Conceptual Design of a Space Power System Based on Combustion of Metals

Open AccessTechnical NotesConceptual Design of a Space Power System Based on Combustion of MetalsEvgeny ShafirovichEvgeny Shafirovich https://orcid.org/0000-0002-3632-2349The University of Texas at El Paso, El Paso, Texas 79968*Professor, Department of Aerospace and Mechanical Engineering. Associate Fellow AIAA.Search for more papers by this authorPublished Online:2 Jan 2023https://doi.org/10.2514/1.B39008SectionsRead Now ToolsAdd to favoritesDownload citationTrack citations ShareShare onFacebookTwitterLinked InRedditEmail AboutI. IntroductionCurrently, there is a growing interest in the so-called chemical heat integrated power systems [1], which are based on exothermic chemical reactions between solid or liquid reactants. Such systems could provide heat and electric power in space missions where the use of sunlight and nuclear energy is impractical and batteries cannot serve as the primary energy source because of their low specific energy, limited storage time, and a relatively narrow temperature range.One approach to the development of the chemical heat integrated power systems involves combustion of lithium with sulfur hexafluoride (SF6). These reactants were used to generate heat for the Rankine power cycle in underwater propulsion systems through a reaction that does not produce any gas products [2]. The SF6 was stored as a saturated liquid (the saturated vapor pressure was 21 bar at 20°C), whereas the lithium (Li) was stored as a solid, which was melted before the operation (the melting point was 180°C). In the Li-SF6 system, the reaction can occur either on the lithium surface or in the gas phase. For the latter, the reactor can include a wick for capillary flow of the molten lithium toward the gas-phase reaction zone [2]. In any case, the reaction produces lithium fluoride (LIF) and lithium sulfide (LI2S): 8Li+SF6→6LiF+Li2S(1)where the LiF and Li2S are liquid (their melting points are 848 and 938°C, respectively). Because they are immiscible with molten lithium and several times denser, they sink, and hence do not inhibit the reaction.Recently, the Li-SF6 system has been proposed to be used as the energy source in missions to Venus [3] and Europa [4]. For Venus missions, it has also been proposed to use in situ CO2 as the oxidizer with Li, magnesium (Mg), or magnesium/aluminum and magnesium/zinc eutectic alloys as the metal fuels [3]. The calculations have shown that both Li-SF6 and Li-CO2 combustors, coupled with Stirling engines, would have superior specific energies (per unit mass of the entire power system) as compared to batteries [3,4]. However, testing the Li-CO2 reactors has revealed a low combustion efficiency due to accumulation of condensed products on the metal–gas interface [5].The Li-SF6 system has a rather high specific energy: 3.9 (kW⋅h)/kg (14.1 MJ/kg) [4]. For comparison, the state-of-the-art primary battery technology provides 0.2–0.3 ( kW⋅h)/kg (0.7–1.1 MJ/kg) [1]. Of course, only part of the generated heat can be converted to electricity, and the masses of the SF6 tank, the reactor, and the energy converter should be accounted for. However, the produced electric power per unit mass of the system may still be rather high, and the unconverted part of the generated heat can be useful in missions to “cold worlds,” such as Europa or the moon’s polar regions. In addition, the system has a long storage life. The successful operation of Li-SF6 combustors in underwater propulsion systems and (on the other hand) the low combustion efficiency of Li-CO2 reactors demonstrate the importance of sinking the condensed products in molten lithium to prevent their inhibiting effect on the combustion. Apparently, this mechanism cannot take place in microgravity or if the gravity is too low.Here, we evaluate an alternative approach, which is based on the combustion of metals with oxygen and is apparently independent of gravity. Specifically, a cylindrical reactor is filled with a combustible metal powder, and gaseous oxygen is fed into the reactor from one end. The combustion wave propagation can be initiated by ignition at either the same end or the opposite one. The former leads to the coflow combustion, where the oxygen infiltrates through the products; whereas the latter leads to the counterflow regime, where the oxygen infiltrates through the initial