Date: 12 December 1994
Subject: Starlite Meeting Minutes, 7-9 Dec 1994 @ UCSD
To: Starlite Team
CC: "Town Meeting" Guests (see below)
From: L. Waganer [Special thanks to many contributors to these minutes.]
Attendees: C. Bathke, M. Billone, L. Bromberg, R. Conn, S. Dean, B. Dove, D. Ehst, L. El- Guebaly, B. Ellis, T. Flynn, B.J. Lee, S. Herring, G. Hofer, S. Jardin, C. Kessel, T.K. Mau, R. Miller, F. Najmabadi, G. Sager, D. Steiner, D- K Sze, M. Tillack, L. Waganer, X. Wang, C. Wong
Visitors: R. Bastasz (SNL), R. Cerbone (TSI), E. Cheng (TSI), M. Day (UCSD), S. Guangzhao (SWIP, China @GA), Y. Hirooka (UCLA), E. Lazarus (ORNL), E. Muraviev (ITER, SD JWS), R. Prater (GA), E. Strait (GA)
In the tradition of interacting with the fusion technical community in the vicinity of the project meeting location, a "Town Meeting" with GA and UCSD personnel was held the morning of 7 December at UCSD. The topics reviewed were advanced physics regimes and operational modes as applied to tokamaks and understanding plasma edge conditions and materials interaction. The Starlite project meeting continued through noon on Friday, 9 December 1994. The highlights of the presentations, technical discussions, and actions are noted below. I have most of the handouts of the presentations - if you wish a copy of a talk you did not receive, let me know.
7 December 1994 - AM
F. Najmabadi welcomed the visitors and the team to the new UCSD facilities in the Engineering Building II. Thanks to the staff of UCSD for their efforts in hosting the meeting! F. Najmabadi also reviewed the purpose of the Starlite Demo project, the progress to date, and the specific purpose of the town meeting; namely, interaction of the Starlite team with the local fusion community.
As the first Town Meeting speaker, S. Jardin explained how the Starlite project is investigating advanced physics options for application on the US Demo. We are in a pre-conceptual project phase, exploring physics and engineering options for minimum and nominal embodiments for Demo. These may be enhancements of prior concepts or adaptations of evolving options.
Ed Lazarus (GA) gave a presentation on "Axisymmetric Equilibrium and Stability Issues for Advanced Tokamak Power Plants." He discussed the advantages of using coils located on the small major radius side of the tokamak for controlling the vertical instability in tokamaks. The goal of an Advanced Tokamak is to reduce the plasma current while maintaining sufficient fusion reactivity and energy confinement for ignition to occur. Configurations for this include (1) the Reversed Shear Configuration, that needs a conducting wall to stabilize the kink modes, (2) the High-li (plasma internal inductance) Configuration, which requires some omega-star stabilization of the internal kink mode to get q(0) < 1, and (3) the Hot Ion VH mode.
E. Strait (GA) discussed "Stability at High Beta and Power Plant Issues." To maximize the ratio of the fusion power to the power lost, you have to maximize H/q, where H is the confinement multiplier over empirical L-mode scalings. To increase beta, you can raise I/aB, increase beta-N, or operate in second stability. DIII finds that the maximum stable beta-N is equal to 4*li. However, efforts to increase beta-N by increasing li are stymied since a bootstrap current profile develops that is incompatible with the equilibrium current, thus lowering li. In DIII, the highest beta discharges have been stabilized by wall stabilization and by toroidal plasma rotation with a frequency in the range 1-2 KHz.
R. Prater (GA) gave a talk on "Heating and Current Drive Issues for Advanced Fusion Power Plants." He discussed scalings for the H-mode threshold and also for current drive and bootstrap fraction. He also gave a list of engineering issues for heating and current drive systems. He estimated the RF system level cost would be in the range of $4/W for first of a kind systems.
M. Schaffer (GA) made a presentation on "DIII-D Divertor Program". He showed results on how the DIII pumped divertor can actually work to de-gas the (carbon) walls. He also showed results illustrating naturally-occurring radiation coming from the scrapeoff plasma between the X-point and the walls. A combined approach of neon injection to radiate from the edge of the main plasma and divertor puffing may offer advantages.
