Brookhaven National Laboratory was a leading proponent (along with Babcock & Wilcox) of the Particle Bed Reactor, which has been evaluated for SDI(1) and SEI applications. This concept was developed in 1981 on the basis of earlier work done on the Rotating Bed Reactor.
Both unmoderated fast spectrum and moderated thermal spectrum versions of the particle bed reactor have been suggested. In the unmoderated version, the fuel-element particles are contained between porous cylindrical screens referred to as frits, with the outer cold frit composed on Inconel stainless steel, and the inner hot frit fabricated from Tungsten and Rhenium. In the moderated version of the reactor, the fuel particles are contained between hot and cold frits, as with the unmoderated reactor, with these fuel elements suspended in a monolithic solid moderator. The reactor was controlled with rotating drums in the outer reflector. Hydrogen propellant flow moves in an inward radial direction, with a nominal exhaust temperature of over 2500 K, and maximum power levels in the range of 1000 MWt.
The Timberwind and SNTP programs are focused on the development of thermal spectrum moderated reactor configurations.
A number of significant technical issues challenges must be resolved if the promise of the PBR concept is to be realized..
+ High temperature particle and coating materials must be developed that are erosion resistant, neutronically benign, and compatible with the hot frit materials.
+ The hot frit material and coating must be erosion resistant, compatible with fuel particles, and have adequate mechanical properties at high temperatures.
+ The potential for propellant flow induced vibration on fuel element components during reactor flight operations requires additional analysis.
+ The potential contributions of slush hydrogen propellants, or the use of low pressure reactor operations to support higher specific impulse operations with disassociated monatomic hydrogen require further analysis.
+ The high exhaust temperature requires either a transpiration or regeneratively cooled nozzle, or a radiation-cooled carbon-carbon nozzle. The carbon-carbon nozzle was used as the baseline, given its low mass and simple design. Carbon-carbon nozzles have been tested in solid-rocket motors at temperatures up to 4000 K. But development of coatings (TaC, ZrC and NbC are candidates) will be required to avoid hydrogen erosion.
+ A hot-bleed cycle turbopump has been selected as the design baseline. Maximizing specific impulse will require maximizing the turbine operating temperature. This necessitates the use of carbon/carbon turbopump blades, which will require coatings to avoid hydrogen erosion. Development of these coated blades pushes the state of the art.
Some initial prototype experiments have been conducted, including electrically heated blowdown fluid dynamics tests, as well as characterization of ZrC coated and UZr-C fuel particles. But significant testing remains to be accomplished.
a - Testing
Development of the Space Nuclear Thermal Propulsion program was envisioned as requiring five major testing efforts:
TOP BOTTOM
Nuclear Element Tests (NET) which are used to validate fuel particle characteristics in a single fuel element are conducted in the Sandia National Laboratory Annular Core Research Reactor (ACRR) (Figure IV-11). The first test, designated NET-0 was a non-nuclear fueled element which was electrically heated to allow testing of cooperative control and test hardware under prototypical conditions without the additional problems that would be posed by equipment failures in a nuclear activated environment. However, this test, conducted in October 1991, failed due to loose graphite particle fragments that blocked the flow of propellant. Resolution of this anomaly was still pending identification of the source of these particles. According to one interpretation, they originated in the electric heaters used to simulate operational thermal conditions. Another interpretation, which would be consistent with prior anomalies, was that the particles resulted from fragmentation of the fuel particles themselves, which would suggest that the present fuel particle design has not resolved the particle structural fragility previously identified in the PIPE-2 test. The subsequent nuclear-heated NET experiments will provide the basis for later tests of fuel elements at typical operating conditions. The first of these, NET-1, is scheduled for no earlier than December 1992, though this test may not be initiated until early 1993, pending resolution of issues identified in the NET-0 test. Subsequent NET tests are planned at a rate to two to four each year over a two to three year period. Each test will include up to 40 seconds of irradiation in the ACRR.(2)
Hot Hydrogen Component Test Facility, experiments will validate non-nuclear components, such as pump machinery and other components, by simulating sub-scale engine operating conditions. A principal activity at this facility will be evaluation of nozzle erosion characteristics. Developed and operated by Allied Signal Aerospace Company's Garrett Fluid Systems Division, it will be located at the remote San Tan test site 40 kilometers southeast of Tempe, Arizona (Figure IV-14). The Preliminary Design Review was completed in March 1992, and the facility will begin operations in the second quarter of CY1993.