powder. For oxygen storage, we propose to use chemical oxygen generators (COGs), such as those used in aircraft, submarines, and space stations. They usually contain sodium chlorate (NaClO3) or lithium perchlorate (LiClO4) as the oxygen source and small amounts of a metal fuel (typically iron), a catalyst, and other ingredients [6].In the present Note, we estimate the energetic and mass characteristics of the heat-generating systems that include the metal combustor, described earlier in this Note, and a COG. We propose a design of a power system that includes heat-generating cartridges and estimate the required masses for two space missions. Based on the recent experimental studies of Mg and Li combustion in oxygen, we make design recommendations for the metal combustor. Finally, we evaluate the feasibilities of using in situ CO2 for heat/power generation in missions to Mars and Venus.II. Results and DiscussionA. Energetic and Mass CharacteristicsIt is important to identify metal fuels that could be used in a reactor where oxygen infiltrates through the powder. Obviously, for its effective operation, especially in the coflow combustion regime, the formed metal oxide should occupy less space than the used metal did; i.e., the metal fuel should have a Pilling–Bedworth ratio lower than 1. Aluminum, although advantageous energetically, has a Pilling–Bedworth ratio of 1.28, which makes its use problematic. For lithium and magnesium, this ratio is equal to 0.57 and 0.81, respectively, and so they are suitable.The reactions of Li and Mg with oxygen are described by the equations 4Li+O2→2Li2O(2)2Mg+O2→2MgO(3)The specific energies of Li-O2 and Mg-O2 systems, determined based on the enthalpies of formation at standard conditions, are 20.0 and 14.9 MJ/kg, respectively, i.e., higher than for the Li-SF6 system (the enthalpies of formation for these and other estimates were taken from Ref. [7]). However, these values include the masses of the reactants only. For the comparison of the power systems, it is necessary to know the masses of all components, such as the oxygen storage unit, the combustor (which also serves for storage of the metal fuel), the heat transfer system, and the energy converter (e.g., a Stirling engine).The effect of using a COG on the specific energy of the system was estimated by assuming that the oxygen-generating mixture contains 90 wt % LiClO4 and the oxidation of its metal ingredient releases the same heat as the decomposition of LiClO4 does (the latter assumption was based on a prior study of sodium chlorate/metal mixtures [8]). The integral reactions of the entire power systems (in both the chemical oxygen generator and the metal combustor) were assumed to be described by the equations 8Li+LiClO4→4Li2O+LiCl(4)4Mg+LiClO4→4MgO+LiCl(5)At these assumptions, the specific energies of the Li- and Mg-based systems are 16.0 and 12.9 MJ/kg, respectively, where the mass includes both the metal fuel and the oxygen-generating mixture.When power needs to be generated for a relatively long time (e.g., 14 Earth days in a lunar night survival mission), the design of the power system can include several cartridges that contain a COG and a metal combustor. The cartridges would be used consecutively. Figure 1 shows a schematic diagram of such a system, where a Stirling engine is used for the conversion of the generated heat into electric power.Table 1 shows the estimated parameters of the power systems with heat-generating cartridges for two missions. In both missions, part of the generated heat would be converted to electric power, whereas the remaining heat would be used to maintain the desired temperature in the lander or spacecraft. In mission A, the cartridges would generate 157 MJ of heat, which correspond to the total energy requirement of the lunar night survival mission [1]: 130 W for 14 days. In mission B, the cartridges would generate 2592 MJ, which correspond to a constant heat rate of 1500 W for 20 days. The parameters of commercially available COGs for aircraft and submarines were used in these estimates. Mission A uses a small (0.45 kg) COG for aircraft, whereas mission B uses a much larger (12 kg) COG for submarines (Molecular Products, MPOG®). In Table 1, the oxygen capacities of the COGs are shown at normal pressure and temperature (1 atm and 20°C). Each mission was analyzed for two fuels: Li and Mg. Based on the oxygen capacity, the