Y. Hirooka (UCLA) explained the importance of plasma materials interaction and presented some experimental results from his PISCES experimental program. He is concentrating on experiments using Be, C, and W as the most likely plasma facing materials given the divertor design constraints. All of these have some problems: beryllium is toxic, has a low melting point, and is difficult to attach to substrates; carbon stores tritium, erodes, and its thermal conductivity is sensitive to radiation; and tungsten needs to be isotopically tailored and the plasma is sensitive to trace amounts of tungsten. He suggested it may be possible to use a lithium coating for startup.
M. Day (UCSD) explained numerical modeling capabilities for studying plasma edge physics (presented later). This modeling is being developed to help guide divertor design approaches. He is using fluid modeling techniques to identify the most promising divertor approaches. His assessment is that the plasma/neutral/impurity interaction is inherently too complex for an analytical solution. The present codes do not converge on a solution at or near the plasma/material interface. (Convergence is defined to be the same solution with a finer grid applied.)
7 December 1994 - PM
R. Conn reviewed the impressions from last week's FEAC meeting in Washington, D.C. FEAC has just been reformed with an emphasis of membership outside the fusion community. The new FEAC panel was briefed by fusion leaders on recent progress and future plans. They appreciated fusion's progress but were concerned about possible future budget reductions. We must effectively communicate why fusion is important to the nation and how our programs are worthwhile in achieving our goal. We must bring our message to our elected leaders so the fusion program has the best possible chance in the budget deliberations.
W. Dove also emphasized DOE's concern that near-term budgets may be reduced unless the representatives and the public clearly understand the importance of developing fusion energy. Bill underscored our project goal to develop a Demo that minimizes risk to proceed to a commercial fusion power plant. Our study will help define national R&D programs to prepare and qualify materials and systems for the Demo. Demo is not really so far away - we must start our planning now to be ready. Starlite is instrumental in helping DOE and other institutions plan for Demo.
F. Najmabadi distributed his latest revision of the Demo Mission and Goals to the team for final review. He will forward a copy to the Starlite Utility Advisory Committee for review. L. Waganer will also solicit the FICUS (Fusion Industry Council U.S.) to help critique the mission statement from the industry standpoint. The next project step is to translate the Demo mission statement and goals down to more detailed requirements to be achieved by the plant as a whole and/or by individual plant systems. [Later, action items were assigned to accomplish this.] F. Najmabadi stated this overall philosophy: The Starlite Project should develop a reference design and provide a solid foundation for planning and design definition for Demo.
R. Miller reported his work in reviewing costing and modeling methodology, data base collection, and collaboration with other future energy technologies. He has been seeking information on the new solar power demo plant but specific cost methods and data are deemed proprietary. B. Dove will check if he can obtain any useful information through DOE channels. In reporting Demo COE values, Starlite should report both the cost for the Demo plant (1st of a kind) as well as a projection to a commercial plant (nth of a kind). R&D costs specific to Demo should also be included in the Demo cost.
T. Flynn presented some North American Electrical Reliability Council (NERC) reliability and availability data for nuclear and fossil power plants. Supercritical steam systems used to be poorer availability performers than standard steam systems, but in recent years they have consistently had the higher availability performance. But he cautioned against having plants which require two developmental systems. Historically, a plant is more likely to have a failure immediately after a startup - thus keep the system running or hot to have high availability.
D. Steiner reviewed our project's licensing and safety tasks and the approaches being employed, including interfacing with DOE, ITER, TPX projects, and various laboratories. He reviewed our perception of how the licensing process may transition from DOE to NRC and the steps that may be necessary to guide that transition. He also highlighted some of the safety issues that need to be addressed. He emphasized that we must begin to start the process now to have processes in place for Demo.
G. Hofer presented the NRC organization chart to identify the people responsible for various decisions and possible points of contact. He reviewed the basic law that establishes the regulation of fusion plants. He highlighted the steps in the issuing of a construction permit. S. Dean asked of our process to evaluate the viability of austenitic and martensitic steels in terms of increased safety and reduced radioactive waste.
S. Herring discussed the development of a new fusion safety standard that will apply to both existing and new fusion facilities. The standard invokes only requirements that currently exist in federal law. The standard is not intended to be prescriptive, rather would allow some flexibility in solutions. The general guideline is to keep the radiation hazard less than one percent additional risk to an individual at the fenceline. Elimination of the requirement for an Evacuation Plan is seen to be a distinct advantage for fusion. The team recommended a change to a Safety Functions chart that should include "Off-Normal Events" in addition to "Normal Operation" and "Accident Conditions."