Engine Integration Tests, would be conducted to demonstrate propellant management systems without an operating reactor in the loop. Approximately 50 of these tests, which would last from a few seconds to several minutes, would be conducted over the ten year life of the SNTP program.(3)
Pulsed Irradiation of PBR Fuel Element Test (PIPET) Ground Test Reactor multi-element reactor tests (Figures IV-12, IV-13) will verify prototypical fuel element characteristics at rated operational reactor conditions. This new reactor will provide a more realistic test environment than can be provided by the Sandia ACRR facility. It will permit simultaneous tests of as many as seven fuel elements, and will be capable of accommodating alternative fuel element designs. The preferred site for the PIPET test reactor was at the DOE Nevada Test Site, which is located within the Nellis Air Force Base test range. The Saddle Mountain Test Station in Area 14 was the recommended location within the NTS. This central location was preferred in order to minimize environmental and nuclear safety concerns. Construction of PIPET is slated to run from 1992 through 1995, when powered operations will begin. PIPET will be the first self-sustained power-producing PBR test, with a maximum power level of 550 MWt.(4) At least four and as many as six PIPET cores will be tested prior to testing a full-scale reactor engine, with each core being tested five times. Each test run will last for approximately 500 seconds, with the reactor experiencing multiple starts and stops. Each set of runs will be separated by intervals of three to five weeks, due to safety and environmental considerations.(5) As part of the PIPET testing, one or more fuel elements will be tested to failure to determine design margins.(6)
Mini Ground Test Assemblies (mini-GTA) will integrate fuel elements developed in the PIPET tests with other engine elements demonstrated in System Component Tests. The initial phase of testing at the Ground Test Facility will include the sub-scale PIPET tests. These would be followed by tests of two mini-GTA reactors, which include a seven-element core designed to more closely represent a prototypic full size GTA PBR. Over a period of three or four years up to 10 tests would be conducted, with approximately five test runs on each set of fuel elements.(7)
Ground Test Assembly (GTA) testing will begin upon satisfactory completion of the mini-GTA tests. These will require additional capabilities, including four larger test cells as well as a much larger effluent fluid treatment system (Figure IV-15), which would be added to the Facility to support full-scale testing (Figure IV-16). This upgrade plan assumes that the GTA tests are conducted at the Saddle Mountain Test Site that will be used for PIPET testing (Figure IV-17). An alternative location for the GTA was at the Idaho National Engineering Laboratory (Figure IV-18). Potential sites at INEL include the existing QUEST and LOFT(8) test sites. The first GTA, which will be limited to testing a power level of 1,000 MWt, will conclude Phase II
of the SNTP program. As many as three subsequent cores, beginning with the 2,000 MWt GTA-2, will also be subjected to up to five tests runs apiece.
Qualification Test Assembly will also be tested under Phase 3 of the program. One test run will be conducted on this reactor, which will also include turbopump assembly, valves, an control systems of the type to be used on the flight test article, as well as a truncated nozzle and a liquid hydrogen run tank.(9) This engine will be tested for up to 1,000 seconds at full power of 2,000 Mwt.
Flight Test Experiments include ground-based tests of the Flight Evaluation Vehicle, with Phase 3 of the program culminating in a flight test launched from Vandenberg Air Force Base. Initial plans for this flight called for the vehicle to be placed into a trajectory over the South Pacific Ocean, with powered climb circling over Antarctica. However, this raised concerns over the potential for a launch accident to deposit the reactor in New Zealand, the only major populated area under the proposed flight path, although the odds of such an event were calculated by Sandia Lab to be relatively small (4.3 out of 10,000).(10)
A variety of other tests have been conducted under Phase 2 of the SNTP effort. These include non-nuclear tests have included evaluation of hydrogen thermal hydraulic flow characteristics in a seven-element particle bed configuration.
All of these tests, including flight testing, are intended to be conducted under the provisions of the SNTP Safety Policy, the primary objective of which is:
To ensure the maximum protection of the health and safety of the public and program workers, and to protect the environment from contamination or damage as a consequence of program activities.
The implementation of this objective is:
To be accomplished through the development of environmental, safety and health plans and guidelines used in monitoring and controlling program activities.
