limiter surface directly contacts the confined plasma, which is not the case for divertor target surfaces. In addition, if the co-deposits are formed largely at locations out of plasma contact, as occurred in JET, then the risk of exfoliated debris causing a disruption is further reduced and perhaps eliminated. Nevertheless, the risk that this process will cause a disruption is sufficiently serious that this aspect should be included as a central element in the project. | ||
| Experimental Approach/Plan: | DIII-D hybrid discharges would be ideal for creating thick carbon co-deposits [Luce, et al, NF 43 (2003) 321]. This could be done by operating for 1 - 2 weeks using such discharges, at the end of the 2008 campaign, with the outer strike point placed at the entrance to the lower pumping plenum. During the vessel opening tiles would be removed from this region for analysis of, and lab experiment on, the co-deposits. The tiles at the strike point location might be sufficiently worn that they would have to be replaced in any case.
Recent lab studies in Toronto using thick codeposits from JET have established that thermal oxidation erodes thick carbon co-deposits more rapidly than thin ones. Almost all earlier laboratory studies have employed thin, of order microns, codeposit films, e.g. the ones from DIII-D: Thin (~2 micron) DIII-D divertor codeposits with low impurity content (<5% B): Oxidation can remove > 85% of the initial D content in 2 h at 623 K and >20 kPa O2 pressure; higher B content, however, reduces the total amount of D removed. Thick (10-250 micron) JET MkII-GB divertor codeposits with high impurity content (up to 50% Be/(Be+C)): During oxidation at 623 K and 20 kPa O2 pressure the initial D removal rates increase nearly linearly with initial D content; ~55% of the initial D is removed in the first 15 minutes, independent of thickness and Be content. After 8 h of oxidation > 90% of the initial D was removed. This implies that the structure of the codeposit is highly porous such that erosion occurs throughout the layer, and not from the geometric surface inward. The different effects of B (DIII-D) and Be (JET) on the codeposit removal rates is attributed to different chemistry. Implication for a DT device: assuming codeposits have similar structure and impurity content as the JET codeposits, we would expect that thermo-oxidation at 623 K and 20 kPa O2 pressure would remove > 90% within a day, independent of the codeposit thickness. [Haasz, et al, US-BPO: ITER Summary/WG-1/Task 5/Topic #5, Sept 13 2007]. It will be important to confirm this highly promising JET-tile finding for thick carbon co-deposits formed in DIII-D. It seems likely that the rate of oxidation for exfoliated carbon co-deposit fragments would be still faster than for thick, still adhering ones, and perhaps lower O-baking temperatures will be adequate for tritium recovery. It will be an important objective of the DIII-D project to establish if this is the case. In addition to providing data on thermal oxidative removal of thick, exfoliating codeposits, both in situ and ex situ, the 2008 campaign experiments would also be used to measure the amount of codeposit formation on the hidden region under the lower pumping entrance, providing the information required to design future experiments that used local heating of codeposits at that location. | ||
| Background: | In the JET DTE1 experiment, most of the tritium retained in the vessel was contained in
carbon codeposits which initially formed as adhering layers on the water-cooled louvers in the (plasma-hidden) entrance to the inner pumping duct. When they became thick, these layers exfoliated spontaneously and the tritium-containing flakes/dust then fell to the bottom of the JET vessel. If JET had had a flake collection system in place at the time of the DTE1 campaign to catch this tritiated exfoliated material, then it could have been readily removed from the vessel and the tritium recovered by heating the material in vacuum to ~ 1000C. Various mechanical recovery methods have been proposed, including vibratory conveyor-belt systems that would allow the exfoliated tritiated material to be removed continuously from the vessel [EU TASK No:DV7A-T438 Development of dust detection and removal techniques in tokamaks, GF Counsell, Euratom/UKAEA Fusion Association, 30 November 1999.] JET had not anticipated the specific nature of the carbon codeposition process that occurred in the DTE1 campaign nor the exfoliation of the codeposits. Had it done so, then catchers (troughs) could have been installed in advance of the DTE1 experiment, into which the exfoliated tritiated material could have fallen, for later removal (not requiring anything so complex as a conveyor belt system) and ex situ tritium recovery. While JET did not exploit the opportunity to demonstrate tritium recovery by such a means, DIII-D is equally well placed to develop and demonstrate this method. The key requirement is the existence of a large opening that is itself shadowed from plasma contact but which directly and immediately faces the divertor target strike point. In JET this was the entrance to the lower inner pumping plenum, where water-cooled louvers are located and on which the carbon codeposits formed initially, before exfoliating. In DIII-D the entrance to the lower outer pumping plenum could be used similarly. Reduction of the T/C ratio in carbon co-deposits by strong surface heating: in JT-60U the divertor locations where the plasma heating was so strong that the surface temperatures reached ~1000C had carbon co-deposits with very low hydrogen content. This is potentially an important way to reduce the amount of tritium retained in the first place in carbon co-deposits, reducing the required frequency for applying any recovery process such as O-baking. DIII-D hybrid discharges heat the targets to very high temperatures, ~ 1200C. These discharges will therefore be ideal for reducing the deuterium content in the carbon co-deposits formed in DIII-D and will permit study and quantification of this important tool for tritium control. | ||
| Resource Requirements: | -- | ||
| Diagnostic Requirements: | -- | ||
| Analysis Requirements: | -- | ||
| Other Requirements: | -- | ||
Print this page