Advanced Automation for Space Missions/Chapter 4.2

4.2 Materials Background

A survey of Solar-System resources available to mankind in the near-, mid-, and distant-future is appropriate in evaluating the potential of the SMF mission concept. Such background is necessary to identify terrestrial and lunar resources, asteroidal materials, and various additional sources for space manufacturing feedstock. This section describes existing chemical extraction and materials processing alternatives including one new option identified during the course of the study large-scale electrophoretic lunar soil processing) and expanded possibilities for the metallurgy of native lunar basalts, followed by a consideration of materials transport both from the Moon to low Earth orbit using silane-based propellants derived in part from lunar materials, and from the surface of the Earth to LEO using a ground-based electromagnetic catapult (Mongeau et al., 1981).

4.2.1 Survey of Solar System Resources

A survey of extraterrestrial resources reveals a number of major stores of energy and raw materials within the Solar System. Ultimately the most significant of these is the Sun itself. Total solar output is 4&times;10$26$ W, approximately 6&times;10$13$ times as much as mankind produced on Earth in 1980. An extremely power-intensive (15 kW/person) world society of 10 billion people drawing its materials resources solely from the common minerals of the Earth's crust would require only a trivial fraction (4&times;10$-11$) of the available solar output (Goeller and Weinberg, 1976).

It is especially significant that the mass of capital equipment required to produce a unit of useful solar power in space is very low. It is anticipated that large-scale solar thermal power stations can be built for 0.1-1 metric ton equipment per megawatt (t/MW) and 1-10 t/MW for solar electric power. These estimates are calculated for 1 AU (i.e., Earth-orbital distances) from the Sun. Alternative terrestrial mass/power ratios are much larger - hydroelectric plants, 10$3$ to 10$4$ t/MW; projected nuclear fusion power stations, 10$3$ t/MW; coal-fired plants, 2&times;10$2$ t/MW (with 4000 tons of coal consumed per MW/yr); and terrestrial (ground-based) solar power, more than 10$3$ t/MW. Thus, energy systems in space can grow at much faster rates using nonterrestrial materials than is possible on Earth. Energy payback times (time for recovery of initial energy investment) for construction of heliocentric orbital systems at 1 AU is on the order of 10 days. The intensity (I) of solar power varies inversely with the square of the radius (R) from the Sun (I/l$0$ = R$0$$2$/R$2$; I$0$ = 1.4 kW/m$2$, R$0$ = 1 AU = 1.54&times;10$11$ m), so space energy systems may be operated at least out to the distance of Saturn (about 10 AU) before capital/energy efficiency ratios (measured in t/MW) approach values comparable to alternative terrestrial power systems. This is because very low mass optical reflectors can be used to concentrate the available sunlight.

Other power sources which eventually may become accessible to mankind include the kinetic energy of the solar wind (10$14$ MW); differences in the orbital and rotational energies of the Sun, planets, moons, and asteroids (perhaps allowing payloads to move between these bodies) (Sheffield, 1979); and the thermodynamic energies associated with the differentiation of chemical elements in planetoids across the Solar System. Tidal dams on Earth, terrestrial rocket launches, and space probe gravitational swing-bys have utilized trivial fractions of the potential and kinetic energies of the Earth and Moon, and Mercury, Jupiter, Saturn and their moons.

An appreciation of the magnitude of accessible matter resources in the Solar System is gained by noting that terrestrial industry processed about 20 billion tons (about 10 km$3$) of nonrecoverable materials in 1972. (Annual tonnages of chemical elements used industrially are listed in table 4.1.) It can be estimated that humanity has processed slightly less than 10$12$ t (about 500 km$3$) of nonrenewable materials since the start of the Industrial Revolution four centuries ago, assuming a 3% annual growth rate. For comparison, a 4.3 km-radius spherical asteroid (density 1.5 t/m$3$) also contains 10$12$ t of matter. Thousands of asteroids with masses in excess of 10$12$ t already are known (Gehrels, 1979). Approximate total mass of the known minor planets is 2&times;10$18$ t, the moons 7&times;10$20$ t, and meteoritic and cometary matter roughly 10$12$ t. The planets have a total mass of 2.7&times;10$24$ t.