SNTP Safety Policy elements include provisions that:
Safety and environmental protection will be explicitly considered and incorporated into each activity or system from its outset and continually throughout its lifetime.
The program shall explicitly consider off-normal and accident situations that could occur during the entire life-cycle of the program. These scenarios will be documented and reduced to a set of credible accidents that will form the analytical and technical safety basis of the development effort.
The program will meet all mandated, statutory and legal requirements for safety and environmental protection.
Compliance will be based on the principle of defense-in-depth involving multiple physical, procedural and administrative barriers.
To the extent practicable, risks will be reduced to levels below the mandatory requirements considering project resources, the state of technology, and the benefits to be gained from the program. Risks will be As Low As Reasonably Achievable (ALARA).
Safety is everyone's responsibility. Line management is responsible for compliance with safety policies. Everyone shares the responsibility to achieve, demonstrate and document outstanding levels of safety and environmental protection.
It is intended that the program will meet or exceed all US Nuclear Regulatory Commission safety goals for commercial reactors, where applicable, as well as Energy Department safety goals for DOE reactors. The program will also comply with the International Atomic Energy Agency's principles for commercial reactors, as well as relevant provisions of United Nations treaties covering space systems.
According to the SNTP Environmental Impact Statement, the Ground Test Article PBR engine would have the following characteristics:(11)
Thrust kN
Run Time 100% Thrust seconds 1000
Specific impulse tests seconds 850
goal seconds 900 - 1000
Thrust-to-weight ratio 30:1 - 40:1
Power MWt 2000
Propellant Temperature 3000 K
Propellant Flow Rate kg / sec 20
Design and operation of the Ground Test Facility was anticipated to require a major engineering effort to provide a system capable of removing fission products from the engine exhaust, which would primarily consist of hydrogen propellant, with a flow rate of approximately 50 kg/sec. The facility must be capable of handling both normal operating conditions, as well as off-normal conditions that might arise from a catastrophic engine failure.
Design of the test facility was complicated by the potential for fission products to contaminate the interior surfaces of the exhaust system. Xenon and Krypton will remain in gaseous form, and can be removed by cold traps or by carbon or zeolite absorption beds. Alkali metals such as Cesium and Rubidium, and halogens such as Iodine and Bromine, can react with steel duct liners and condense at relatively low temperatures, but they remain water soluble, and thus surfaces can be decontaminated with wash water. Strontium and Silver will also condense on duct surfaces, and silver contaminants must be removed by mechanical scrubbing.
Two approaches have been evaluated for this objective:
1 - Storage of exhaust gases for slow processing after engine tests are completed. Tunnels at the Nevada Test Site constructed to support nuclear weapons testing could be used for this purpose.
2 - On line cleanup of the exhaust gases with subsequent release to the atmosphere, as was done in the Nuclear Furnace program in the early 1960s.
Although location of the Ground Test Facility underground might provide an additional margin of safety, there are a number of questions concerning the use of existing tunnels at the Nevada Test Site for this purpose. These tunnels were constructed to support underground nuclear tests, and are still required for this application. They are located several dozen kilometers from existing nuclear rocket test facilities at the Nevada Test Site, which complicates the logistics of their use. Given the large volumes of gas that would be released during engine tests, the existing volumes of these tunnels does not appear to be adequate to contain exhaust gases at acceptable pressures. This shortcoming would be magnified by the containment requirements posed by off-normal operations. And during normal operations the tunnels would be contaminated by fission products.
The Nuclear Furnace system included water sprays for cooling engine exhaust, with boilers and condensers to cool the bulk of the resulting water vapor. Filters, driers and hydrogen cooling was used to condense the remaining water vapor. Charcoal beds were used to remove Xenon and Krypton gases from the hydrogen flow, which was then flared to the atmosphere. Construction of a facility using these techniques could cost from $400 and $700 million. The inclusion of capabilities to simulate high-altitude operations could increase this cost substantially. The SNTP Effluent Treatment System was intended to be capable of removing 99.9% of particulates and volatiles, and 99.5% of halogens and noble gases from the propellant effluent stream.(12)
b - Fuel Particle
The PBR fuel particles consist of a Uranium Carbide kernel, surrounded by a series of layers of other materials intended to provide fission product containment, structural integrity, and fuel kernel protection from hydrogen propellant flow.