mass of fuel and the chemical energy contained in one cartridge were calculated according to the stoichiometry of the reactions described by Eqs. (2) and (3). Next, for the total energy required and the energy of one cartridge, the number of cartridges was determined; and the actual energy generated was calculated by multiplying the energy of one cartridge and the number of cartridges. The heat releases of the COGs (approximately 2.5 MJ for MPOG according to the manufacturer) were not taken into account. The time between the two starts of the operation was determined for the total duration and the obtained number of cartridges.Fig. 1 Schematic diagram of the power system with heat-generating cartridges for supplying electric power (black arrow) and heat (red arrows) to a lander or spacecraft.It is seen that mission A would require 40 COGs with an oxygen capacity of 80 liters, whereas mission B would need 20 COGs with an O2 capacity of 2600 liters. The required mass of Mg is larger than the mass of Li by 75%. However, the densities of solid Li and Mg are 534 and 1738 kg/m3, respectively; i.e., Mg is 3.25 times denser than Li. Thus, Mg would occupy a smaller volume. It should be noted that Table 1 does not include the total mass of the metal combustor because it is hard to estimate the mass of its structure at this stage. It is also worth mentioning that the masses of the COGs and metal fuel (Li or Mg) in mission A are comparable with the estimates for the SF6 tank (15.84 kg) and Li fuel (4.58 kg) for such a mission [1].Note that for space power systems, COGs with higher oxygen contents could be developed. For example, COGs for aircraft are designed for the oxygen flow rate that follows the altitude profile of the descending (after decompression) airplane, which is unnecessary in the proposed application. Furthermore, both considered COGs use NaClO3 as the source of oxygen, but LiClO4 contains more oxygen per unit mass (60 wt % vs 45 wt %), and it has been used in the backup oxygen generators for space stations [9]. Finally, in a thorough design, the heat released by the COG could be added to the heat released in the metal combustor.It should also be noted that the performed calculations assumed 100% combustion efficiency, i.e., full oxidation of the metal fuel during the combustion. The actual combustion efficiency should be determined experimentally. The next section presents the results of laboratory experiments with small samples of Mg and Li powders ignited in a closed chamber filled with oxygen gas. The goal of these experiments was to verify that a self-sustained combustion wave can propagate over the metal powder bed and result in a high extent of metal-into-oxide conversion. The experimental results are discussed with the respect to the proposed design of the power system.B. Experimental Verification and Recommendations for the Combustor DesignThe combustion of metal powders with infiltration of the reactive gas has taken place during the self-propagating high-temperature synthesis of nitrides and hydrides [10,11]. The synthesis was usually conducted at pressures over 1000 bar in order to have a sufficient amount of the gaseous reactant in the pores. In the proposed combustors, the gas pressure will be much lower, which requires a rapid infiltration of oxygen through either the initial metal powder or the formed metal oxide.Recently, the combustion of Mg and Li powders at the natural infiltration of oxygen has been studied experimentally [12,13]. The natural infiltration means that there is no pressurized (forced) feed of oxygen (i.e., the powder sample is located in a quiescent gas environment), and the oxygen flow through the powder is caused by the pressure difference between the environment and the reaction zone. The powder sample in a vertical quartz tube with oxygen access from both ends was ignited at the top by a laser beam inside a vacuum chamber filled with oxygen at atmospheric or lower pressure. For both Mg and Li, a counterflow combustion wave propagated downward owing to the infiltration of O2 from the open bottom end to the reaction zone. The conversion of both powders in the downward combustion wave was incomplete. For Mg, this led to a second ignition at the bottom and a coflow propagation of the combustion wave up. For Li, the reaction typically continued over the entire sample through the upward propagation