ARIES Documents -- Meetings ArchiveARIES Project Meeting, 22-23 April 2002
University of Wisconsin, Madison
|(DOE)||Ku, Opdenaker, Sauter|
|(LBNL)||Debonnel, Logan, Yu|
|(UCSD)||Miller, Najmabadi, Raffray|
|(UW)||Anderson, El-Guebaly, Haynes, Henderson, Kulcinski, Mogahed, Peterson, Santarius, Sawan, Sviatoslavsky, Varuttamaseni, Wilson|
Agenda - Les Waganer explained the agenda noting a few last minute modifications.
Welcome - Jerry Kulcinski summarized the excellent fusion program at the University of Wisconsin, Madison. The funded research is around $11M/y with the current experiments of RFP, HSX, Pegasus, and IEC. Jerry noted a tour is planned at the conclusion of the first day of the meeting to see the Shock Tube Facility and the operation of the IEC facility.
Budget and DOE Direction - Al Opdenaker, our new DOE Program Manager, emphasized the importance of fusion by noting the Secretary of Energy strongly endorsed exploration and development of new energy sources, such as fusion. As evidence of the perceived importance of the ARIES Systems Studies to Anne Davies, the Associate Director of Fusion Energy Sciences, she has placed ARIES at a high level in her organization under Al Opdenaker. Al referenced a slight increase in the FY03 fusion budget; however, he noted that the ARIES budget is reduced from last year and efforts are underway to increase our budget over the FY02 value to enable important MFE and IFE studies.
Al discussed the three new FESAC charges: recommend a burning plasma strategy (Snowmass), recommend a roadmap for a joint OFES and OASCR (Advanced Scientific Computing Research) Initiative, and recommend if the DOE program scope should be broadened to include non-electric fusion applications.
Program Status - Farrokh Najmabadi emphasized that this meeting will conclude the reporting on analyses of the thin liquid protected wall chambers. The ARIES team should begin to document the thin wall option results. This meeting will also initiate the thick wall liquid protected chamber evaluation, starting with the Wayne Meier summary of current design approaches and technology requirements.
Regarding the planned effort for FY03, the IFE effort would be concentrated on heavy ion driver related activities. We will perform detailed and in-depth study of selected high-leverage technical areas that have been identified by ARIES-IFE study. Specific topics would include:
In MFE, the ARIES team would review the assessments of a compact stellarator as a power plant, in particular, the advanced physics and technology requirements for the compact stellarators and proof of principle experiments. This effort would involve:
Next Meeting/Conference Call – The next ARIES meeting is scheduled to be held at GA, San Diego on Monday, July 1 and the morning of Tuesday, July 2. The next conference call was scheduled to be on Tuesday, 14 May 2002, but due to a conflict, it has been moved to 21 May at the usual time. Les Waganer will provide a call-in number.
ARIES-IFE Study Presentations
Systems Analysis and Integration
ARIES-IFE Systems Issues - Ron Miller discussed the tools for assessing uncertainties in the conceptual IFE power plant. He summarized the power balance cases he has considered. He mentioned the areas of uncertainties, noting areas that strongly affected the results (gain curves, driver efficiencies, recirculating power fractions, thermal efficiencies, and target costs. As an example, the recirculating power fraction for the thick wall option strongly affects net power output and the COE.
Dan Goodin offered that the DD targets are predicted to cost $0.154 each in quantities sufficient for a commercial IFE power plant. It was suggested Dan assess the cost for a distributed radiator target for comparison and inclusion in the systems code.
Heavy Ion Fusion Modeling Update - Spot Size Model Changes - Wayne Meier provided an update on his spot size modeling of the HI driver code that included emittance growth in the accelerator due to voltage errors in the gaps and field errors. He now plans to extend the modeling to include beam transport into the chamber. The main contributing factors to spot size are emittance growth in the chamber and photo-ionization of the prepulse and the main pulse of the beams. He is planning on completing the code modifications for spot size scaling, run a few cases to verify spot size requirements, determine target scaling, and prepare a new reference point.