In principle, the coated fuel particles used in PBR systems could provide enhanced environmental and safety characteristics:(13)
"The requirements imposed by modern safety and environmental standards deserve more discussion. The current regulatory and public awareness climate makes the open-cycle ground testing and flight qualification objectives orders of magnitude more dependent on fuel integrity and fission product retention than was the case during Project Rover... Additional fission product release barriers will be required in a program conducted in the 1990s and beyond. The coated particle fuel concept was particularly responsive to the more demanding requirements because it provides multiple barriers to fission product release. SEI applications of NTP will be open to public review and debate, unlike Rover, which was a classified project."
However, meeting safety and environmental goals will prove challenging, since even system proponents concede that, although:(14)
"The goal of the fuel development program is to retain the fission products within the particle by using multiple coatings -- one of the coatings being investigated is Niobium Carbide. Each coated fuel particle is its own fission product containment vessel.... at very high operating temperatures, there may be some limited fission product diffusion through the particle coatings."
Stress on fuel particles was a product of thermal stress, coupled with internal stress due to buildup of fission gases, volatiles and solid products. The accumulation of solid fission products reduces matrix porosity, while fuel consumption increases porosity. Thermal stresses can be minimized by application of coatings at temperatures near operating levels.
Actual operating temperature limits of fuel particles will be lower than the theoretical maximums, to account for factors such as uncertainties in heat transfer and thermal hydraulics, as well as to provide performance margins to enhance reliability. Some applications, such as piloted missions, may result in further decrements in operating temperature to provide further safety margins.
BISO particle fuel (Figure IV-19) is well characterized and already available, based on use in commercial power reactors. These particles have been documented to have high fission product retention (greater than 99.99%) with burnup fractions up to 50%.(15)
TRISO particle-fuels (Figure IV-20) in the HTGR program had an outer carbon coating, and a containment layer of SiC. The PBR concept draws on the experience accumulated in the High Temperature Gas Reactor (HTGR) program, which included extensive work on TRISO coated-particle fuels.(16) In tests, these particles have demonstrated radionucleide gas containment fractions of 10-5 at temperatures of up to 2350 K, though containment fractions rapidly fall to about 10-1 at temperatures of up to 2500 K.(17)
SNTP Baseline fuel particles (Figure IV-21) are approximately 0.5 mm (500 microns) in diameter.(18) Unlike the TRISO particles, they do not have an outer coating. The particles have a 40 micron containment layer of ZrC, covering a 40 micron sealant layer of pyrolytic Carbon, overlying a 40 micron buffer layer of porous 50% Carbon. As with the previous particle concepts, the 260 micron diameter fuel kernel would consist of Uranium Carbide (UC2), enriched to 93%. This baseline fuel particle is limited to an operating temperature of 2,650 K. A pilot line for the production of these particles was established in support of the PIPE tests, which generally validated the robustness of this design, although some particle fracturing was observed.
Eutectic fuel particles (Figure IV-22) consist of an outer carbide coating, surrounding a fuel kernel of either pseudo-binary (Zirconium/Uranium Carbide or Niobium/Uranium Carbide) or pseudo-ternary (Zirconium/Niobium/Uranium Carbide).(19) These solid solutions are fabricated by extensive ball milling of graphite with appropriate mixtures of Uranium Oxide, Zirconium Carbide and Niobium Carbide. Binary ratios range from 30% Uranium and 70% Niobium or Zirconium to 10% Uranium and 90% Niobium or Zirconium. Ternary ratios are typically 30% Uranium, 30% Niobium and 40% Zirconium.(20)
Variables in the preparation of the kernels include Zirconium, Niobium and Uranium ratios, as well as the carbon to metal ratio. The homogeneity of the final product is influenced by the initial carbon content. Single phase solid solutions resulting from initial carbon to metal ratios of less than one, as with a 30% UC, 30% NbC and 40% ZrC mixture which yields a carbon to metal ratio of 0.98. Higher carbon contents, such as the carbon to metal ratio of 1.06 of a 30% UC, 30% NbC and 40% ZrC mixture, producing heterogenous kernels with discrete isolates of tetragonal phase Uranium Carbide. Preliminary tests suggest that this fuel particle configuration may permit operations at temperatures as high as the 3,300 K melting point of its Uranium Carbide kernel.