of the coflow combustion wave was observed in some tests. For both metals, the conversion was still incomplete after the entire process: up to 74% for Mg [12] and 63±10% for Li [13]. It is worth mentioning that the extent of conversion was higher at lower velocities of the coflow combustion wave [12].The findings [12,13] allow for recommendations to be made for the development of combustors based on the forced infiltration of oxygen through a bed of Mg or Li particles. The coflow combustion scheme is preferred because it can operate at a lower porosity of the metal powder. The ignition energy and duration should be thoroughly determined for the combustor to prevent sintering of the products and ensure the effective infiltration of oxygen. The optimal configuration of the reactor may include two open ends. This configuration may result in counterflow propagation of the combustion wave at incomplete conversion followed by self-ignition at the other end and coflow propagation of a second combustion wave backward. It is expected that the use of forced infiltration in the power systems will increase the combustion efficiencies of Mg and Li powders.For a successful implementation of the proposed concept, the velocity of the combustion wave in the reactor should be relatively low. Indeed, an increase in the front velocity at the same length of the reactor will shorten the time of propagation and increase the required rate of heat removal, which is undesirable. It is remarkable that the axial velocity of the front in the experiments [12,13] was rather low: on the order of 0.1 mm/s. Such low velocities are rarely seen in combustion systems, but they are typical for COGs [8]. It is expected that in the reactor with forced infiltration of oxygen, the velocity of the combustion front can be of the same order. This allows for the reactor dimensions to be estimated. Such estimates were made for the Mg-COG cartridges described earlier, assuming for simplicity complete combustion in a single combustion wave with a velocity of 0.1 mm/s. The combustion duration was assumed to be equal to the operation time of the used COG, which is typically around 20 min for the aircraft COGs and 75 min for the MPOG. For the Mg masses shown in Table 1 and 71% porosity of the powder (the average porosity of Mg power in Ref. [12]), the cylindrical combustors have the following dimensions: diameter of 5.8 cm and length of 12 cm in mission A, and diameter of 17.2 cm and length of 45 cm in mission B. These dimensions appear to be reasonable, and they are close to those of commercial chemical oxygen generators. Many COGs used in aircraft have a diameter of around 5 cm and a length of 10–12 cm, and the MPOG’s dimensions are 13.3×13.3×40 cm. These results are favorable for the metal combustor design.One additional advantage of the system where a chemical oxygen generator is coupled with a filtration combustion reactor is its potentially autocontrolled operation. During a steady mode, the oxygen generation rate is equal to the oxygen consumption rate in the metal combustor. In the coflow regime, if gas pressure at the inlet of the combustor is constant, the propagation of the reaction front will decrease the pressure gradient, and hence the infiltration rate and the oxygen consumption rate. However, the oxygen generation rate is essentially independent of the pressure in the chemical oxygen generator. Therefore, if the oxygen consumption decreases, the pressure will increase, which in turn will increase the pressure gradient and the infiltration rate; i.e., the system will be stabilized. Similarly, in the counterflow regime, the acceleration of the combustion front, normally observed with approaching the sample end, will be hindered by decreasing pressure due to the faster consumption of oxygen.The challenge in the proposed concept is the relatively short time of O2 generation. For example, as mentioned earlier, it is around 20 min for the aircraft COGs and 75 min for the MPOG. Because the metal combustor should be designed for the same duration of the process, it is necessary to transfer heat from the combustor at a rather high rate. Assuming 20 min for the combustor in mission A (where the aircraft COGs are used) and 75 min for that in mission B (the MPOGs are used), the heat transfer rates are 3.3 and 29 kW, respectively. One option for cooling the combustor at such high rates is