- Количество слайдов: 15
Reducing the Costs of Targets for Inertial Fusion Energy G. E. Besenbruch D. T. Goodin, J. P. Dahlburg, K. R. Schultz, , A. Nobile 1, E. M. Campbell General Atomics, P. O. Box 85608, San Diego, California 92186 -5608 1 Los Alamos National Laboratory, Albuquerque, New Mexico HAPL Project Review Pleasanton, California November 13 -14, 2001 (IFSA 2001 Paper #1113)
Feasibility of economical target fabrication is a critical issue for IFE power plants · A number of power plant conceptual designs are available - pulsed power systems that operate at ~6 -10 Hz · Must supply about 500, 000 targets per day with: - precision geometry, and cryogenic, layered DT fill Concept for “HILIFE-II” IFE 1000 MW(e) Power Plant (Chamber radius = 3 meters) . . Cost reductions from about $2500 to about $0. 25 per target are needed for economical electricity production
Preliminary target designs have been identified Some Expected Direct Drive Specifications Capsule Diameter 4 mm Shell Wall Thickness 200 m Foam shell density 20 -120 mg/cc Out of Round <1% of radius Non-Concentricity <1% of wall thickness Shell Surface Finish 500 Angstroms RMS Ice Surface Finish The heavy-ion driven target has a number of different regions <2 m RMS NRL Radiation Preheat Target Other Potential Direct Drive Target Concepts Empty Outer Foam 0. 25 g/cc foam Nuclear Fusion 39(11) D. A. Callahan-Miller and M. Tabak Regions of lowdensity foams and unique materials Seal, DT Thick Outer Capsule Dense ablator Seal, DT LLNL Close-Coupled HI Target
Cost reductions of four orders of magnitude are challenging - but feasible Current cost ~$2500/target ~3500 µm ~1000 µm GDP Gas cooled reactor fuel particle with 4 PAMS coating layers Fuel particle scaleup experience is encouraging for IFE Inertial fusion energy target . . GA has previously used fluidized bed technology to reduce costs of 11 coated nuclear fuel particles and produced over 10 particles!
Technological improvements lead to dramatic changes in products (i. e. Moore's Law) Technology Review, C. Mann, May/June 2000 . . The number of transistors on a chip increased 4 orders of magnitude from 1971 to 1999
Moore's law analogies can be applied directly to cost reductions Main memory cost per byte (pence) Ref: http: //www. cse. dmu. ac. uk/~cfi/Networks/Work. Stations /Workstations 5. htm Year The cost of computer memory decreased by 106 between 1970 and 1990. This was achieved through reductions in process costs and improvements in manufacturing technology.
One can estimate IFE target production costs beginning with current experimental-target costs · One can find the approximate cost per current-day target by Total Project Cost/ Number of Delivered Targets = ~$2500 (capsule only) · However, there are tremendous differences in the program requirements - and in the consequent approaches to manufacture Item Experimental Program IFE Program Production Rate Relatively Small (~2500 targets per year by GA) 500, 000 per day FOAK Costs Characterization Product Yield Batch sizes Very high - targets always vary Extensive - individual details needed Low - product varies, small amounts needed Small - small amounts needed (<100) Essentially none Statistical sampling High Large … IFE target cost reductions will be achieved by Eliminating FOAK Costs Increasing Batch Sizes Reducing Characterization Increasing Yield
Costs will be dramatically lower when targets are identical - eliminates First of a Kind (FOAK) costs Today, few targets are made more than once! M=Metal M-GDP Wall thickness, µm · Currently delivered targets are nearly always unique- with most of the labor going to development and trial runs · We estimate the average FOAK labor now as hundreds of hours · These costs will be minimal for IFE production 20 18 16 14 12 10 8 6 4 2 0 X=Halogen GDP M-X-GDP 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 target batch # Example - Dopants and wall thicknesses vary on each batch ordered for experiments. . . For IFE, a single type of target is repeatedly produced, and FOAK development costs are essentially eliminated
Large savings can be achieved in characterization and QC Currently, shot-quality targets are highly characterized before delivery “pedigree” with detailed data on individual targets. Current manual characterization - ~8 hours per shell For the IFE Target Fabrication Facility, the cost of QC is reduced by: - reduced precision in IFE target designs - statistical sampling for process control - only periodic in-depth checks - automated characterization equipment. . Major characterization cost reductions can be achieved Future automated system for dimensional inspection of IFE target foam shells
Process development focusing on routine production will result in high product yields First-of-a-kind thin walled capsules have low yield (imploded during solvent extraction) FOAK batches: low yields (15%) Target Fabrication Process Development Programs After R&D and applying the science to process conditions, implosions are almost eliminated. High Yields (like chemical industry processes) of >95% but same operations cost
IFE target development programs must provide the technology basis for batch size increases and high yields Scaleable Processes Microencapsulation (shells) Fluidized bed coatings (shells) Interfacial polycondensation (seal coats) Sputter-coating (high-Z coatings) Casting (foams, hohlraum cases) Assembly (hohlraums, cryogenic, remote) Microencapsulation is inherently a high-volume production process Coating Example bounce-pan holds 4 -100 shells for coating Bounce Pan Example - two 9" diameter fluidized bed coaters can produce 500, 000 particles per day 9" ID nuclear fuel coater
Target filling and layering methods must be scaled to Fluidized Bed high throughputs The first full target supply system is at OMEGA 4 filled/layered targets/day Concept for Capsule Layering Tube Layering Concept for Hohlraums INJECT IR FLUIDIZED BED WITH GOLD PLATED (IR REFLECTING) INNER WALL ASSEMBLED HOHLRAUMS ARE STAGED IN VERTICAL TUBES WITH PRECISE TEMPERATURE CONTROL Pressure cell with trays 36 " I. D. X 40 " Tall, 8 trays, 290, 000 targets COLD HELIUM . . Basic premise: develop processes so small crews can operate
Anticipated target injection and tracking costs are low Target injection critical issues 1) Withstand acceleration during injection 2) Survive thermal environment 3) Accuracy and repeatability, tracking Must supply about 500, 000 targets per day for a 1000 MW(e) power plant 1) Injection placement accuracy to ± 5 mm 2) Indirect/direct drive tracking and beam steering to less than ± 200/20 m HYLIFE-II power plant concept showing basic injector components Direct drive target sabot . . Additional work will be needed to define injection costs
Major steps to reduce IFE target manufacturing costs Cost Item Current Cost Per Shell ($) Production Cost ($) Comment Total Cost ~$2752 $0. 083 Per "shot-quality target " Eliminate FOAK (R&D) $1200 ~0 Produce a fixed target design Reduce Characterization - Support R&D - Pedigree 225 1200 ~0 <$0. 05 No R&D support Process control Manufacturing Cost -Labor (yield, batch size) -Materials Cost $0. 013 125 2 $0. 02 The vast majority of the cost reductions come from eliminating R&D and the QC “pedigree” for each target. . . Additional work will be needed to define filling, layering, and injection costs
Summary and conclusions Current experimental-target fabrication costs need to be reduced about four orders of magnitude for economical IFE power production 4 Cost reductions of 10 or more from early fabrication to mass-production are common in high-tech industries Reductions from the current cost will be achieved by: - eliminating first-of-a-kind and development efforts inherent in today's experimental-targets - reducing the cost of QC by implementing statistical process control and automating inspection processes - developing equipment and processes for large batch sizes and/or continuous production - conducting the development programs necessary to achieve high product yields . . A significant development program is needed to provide low-cost mass-production of IFE targets