8 December 1994 - AM
S. Jardin started the morning talks with a discussion of proposed physics risk and economics analysis of Demo concepts and the assessment of physics innovation options. A method was presented for quantifying the technical risk associated for each demo concept. Everyone agreed with the need for the analysis and his general technique. Most of the ensuing discussion involved formulating the risk analysis technique to differentiate non-acceptable options from less attractive options. It was suggested that instead of Risk, we use "Confidence Factor" = 1/Risk. S. Jardin accepted an action to rework the methodology to address the project's concern. R. Conn suggested Steve review the techniques employed by the fission industry in risk analyses and perhaps adopt some of the formalism.
The use of the available "Low Aspect Ratio" concept data for illustrative purposes was a concern. A modified low aspect ratio design parameter set will be developed by the Starlite Physics Group and used in the comparative analysis.
S. Jardin presented the Physics Group's proposed five physics' figures of merit. The figure of merit methodology has met with acceptance by the fusion physics community. These figures of merit will be used in conjunction with the Physics Risk (Confidence?) technique to rank the candidate reactor concepts. The following physics figures of merit were proposed:
* MHD: 2D Graph of beta/(S*epsilon) vs epsilon * betap [where S = (1+kappa**2)/2 ]
* Current Drive: (Ip)(Ne)R / P
* Heat Exhaust: 2D graph of P/R vs fRAD
* Confinement: 2D graph of beta/epsilon vs H/q
* Ash Removal: Tau-p* / Tau-E
Dave Ehst reviewed the recent US/Japan workshop on bootstrap current. One of the concepts proposed by the Japanese was a large 24-hour burn, pulsed commercial reactor. There was a discussion of TFTR tearing modes with nonlinear bootstrap current generation which limits their deleterious effects on confinement.
D. Ehst also presented results, "Bootstrap Current in RS and Other AT Equilibria". The RIP code has been used to compare several different advanced tokamak equilibria. Improvements have been made in the RIP code in modeling the fast alpha pressure and in modeling density profiles with two parameters. The conclusions of the study were that the negative shear (NS) mode (also referred to as Reverse Shear) is significantly better than other advanced tokamak modes because of simultaneously achieving high beta, high normalized beta, and well-aligned bootstrap current. D. Ehst used a simplified costing algorithm to evaluate the relative performance of several options, but the team cautioned that the simplified algorithms may mask or not address other important factors.
C. Kessel examined the equilibrium results in the Reverse Shear Reactor Study. The reverse shear equilibrium from the RIP code used by Ehst is unstable because the minimum in q is not far enough from the axis. A more reactor relevant version of the reverse shear mode would not use LHCD, but would use bootstrap current to provide the correct safety factor profile and just some central current drive using ICRH. An analysis was made of the requirements for resistive wall stabilization of the external kink mode for the RS configuration. The plasma must rotate at 5% of the Alfven velocity, or at about 5 * 10**6 m/sec. The wall separation should be 30% of the minor radius. The wall resistivity should be such that the wall time constant is longer than 2 ms.
S. Jardin spoke about the stability of high bootstrap fraction, high beta, low aspect ratio tokamaks A systematic study is underway of both the stability and bootstrap fraction properties of low-A tokamaks. Cases have been found with beta-N up to 5.7 which are stable with no wall. With a wall, cases with beta-N up to 7.5 have been found to be stable. There is a technical difficulty in resolving the kink mode stability of configurations with edge-q above about 15. High bootstrap fraction (fBS) equilibrium have been found. There is a case with fBS=86%, beta-N=5.8, and beta=12% stable to ballooning and without any edge current drive. There is also a case with fBS=84%, beta-N=7.8, and beta=28% which is stable to ballooning, but which requires edge current drive. The kink stability of these two cases are unknown.
T.K. Mau looked at current drive on low-aspect-ratio tokamaks. A preliminary investigation of the fast wave current drive on low aspect ratio tokamaks was conducted. An A=1.45 tokamak plasma with fBS = 0.30 was used for the preliminary study, which provided insight into FWCD in an A=1.2-1.3, high fBS reactor relevant configuration. It was found that fast wave propagation and damping properties are more sensitive to launch location and spectrum than in a conventional tokamak. A reactor relevant equilibrium, in EQDSK format, is required to carry out a more definitive CD assessment using CURRAY. Possibly, both fast wave and lower hybrid wave power will be needed. An approach being looked at is to write a routine that puts a JSOLVER equilibrium into an EQDSK file.