the use of an alkali metal, such as sodium or sodium- potassium eutectic, as a coolant and feeding it with an electromagnetic pump. This option has been used, for example, in the TOPAZ-II space nuclear reactor, which generated a thermal power of 115 kW [14]. Another option is to use heat pipes with an alkali metal as the working fluid. This option is attractive because no pump is needed. Heat pipes with sodium as the working fluid for space nuclear reactors that generate thermal power in the range of 4 to 40 kW have been developed and tested [15]. If an effective system for heat transfer from the cartridge can be built, significant advantages of the proposed concept over the Li-SF6 system will be realized, such as no need for gravity and for melting the metal fuel before the ignition. The metal fuel in the cartridge could be ignited by placing a small amount of a pyrophoric powder near the oxygen inlet.Table 1 shows that the mass of COGs is several times larger than the mass of Mg or Li fuel. Therefore, it is expected that the use of an in situ oxidizer can significantly reduce the total mass of the system, provided that its energetic and reactive characteristics are not much worse than those of oxygen. In the next section, the use of in situ CO2 is analyzed.C. Use of In Situ CO2In missions to Mars and Venus, it would be attractive to improve the performance characteristics of the proposed power system by adding in situ CO2 to the oxygen flow. The potential improvement can be evaluated by thermochemical calculations. Combustion of Mg in CO2 produces MgO, CO, and C via parallel reactions: Mg+CO2→MgO+CO(6)2Mg+CO2→2MgO+C(7)where the CO/C ratio depends on the process conditions such as the temperature and the availability of a sufficient amount of CO2 for complete oxidation of Mg via the reaction described by Eq. (6) [16–18]. The specific energies of these reactions are 13.1 and 16.6 MJ/kg (where the mass includes only Mg), respectively; i.e., they slightly exceed the estimated specific energy of the Mg-COG system: 12.9 MJ/kg.Combustion of Li in CO2 leads to the formation of lithium carbonate (Li2CO3), CO, and C, where the CO/C ratio is again dependent on the process conditions [5,13,19]. Therefore, the overall reaction can be described as a certain combination of two parallel reactions [13]: 2Li+2CO2→Li2CO3+CO(8)4Li+3CO2→2Li2CO3+C(9)The specific energies of these reactions are 38.9 and 45.1 MJ/kg (where the mass includes only Li), respectively; i.e., both are much higher than the estimated specific energy of the Li-COG system: 16.0 MJ/kg.Based on these considerations, Li is clearly advantageous over Mg in the power systems that use in situ CO2. Unfortunately, in the experiments on combustion of Li and Mg powders at natural infiltration of CO2, a self-sustained propagation of the combustion wave was not achieved: apparently because the formed products hindered transport of CO2 [13]. It is possible that forced filtration will help overcome this problem. The metal-CO2 combustion could also be organized in the configurations where the products may not inhibit the process. For example, combustion of powdered Mg with CO2 in laboratory-scale rocket engines has been demonstrated [20–25]. Combustion of Mg and Li powders mixed with solid CO2 could be studied for applications to power systems for Mars missions.III. ConclusionsLithium and magnesium are promising fuels for the power system where heat is generated by combustion of a metal powder with oxygen supplied by a chemical oxygen generator. Thermochemical estimates have shown that the masses of metal (Mg or Li) fuel and chemical oxygen generators are comparable with those of Li fuel and the SF6 tank in the power system that has been proposed recently for the lunar night survival. The design with multiple heat-generating cartridges could be used to provide heat and power for durations that are much longer than the combustion process, such as lunar night.Based on recent combustion experiments with lithium and magnesium powders, recommendations have been made for the design of the metal combustor. Because the ignition of the top layer resulted in sintering of the products, the optimal configuration may include two opposite inlets of the oxidizer. In this case, the ignition may trigger a counterflow combustion wave followed by backward propagation of a coflow combustion wave.Thermochemical estimates