عملکرد تجزیه و تحلیل از سلول تبدیل فلزقلیایی آند چندلوله ای بخار Nb-1Zr/C-103 از حرارتی به برق

The results of performance analyses of a refractory Nb–1Zr/C-103 vapor anode multi-tube alkali-metal thermal-to-electric conversion (AMTEC) cell are presented and discussed. This cell could be used with a radioisotope heater unit to provide electric power from tens to a few hundreds of watts. In the tens of kilowatts electric range, the AMTEC cells could be used with a parabolic solar concentrator or a nuclear reactor heat source. The present cell measures 41.27 mm in diameter and is 125.3 mm high and has eight sodium beta′′-alumina solid electrolyte (BASE) tubes, which are connected electrically in series to provide a load voltage in excess of 3 V. The hot structure of the cell, including the hot plate, the BASE tube support plate, the hot plenum wall and conduction stud, the evaporator standoff and porous wick and the side wall facing the BASE tubes, is made of Nb–1Zr. The cell's colder structure, which includes the condenser structure, the interior thermal radiation shield, the casing and wick of the liquid sodium return artery and the side wall above the BASE tubes, is made of C-103. This niobium alloy is stronger and has a lower thermal conductivity than Nb–1Zr, reducing the parasitic heat conduction losses in the cell wall, hence enhancing the cell's performance. The base cell weighs 163.4 g and delivers 7 We at 17% conversion efficiency and load voltage of 3.3 V (cell specific mass of 23.4 g/We). These performance parameters were for TiN BASE electrodes characterized by B=75 A K1/2/m2 Pa and G=50, assuming a BASE/electrode contact resistance of 0.06 Ω cm2 and a BASE braze structure leakage resistance of 3 Ω. Also, the inner surfaces of the thermal radiation shield and the cell wall above the BASE tubes were covered with low emissivity rhodium. The temperatures of the BASE brazes and the evaporator were below the recommended design limits (1123 and 1023 K, respectively), and the temperature margin was ⩾+20 K to avoid sodium condensation inside the BASE tube, shorting the cell. When high performance electrodes, characterized by B=120 A K1/2/m2 Pa and G=10, were used, the cell's electric power increased to 8.38 We at 3.5 V, and the efficiency increased to 18.8%, decreasing the specific mass of the cell to 19.7 g/We without exceeding any of the design temperature limits.

During the last four to six years, extensive advances have been made in the fabrication, testing and performance of vapor anode multi-tube alkali-metal thermal-to-electric conversion (AMTEC) cells for potential use in space and terrestrial applications. The driving force behind the recent advances in the technology of these AMTEC cells has been to demonstrate the technology readiness for potential space missions with radioisotope heat sources. AMTEC cells have been considered for use in radioisotope power systems for spacecraft scheduled for launch in the year 2003 to explore Jupiter's moon, Europa, and the Pluto-Express (PX) flyby spacecraft to be launched in the year 2005 [1]. The power systems for these missions consist of one or two generators connected electrically in parallel. Each generator uses three to four standard general purpose heat source (GPHS) modules and 8–16 AMTEC cells connected in series in two parallel strings for redundancy. A GPHS module generates 250 and 230 Wth by radioactive decay of 238Pu at the beginning-of-mission (BOM) and end-of-mission (EOM), respectively. Because of the long half life of 238Pu (86 years), the thermal power output of a GPHS module decreases by only 11% at the end of a 10 year space mission. The radioisotope/AMTEC power systems for the Europa and PX spacecraft were designed to provide BOM electric power of 141 We and EOM power of 98.5 and 112 We, respectively [2]. The projected operation lifetime for these missions is seven years and 10–15 years, respectively. Vapor anode multi-tube AMTEC cells could also be used in conjunction with a parabolic solar concentrator or a nuclear reactor heat source to generate tens or even hundreds of kilowatts of electric power. For these power systems, reducing the mass of the AMTEC cells and operating at a high bus voltage of 100–200 V are important considerations for reducing the mass of the power conditioning subsystem and, hence, that of the spacecraft. The relatively high operating voltage of vapor anode multi-tube cells (>3 V) is more than an order of magnitude higher than other static energy conversion options. These include thermoelectric, thermionic, thermophotovoltaic or solar photovoltaic cells. A typical vapor anode multi-tube AMTEC cell comprises six to nine sodium beta′′-alumina solid electrolyte (BASE) tubes connected electrically in series. The cell operates at typical hot and cold side temperatures of 1150–1200 and 550–620 K, respectively. The cell could provide 6–10 We at a load voltage >3 V and a conversion efficiency of 14–18% [3] and [4]. With further advances in technology, the cell's conversion efficiency can potentially reach 25–30% in the next 5–10 years, making vapor anode multi-tube AMTEC cells very attractive for a host of space and terrestrial applications. During the last four years, several stainless steel, PX-series cells (Fig. 1), fabricated by advanced modular power systems (AMPS), have been tested in vacuum at the Air Force Research Laboratory in Albuquerque, NM. A few of these cells were fabricated of nickel and Haynes-25 to improve compatibility with the sodium working fluid and reduce heat conduction losses in the cell wall. Some of the PX-series cells have operated continuously for more than 8000 h without failure [3], [5] and [6]. The results of the tests, however, have indicated some early failures and gradual decreases in the cell performance of more than 30% with operation time. The gradual decreases in performance were attributed to the release and transport of volatile