$a$Major elements (&gt;0.1%) are reported first as both the usual oxide notation and elements. Data compiled from the Data Base Compilation of the Lunar Sample Curator, NASA Johnson Space Center, Houston, Texas.

Mankind has launched about 5000 t into LEO since 1959. Most of this was propellant for Apollo lunar missions and for satellites hurled into geosynchronous orbits or into deep space. Approximately 1000 t was hardware. Averaged over the last 10 years, humanity has ejected mass from Earth at approximately 0.05 t/hr or 400 t/yr. Waldron et al. (1979) estimate that oxygen and possibly most of the fuel (silane based) for liquid-propelled rockets can be produced from lunar soil using chemical processing plants with intrinsic capital mass of 100 t/(t/hr) of output. Thus, a LEO propellant production plant weighing about 10 t could service all current major needs if provided with lunar materials. At some point in the future, major mass fractions of space facilities may be constructed of nonterrestrial matter. Space hardware should be produced in orbit at the rate of 10 kg/hr to match the 1970s and anticipated 1980s launch rates.

The United States Space Transportation System (STS), popularly known simply as the "Shuttle," is expected to establish approximately the same mass/year launching ratio during the 1980s at a cost of about $1000/kg to LEO. Energy represents only a small fraction of this expenditure. Perfectly efficient conversion of $0.05/kW-hr electricity into LEO orbital energy (about 10 kW-hr/kg) would cost roughly $0.50/kg for materials transport to orbit, a factor of 2000 less than near-term STS lift prices. Projected bulk transport versions of the STS may lower Earth-to-LEO expenses to $100/kg; still some 200 times greater than the equivalent cost of electrical energy at present-day rates. When launch charges reach $1/kg a large Earth-to-LEO traffic becomes reasonable, since most terrestrial goods are valued at $1-2/kg (Ayres et al., 1979). However, if space industry someday is to approach cost distributions typical of terrestrial industries, then the supply of bulk or raw materials from the Moon and the asteroids must fall to a few 4/kg (Criswell, 1977a, 1977b). On an energy basis alone, this goal appears achievable using high throughput lunar-mass drivers and relatively cheap solar energy (O'Neill, 1974).

Atmospheres of the various planets and moons are valuable as sources of materials and for nonpropellant braking of spacecraft (Cruz et al., 1979). Deliveries of propellants and fabricated parts to space from Earth may be sharply reduced by making full use of local (nonterrestrial) materials, energy, and linear and angular momentum.

Terrestrial materials. Progressive developments of more efficient Earth-to-LEO boosters are expected to reduce transport costs eventually to at least $10-20/kg, comparable to the price of transoceanic air travel (Akin, 1979). The major tradeoff is between development costs of new launch systems and rates of transport in t/yr. Thus, Earth-to-LEO shipment of higher-value products (above $10/kg) needed in low annual tonnages is acceptable and should not seriously restrict the growth of space industries (Criswell, 1977a, 1977b). Space manufacturing directly leverages the effectiveness of any system for transporting goods and materials off-Earth if the value added to the space products is less than the value added by launch of functionally similar goods from Earth (Goldberg, 1981).

STS components such as exhausted hydrogen/oxygen propellant tanks call be used for raw materials. Shuttle external tanks could provide approximately 140 kg/hr of alumina and 10 kg/hr of other elements (e.g., plastics, residual propellants) for early development of manufacturing procedures and products, assuming 30 Shuttle flights per year. (See sec. 4.4.2.)

Earth's upper atmosphere also may prove a valuable source of nitrogen and oxygen for use at LEO and beyond. At 200 km altitude a scoop 1 km in radius oriented perpendicular to the orbital motion intersects approximately 4 t/hr of molecular nitrogen and 3 t/lir of atomic oxygen. Physical convergent nozzles might be used to collect either N$2$ or 0$3$, and a convergent magnetic field might be employed to recover O$2$. Power must be supplied to liquefy the gases and to accelerate a portion of the gathered material to maintain orbital velocity.

