ARIES Program
Public Information Site

ARIES Documents -- Meetings Archive

ARIES Project Meeting, 9-10 August 1999

University of Califonia, San Diego
Documented by L. Waganer

Agenda and Presentations

Attendees: C. Baker, L. Bromberg, V. Chan, E. Cheng, L. El-Guebaly, S. Jardin, C. Kessel, T.K. Mau, Ron Miller, F. Najmabadi, D. Petti, R. Raffray, D. Steiner, I. Sviatoslavsky, D-K. Sze, M. Tillack, L. Waganer, X. Wang

Administrative and General Topics

Possible October dates for the next meeting were proposed, but several conflicts were noted with other meetings and symposia. Instead, the team opted for an e-meeting in September, thus postponing the site meeting until November. A tentative e-meeting date of 7 September was proposed. Les Waganer would try to obtain a conference number. [The conference phone number for 7 September, 9:00 to 12:00 PST is 206-655-0054 or 1-800-764-2618 and pass code 3422#.]

The ARIES-ST study results are to be published in a special issue of Fusion Engineering Design Journal. All chapters have been received and formatted for review except for Physics and Systems Studies. Final drafts of these two chapters are being prepared for submittal to F. Najmabadi, who will distribute a complete set for final review.

Farrokh Najmabadi reviewed the future plans for the ARIES program. The original plan included a second phase for the Non-Electric Applications (Neutron Source Study) and an extended conceptual study of the Advanced Tokamak Power Plant (ARIES-AT). The current thinking is that the ARIES group should shorten the ARIES-AT assessment, eliminate the second phase of the Neutron Source Study, and begin an independent assessment of a new IFE power plant. If any new MFE study commences, it would probably be for a proof of principal (PoP) experiment of an emerging concept.

All abstracts and papers for the upcoming conferences should be submitted to F. Najmabadi for review and incorporation in the ARIES Web archives.

Large Fusion Power Plant

Status of Large Power Plant Study - Les Waganer explained the purpose of the Large Fusion Power Plant study, the team conducting the study, and some of the early study results. Duke Engineering and TVA have been under contract for a few months. They have found that a large facility or independent power operator (IPO) should be able to integrate a 2 to 3 GW into their system. Larger IPOs might also be able to accommodate even larger facilities. Joan Ogden of Princeton University and Les Waganer will examine the facilities and economics to co-generate hydrogen and electricity. Les discussed why hydrogen generation and use is being considered. He also explained the techniques being considered to produce the hydrogen. The study will compare the costs to generate electricity, hydrogen, or co-generate electricity and hydrogen. Other possible options would include electricity demand load-following (with off-peak hydrogen production) and using hydrogen combustion turbines to provide extra peaking power.


  • Waganer - Check on energy required to compress and liquefy hydrogen.
  • Waganer - Check with Clement Wong on GA's concept to produce hydrogen with HTGR.
  • Sze - Verify the data Waganer presented on the amount of power needed to provide hydrogen to a vehicle fleet equal to 20% of those in LA basin.

Palladium Membrane Reactor for Hydrogen Production - D-K Sze presented a schematic and theory of operation of the palladium membrane reactor process to convert CO and H2O using a Pd catalyst into CO, CO2, and H2 at a process temperature of 450°C to 550°C. Leslie Bromberg thought this was essentially the water shift process, which could be done at much lower temperatures.


  • Sze - Determine the chemical reaction and energy required to convert CO2 back to CO, thus completing the chemical cycle.

Neutron Source Study

Overview of Neutron Source Study - Don Steiner reviewed the Snowmass results, which concentrated mainly on a range of fusion-fission applications. Proposed metrics include: cost of neutrons, total number of neutrons produced, capital cost, operating cost, cost of product, value of product, environmental/safety/health metric, and licensing metric. To help compare different approaches to transmute nuclear fuels or waste, it was suggested to use cost per transmutation rather than cost per neutron.

Summary of Neutron Source Products - Les Waganer reviewed the possible neutron applications in the general categories of transmutation, direct usage, and thermal conversion. The more likely applications were rank ordered based on a previous evaluation. He suggested focusing the ARIES Neutron Source study on applications using the mid-level fusion reactors (50 to 500 MW). Applications using larger plants have been or currently are being studied elsewhere. Applications involving smaller fusion reactors are best left to the concept advocates for evaluation. Hence, the applications are likely to be transmutation of nuclear waste, radioisotope production, activation analysis, and tritium production. Selection of the best one for further assessment will involve evaluation of the corresponding metrics.