C.G. Bathke presented some results from an initial systems level comparison between the five Demo candidates. The Reversed Shear (RS) option has been added to the ASC. (Note: The title of the systems code has been changed to DSC for Demo Systems Code). For the Low-A configuration, results were taken from a 1986 Miller, et al. report. The preliminary assessment is that the RS design will have a very attractive COE because of its high Mass Power Density, high Q , high beta, and high gamma-*. The RS blankets must be re-examined for the effect of the higher neutron wall loading. The cost of electricity must be reported for both the first of a kind Demo plant and for the nth of a kind commercial plant using this technology.
B.J. Lee discussed "Divertor Modeling for Starlite." A scoping study was done for the divertor needs for the RS configuration. It was concluded that the radiating mantle concept may work, but there is concern about the harmful effects on the core plasma due to fuel dilution. The high pressure gas target divertor may work with bulky divertor legs, but there is a need to watch the distance between the target and the ionization front.
A fluid equation for the neutrals, assuming that charge exchange is dominant over ionization and ignoring neutral-neutral elastic collisions, has been developed by Catto, Helnader, and Krasheninnikov. Comparisons were presented of the 1D plasma-neutrals simulation and PISCES data. Work is in progress to apply the new equation set to reactor grade tokamaks for the high pressure deuterium gas target divertor. Impurity radiation physics is being added, and a 2D code is being planned.
8 December 1994 - PM
Mark Tillack discussed preliminary Engineering Group assignments to various systems and tasks. Further definition is needed on Configuration and Maintenance, Safety and Licensing, and RAMI. Interfaces need to be considered to assure system designs are compatible and meet the overall plant requirements. Design alternates will be carried until April 1995 and then a preferred option will be selected. An assessment process was described. We will not prejudge any option at this point, prior to determination of the lower tier requirements.
C. Wong described the process of identifying critical issues to help identify and quantify the necessary tasks to prepare for and design a Demo plant. Clement listed the goals of the Critical Issues task which included collection of issues, identification of capability to achieve the goals, and proposal of solutions. He is presently soliciting and collecting critical issues.
Glenn Sager outlined the general functions of the PFC systems. He will accomplish a rolldown of the upper-tier mission requirements to the PFC-system level requirements.
D-K Sze discussed the ability of the candidate materials to handle the anticipated heat load. He is doubtful if the proposed materials (V or SiC) can handle greater than approximately 3 MW/m2 heat flux with reasonable wall thicknesses (2 mm). Waganer volunteered to have a stress analysis conducted on tubes with a range of heat fluxes and thicknesses.
D-K Sze also presented the blanket issue comparisons of the possible structural materials. Vanadium and SiC were discussed in detail with lesser detail on ferritic steels and TZM. Again, the options will be carried until specific requirements can be adopted and used as a selection criteria.
L. El-Guebaly presented the list of breeders available for Demo and discussed their relative merits. Both solid and liquid breeders were described. The use of a beryllium multiplier was discussed, along with various structural materials at varying structural fractions. Laila displayed the options relative to TBR and energy multiplication. She also presented an improved shield design which provided better shielding at a lower cost by using borated stainless steel and WC bulk shielding. The cost reduction was on the order of a 10% lower COE. Laila also discussed the impact of radiation out the back of the shield on remote maintenance equipment. It turns out that the baseline she has been using did not meet hands-on maintenance within 24 hours after shutdown. The project recommended that additional shielding be added to meet the hands-on maintenance requirement.
Mike Billone discussed the interface between materials knowledge and the evolution of a design. Many new materials are being considered for the next generation experimental machines and then for Demo. We need to understand the needed requirements so we can properly plan for development of the material databases. He gave examples of the scatter of measured data and areas where data was lacking. He handed out a listing of ongoing materials programs and points of contact for information. He also presented some data on the assessment of ferritic steels.
C. Wong described the thermal conductivity issues associated with irradiated SiC and the related design implications. They are investigating the thermal conductivity of doped SiC. The data after irradiation indicates much lower thermal conductivity values. Pure SiC may be a limiting case for the thermal conductivity data.
T.K. Mau discussed the role of heating and current drive in Demo. Key design issues are coupling efficiency to plasma, system electrical efficiency, unit power cost, and several other operational and maintenance requirements. He proposed extrapolating from presently envisioned systems for ITER and TPX. The group suggested he broaden his choices to more closely align with the Demo needs and then see how the developmental program needs to be structured to meet those needs. He projects the system cost to be in the range of $1-2/W based upon an ITER cost estimate. Antennas were described, but it is unlikely they would be used in a commercial device due to wall coverage. The stated reliability of the RF amplifier tetrode was quoted as > 8000 hr which was deemed to be too short. T.K. was requested to determine the cost to refurbish the tetrode.