have shown that the use of in situ CO2 would increase by 2.4–2.8 times the specific energy of the system where lithium is used as the metal fuel. The potential increase is much smaller in the case of magnesium (by 1.0–1.3 times).J. RoveyAssociate EditorAcknowledgmentsThe material presented in this work is based upon the work supported by National Aeronautics and Space Administration under grant no. 80NSSC20K0293. The author thanks Sergio Cordova (currently at Northrop Grumman) and Steven L. Rickman of NASA Engineering and Safety Center for helpful discussions. References [1] Hendricks T. J., Brandon E. J., Lam R. L., Peterson D. E., Anderson K. R. and Carroll B. C., “Chemical Heat Integrated Power Systems (CHIPS) to Survive Lunar Night Environments,” 2020 Conference on Advanced Power Systems for Deep Space Exploration, Jet Propulsion Lab., Pasadena, CA, Oct. 2020, https://trs.jpl.nasa.gov/handle/2014/53758 Google Scholar[2] Hughes T. G., Smith R. B. and Kiely D. H., “Stored Chemical Energy Propulsion System for Underwater Applications,” Journal of Energy, Vol. 7, No. 2, 1983, pp. 128–133. https://doi.org/10.2514/3.62644 LinkGoogle Scholar[3] Miller T. F., Paul M. V. and Oleson S. R., “Combustion-Based Power Source for Venus Surface Missions,” Acta Astronautica, Vol. 127, Oct.–Nov. 2016, pp. 197–208. https://doi.org/10.1016/j.actaastro.2016.05.006 CrossrefGoogle Scholar[4] Greer C. J., Paul M. V. and Rattner A. S., “Analysis of Lithium-Combustion Power Systems for Extreme Environment Spacecraft,” Acta Astronautica, Vol. 151, Oct. 2018, pp. 68–79. https://doi.org/10.1016/j.actaastro.2018.05.039 CrossrefGoogle Scholar[5] Greer C. J., Peters J. A., Manahan M. P., Cor J. J. and Rattner A. S., “Experimental Characterization of Lithium-Carbon Dioxide Combustion in Batch Reactors for Powering Venus Landers,” Acta Astronautica, Vol. 181, April 2021, pp. 235–248. https://doi.org/10.1016/j.actaastro.2021.01.010 CrossrefGoogle Scholar[6] Mausteller J. W., “Oxygen-Generation Systems,” Kirk-Othmer Encyclopedia of Chemical Technology, Wiley, New York, 2000, pp. 1–12. https://doi.org/10.1002/0471238961.1524250713012119.a01 Google Scholar[7] Chase M.W., “NIST-JANAF Thermochemical Tables, Fourth Edition,” Journal of Physical and Chemical Reference Data, Monograph 9, 1998, pp. 1–1951, https://srd.nist.gov/JPCRD/jpcrdM9.pdf. Google Scholar[8] Shafirovich E., Mukasyan A. S., Varma A., Kshirsagar G., Zhang Y. and Cannon J. C., “Mechanism of Combustion in Low-Exothermic Mixtures of Sodium Chlorate and Metal Fuel,” Combustion and Flame, Vol. 128, Nos. 1–2, 2002, pp. 133–144. https://doi.org/10.1016/S0010-2180(01)00340-6 Google Scholar[9] Graf J., Dunlap C., Haas J., Weislogel M., Lewis J., Meyers K. and McKernan A., “Development of a Solid Chlorate Backup Oxygen Delivery System for the International Space Station,” 30th International Conference on Environmental Systems, SAE TP 2000-01-2348, Warrendale, PA, 2000. https://doi.org/10.4271/2000-01-2348 Google Scholar[10] Merzhanov A. G., “Historical Retrospective of SHS: An Autoreview,” International Journal of Self-Propagating High-Temperature Synthesis, Vol. 17, No. 4, 2008, pp. 242–265. https://doi.org/10.3103/S1061386208040079 Google Scholar[11] Varma A., Rogachev A. S., Mukasyan A. S. and Hwang S., “Combustion Synthesis of Advanced Materials: Principles and Applications,” Advances in Chemical Engineering, Vol. 24, 1998, pp. 79–226. https://doi.org/10.1016/S0065-2377(08)60093-9 CrossrefGoogle Scholar[12] Cordova S., Estala-Rodriguez K. and Shafirovich E., “Infiltration-Controlled Combustion of Magnesium for Power Generation in Space,” Combustion and Flame, Vol. 238, April 2022, Paper 111950. https://doi.org/10.1016/j.combustflame.2021.111950 Google Scholar[13] Estala-Rodriguez K., Cordova S. and Shafirovich E., “Oxidation and Combustion of Stabilized Lithium Metal Powder (SLMP),” Proceedings of the Combustion Institute, advance online article, 17 Aug. 2022. https://doi.org/10.1016/j.proci.2022.07.051 Google Scholar[14] Zhang W., Zhang D., Tian W., Qiu S. and Su G. H., “Thermal-Hydraulic Analysis of the Improved TOPAZ-II Power System Using a Heat Pipe Radiator,” Nuclear Engineering and Design, Vol. 307, Oct. 2016, pp. 218–233. https://doi.org/10.1016/j.nucengdes.2016.07.020 CrossrefGoogle Scholar[15] Gibson M. A., Poston D. I., McClure P. R., Sanzi J. L., Godfroy T. J., Briggs M. H., Wilson S. D., Schifer N. A., Chaiken