EXPLORATORY RESEARCH ON RADIOISOTOPE THERMOELECTRIC GENERATORS FOR DEEP SPACE MISSIONS

To support the development of a Radioisotope Power Source (RPS) for exploration missions into deep space based on 241 Am, the European Commission’s Joint Research Centre (JRC) has recently launched an exploratory research project, scheduled for a period of two years. It aims at finding innovative solutions for higher performance, safety and reliability of such a device. In addition, it will serve as a start-up platform for this field of research and enable identification of research areas, where JRC can significantly complement and support the already existing activities in Europe. The project is divided into several subtasks and will investigate The new exploratory research project will be introduced together with an overview on the available facilities and capabilities of JRC in this domain. Alternative americium forms with potential improved stability versus the oxides are discussed and innovative thermoelectric materials based on actinides are introduced.


INTRODUCTION
The European Space Agency (ESA) has decided to develop a Radioisotope Heating Unit (RHU) and a Radioisotope Power Source (RPS) for future exploration missions into deep space and contracted several research organisations, mostly in UK, with the implementation [1]. Unlike past ESA missions, which utilized solar panels as energy source, the new device will be powered by 241 Am, a transuranic isotope, which generates heat by decay. The heat will be either used directly to keep sensitive equipment at a required temperature or converted to electricity by means of thermoelectric converters (thermocouples). Future concepts might also include more efficient conversion mechanisms, such as Stirling motors [1], [2].
Most of the historical nuclear power sources for exploration missions were either deployed by the United States or the former Soviet Union and were based on 238 Pu [2], [3], a radionuclide with a high isotopic power of 0.567 W/g and a half-life of 87.7 years [4]. Unfortunately, there is a global shortage of this isotope and the costs associated with its production are high. Therefore, ESA has decided to alternatively use 241 Am, which has a half-life of 432.2 years. 241 Am is a decay product of 241 Pu (T1/2 = 14.35 y) and can be extracted chemically from existing plutonium stocks in Europe. 241 Am, however, has some disadvantages compared to 238 Pu, such as a relatively low power density of 0.114 W/g, chemical instability at high temperatures and higher radiation levels.
In order to support the development of a European RPS the European Commission's Joint Research Centre (JRC) has launched an exploratory research project. Within the frame of this project JRC will perform an assessment of the safety and reliability relevant properties of americium oxide, which is the currently considered chemical form for application in space batteries, (e.g. measurement of thermal properties, investigation of helium release behaviour, swelling behaviour and oxygen release kinetics, ageing studies, examination of self-irradiation damage) and its compatibility to cladding and structural materials. In addition, alternative americium compounds will be investigated for their thermal and mechanical properties as well as their chemical stability under space mission conditions. Innovative candidate materials for thermoelectric conversion containing actinides will be assessed at high temperatures, utilizing the unique facilities and competences at JRC-ITU and JRC-IET. 241 Am is a transuranic element which is created in nuclear reactors by neutron capture and subsequent decay. Despite the half-life of "only" 432.2 years 241 Am is one of the main contributors to the long-term radiotoxicity of spent nuclear fuel, due to its decay product 237 Np, which has a half-life of 2.144 million years. It is, therefore, a key isotope for research on partitioning and transmutation with the aim to destroy the long-lived radioactive waste in dedicated irradiation facilities [5]. However, up to now 241 Am was never utilized as power providing isotope for space applications (to our knowledge), and some of its characteristics are different to 238 Pu and need to be considered.

241 AM AS POWER PROVIDING ISOTOPE
Radiation: 241 Am decays by alpha decay to 237 Np with a specific activity of 1.271·10 11 Bq/g and it emits significant gamma radiation (Tab. 1). Therefore, it has to be handled under strict radiation protection measures, and special equipment, such as shielding and remote operation tools, is required. Due to the low energy of the emitted gamma rays (main line at 59 keV) it can be shielded relatively easy with few millimetres of lead (or comparable), but in opposite to 238 Pu at least some shielding has to be provided. Therefore, additional payload for a RPS powered by 241 Am might be necessary. Table 1. 241 Am Gamma Emission [4].

