A more recent, higher power design specifically for lunar base power is the multi-megawatt system designed by Panchyshyn, Presentin, and Trueblood at the University of Washington [Pan86]. This design uses an efficient falling-drop radiator and is baselined at 10MWt, 3MWe power output.

This reactor is an inert-gas cooled system, using a Brayton cycle convertor to make best use of the heat produced. Mass balance is some 18,100kg for the reactor and shield; 20,420 kg for the radiators; 12,221 kg for the Brayton convertors and generators; and 615 kg for miscellaneous structure. The 3.366 MWe output power divided by the total mass of 51,356 kg gives a specific power density of 66 We/kg. Operating life of the reactor is ten years, and it is designed to run without human intervention for that time.

Fission reactors are complex machines to build, requiring a large industrial infrastructure with access to a variety of specialized materials, not limited to the fissionable fuels. There are no known ore-grade uranium deposits on the Moon, and no infrastructure to create and machine the reactor components. Hence a reactor-based power system can neither be expanded nor refueled from indigenous lunar materials. However, the large total power output for the initial mass investment offsets this. Nuclear reactors can run continuously, providing power throughout the lunar day and night.

Decay

There are two methods of extracting useful energy from material undergoing radioactive decay: conversion of the produced heat, usually by thermoelectric means; and direct conversion of emitted beta particles by solid state electronics.

The thermoelectric conversion technique has been widely used to power spacecraft (the Pioneer, Voyager and Galileo probes) and for lunar applications. The Apollo Lunar Scientific Experiment Packages (ALSEP) left by the later Apollo missions were all powered by radioisotope thermal generators (RTGs) using plutonium-238.

Taking the ALSEP unit as a Moon-proven example, it was designated the SNAP-27 and built by General Electric. Fueled by Pu-238 it was designed to produce about 60 watts continuously for over a year, with a power density of about 3.3 W/kg. [Hal69, pg. 132]

These systems provide a very reliable power source with reasonable p/w ratio. The power output does reduce over time as the isotope decays, typically producing usable power for 10 years. The instantaneous power output is not great, however, limiting the applications. Further, refueling an exhausted RTG is likely a messy process, although disposal of radioactive wastes is much less a problem on the Moon than on Earth.

Fusion

No practical fusion power sources yet exist, so the following discussion is hypothetical.

The lunar regolith contains small but significant quantities of helium-3 (3He), a rare light isotope of helium, implanted by the solar wind. This has long been proposed as a fusion fuel (in combination with, for example, deuterium) because the reaction process does not involve the emission of neutrons (which can cause induced radioactivity in the reactor walls, and require shielding). Indeed, 3He has been suggested as an economic export for a lunar base, to fuel such fusion reactors on Earth [Wit87]. Given the existence of such reactors, exporting the helium may be more attractive than using it on the Moon for fuel.

Other hot fusion processes require deuterium and/or tritium, neither of which is particularly abundant on the Moon but may be obtainable, as is 3He, from the lunar regolith. Even if that is not the case, only very small quantities would need to be imported. However, the technology for hot fusion reactors has not yet reached the point where they are practical power sources.

The existence of solid-state or so-called cold fusion is still being investigated. If this phenomenon does occur at useful power levels, it could vastly simplify fusion reactor construction. Lunar sources of deuterium, the apparent solid-state fusion fuel, are scarce, but again the material is light and imports are feasible. Such fusion reactions appear to require a host metal such as palladium or titanium. The lunar regolith is relatively rich in titanium (as ilmenite, TiFeO3, as much as 15%-20% by volume of some mare basalts [McK79]). Palladium is generally found associated with platinum group metals, which are known to occur in relatively high abundance in metallic meteorites (siderites). We can assume that meteoric deposits of these metals may be found on Luna.

Solar

Solar irradiation on the Moon is about 1.3 kW per square meter, unimpeded by atmosphere or weather. However, over most of the Moon the sun is only visible for about two weeks out of every four. We will discuss here methods of generating electrical power from sunlight, and then examine some ways of storing power for lunar nighttime use.