Thin Liquid Chamber Design Approaches and Analyses
Description of thin Liquid Wall Protection with Porous Walls - Based on Laila El-Guebaly’s need to define the primary coolant inventory and the project need to have a strawman design approach, Les Waganer described two approaches for the thin liquid wall protection using porous walls. One was a stand-alone liquid cooled porous first wall and supply channel paired with a solid breeder cooled with helium (like the Prometheus blanket). The second approach was an integrated liquid cooled first wall, supply channel, and liquid breeding zone (more like the ARIES-AT blanket design). A common geometry of a right circular cylinder with hemispherical ends was shown. A generic liquid cooled chamber heat transport system was presented as a baseline approach. Both geometries were shown and compared for thermal hydraulic parameters and inventories. It was suggested a common energy source term should be used for comparative purposes.
Thin Liquid Wall Protection Concepts for IFE Reactors - Elsayed Mogahed described a series of thin walled protection concepts dating back to a LASL (now LANL) concept in 1972. The range of design approaches used liquid metal or metal salts and a porous, woven, or perforated first wall to protect the physical wall. The issues that still remain to be addressed are:
Numerical and experimental studies of the liquid wall protection concepts - Said Abdel-Khalik presented the results of three studies dealing with thin liquid film protection schemes: (1) Numerical simulation of downward facing porous wetted wall, (2) Experimental study of porous wetted walls, and (3) Experimental study of forced thin liquid flow on downward facing surfaces.
In the first study, the transient conservation equations were solved to follow the evolution of the liquid film surface with time for different values of the initial film thickness, liquid injection velocity through the porous wall, surface disturbance amplitude and configuration, surface disturbance mode, and surface inclination angle. To date, Said has only analyzed inclination angles of zero, 5 and 10 degrees. Results showing the drop detachment time as a function of these variables were presented for lead at 700 K. Generalized non-dimensional charts were also presented to allow other coolants at different temperatures to be assessed. The results imply that liquid film stability may impose a limit on the minimum repetition rate to avoid liquid "dripping" into the cavity between shots.
Said explained the experimental setups for both porous walls and tangential injection use water as the fluid. A few results were shown on the porous media experiment. These experiments are aimed at validating the results of the numerical simulations. Experiments have also been conducted with high velocity jets (with large aspect ratio flows with a rectangular cross-section) impinging tangentially onto flat plates to simulate forced flow thin film protection schemes. The intent is to see when the liquid film separates from the flat plate for different values of the surface inclination angle, liquid film velocity, and initial thickness. Photographs documented the results to date on both experimental studies.
Parametric Study Results of Thin Liquid Wall Responses Don Haynes presented BUCKY simulation results for the response of a thin liquid wall chamber using the threat spectrum from the close-coupled HIB target. The surface material was either Pb or FLiBe on a 4.5 m sphere. Other assumed parameters were the vapor composition (Xe or vapor from the wall liquid) and the vapor pressure. After the analysis, all combinations of the input parameters were found to be acceptable in terms of ion deposition, vaporization thickness, and condensation rates. FLiBe is more neutron transparent, thus Flibe has more volumetric heating and less shielding than Pb. Current results indicated recondensation should be adequate to achieve a 5-Hz repetition rate. More vapor material does not provide additional protection due to soft re-radiation with zero time of flight.
Concluding Sacrificial Liquid Film Activities Rene Raffray informed the group that the condensation flux and characteristic time to clear the chamber is a function of the Pb vapor and film conditions. Condensation of the Pb vapor is also affected by the presence of a non-condensable gas and its vapor pressure. To illustrate this effect, René showed a graph of Pb diffusion flux over a range of Xe pressures for selected Pb vapor pressure values. Depending on the Xe and Pb pressures, the characteristic condensation time is diffusion-controlled at higher pressures and ultimately condensation-controlled by the Pb partial pressure.
The processes leading to aerosol formation from the protecting surface were discussed. The equation controlling vaporization for the free surface and the heterogeneous nuclei were shown. René explained the phase explosions at and near the surface that occur with high heating rates.
Rene proposed an outline for the thin liquid film paper along with author assignments and page allocations. Comments are welcome.