L. Bromberg described the critical issues related to the magnets. The conductor in a cryogenic case is not yet a well understood technology. Currents in S/C strands are non-uniform and are time varying. No common magnet design codes have been adopted; rather, each project defines their own. No acceptable, timely maintenance approaches have been proposed. Magnet out-of-plane loads dominate design approach - ITER is in the process of re-evaluating their design approach for handling out-of-plane loads.
R. Schleicher revisited the thermal conversion options available for a commercial fusion plant with the goal of selecting the most attractive option(s). Possible power cycles included Rankine, Brayton (gas turbine), and dissociative. Blanket materials were ferritic steel, vanadium, and SiC composites and working fluids were water, helium, liquid metals, and organic and inorganic salts. Ferritic steel with lithium yielded an efficiency of 37%. The vanadium case with helium coolant in a Brayton cycle yielded 41% whereas lithium in an indirect Rankine yielded 45%. The SiC composite had 52% for the direct Brayton helium cycle and 46% in the direct H20 Rankine cycle. He thought the V/Li and the SiC/He were the better options.
9 December 1994 - AM
M. Tillack explained the purpose, content, and access procedure to the ARIES Archive that is now installed at UCSD. All meeting and conference call minutes, reports, files, documents, lists, and enabling software are contained within the archive. Access and document retrieval can be accomplished with a program such as Fetch. Mark has also set up a World-Wide Web Server home page for Starlite. It contains, in native language, most of the archive documents and graphical information. There will be both public and project-private access areas for viewing or retrieval of information.
L. Waganer reviewed the Action Item List for completed or delayed items. It was recommended L. Waganer and F. Najmabadi reformat the list with more immediate actions and eliminate the distant or continuing resolution items. The following list of next actions were compiled from summary presentations and the team discussion. Some may be incorporated into the Action Item List:
* (All) Demo I configuration will be defined June 95 - Demo II will be defined later
* (Ehst) Propose Current Drive Figure of Merit and perform database assessment
* (Najmabadi) Contact Sigmar about engaging divertor modeling community on Starlite
* (Mau) Contact Mandrekas about radiating mantle modeling for Starlite - propose experiments on TFTR
* (Ehst & Kessel) Finish Reverse Shear Studies by providing aspect ratio scan (A=3,4.5 + one more (TBD) and Beta & Icd
* (Kessel) Develop interface between JSolver and CURRAY for low-A configuration while keeping A=1.2, tradeoff fBS for Beta
* (Mau) Perform current drive assessment for low-A configurations
* (Jardin) Refine Risk/Confidence level assessment of demo candidates
* (Bathke) Develop Reverse Shear and Low-A strawmen
* (Waganer,Sze,Tillack) Quantify thermal stress on high heat flux first wall tubes
* (Engr Grp) Ferritic Steel Assessment - (Billone) Assess impact of DBTT ; (Sze) assess temperature and density effects[my notes don't make sense here]; (El-Guebaly) assess activation of proposed FS alloys
* (Engr Grp) Assess PFC material options for armor, substrate, wall, and structure [specific assignments?? Mark??]
* (El-Guebaly) Assess impact of adding shielding to enable hands-on maintenance 24 hr after shutdown
* (Hofer) Complete tritium handling document by end of Jan 95.
* (All) Review mission statement (comments due to Najmabadi Jan 4)
* (Najmabadi) Send Mission statement to UAC for comment
* (Waganer) Send mission statement to FICUS for comment
* (All Group Leaders) By next CC (Jan 4), formulate draft list of top level requirements
By following CC, draft of quantified requirements
By 3rd CC, finalize requirement values (both minimum and nominal #s)
* (All ) Write up quarterly progress, send to task leaders for summary. Leaders to submit electronic draft at next project meeting (3-5 pg/area).
Next Utility Advisory Committee Meeting is all day Feb 16 and the morning of 17th at UCSD. Therefore, the next Starlite project meeting will be held immediately prior to the UAC meeting. It was suggested that the Starlite meeting run from Monday @ 1:00 (2/13) through noon Wed (2/15). D. Steiner offered an alternative start date of 8:00 am on Tuesday. We can discuss meeting arrangements on the next CC Jan 4. (same phone number).