Lunar resources. Table 4.1 lists major oxides and elements found in samples of the mare and highland areas of the Moon and returned to Earth during the Apollo and Soviet programs. Table 4.2 summarizes the major lunar minerals and the general uses to which each could be put (Arnold, 1977). The Moon is extremely rich in refractories, metals (Fe, Mg, Ti, Al), oxygen and silicon. Extensive knowledge of lunar resources permits the immediate investigation and development of processing techniques to be employed at an early time in space or on the Moon (Criswell, 1978, 1979; Green, 1978; Inculet and Criswell, 1979;Pomeroy and Hubbard, 1977). Further lunar exploration from orbit (European Space Agency, 1979; Minear et al., 1976) and on the surface using machine intelligence techniques (Duda et al., 1979) almost certainly will reveal additional resources. Discovery of volatiles, such as icy-dirt in permanently shadowed craters at the poles (Amold, 1978; Watson et al., 1963), certainly would expedite the growth of space industries.

The major components of the dark maria surfaces are basalt in the form of lithified or basalt-derived lunar soil and anorthositic plutonic rocks. "Granitic" glass is present in the light highlands.

On the basis of data and samples gathered by the Apollo and Luna missions it has been established that lunar surface basalts can be divided into two classes - high Al/Si (highland basalts) and low Al/Si (mare basalts). The major difference is in feldspar content, which is high in highland basalts and low in mare samples. Major minerals, and others found as minor constituents or traces in lunar basalts, are tabulated in table 4.3.

Pyroxenes occur as enstatite (MgSiO$3$), wollastonite (CaSiO$2$), ferrosilite (FeSiO$2$), and mixtures of all three. Olivines are found as solid solutions of forsterite (Mg$2$SiO$5$) and fayalite (Fe$2$SiO$2$), with most falling in the range of 50-75 mole-percent forsterite. Plagioclase feldspars occur as solid solutions of anorthite (CaAl$a$Si$2$O$+$) and albite (NaAlSi$+$O$2$), with most in the range of 80-100 mole-% anorthite.

A normative chemical analysis of "typical" lunar basalts is shown in table 4.4. It must be remembered that these values are for only two samples of basalt, and therefore, may not represent all lunar basalts. The composition of lunar soil is essentially the same as for lunar basalt, with grain constituents including agglutinates, basalt clasts, anorthite clasts, plagioclase, olivine, ilmenite, and glass. The average grain size is approximately 40 um, but lunar soils often display bimodal size distributions.

Plutonic anorthosites are present in the highland areas. The mineral distributions in three anorthosite samples collected during Apollo missions are given in table 4.5.

Lunar glasses occur in two forms, basaltic and "granitic." Basaltic glass has roughly the same normative chemical distribution as lithified basalt. "Granitic" glass is somewhat anomalous, and may represent the quenched product of magma fractionation. The normative chemical composition of lunar glasses is shown in table 4.6.

Asteroidal materials. Asteroids, especially those with near-terrestrial orbits, are expected to offer a wider range of useful minerals and elements than is available on the Moon (Gehrels, 1979). These bodies may be able to supply many minerals rare or absent on the Moon. For instance, spectroscopic analysis of outgassed volatiles suggests that some asteroids may have abundant water-ice (Degewii, 1980). Those bodies with carbonaceous chondritic composition should contain abundant carbon (up to a few percent by weight), an element which is comparatively rare in lunar soil. The water and carbon expected to be obtainable from asteroids could allow use of water-based and organic chemistry in space factories, techniques otherwise infeasible on a dry, carbonless Moon (though careful recycling will still be necessary). Asteroidal iron-nickel fractions should contain metals in the reduced state and may be rich in platinum-group elements. These resources complement those already found on the lunar surface.

It is conceivable that small quantities of meteoritic material have been trapped in the "gravity wells" (Lagrangian points L4 and L5) of the Earth-Moon (Freitas and Valdes, 1980) and Earth-Sun (Dunbar, 1979) systems. Should such materials exist, very little energy would be required to retrieve them to LEO.