Thick Liquid Chamber Design Approaches and Analyses
Introduction to Thick Liquid Wall Chambers - Wayne Meier outlined the key features and issues associated with the thick liquid wall chambers, such as HYLIFE-II. The key features are:
Approximately 58 cm of FLiBe is required to protect the structural wall and coolant plumbing against neutron damage and ensure Class C requirements at decommissioning. The thick central region jets are now considered to be porous liquid structures of a set of close-packed jets to better absorb the target shock. A set of crossed, fixed jets protects the end regions while allowing transit of HI beams and targets. The interior of the beam lines is protected by liquid metal vortices that also may enhance vacuum pumping of the chamber gases and debris. The currently predicted magnet lifetime is in excess of 30 years although that is being reviewed.
Issue Identification of Thick Liquid Walls René said that most issues are design dependent. Thus the team should adopt an existing design approach, such as HYLIFE-II. René thought the major issues for this design were:
Chamber/Beam Physics and Chamber Clearing
Final Focus Beamline Design - Simon Yu said to assure a high beam quality throughout the plant lifetime, the beamlines must:
FLiNaBe (fluorine/lithium/sodium/beryllium) is a good choice for the liquid metal on the internal surface of the beamline magnets as it is a good vacuum gettering material.
Strategies to Control the HIB Line Gas Density and Pressure in the HYLIFE Thick Liquid Chamber Christophe Debonnel from LBNL discussed methods to control debris contamination on the 162 heavy ion beamlines (two 9 x 9 arrays of beams). The TSUNAMI code provided gas dynamics behavior during the venting process of the target explosion within the liquid flow regions. The HYLIFE liquid flow is designed to be closed at the top and bottom of the liquid chamber, but there are some planned openings in the liquid streams to allow the target gas to vent. Both pressures and gas density plots were shown for small time steps. This allowed calculation of the impulse load (~3 x 103 Pa•S) on the liquid structure and the mass and energy flux to the wall and ports. These data also provided the starting point for the design of the beamline elements to reduce the debris contamination and pressure down the beamlines.
Diversion of Plasma in Beam Port with a Vertical Magnetic Field Dale Welch summarized the expected plasma conditions at the beam port (10 14 cm -3 density, 10-100 eV, and a velocity of 3-9 cm/µs (30-90 km/s)). He used the Lsp code to simulate a no-field case and another case with a magnetic field to divert the plasma to the beamline wall. With the 1 kG By field, the plasma stagnates around 18 cm and drifts to the wall within 250 ns. Thus a 1 kG field applied over a few centimeters is adequate to divert the plasma in the beam tubes.
Effect of Pre-Neutralization and Photo-Ionization on Transport Dave Rose showed his results on the effect of pre-ionization and photo-ionization on the beam transport. He is progressing on an integrated calculation for both the foot and the main pulses. He presented the nominal beam and chamber parameters. He said the plasma electrons are effectively neutralized even with ion stripping. Plasma pre-neutralization significantly improves beam transport and also works at higher perveance for both the foot and the main pulses. With this technique, Dave was estimating a focal spot size of 2.5 mm. Photo-ionization reduces the focal spot size by 15%. Inclusion of multiple gas ionization should improve the spot size further.
HEIGHTS Integrated Models for Liquid Walls in IFE Ahmed Hassanein discussed his hydrodynamic modeling of the phenomena of liquid walls in IFE chambers. He explained he is using a time variation-diminishing scheme to obtain stable numerical solutions with no generation of non-physical oscillations along shock waves. He presented results from both rarefied (0.21 torr) and dense (23 torr) Xe chamber atmospheres for chamber radii of 6.5 m and 3.0 m, respectively.
Ahmed explained the modeling of the atomic physics of the chamber. The HEIGHTS code calculated the time-dependent re-emitted radiation flux, which in turn evaluated the surface temperature for chamber gas pressures from 0 torr to 0.5 torr Xe. He showed the results of calculating the radiation of lead using the different models. He compared the surface temperature of lead and lithium assuming a high-yield DD target with 401 MJ of yield as well as high and low yield targets for a DD target.