M. F. and Lugasy N., “Heat Transport and Power Conversion of the Kilopower Reactor Test,” Nuclear Technology, Vol. 206, No. 1, 2020, pp. 31–42. https://doi.org/10.1080/00295450.2019.1709364 Google Scholar[16] Shafirovich E. Y. and Goldshleger U. I., “Combustion of Magnesium Particles in CO2/CO Mixtures,” Combustion Science and Technology, Vol. 84, Nos. 1–6, 1992, pp. 33–43. https://doi.org/10.1080/00102209208951843 Google Scholar[17] Shafirovich E. Y., Shiryaev A. A. and Goldshleger U. I., “Magnesium and Carbon Dioxide: A Rocket Propellant for Mars Missions,” Journal of Propulsion and Power, Vol. 9, No. 2, 1993, pp. 197–203. https://doi.org/10.2514/3.23609 LinkGoogle Scholar[18] Shafirovich E. and Varma A., “Metal-CO2 Propulsion for Mars Missions: Current Status and Opportunities,” Journal of Propulsion and Power, Vol. 24, No. 3, 2008, pp. 385–394. https://doi.org/10.2514/1.32635 LinkGoogle Scholar[19] Fischer P., Schiemann M., Scherer V., Maas P., Schmid G. and Taroata D., “Experimental Characterization of the Combustion of Single Lithium Particles with CO2,” Fuel, Vol. 153, Aug. 2015, pp. 90–101. https://doi.org/10.1016/j.fuel.2015.02.098 CrossrefGoogle Scholar[20] Wickman J. H., “In-Situ Mars Rocket and Jet Engines Burning Carbon Dioxide,” AIAA Paper 1999-2409, June 1999. https://doi.org/10.2514/6.1999-2409 Google Scholar[21] Zubrin R., Muscatello T., Birnbaum B., Caviezel K. M., Snyder G. and Bergren M., “Progress in Mars ISRU Technology,” AIAA Paper 2002-0461, Jan. 2002. https://doi.org/10.2514/6.2002-461 Google Scholar[22] Szabo J., Miller T., Herr J. and Peters J., “Metal Bipropellant Rockets for Martian Ascent Vehicles,” AIAA Paper 2011-5834, July–Aug. 2011. https://doi.org/10.2514/6.2011-5834 Google Scholar[23] Li C., Hu C., Zhu X., Hu J., Li Y. and Hu X., “Experimental Study on the Thrust Modulation Performance of Powdered Magnesium and CO2 Bipropellant Engine,” Acta Astronautica, Vol. 147, June 2018, pp. 403–411. https://doi.org/10.1016/j.actaastro.2018.03.029 CrossrefGoogle Scholar[24] Li Y., Hu C., Zhu X., Hu J., Hu X., Li C. and Cai Y., “Experimental Study on Combustion Characteristics of Powder Magnesium and Carbon Dioxide in Rocket Engine,” Acta Astronautica, Vol. 155, Feb. 2019, pp. 334–349. https://doi.org/10.1016/j.actaastro.2018.11.006 CrossrefGoogle Scholar[25] Hu J., Hu C., Che Y., Zhu X., Yang J., Li M. and Li C., “Experimental Study on the Working Performance of Powdered Magnesium and Liquid Carbon Dioxide Bipropellant Rocket Engine for Mars Missions,” Acta Astronautica, Vol. 184, July 2021, pp. 274–285. https://doi.org/10.1016/j.actaastro.2021.04.025 CrossrefGoogle ScholarTablesTable 1 Energetic and mass characteristics of power systems with heat-generating cartridges Mission A; LiMission A; MgMission B; LiMission B; MgTotal energy required, MJ15715725922592Duration, days14142020Mass of COG, kg0.450.451212O2 capacity of COG, L80802,6002,600Mass of oxygen in one COG, kg0.10640.10643.4593.459Mass of fuel in one cartridge, kg0.09230.16163.0015.253Energy of one cartridge, MJ3.9833.998129.4130.0Number of cartridges40402020Total energy generated, MJ15916025882600Time between two starts, h8.48.42424Mass of all COGs, kg1818240240Mass of fuel in all cartridges, kg3.6926.46460.02105.06 Previous article FiguresReferencesRelatedDetails What's Popular Volume 39, Number 3May 2023 CrossmarkInformationCopyright © 2022 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. All requests for copying and permission to reprint should be submitted to CCC at www.copyright.com; employ the eISSN 1533-3876 to initiate your request. See also AIAA Rights and Permissions www.aiaa.org/randp. TopicsCombustionCombustion ChambersCombustorsElectric PowerEnergyEnergy ConversionEnergy FormsEnergy Forms, Production and ConversionEnergy ProductionFuelsHeat EnginesPropulsion and PowerRocketryThermophysics and Heat Transfer KeywordsPower SystemCombustor DesignMetal FuelsSpace MissionsChemical EnergyCombustion EfficiencyElectric PowerStirling EnginesHeat TransferAcknowledgmentsThe material presented in this work is based upon the work supported by National Aeronautics and Space Administration under grant no. 80NSSC20K0293. The author thanks Sergio Cordova (currently at Northrop Grumman) and Steven L. Rickman of NASA Engineering and Safety Center for helpful discussions.PDF Received31 August 2022Accepted12 December 2022Published online2 January 2023