Infiltrated Kernel fuel particles (Figure IV-23) could permit higher operating temperatures by confining liquid uranium fuel within a porous carbon kernel.
Maintenance of fuel particle integrity during reactor operations has consistently proven a major technological challenge for particle bed reactors. The shift from BISO to TRISO particles was motivated by the superior operating lifetime of the more robust TRISO design for long-life high temperature gas-cooled power reactors. While current PBR operating requirements are denominated in hours rather than years, similar operating lifetime challenges remain. Although these shorter operating lifetimes obviate some of the concerns posed by earlier power reactor requirements, such as uranium kernel migration, coating
integrity has become a more serious issue. Differences in thermal coefficients of expansion place considerable strain on complex multi-layer particles. Particle configurations which are theoretically capable of withstanding projected PBR operating environments have been designed. But it has proven difficult in practice to fabricate large numbers of such particles with adequate levels of quality control to avoid unacceptable rates of particle fragmentation during reactor operations. The appropriate mix of configuration modification and process control needed to resolve these problems awaits identification.
c - Fuel Element
The PBR fuel particles are contained between a pair of concentric porous cylinders, called frits. Propellant flows through the particle bed in an inward radial direction, minimizing the pressure drop between the cold frit inlet annulus and the hot frit outlet channel. Although fast reactor configurations have been considered using particle fuels in the past, the SNTP engine was based on a thermal reactor configuration, with hexagonal moderator blocks surrounding each fuel element.
The exterior surface of the cold frit is structured to properly distribute the propellant flow along the length of the fuel element. Radial inflow of propellant through the fuel element results in lower pressure drop, improving overall performance. As a result, the fuel element response to transients in propellant flow restores nominal operating conditions on time scales of a few seconds.
Improved computer simulation capabilities, as well as more capable advanced materials, facilitate the customized arrangement of particle bed fuel elements to optimize performance for specific missions. The fuel element would be supplied with a propellant or coolant flow that was optimized for nuclear effects, heat transfer characteristics, or power conversion efficiency.
The initial PIPE fuel element configuration (Figure IV-24) required development of a one-piece slotted rhenium metal hot frit, as well as a stainless steel cold frit of variable thickness configured for power and propellant flow matching. These tests demonstrated the integrity of the packed particle bed under high temperature conditions. The hot frits of this configuration used Rhenium wire, which is mechanically strong, ductile, and not subject to high temperature hydrogen corrosion, as is graphite. Rhenium hot frits were fabricated by Babcock and Wilcox for the PIPE tests. The surrounding hexagonal moderator block consisted of a aluminum and polyethylene matrix. However, this moderator design had a relatively low operating temperature and short operating lifetime that could support interceptor missions, but not other applications.
The next configuration, proposed for the SNTP program (Figure IV-25), substituted a sintered aluminum cold frit for the stainless steel cold frit of the PIPE fuel element. This cold frit was fabricated using a micro-peening process which was computer controlled in order to produce the desired type of porosity mapping. This selectively closes off porosity along the length of the cylinder in order to regulate the longitudinal temperature and propellant flow distribution. The hot frit consists of a porous carbon-carbon composite coated with ZrC to protect the carbon substrate from erosion by hot hydrogen.(21)
In contrast to the symmetrical configuration of the PIPE fuel element, the volume of the particle bed itself was tapered in this design, with the greatest thickness at the inlet end, and minimum thickness at the outlet end, to provide improved power and propellant flow matching. Material options for the hexagonal moderator block include water or lithium hydride and beryllium combinations, which would provide for higher operating temperatures and longer operating lifetime than the previous configuration.
The third configuration considered (Figure IV-26) was based on the results of the PIPE tests, during which it was determined that low propellant flow rates produced uneven heating in the particle bed, contributing to particle fracturing during reactor power ramp-up and ramp-down, as a result of uneven expansion and contraction. As a result, a liner was added between the cold frit and the particle bed to maintain particle compliance, allowing the particles to breathe during ramp-up and ramp-down. The liner was designed to come into contact with the cold frit at operating temperatures. A niobium coated graphite hot frit, which will be tested in a future Nuclear Element Test, will be fabricated using a numerically controlled development process. This design permits the hot frit to operate at higher temperatures while reducing hydrogen erosion and neutronic penalties.