elements from the cell structural materials and the degradation of the TiN electrodes of the BASE tubes (Fig. 1). The test results of the PX-type cells agreed well with predictions of the AMTEC performance and evaluation analysis model (APEAM) [7] and [8]. Full-size image (18 K) Fig. 1. Cutaway and plan views of a PX-type cell. Figure options El-Genk et al. [4] performed parametric analyses of nickel/Haynes-25 AMTEC cells. One of these cells, designated cell D, was 41.25 mm in diameter and 127 mm high and had eight BASE tubes, 0.4 mm thick and 50.8 mm long, connected electrically in series. The hot structure, including the cell wall facing the BASE tubes (∼200 μm thick), the BASE tube support plate, the evaporator stand off, evaporator wick and the interior cylindrical thermal radiation shield, was made of nickel. The thin (∼100 μm thick) cell wall above the BASE tubes and the casing and wick of the liquid sodium return artery were made of Haynes-25 to take advantage of this alloy's low thermal conductivity and high strength. The cell was insulated on the outside by an inch of Min-K, separated from the cell wall by a small gap to minimize heat losses through the wall. When operated at a condenser temperature of 640 K and heat input of 51.2 Wth, the predicted cell efficiency was 19% at 3.5 V. The calculated temperatures of the BASE brazes and the evaporator were 1171 and 1058 K, respectively, and the temperature margin in the cell was +29 K. While the margin is quite sufficient to ensure that no sodium vapor condensation occurs inside the BASE tubes, the temperatures are slightly higher than the recommended design values of 1123 and 1023 K, respectively. This cell, which had a design similar to that shown in Fig. 1, weighed 192 g for a projected specific mass of 19.7 g/We (Table 1) [4]. However, because of concerns regarding the compatibility of sodium with nickel, attention has been focused on replacing the stainless steel, nickel, and Haynes-25 with refractory alloys. Recently, AMPS has designed and conducted performance optimization of Nb–1Zr cells [2]. In this paper, the Nb–1Zr alloy is used only for the hot structure, with the niobium alloy C-103 being used for the cold structure of the cell. The cell's hot structure includes the hot plate, BASE tube support plate, hot plenum wall, conduction stud, evaporator standoff, evaporator wick and the side wall facing the BASE tubes (∼200 μm thick). The cell's cold structure includes the condenser assembly, interior thermal radiation shield, the casing and wick of the liquid sodium return artery and side wall above the BASE tubes (∼100 μm thick). The C-103 alloy is stronger and has a lower thermal conductivity than Nb–1Zr, decreasing the parasitic heat conduction losses in the cell wall and enhancing its structural strength [9]. The AMPS Nb–1Zr cell (designated EPX-1) was designed for meeting the power requirement of the advanced radioisotope power system (ARPS). The cell was 50.5 mm in diameter and 101.6 mm high. It also has one cylindrical and 21 conical thermal radiation shields for decreasing the parasitic heat losses by thermal radiation in the cell [2] and eight BASE tubes, 10.16 mm in diameter and 25.4 mm in active length. When modeled using electrodes characterized by B=120 A K1/2/m2 Pa and G=10, the projected cell power was 8.94 We at 3.5 V and 16.4% efficiency. The values of the BASE/electrode contact resistance and the BASE braze leakage resistance were not reported. When electrodes characterized by B=80 A K1/2/m2 Pa and G=50 were used, the cell power and efficiency at 3.5 V decreased to 8.56 We and 15.7%, respectively [2]. The calculated maximum BASE tube (or braze joint) temperature was well below the 1123 K design limit, while the evaporator temperature, 1061–1063 K, was higher than the recommended value of 1023 K ( Table 1). The estimated total cell mass and specific mass were ∼350 g and ∼39 g/We, respectively. The objective of this paper is to design and investigate the performance of AMTEC cells that have an Nb–1Zr hot structure and a C-103 cold structure (Fig. 2a and b). The effect on the cell performance of applying a low emissivity rhodium coating on the inner surfaces of the side wall above the BASE tube and the thermal radiation shield (Fig. 1 and Fig. 2) is also investigated. In addition, parametric analyses are performed to assess the effect of reducing the thicknesses of the hot plate, BASE tube support plate, conduction studs and the hot plenum on the performance and total mass of the cell. The performance of the Nb–1Zr/C-103 cell is compared with that of a nickel/Haynes-25 cell of similar design and dimensions to compare the cell's specific mass. Full-size image (38 K) Fig. 2. (a) An enlarged trimetric view of the present Nb–1Zr/C-103 cell. (b) A trimetric view of the conduction stud and the hot plenum of the present Nb–1Zr/C-103 cell. Figure options The present analysis is conducted using the APEAM, a cell model developed at the University of New Mexico's Institute for Space and Nuclear Power Studies. This comprehensive model has been benchmarked successfully against PX-type cell test results generated at the Air Force Research Laboratory in Albuquerque, NM [7] and [8]. APEAM has been used to perform parametric and performance analyses of vapor anode multi-tube AMTEC cells without requiring long computation time. A typical single case run of APEAM on a Pentium class PC takes only minutes of CPU time. The interactively coupled building blocks in APEAM are (a) a vapor flow model [10], (b) a thermal radiation model [11], (c) an electrochemical model for the expansion of sodium vapor through the base tubes [12], (d) an electric model, including the BASE electrodes and electric current collectors on the cathode and the anode sides of the BASE tube [12], (e) a liquid sodium return artery and evaporation wick model [13], (f) an overall momentum and energy balance model of the entire cell.