Planning International Activity on Pu and Actinide Transmutation in Fusion Reactors - Ed Cheng noted that there are 760 tons of radioactive actinides from fission power plants in the US. For every three 1-GWe fission reactors, a fusion (or equivalent) plant will be needed to burn Pu and activated actinides (3 GWth ~ 1 MT/y of Pu or U235). Accelerators and fast fission reactors are transmutation competitors for fusion. Licensing and public perception are major hurtles for transmutation, regardless of the method or power source. Technically, fusion is seen to have advantages to the other two approaches, but is less well developed. International cooperation is being proposed to help promote fusion as a transmutation option.

Neutron Source Metrics and Systems Modeling - Ron Miller noted that Don Steiner had presented the list of metrics that reasonably characterized the neutron source products and applications. It was recommended that cost per transmutation be used in certain applications where more appropriate.

Regarding modeling of fusion power plants in general, Ron said that he had gotten the systems code to reproduce the 1996 ARIES-RS results with the new algorithms implemented. The cost basis is being updated to a 1998/1999 reference to facilitate economic comparisons. Physics inputs and modeling for the ARIES-AT are in progress to produce a higher fidelity model. The proposed ARIES-AT blanket and shield options have been incorporated along with the HTSC TF coil system. Responding to a request from Les Waganer, Ron calculated the projected plant and COE for a range of plant sizes up to 4 GW with the separate constraints of 16 Telsa TF coil field strength and 4 MW/m2 neutron wall load.

Results for the ARIES-ST (06/98 strawman in 1992$) are:

  • Conventional TF coil fabrication 87.2 mill/kWh
  • Low cost TF coil fabrication 80.8 mill/kWh (7.3% reduction)
  • No stabilizing wall (beta = 0.27) >102 mill/kWh

Engineering and Nuclear Performance Results - Mark Tillack is compiling engineering and nuclear performance data on the selected neutron source applications. He is displaying the data on the ARIES Neutron Source web page. The PbBi accelerator blanket has similar characteristics to those of ARIES-RS and ST. The difference is the particular driver. He would recommend using the same blanket concept as the ATW blanket.

Summary of Neutron Source Work - Don Steiner summarized the neutron source work necessary to conclude the work and reporting by next meeting. Les Waganer should summarize the selection and prioritization of the applications. The focus should be on the transmutation of actinides and plutonium. The metrics should be applied to each of the applications and the systems analysis completed. Mark should concentrate on the assessment of the engineering issues and nuclear performance. The team should try to obtain a neutronics analysis of the approaches being evaluated. Status will be evaluated at the next e-meeting.

Physics Assessment

Beta Optimization and Transport Calculations - Vincent Chan presented results concerning GA analysis of stability, divertors, and transport. Stability results were in general agreement with Kessel's results. Progress reports were given on NTM, RWM, CD, divertor, and transport analysis.

Optimization and Aspect Ratio Effects - Kessel reported on a continuation of his optimization activities. By extending his analysis to the 99% flux surface, increasing triangularity to 0.9, and expanding the functional form of the pressure profile description, he was able to obtain stable configurations with A=4 with betaN about 6.85. These required a current drive value in excess of 20%, using a banana/plateau/collisional bootstrap model. If betaN is dropped to 6.5, the CD requirement goes down to about 10%. This still corresponds to about beta = 8.6% for A = 4. BetaN is almost constant as A is reduced to 3.

Current Drive Issues - ARIES-AT Studies - T.K. Mau presented analysis aimed at reducing the number of CD systems to two and minimizing CD power requirements. He successfully matched the profiles obtained from detailed calculations with the seed profiles assumed. He found about 100 MW of LH power and 10 MW of ICRF power are needed to drive the 10% current in the latest Kessel strawman with betaN = 6.5.