He also modeled the behavior (pressure waves and velocities) of thick liquids with intense neutron energy. He described the modeling of the aerosol produced on the surface of a liquid from intense heating of a local spot. Macroscopic droplets are formed given these conditions.
Dynamics of Liquid Wall Chambers Bob Peterson explained that the more complex liquid wall concepts are superior to the simple dry wall concepts for certain laser or beam transport mechanisms, supports higher target yields at smaller chamber radii, has a renewable first wall surface, and potentially offers a lower pressure and less-contaminated chamber atmosphere. However, the liquid wall concept is dominated by liquid phenomenology - chamber vapor pressures, liquid response to x-rays, wall vaporization, vapor recondensation, and aerosol production/dynamics.
Bob showed how a Z-pinch power plant might be designed with liquid wall protection. He provided BUCKY results for the ZP3 design, both with and without liquid wall protection. Both cases had Ľ atmosphere of helium chamber gas. The case with liquid wall assumed the lithium at a spherical radius of 0.5 m. The second case had a steel spherical shell at 2.5 m. The BUCKY code simulated an X-1 target yield at 400 MJ and scaled it to 4 GJ, both with and without hohlraum cases. At a target yield of 400 MJ, the explosion evaporated 3.5 mm of lithium or 5.6 kg (at 0.5 m radius), whereas the same explosion on the second case evaporated 0.02 mm of steel or 10.6 kg (at 2.5 m radius) per shot. A recoil pressure wave is propagated into the liquid or solid wall from the intense vaporization of the wall surfaces. This pressure is quickly dissipated within the wall.
Chamber and Liquid Coolant Nuclear Analysis
Radiological Issues for Thin Liquid Wall Options - Laila El-Guebaly outlined the injection approaches and the materials to be considered in her activation analysis. She considered radioactivity for the safety analysis, decay heat for the LOCA and LOFA analysis, and waste minimization and disposal. Two extreme cases were assumed; the worst case is no mixing of the first wall material with the breeder, and the best case is the mixing the two coolant streams. She noted data supplied by L. Waganer to help quantify the volumes and residence time both inside and outside the high radiation zone of the power core. She then illustrated the flow parameters for the different injection and processing options. For tangential injection with separate flows, the liquid wall will generate tons of high level waste unless the in-chamber residence time and/or exposure time is limited. Lead or any lead compound is the worst.
For porous wall injection with mixing of first wall coolant and breeder, the waste disposal rating is not sensitive to the in-chamber residence time of the liquid wall fluid, however LiPb is worse than FLiBe. In this case, small amounts (several cups) of transmutation products in LiPb would be filtered out and then the LiPb can be reused or disposed of as LLW. Without mixing of the flows, the results are slightly higher activation levels, but disposal can be accomplished in the same manner.
Feasibility of Recycling Hohlraum Wall Materials Laila El-Guebaly addressed the recycling of candidate hohlraum target materials. For this assessment, she examined two extremes; one-shot use and disposal in repositories or recycle continuously without removal of transmutation products. Hohlraum materials represent a very small waste stream, less than 1% of the total IFE waste, meaning recycling is not a must requirement. Laila showed a schematic of the target materials flow stream in the power plant and target plant including holding tanks for a cooling period. Cooling periods could vary from a day to several years. Recycling introduces problems as it produces high-level waste, increases activity and decay heat, requires remote handling, adds radioactive storage facilities, and increases the cost and complexity to the plant. The one-shot approach solves the WDR problem, but some materials may have cost and resources problems. The solution may be to remove some activation or transmutation products on-line (complexity and cost), store materials for a cooling off period (cost impact), and/or limit exposure time by using new materials (cost). Laila offered a lot of suggestions to minimize the activation and/or cost of hohlraum materials, but without a final design, solid material costs, and firm regulation guidance, no significant conclusion can be made. Laila proposed a new strategy to avoid the deep geological burial of the fusion high-level waste. The main idea involves the separation of the high level waste followed by a transmutation process in a burning module in fusion devices
Target Fabrication, Injection, and Tracking
Aerosol Limits for Target Tracking - Ron Petzoldt noted that direct drive target fabrication specifications (surface finish and thickness uniformity of high-Z layer) affect acceptable aerosol size and density. The 50-nm RMS surface finish suggested that aerosols should not be larger than 50 nm. The requirement for thickness uniformity translates into a build-up on the surface of less than 3 nm while traversing the ~6-m chamber radius.