The most advanced fuel element configuration considered (Figure IV-27) incorporates a more sophisticated cold frit design which maintains porosity with a stainless steel platelet surface, in place of the porous aluminum of prior configurations. This platelet surface was separated from the particle bed by a compliant stainless steel liner. It also uses a more advanced hot frit fabrication technique which substitutes monolithic three-dimensional carbon-carbon as the substrate for either a niobium carbide or more capable tantalum carbide coating.
1. Marshall, A.C., "A Review of Gas-Cooled Reactor Concepts for SDI Applications," SAND87-0558, (Sandia National Laboratory, Albuquerque, NM, August, 1989).
2. Department of the Air Force, Space Nuclear Thermal Propulsion (SNTP) Program -Final Environmental Impact Statement, 19 September 1991, partially declassified 11 March 1992, page 2.3-7.
3. Department of the Air Force, Space Nuclear Thermal Propulsion (SNTP) Program -Final Environmental Impact Statement, 19 September 1991, partially declassified 11 March 1992, page 2.3-38.
4. ibid.
5. United States Air Force, "Scoping Meeting on the Environmental Impact Statement for the Space Nuclear Thermal Propulsion Program," 7-9 April 1992.
6. Department of the Air Force, Space Nuclear Thermal Propulsion (SNTP) Program -Final Environmental Impact Statement, 19 September 1991, partially declassified 11 March 1992, page 2.3-32.
7. ibid, page 2.3-13.
8. Loss Of Fluid Test.
9. Department of the Air Force, Space Nuclear Thermal Propulsion (SNTP) Program -Final Environmental Impact Statement, 19 September 1991, partially declassified 11 March 1992, page 2.3-28.
10. Smith, Jeffrey, "US Developing Atom-Powered Rocket," The Washington Post, 3 April 1991, pages A1, A6.
11. Department of the Air Force, Space Nuclear Thermal Propulsion (SNTP) Program -Final Environmental Impact Statement, 19 September 1991, partially declassified 11 March 1992.
12. Department of the Air Force, Space Nuclear Thermal Propulsion (SNTP) Program -Final Environmental Impact Statement, 19 September 1991, partially declassified 11 March 1992, page 4.3-22.
13. Horman, F.J., et al, "Particle Fuels Technology for Nuclear Thermal Propulsion," AIAA/NASA/OAI Conference on Advanced SEI Technologies, Cleveland, Ohio, 4-6 September 1991, Paper AIAA 91-3457.
14. United States Air Force Systems Command Phillips Laboratory, "Phillips Laboratory Announces Program in Space Propulsion," Office of Public Affairs release 92-02, 13 January 1992.
15. Botts, T.E., et al, "Nuclear Reactors Using Fine Particulate Fuel for Primary Power in Space," 17th Intersociety Energy Conversion Engineering Conference - 1982, paper IEEE 82-9237.
16. Powell, J.R., and Botts, T.E., "Particle-Bed Reactors and Related Concepts," in Advanced Compact Reactor Systems, (National Research Council, Washington, DC, 1983).
17. Kania, M.J., et al, "Coated Particle Fuel Performance," NASA Nuclear Propulsion Workshop, NASA Lewis Research Center, Cleveland OH, July 10-12, 1990.
18. Lee, C.C., et al, "Propellant-Neutronic Stability of the Rotating, Fluidized Bed, Space-Propulsion Reactor," Proceedings of the 26th Intersociety Energy Conversion Engineering Conference, Boston, Massachusetts, 4-9 August 1991, Volume 1, pages 403-408.
19. El-Genk, Mohamed, et al, "Pellet Bed Reactor for Nuclear Propelled Vehicles," NASA Nuclear Propulsion Workshop, NASA Lewis Research Center, Cleveland OH, July 10-12, 1990.
20. Bremser, A.H., et al, "Preparation of Mixed (U, Zr), (U, Nb) and (U, Nb, Zr) Carbides," 9th Symposium on Space Nuclear Power Systems, Albuquerque, NM, 12-16 January 1992.
21. Powell, J.R., et al, "Nuclear Propulsion Systems for Orbit Transfer based on the Particle Bed Reactor," chapter 28 in Space Nuclear Power Systems 1988, (Orbit Book Company, Malabar, FL, 1989), pages 185-198.