In physics group discussions, the following action items were adopted:

  1. GA and PPPL would make their bootstrap current models available to each other and exchange "interesting" configurations for independent analysis.
  2. GA will perform free-boundary solutions using EFIT, and the ballooning calculation will be carried to the separatrix. They should try to get an independent solution for the PF coils' locations and currents.
  3. The baseline configuration will have shape parameters of kappa = 2 and delta = 0.9.
  4. The GA group will explore equilibrium with finite p-prime at the edge to allow finite bootstrap current and hence possible access to 2nd stability.
  5. Ming Chu is developing a new technique for using GATO to calculate intermediate-n modes using a ballooning representation.
  6. In calculating the resistive wall mode, it will be assumed that the PbLi in the blanket has the same stabilizing properties as a solid (conducting) wall.
  7. Both PPPL and GA will continue to optimize betaN versus ICD.
  8. The rotation profile from NBI will be calculated.
  9. The use of two or three CD systems is still an open question.
  10. A goal will be to remove the 5/2 mode by raising qmin above 2.5.
  11. GA will look at plasma startup, as they did with ARIES-ST.
  12. GA will use the Waltz-model to calculate transport.
  13. GA will examine the edge density with respect to divertors, bootstrap current, and current drive and make a recommendation.

ARIES-AT Engineering

Blanket Design Issues - Laila El-Guebaly summarized the general design issues that related to both the self cooled LiPb/SiC and the dual cooled LiPb/SiC/He blanket designs. For the moment, both designs will use the fusion power of the ARIES-RS 8/96 design basis, namely 2170 MW. The lifetime of the SiC structural material is assumed to be a nominal value of 3% burnup until more definitive irradiated data are available. The 3% burnup value results in a first wall lifetime of 2.8 years for a 6.6 MW/m2 peak outboard neutron wall load. The tritium breeding ratio is 1.1, and the neutron multiplication is 1.1 for both designs.

There will only be one radial blanket cell on the inner surface, whereas on the outboard there will be two cells. The second cell on the outboard region will be designed to be a lifetime component. This reduces the cumulative FW/B radwaste volume and the replacement cost by a factor of two. All first wall, blanket, shield, and divertor components cannot be classified as "cleared." The vacuum vessel will depend upon the shielding capabilities of the first wall, blanket, and shield. The magnets are capable of being classified as "cleared."

There was a question as to the relative cost of the heat transfer and transport systems for the two approaches. Ron Miller will work with Igor Sviatoslavsky and Rene Raffray to better quantify the cost of the two approaches.

Les Waganer will evaluate the feasibility of applying the low cost fabrication approaches to the low temperature shield and the vacuum vessel. The shield can either be solid low activation ferritic steel or a thinner structure filled with WC spheres. L. Waganer will develop costs for both approaches after the designs are better developed. Laila will assess the impact of the inboard shielding options on the radial build.

Self-Cooled Blanket Geometry and Engineering - Igor Sviatoslavsky described the self-cooled LiPb first wall and blanket design. He is assuming a SiC coefficient of heat transfer of 20 W/m·K, an allowable operating temperature of 1000°C, and a maximum primary stress of 140 MPa. The maximum LiPb coolant pressure is 0.75 MPa, of which 0.5 is hydrostatic. The baseline design produces an outlet LiPb temperature of 1000°C, achieving a thermal conversion efficiency of 55.8%. An option for 1100°C outlet temperature is also being evaluated which produces an efficiency of 58.8%. Igor described the coolant flow routes through the first wall and the blanket, along with the related thermal-hydraulic results.

Dual-Cooled Blanket Geometry and Engineering - Rene Raffray explained the dual-cooled LiPb/He first wall and blanket design. He is using the same SiC coefficient of heat transfer value, but he assumed a nuclear heat load 50% greater than ARIES-RS. The dual coolant approach allows the maximum temperature of SiC and the He (in the blanket) to remain at less than 1000°C, but the LiPb is heated slightly higher than 1100°C. The helium cools all first wall and blanket structural members. The CVD SiC surface coating on the first wall is expected to have a maximum temperature in the range of 1100°C. The exiting LiPb superheats the helium coolant in an intermediate heat exchanger to yield helium at 1080°C going to the Brayton turbine, thus achieving efficiencies in the range of 60%. Xueren Wang described the thermodynamic and hydraulic stress calculation results, which stay below the SiC design allowables.