Photovoltaic

Photovoltaic (PV) cells, or solar cells, have a long tradition of space use within the orbit of Mars. Panels of photovoltaic cells (solar panels) have been used to power all unmanned lunar exploration craft, including the soft-landing Surveyor series and the roving Lunokhod.

Solar cells rely on the photoelectric effect, converting photons of appropriate energy to electric current. The bandwidth of allowable photon energies puts an inherent limit on the efficiency of PV cells, which can be further reduced by heating effects of the cell. Early silicon cells had an efficiency of about 12%-15% (about 100W/m²). Better fabrication techniques and design have raised this to the 15%-20% range [Lan89b]. Recently GE announced a new PV cell based on a sandwich of gallium and germanium (thus effectively widening the acceptable photon bandwidth) with efficiencies as high as 28%. Spacecraft solar arrays (silicon) are typically in the 45 W/kg range for power density, including supporting structures etc. [Rau80]. Lighter-weight designs can push this to 200 to 300 W/kg.

Photovoltaic cells have no moving parts (although for better efficiency a panel may be gimballed to track the sun) and are very reliable, degrading slowly over the years with exposure to radiation. (Amorphous silicon is more radiation resistant than single crystal, but somewhat less efficient.) Since silicon is abundant in lunar soil and rock (which is mostly silicates) Si PV cells could be readily fabricated from lunar material with only a small capital (mass) investment. Landis and Perino present a brief discussion of such fabrication in [Lan89a]. The equipment for extracting the silicon and producing cells is not overly complex, and may be integrated with machinery for extracting oxygen. Lunar sources of gallium and germanium, however, are scarce to nonexistent.

Solar Thermal

There are two ways of using solar thermal energy: directly, by thermoelectric means, or indirectly, using the sun to heat a working fluid which then drives machinery which turns a generator. The latter method can be broken down into methods which use high temperatures and those that use lower temperatures.

Solar thermoelectricity simply requires thermoelectric elements placed at the focus of a reflective concentrator. While simple to construct, and constructable from indigenous materials (e.g. iron and titanium), the power density is low.

High temperature solar dynamic power production also requires the use of reflective concentrators. The working fluid is heated to high temperature (vaporized) and the vapor then drives a heat engine which turns a generator. Different heat-engine systems have been proposed, including both Stirling cycle and Brayton cycle. Various working fluids have been suggested.

Low temperature dynamic systems operate in a manner similar to high-temperature dynamic above, but use a working fluid with lower boiling point (e.g. ammonia). The energy density is somewhat less, but the system would not require parabolic concentrators with tracking mechanisms. Further, due to the efficiency of such systems at extracting energy from low grade heat sources, they could be used to provide energy at lunar night by storing solar heat in, for example, bulk rock. (See below).

Low temperature dynamic generator technology has been extensively explored to extract energy from ocean temperature differences (Ocean Thermal Energy Conversion or OTEC, see [refs]) but to this author's knowledge this low-temperature technology has not been previously suggested for Lunar application.

Power Storage

Reliance on solar power as the primary means of electrical generation means that some method of energy storage is used during the 2-week lunar night. (This implies that excess capacity is built into the primary sources to charge the storage system during the day). These storage methods may be chemical, mechanical, thermal or electrical.

Chemical

Chemical storage devices in the form of nickel-cadmium (Ni-Cd) or nickel-hydrogen (Ni-H) batteries are commonly used on solar-powered spacecraft. However, the prolonged time over which such would be required to produce power on the Moon means that the mass needed is high. The power density of sealed Ni-Cd cells is about 30-40 Wh/kg [Wis84], while the power density of Ni-H is currently about 40 Wh/kg, with nearly 60 Wh/kg in laboratory tests [Dun84]. (For comparison, the energy density of sealed lead-acid cells is 25-30 Wh/kg [Ham84].)

An attractive alternative is the regenerative H₂-O₂ fuel cell, or more simply separate electrolysis and fuel-cell systems. During the day excess power is used to electrolyse water and liquify the H₂ and O₂. At night these gases are recombined in a fuel cell to produce electricity and water. Fuel cell technology is mature, having been used in manned spacecraft since the Gemini missions. Fuel cells used in the NASA Space Shuttle mass 91 kg (plus fuel and tankage) and have a peak power output of 12 kW, 7 kW average, for 0.076-0.13 kW/kg. [Roc82]. Efficiency is high, about 75%, producing about 1.9 kWh/kg fuel. Efficiency of electrolysis is also high.