As of 1978, 40 asteroids were known to have trajectories passing close to or inside of the Earth's heliocentric orbit. It has been estimated that 500-1000 Apollo and Amor objects have diameters in excess of 700 m (mass about 1-5&times;10$2$ t), together with more than 100,000 objects greater than 100 m diam with a mass of about 10$2$ t each (Amold and Duke, 1978; Gehrels, 1979). Although most of these asteroids have high velocities and inclinations with respect to Earth's motion around the Sun, a few percent have low inclinations and perihelion near Earth orbit. One, Anteros, can be reached from LEO with less delta-v than is required for transfer to the Moon (Hulkower, Jet Propulsion Lab, private communication, 1980). Several detailed studies have been conducted to examine the possibility of returning one or more of these objects to the vicinity of Earth for use in space manufacturing (Bender et al., 1979; Gaffey et al., 1979; O'Lealy et al., 1979). Methods considered for retrieval have included mass drivers, pellet launchers, solar sails, or detonation propulsion, perhaps expedited by gravitational swing-bys of Mars, Venus, Earth, or the Moon, as required. Extensive increases in ground-based searches and exploration missions to favorable objects should be initiated to fully characterize these resources in preparation for utilization. Table 4.7 summarizes the compositional information now available on Apollo/Amor asteroids, some of which are expected to be a far richer source of volatile materials than low-latitude lunar soils.

$2$Where adequate spectral data are available, mineralogical characterizations and meteorite equivalents are given (from work by Gaffey and McCord, 1977). Where only UBV colors (i.e., C, S, O, U) are available, the Chapman-Morrison-ZelIner classification of the object as summarized by Zellner and Bowell (1977) is given. Underlined classification symbols indicate those based on a single classification criterion. Probable mineral assemblages are indicated.

$2$Albedos and diameters as summarized by Morrison (1977). The diameters in parentheses were derived assuming an average albedo for the "O-S" class of the object and should be considered as indicative only.

Between the orbits of Mars and Jupiter lie thousands of asteroids. These range in diameter from 1000 km down to the limits of telescopic visibility - a few kilometers (Gehrels, 1979; Morrison and Wells, 1978). Certainly still smaller bodies exist but cannot be seen from Earth. Table 4.8 summarizes available information on the widely variable surface compositions of asteroids (Lunar, 1978). The predicted large quantities of rare elements, such as chromium and vanadium, and common metals such as iron and nickel might ultimately have great importance to terrestrial markets and space industries (Gaffey and McCord, 1977; Kuck, 1979). Industrial facilities and habitats constructed from asteroidal materials would make possible the rapid spread of humanity throughout the Solar System.

$2$TRIAD = Tucson Revised Index of Asteroid Data is the source of all data, except as noted in subsequent footnotes. Contributors to this computerized file are: D. Bender (osculating orbital elements), E. Bowell (UBV colors), C. Chapman (spectral parameters), M. Gaffey (spectrophotometry), T. Gehrels (magnitudes), D. Morrison (radiometry). E. Tedesco (rotations), and B. Zellner (polarimetry). TRIAD is described in Icarus 33, 630-631 (1978). To use TRIAD, contact: B. Zellner, Lunar and Planetary Laboratory. University of Arizona, Tucson, Arizona 85721.

$2$Albedos are geometric albedos from radiometry. They are not always consistent with tabulated diameters.

$2$Except at noted, diameters are from Bowell et al. (Icarus, Sept 1978). Values are less reliable for asteroids far which no albedo is listed in previous column. Especially unreliable diameters are listed in parenthesis.

$2$Taxonomic type, related to surface composition, is from Bowell et al. (Icarus, Sept. 1978) wherein the types are defined.

$2$From Gaffey and McCord (Proc. Lunar Sci. Conf. 8th, p. 113-143, 1977), here augmented in several cases by C. Chapman. Refers only to optically important phases.

$2$Stellar occultation diameter (Wasserman et al. and Elliot et al., Bull. Amer. Astron. Sec., Oct. 1978).

$2$Stellar occultation diameter (Bowell et al., Bull. Amer. Astron. Sec., Oct. 1978).

$2$Hartman and Cruikshank (Icarus, in press, 1979).

(Lunar, 1978)

Investigation and development of asteroidal resources will require at least a three-phase approach. First, it is important to find and catalogue the populations and spectral classes of near-Earth asteroids. This could begin at once with a modest investment in a dedicated automated telescope and television camera system which, it is estimated, should be able to find approximately one new Earth-crossing asteroid every night (Gehrels. 1979).