There was a lot of concern about leakage (and subsequent pressurization of the blanket module) of the 20 MPa helium. Igor Sviatoslavsky showed a design capable of self cooling the entire blanket module, so the helium is not needed to cool the blanket structure. It was suggested that the two designs be merged to have a helium-cooled separate first wall (and divertor) and a self-cooled blanket structure. [Action: Raffray, Sviatoslavsky, and El-Guebaly] The temperature of the SiC blanket structure may exceed the recommended limit of 1000°C, but perhaps the approach used by the 1100°C self-cooled design could be adopted to maintain an allowable structural temperature.

Shielding and Activation Issues for ARIES-AT - Laila reviewed the shielding requirements for the ARIES-AT along with the main features of the shield and vacuum vessel. The radial builds for the inboard region, divertor region, and outboard regions were presented. The thickness of the vacuum vessel (20 cm inboard, 20 cm divertor, and 30 cm outboard) was highlighted. The outboard and divertor shield and vacuum vessel were assumed to be of borated ferritic steel while the inboard low temperature components employ tungsten carbide (WC) filler to reduce the size of the overall machine. These design choices result in radiation damage levels at the magnet that are considerably lower than Li/V ARIES-RS design.

Substitution of a WC high temperature (HT) shield in place of the borated, ferritic steel HT shield decreases the inboard radial build by 5 cm but generates higher afterheat by an order of magnitude. Correspondingly, the temperature rise during LOCA is also higher for the WC HT shield by 1 to 2 orders of magnitude depending on the time period after the onset of the accident.

Laila also discussed the ability to design the components so that they meet the "clearance" criteria. It is possible to increase the shield thickness to allow a thinner (~10-cm) vacuum vessel to meet the clearance criteria.

Costing of Superconducting TF and PF Magnets - Leslie Bromberg noted that the magnet community, along with the emphasis in Snowmass, is recommending that the low temperature superconducting magnet systems be reduced by a factor of 3 from the cost data predicted by the ITER community (no time period specified.) The superconducting strands (and sheath) are the major portion of the cost. Leslie predicted these elements could be reduced by a factor of 6 in perhaps 10-20 years. This might be accomplished by switching the sheath materials from Inconel to a steel with a similar coefficient of expansion to the superconduction strands, better manufacturing methods, and applying 10th of a kind learning curves.

Prospects for Obtaining a Cost-Competitive Fusion Reactor - After hearing some significant incremental improvements in a few key areas and the promise of several more, Les Waganer summarized his optimism that an advanced (physics and technology), commercial tokamak power plant might be able to favorably compete in the electrical (and hydrogen) energy market. The enabling factors include lower cost vacuum vessel and low temperature shields (~5% improvement), higher thermal conversion efficiency (~25%), higher availability (~12%), and larger plant sizes (~25% for a 2 GW size). If all of these promises are realized, the cost of electricity could be cut in half as compared to the 1996 ARIES-RS COE of 76 mills/kWh to achieve a COE in the mid 30s. This would really put fusion in the competitive range. But we must make a credible design basis for all of these (and other) improvements.

Snowmass Topics and Conclusions

The 11-23 July 1999 Snowmass fusion workshop was attended by several people on the ARIES team. It was felt that certain issues discussed in the workshop should be reported to the entire team to help guide and structure future ARIES research direction and emphasis.

Plasma Science Issues - Steve Jardin summarized the integration and physics performance measures identified for pulsed conventional tokamak (e.g. ITER-RC), advanced tokamak (e.g. ARIES-I and ARIES-RS), spherical torus PoP (NSTX), low-aspect ratio stellarator PoP, reversed field pinch PoP, and spheromak concept exploration. For each, the required developments, benefits, issues and weaknesses, opportunities to reduce development costs, development metrics for the next stage, international interaction, and key issues were identified.

Visions for Power Applications (Link to Snowmass site) - Don Steiner stated the four general topic areas that were addressed:

  • What are the currently projected economic, environmental and social characteristics of the alternative energy systems for the 21st century? (Fusion, fission, fossil, and renewables)?
  • What are the desired/required characteristics of an attractive fusion energy/non-energy system? Can current concepts evolve into such an end point(s)? If so, what are the required developments? If not, what are the required breakthroughs/innovations that will be required to make fusion systems commercially attractive?
  • What is the potential role of fusion for applications other than energy (e.g., H-generation, fissile fuel production, tritium production, (nuclear) waste burning, etc.)?
  • Is there a potential role for advanced fusion fuels (such as D3He)? If so, what is that role and how might it be exploited?