For indirect drive targets, the droplet density should not exceed 1 g/m3 to allow external target tracking. Ron calculated that if the droplet density and size are not excessive, the accumulation on the target should not cause an unacceptable energy loss (< 1%). Scattering of the beam by droplets in the chamber may cause more losses.
Ron calculated the absorption and scattering of light by Pb aerosols. He showed a table of aerosol particle size and density (mg/m3) for laser direct drive, ion beam indirect drive, and tracking system limit.
Aerosol Production in the Post-Shot IFE Chamber Environment Phil Sharpe discussed four models he is using to simulate aerosol production:
The chamber geometry he assumed was a 6.5m sphere with a thin Pb liquid wall. The target was the HIB ID target with a 458 MJ yield. As a conservative case, he assumed all the energy is deposited on and vaporizes the wall material. In this case, all nucleation ceases by 1 ms, though the aerosol population is convected farther into the chamber. The average particle size in a region up to 1 m from the wall is 2.0 µm, and the number density approaches 1e13 #/m3.
A second case with equivalent chamber geometry was examined using singular (2µs) chamber conditions predicted with the BUCKY code. The net vaporized mass at this time was 37 kg. Phil's code then simulates the expansion and cooling of the vapor, with subsequent aerosol formation. In this case, nucleation ceases after 5 ms, and after 100 ms the aerosol population is roughly uniform in the chamber with a density of ~ 1e10 #/m3 and an average size of 6 µm (corresponding to ~ 15 kg of aerosol mass in the chamber).
Future work may include aerosol model refinements (addition of ion-induced nucleation, and model of multi-component aerosols) and coupling the aerosol model to more suitable chamber radiative gas dynamics codes (e.g. BUCKY).
S&E Considerations for IFE Thick Liquid Wall Concepts - Susana Reyes summarized the HYLIFE-II safety and environmental attractive features. Simulated LOCA/LOFA with simultaneous loss of confinement situations have been analyzed and show the radioactive afterheat low enough to allow cooling of structures during the transient conditions. The steel walls remain below melting temperature. Tritium retained in the structures dominates the accident dose, which is about 5 rem at the site boundary.
Alternate waste management scenarios have been evaluated. All structures would qualify for shallow land burial (WDR<1). The total life-cycle waste volume is dominated by 5300 m3 of concrete from the building. Clearance indices have been calculated for all the HYLIFE structures. Clearance would be possible after a cooling period of 1 year, which is less than the decommissioning period.
Magnet System Design and Analysis
ARIES-IFE Final Focus Magnet Design and Analysis - Leslie Bromberg discussed the advantages of using the high temperature superconductor, magnesium diboride, given moderate field requirement. It would be a good replacement for the Nb3Sn low temperature in the field region. Leslie also talked about a different magnet geometry for the quadrupole fields that is simpler (Sin 2 Theta) with minimal field errors as opposed to the traditional (Cos 2 Theta) geometry. He also discussed some ongoing work on analyzing and minimizing end effects.
Aerosol Protection of Laser Optics by Electrostatic Fields Leslie Bromberg explained how the charged and moving (a few cm/s) aerosol could be cleared by an electrostatic field of 100 V/cm.
Plasma Window Performance Leslie Bromberg explained the operating principles of a plasma window to isolate regions with different vacuum conditions. The HIB driver requires high vacuum conditions (10-6 to 10-9 torr) as compared to the power chamber of 10-3 to 10 torr. Leslie described an experimental setup to evaluate the capability of a plasma window to provide the necessary vacuum isolation with a sufficiently large beam port area. He has demonstrated that the plasma window was able to hold a low-pressure region at 30 mtorr from an argon gas at atmospheric pressure. More experiments will continue this summer.
Action Items - René Raffray assembled a list of action items to help evolve a viable design window and characterization of the thick liquid wall design constraints, shown below.