The performance of a number of Nb–1Zr/C-103 multi-tube vapor anode AMTEC cells is analyzed. The cell's hot structure (hot plate, BASE tubes support plate, hot plenum wall, conduction stud, evaporator standoff, evaporator wick and side wall facing the BASE tubes) is made of Nb–1Zr. However, the cold structure of the cell (condenser, interior cylindrical thermal radiation shield, the casing and wick of the liquid sodium return artery and condenser and the side wall above the BASE tubes) is made of the niobium alloy C-103. This alloy is stronger and has lower thermal conductivity than that of Nb–1Zr for maintaining the structural strength of the cell's cold wall (∼100 μm thick) and decreasing the parasitic heat conduction losses to the condenser. The performance of a nickel/Haynes-25 cell of similar design is also evaluated and compared with that of Nb–1Zr/C-103 cells. The performance contours and operation envelopes of the present nickel/Haynes-25 and Nb–1Zr/C-103 cells are calculated and compared. The cell's performance envelope is bounded by the isotherms corresponding to the recommended design temperature limits for the BASE braze joints and the evaporator (1123 and 1023 K, respectively) and the temperature margin ΔT=+20 K. The results indicate that the performance of the Nb–1Zr/C-103 cell is not only superior to the nickel/Haynes-25 cell, but also more compatible with the environment within an operating AMTEC cell. The mass of the Nb–1Zr/C-103 reference cell (194 g) is only 2 g heavier than the nickel/Haynes-25 cell. Because of its high corrosion rate and volatility, nickel should be excluded as a structural material in vapor anode AMTEC cells for space or terrestrial applications. The zirconium and hafnium in the niobium alloys in the present refractory cell stabilize oxygen, enhancing the corrosion resistance of the alloys. Therefore, the selected Nb–1Zr and C-103 alloys should be compatible with the sodium working fluid at the typical hot side temperatures of 1150–1200 K in vapor anode AMTEC cells. When the surfaces of the thermal radiation shield, casing of sodium return artery and cell wall above the BASE tubes were not covered with rhodium the calculated cell's specific mass was 28.5, and the total mass was 194 g. This specific mass is calculated at the cell's maximum load voltage within the operation envelope. Covering the surface of the C-103 wall, the casing of the liquid sodium return artery and the thermal radiation shield with a low emissivity rhodium coating improved the cell's performance. The maximum load voltage within the cell's operation envelope increased from 3.35 to 3.65 V, the electric power remained unchanged at 6.8 We, but the cell efficiency increased from 15.8% to 17.3%. These performance parameters are calculated with BASE electrode characteristics of B=75 A K1/2/m2 Pa and G=50. The mass of the Nb–1Zr/C-103 cell is decreased by ∼15.77% (or 30.6 g), by reducing the thicknesses of the structures in the cell's hot plenum, to 163.4 g, with little change in performance. The calculated maximum electric power of this cell was 7.2 We at a load voltage of 2.8 V and an efficiency of 16.2%. The corresponding specific mass of the cell is 22.7 g/We. For a load voltage of 3.5 V, the cell's maximum cell power within the operation envelope is 6.89 We, and the corresponding efficiency is 17.3%. When BASE electrodes having characteristics of B=120 A K1/2/m2 Pa and G=10 are used, the performance of the low-mass Nb–Zr/C-103 cell is improved significantly. The cell's maximum electric power within the operation envelope increased by 15% to 8.3 We, the cell efficiency increased by 1.10 percentage points to 17.3% and the load voltage increased by 7.14% to 3 V. At this performance level, the cell specific mass decreased by 15% to 19.7 g/We. For a load voltage of 3.5 V, the cell's maximum electric power within the performance envelope increased by 13% from 6.89 to 7.97 We, and the cell efficiency increased by 6.35% from 17.3% to 18.4%. The cell's performance parameters at a load voltage of 3.5 V is important to integration of AMTEC cells in the radioisotope power systems for Europa and Pluto. At the current bus voltage of 28 V, exactly eight of the 3.5 V AMTEC cells would be connected in series in each of the parallel strings of the power system. Future spacecraft buses, however, are being designed to operate at much higher voltage (120–250 V) for reducing the electric energy losses in the connecting cables and the total mass of the power conditioning subsystem. The mass of the current power conditioning subsystem represents ∼30–40% of the total mass of the space electric power system.