Electrolysis and fuel-cell systems require a supply of oxygen and hydrogen, and platinum wire is used for the electrochemically active surfaces. Potassium (K) salts provide the electrolyte. O₂ and K are certainly and platinum very likely available on Luna. H₂ must be imported, but is of low mass.

Mechanical

Mechanical storage systems on Earth generally consist of a reservoir above a generating station to which water can be pumped. This is obviously impractical on the Moon. An alternative storage method is the kinetic energy of a rotating flywheel. In the lunar vacuum and with magnetic levitating bearings, a flywheel loses very little kinetic energy unless tapped to drive, say, a generator. Flywheel rotational velocities are limited by the strength of the material the disc is constructed of, but with modern composites this can be very high indeed. A 5600 kg carbon fiber wheel 2.36m in diameter (inside diameter 1.18m) can store >700 kWh as kinetic energy [Ols83, Jon85]. Five of these would provide 10 kW continuously through the lunar night. Energy storage density of the system, when overhead for supports and power takeoff is included, is about 0.1 kWh/kg, but peak power output can be as high as 100 MW for short pulses.

Thermal

Thermal storage can include both bulk heat storage as temperature, and as latent heat in a phase-change material. In general this is a relatively low grade heat suitable for extraction by a low-temperature dynamic system as above.

Electrical

Electrical storage on this scale tends to preclude devices like capacitors. Storage density is typically <1 Wh/kg, although it is possible to achieve 10 Wh/kg. However the capacitance of a storage device intended to provide power throughout the lunar night would be measured in kilofarads! A better candidate appears to be superconductive storage rings. At present only the low-temperature (e.g. Nb-based) superconductors have suitable energy density, which is related to the current density at which the material loses superconductivity. Niobium-tin storage rings range in power density from 0.5 to 18 Wh/kg [Lan89b]. These require additional cooling below what is naturally available on the Moon. However progress is being made in improving the energy density of relatively high-temperature (90a1K) and lighter weight ceramic superconductors.

Other Sources

Special Locations

Certain locations on the Moon may be conducive to particular power generation techniques. For example, the Moon's low axial tilt means that a sufficiently tall mountain or tower at the lunar north or south pole would be in continous sunlight. This in turn means that any of the solar power generation techniques could be used without requiring nighttime storage [Gre79, Bur85]. Another alternative for a lunar polar base is to run cables to 3 or more solar panels arrayed evenly around the pole at about 50 km (30 miles) distance from it. One or more panels will always be in sunlight.

The Moon is generally regarded as geologically (selenologically) dead, with low heat transfer rates from the core. This seems to rule out geothermal (selenothermal) power sources. However, there are numerous reports of astronomers sighting unusual glows or clouds at some locations, for example near the crater Alphonsus and in the Aristarchus region [Hal69 pp. 104,105; Cam91]. These phenomena may indicate some local geological hot-spot, where such thermal energy could be tapped. The advantages include not only continuous power, but also the possible availability of volatiles that are otherwise rare on the Moon.

Powersats

Solar power satellites, collecting power using vast PV arrays and beaming it via microwave, have been studied for providing power to Earth [Gla77]. They have also been proposed for providing power to locations on the Moon. In the latter case the satellites might best be positioned at the L1 or L2 points. However, the L2 point might well be ruled out since this could interfere with observations from an astronomical facility on the Lunar farside. If the powersats were fitted instead with solid-state lasers tuned to the absorption band of photovoltaic receptors, the satellites could be positioned further away at the L4 or L5 point, or even Earth. [Lan89b]

Future Research

This survey has shown that there is already a vast body of information regarding power sources for lunar bases. However there are a number of areas in which additional research can prove useful. Improving the efficiency of silicon PV cells would pay off both in space and on Earth. Methods for long-term energy storage, especially high density storage, would similarly have benefits both on Earth and off. Research into the cold fusion phenomenon is called for to determine whether this is a usable power source. Continued work on improving the energy density of ceramic superconductors is needed. Finally more extensive geological exploration of the Moon is needed, to answer the questions of geological hotspots, availability of continuous sunlight at the poles, and cataloging of indigenous resources.