Type
Energy(keV) Emission Probability 69.760 (± 0.020) 2.88e-5 (± 3.0e-6) Ȗ 28.510 (± 0.050) 2.77e-5 (± 3.0e-6) Ȗ 123.000 (± 0.040) 1.080e-5 (± 3.9e-7) Chemical stability: In order to avoid burnup and potential dispersion into the atmosphere during launch abort or re-entry accidents a chemical and mechanical stable form of americium under oxidizing conditions is desired. Therefore, the currently existing RPS designs include dioxide pellets encapsulated in an iridium cladding [2]. Unfortunately, americium exists in different oxidation states and the dioxide is prone to phase changes at elevated temperatures and in vacuum [6]. Fig. 1 illustrates the phase change in vacuum at 1200°C from the dioxide (AmO2) to the sesquioxide (Am2O3) as well as the associated phase change from cubic (alpha) to hexagonal structure.
Because of this behaviour, the oxide form might be not the optimal solution with respect to long term stability of the pellets in the vacuum of space. Alternative compounds, such as AmAlO3, could show a more stable behaviour at high temperatures under both, oxidizing and reducing conditions. Within the frame of the presented exploratory research project JRC will investigate a number of candidate americium compounds for their chemical stability and resistance towards self-irradiation effects. Figure 1. Americium-Oxygen Phase Diagram [6].
Thermal power: Tab. 2 shows a comparison between the thermal properties of the oxides of 238 Pu and 241 Am. 241 AmO2 has a power density of 0.101 W/g, which is about one fifth of 238 PuO2 (0.493 W/g). As a consequence it is necessary to use much more material in order to achieve the same power if americium is chosen. Therefore, it would be very interesting for an americium driven RPS to find possibilities to partly compensate the low power density of 241 Am by increased efficiency of the electricity conversion. This could be f.i. achieved by higher operating temperatures, thereby increasing the Carnot efficiency. Therefore, the exploratory research project will look into alternative americium forms, which are long-term resistant to high temperatures, but also candidate thermos-electric materials with high temperature resistance based on actinides.
longer half-life, which makes it the ideal power providing isotope for very long lasting missions into the outer region of the solar system or even into the interstellar space and the Oort cloud. Fig. 2 shows the decrease of thermal power of 241 Am compared to 238 Pu over an period of 200 years. While the plutonium has decayed to circa 20% of the power at BOM, the americium source is still producing about 80% of the initial power. However, it should be noted that a large fraction of the power decrease of an RPS comes from the degeneration of the employed thermoelectric material. Therefore, the exploratory research project is looking into very long-term stability of potential americium compounds as well as alternative actinide based thermo-electric materials.

Figure 2. Decrease of 241 Am and 238 Pu power over 200 y
Thermal conductivity: The thermal conductivity of americium dioxide was found to be slightly lower compared to plutonium dioxide [7]. However, it should be noted that it decreases dramatically with deviation from stoichiometry and the sesquioxide shows a very low thermal conductivity (~ 1 W/m K at 1500 K, see Tab. 2). Since surface temperature, power density and thermal conductivity determine the central pellet temperature during operation, a low thermal conductivity can represent a limiting factor for a desired hot leg temperature of the thermocouples. In addition, a significantly increased thermal conductivity can increase the design flexibility of the RPS and enable relevant space and weight savings. Therefore, we will also investigate alternative americium compounds, and composites, with respect to their thermal properties (including thermal conductivity) at high temperatures.

THE MINOR ACTINIDE LABORATORY AT JRC-ITU
Due to strong gamma radiation, working with 241 Am can result in high dose rates for the operating personnel. Therefore, remote operated and shielded equipment is advantageous or even necessary in order to prepare americium based pellets for RPS. JRC-ITU possesses a unique infrastructure for handling of highly radiative actinide materials, the so called Minor Actinide Laboratory (MA-Lab) [8]. It is of high importance for safety research on Transmutation fuel in Europe, as it is one of the only dedicated facilities for the synthesis of minor actinide containing materials, either for property measurements or for the preparation of test pins for irradiation experiments. The MA-lab consists of seven glove-boxes with protection walls forming two separate chains. A schematic lay-out of the Ma-Lab is shown in Fig. 3. The glove boxes are shielded by 50 cm water and 5 cm of lead. Based on the thickness of the water and lead wall, the mass limits have been calculated to 150 g of 241 Am or 5 g of 244 Cm. The glove boxes can be accessed from behind manually if radiation levels are low, for experiment preparation or maintenance. In addition, tele-manipulators and remote operated automated equipment can be used for operation at high dose rates.
The glove boxes of the minor actinide laboratory are configured as complete preparation chain for minor actinide containing samples from the base material to the fully encapsulated sample, and the MA-Lab represents an ideal infrastructure for preparation of highly radiating americium pellets and fully qualified fuel pins.
The synthesis of the base material (powder) is performed in the glove box named "infiltration". The process is dust-free, based on the so-called gel supported precipitation [9] and the porous bead infiltration technique [10]. This process is highly flexible and easily adapted to the requirements and specifications of new sample compositions. The next glove box contains a calcination furnace and other equipment for powder preparation. The prepared powders are dust free, the individual beads typically show heterogeneous size distributions between 30 μm to 120 μm and are ideal for pressing pellets. The ready to press powder can be transferred via an automated channel to the next glove box, where it can be pressed to pellets. After sintering in reducing or oxidizing atmosphere (glove box "Sintering") the pellets are fully characterized and inserted into a cladding. Finally, pin welding and non-destructive weld examination are performed in the two last, alpha free glove boxes.

CONCLUSIONS
The European Commission's Joint Research Centre has launched a new exploratory research project in order to find innovative solutions and to improve safety and long-term reliability of 241 Am based power sources for future ESA missions into deep space. The project utilizes the unique facilities of JRC for preparing, handling and characterising highly radiating americium compounds. It investigates particular aspects of americium oxide, such as its chemical and selfirradiation stability and looks into promising alternative candidate compounds for americium and new thermoelectric materials containing actinides.