Combining first and second items, there is consensus that in the near future there will be a need to stabilize CO2 in the atmosphere; hence, there will be reduced use of coal, oil, and gas and/or CO2 sequestration (at some increased consumer cost). The fusion community should strive to establish a target date for the introduction of fusion because, at present, fusion is not even considered a potential energy source in the planning process. It was thought that if fusion could produce electricity for 5 to 6 mills/kWh this would be acceptable. Tokamaks are considered to be capable of leading to a commercial power plant, but other alternative MFE concepts are not defined well enough to determine their viability as a power plant energy source.

Alternative applications for fusion focused on applications involving neutron sources and propulsion. The neutron source applications involved transmutation of nuclear wastes, burning Pu, breeding fissile fuel, and other related applications. The competition for neutron sources is generally fission reactors and accelerators, with fission being the stronger competitor. NASA is studying fusion propulsion systems for deep space probes. For this application, fusion is the favored propulsion system, ahead of matter-antimatter and wind sail propulsion systems.

Advanced fuels, mainly D3He and catalyzed DD, were viewed as less developed and more speculative. Advanced fuels offer the distinct advantages of a higher direct energy conversion efficiency and reduced neutron flux. With D3He, the availability of the helium-3 is a concern.

Chamber Technology - Mark Tillack noted that, in addition to detailed technologies addressed, there were several common questions that applied to all technologies. What are the most important contributions that technology can make over the next 10 years to improve the vision for an attractive and competitive fusion product and reduce the cost of R&D for fusion? What contributions will technology make to advancing science? What research areas will be pushing the frontiers of sciences? What constitutes concept exploration, engineering proof of principle, and engineering performance extension for fusion energy systems?

It was recommended that the cost of superconducting magnet systems be lowered by a factor of two. Higher power density plasmas must be developed with more efficient current drive and plasma heating. Solid walls should handle up to 50 MW/m2 heat flux with low tritium retention. Liquid walls should be assessed for both IFE and MFE. Safety and environmental aspects will be improved by reducing both the volume and the hazard potential of the generated waste. Low activation materials should be pursued as well as possible metal structural systems. RAMI requirements should be evaluated on the next generation experiment.

Magnet Systems - Leslie Bromberg explained the goals of reducing the cost of superconductor and structure as well as optimizing the entire magnet system to achieve both physics and engineering goals. In MFE, there is a goal to reduce the cost of the TF and PF magnet systems by two. In IFE, the goal is to reduce the heavy ion driver quadrapoles by a factor of two. In resistive magnets, such as ARIES-ST, the goal is to verify that ultra-low cost manufacturing techniques are credible. In high temperature superconductors (HTS), the challenge is to develop high current and large area joints with a reasonable fabrication cost and to develop long wire lengths. A test facility is recommended to validate the new magnet materials and technologies.

Comments on Liquid Wall Waste Reduction and Tritium Self-sufficiency - Dai-Kai Sze responded to a Snowmass position paper that postulated a waste reduction factor of 1500, comparing a solid wall and a liquid wall chamber. He suggested more realistic engineering assumptions and the waste reduction factor dropped to a range of 10-50.

Dai-Kai also discussed a chart of the expected tritium availability with tritium supplies from CANDU reactors and several using facilities including some ITER options and a VNS option. The CANDU reactors were assumed to stop producing tritium in the 2025 timeframe. Some ITER options depleted the tritium supply inventory, except the ITER-RC, which retained some small inventory with nominal operation. When VNS is combined with ITER-RC, the tritium supply is marginally exhausted. Based upon these data, it was recommended ITER have a tritium-breeding blanket from the start of operation.

Waste Minimization - Dave Petti noted the emphasis on nuclear waste volume minimization. Fusion is predicted to have three times the low-level nuclear waste of a comparable fission reactor. Even Class C waste has negative connotations. He described a new concept that is being promoted in Europe and is being discussed in the US. This concept is "Clearance" and is quantified to be material with a low enough hazard potential to be released or cleared for reuse. It would be 100 to 10,000 times lower radioactivity than Class C waste. Laila El-Guebaly noted that the blanket and shield could never be cleared, the vacuum vessel may or may not be cleared, and the magnets probably will always be capable of being cleared.

Top of the page.