References

[And85] Anderson, R.V., "The Rockwell SR-100G Reactor Turboelectric Space Power System" in Space Nuclear Power Systems 1984 (Vol. I), Orbit Book Co., Malabar, 1985

[Avd85]Avduyevsky, V.S., ed., Manufacturing in Space: Processing Problems and Advances, Mir Publishers, Moscow, 1985.

[Bur85]Burke, J.D., "Merits of a Lunar Polar Base Location"

[Cam91]Cameron, W. S., "Transient Lunar Phenomena", in Sky and Telescope, V81, N3 (March 1991), pp. 265-268.

[Cri82]Criswell, D.R., and Waldron, R.D. "Lunar Utilization", in Space Industrialization (Vol. II), CRC Press, Boca Raton, 1982.

[Dun84]Dunlop, J.D. "Nickel-Hydrogen Batteries", in Handbook of Batteries and Fuel Cells, McGraw-Hill, New York, 1984

[Gla77]Glaser, P.E., "Solar Power from Satellites", in Physics Today, (February 1977), pp. 30-38.

[Gre79]Green, J., "The Lunar Polar Base - A Viable Alternative to L-5"

[Hal69]Halacy Jr, D.S., Colonization of the Moon, D. Van Nostrand, Princeton, 1969.

[Ham84]Hammel, R.O., "Sealed Lead-Acid Batteries", in Handbook of Batteries and Fuel Cells, McGraw-Hill, New York, 1984

[Hei91]Heiken, G.H., Vaniman, D.T., and French, B.M., eds., Lunar Sourcebook. Cambridge University Press, New York, 1991

[Jon85]Jones, J.E. Jr, MacPherson, R.E., and Nichols, J.P., "Multimegawatt Space Nuclear Power Concept" in Space Nuclear Power Systems 1984 (Vol. I), Orbit Book Co., Malabar, 1985

[Kat85]Katucki, R., Josloff, A., Kirpich, A., and Florio, F., "Evolution of Systems Concepts for a 100kWe Class Space Nuclear Power System" in Space Nuclear Power Systems 1984 (Vol. I), Orbit Book Co., Malabar, 1985

[Lan89a]Landis, G.A., and Perino, M.A., "Lunar Production of Solar Cells: A Near-Term Product for a Lunar Industrial Facility" in Space Manufacturing 7, Proceedings of the 9th Princeton/SSI/AIAA Conference, AIAA, Washington, 1989

[Lan89b] Landis, G.A., "Solar Power for the Lunar Night" in Space Manufacturing 7, Proceedings of the 9th Princeton/SSI/AIAA Conference, AIAA, Washington, 1989

[McK79]McKay, D.S., and Williams, R.J, "A Geologic Assessment of Potential Lunar Ores" in Space Resources and Space Settlements, NASA SP-428, 1979.

[Ols83] Olszewski, M. and Steele, R.S., Fixed-Base Flywheel Storage Systems for Electric Utility Applications, ORNL/TM-8391 Oak Ridge National Laboratories, Oak Ridge, 1983

[Pan86]Panchyshyn, M., Presentin, R., and Trueblood, B., "A Multi-Megawatt Nuclear Reactor for Lunar Base Applications", AAS 86-308, inProceedings of the 33rd Annual AAS Meeting, Univelt, San Diego, 1987.

[Rau80]Rauschenbach, H.S., Solar Cell Array Design Handbook, Van Nostrand Reinhold, New York, 1980

[Roc82]Rockwell International, Press Information Space Shuttle Transportation System, Rockwell International, California, 1982

[Wis84]Wiseman, J.A., "Sealed Nickel-Cadmium Batteries" in Handbook of Batteries and Fuel Cells, McGraw-Hill, New York, 1984

[Wit87]Wittenberg, L.J., Santarius, J.F., and Kulcinski, G.L., "Lunar Source of 3He for Commercial Fusion Power" in Fusion Technology, V10, pp. 167-178.