Second is the necessity for direct exploration and sample-return missions. Although there is evidence suggesting that asteroids are equivalent to terrestrial meteorites in composition, the precise physical structures of these bodies are unknown. They may be solid, "fluffy," or more like "raisin bread" with rocks and metals distributed in some matrix. Refining and processing system designs would be significantly affected by the structural configurations of asteroids.

The third and final phase involves large-scale utilization of asteroidal materials either on-site or following transport into near-Earth space. There is a need to develop systems for despinning asteroids, emplacing powerful thrusters, then returning the body to near-Earth space. Ultimately, whole factories might be delivered to or evolved upon individual asteroids. One unusual possibility is that automated factories sent to asteroids could "blow" local materials (metals, glasses, composites) into large, thin, glass-like bubbles many kilometers across, or into metal-coated film for use as solar sails (Drexler, 1980; Nichols, 1979, unpublished report, CIT, Pasadena, Calif.), or as mirrors.

Each of the three asteroid resource development phases is an excellent driver for machine intelligence, robotics, and teleoperation technologies. Long mission times to asteroids favor automation over manned missions. However, it appears that emplacement of thrusters, large-scale bubble-blowing and processing are beyond state-of-the-art, especially in the absence of teleoperation.

Other Solar System resources. Eventually, the resources of the planets (Greeley and Carr, 1975) and their major satellites may become accessible to mankind. Initial attempts at utilization probably will focus on the moons of Mars to support permanent exploration of that planet as well as travel between Mars and the Earth, and beyond. Much of the technology needed for maintaining permanent occupancy of LEO and the Moon should help make extensive exploration of Mars economical. The atmospheres of Venus, Jupiter, Saturn, and their moons and rings are likely early-target resources within these planetary systems (Table 4.9, Lunar, 1978). Later the surface materials of many of the moons may be accessed (Burns, 1977). The radiation belt of Jupiter constitutes a major impediment to utilization of that diverse system. (Access to inexpensive mass for shielding would permit both manned and unmanned penetration of the Jovian magnetosphere for extended periods of time.) Methods have been suggested for extracting energy directly from the particle radiation of the belts by means of secondary emission of charged particles.

Error bars are one standard deviation.

$2$Too uncertain for quantitative entry.

$3$For midlatitudes in spring or autumn where global and seasonal effects are present.

$4$Where distinguishable from surface.

(Lunar, 1977)

Comets, the solar wind, and the Sun are the last major material resources within the Solar System. Most comets pass through the inner Solar System at very high velocities and inclinations (table 4.10, Lunar, 1978). To dependably retrieve large quantities of cometary material it may be necessary to locate and intercept these bodies in the outer Solar System or beyond and provoke repeated gravitational encounters with various planets to effect capture for near-Sun use. These bodies should be exceptionally rich sources of C, N, H, Na, and other volatile elements. Wetherill (1979) estimates that comets, on a 100,000-year timescale, become new Apollo/Amor objects at the rate of 10$3$ t/yr. Deliberate capture probably could increase this rate by several orders of magnitude.

$4$Comets are normally named after their discoverers, but no more than the first three independent observers are so recognized. A few comets have been named after those who made extensive studies of their motion (e.g., Halley and Encke). Comets with periods less than 200 years are arbitrarily called "short-period" and are often written with a preceding P/ for periodic (e.g., P/Tempel 2). Long-period comets are sometimes preceded by C/ just to designate they are comets (e.g., C/West). When one observer or combination of observers discovers more than one short-period comet, the names are followed by an Arabic numerical in order of their discovery to tell them apart (e.g., P/Tempel 1 and P/Tempel 2).

In addition to their discoverers' names, comets are given a temporary designation of the year followed by a lower case letter indicating the order of discovery or recovery. Thus new comet C/Meier is 1978f, while 1978h is P/Giacobini-Zinner making its tenth observed appearance. A few comets such as P/Encke can be observed completely around their orbits. These are designated "annual" comets and are not given temporary letter designations. After about two years, when it is reasonably certain that all of the comets with perihelion passage in a given year have been discovered, each comet is given a permanent designation of the year and Roman numeral indicating the order in which it passed perihelion. Thus P/Encke was 1971 II and 1974 V, and sometime in 1979 it will receive a 1977 permanent designation. These permanent designations are often used together with the name for long period comets [e.g., C/West (1976 VI)] since many discoverers have more than one long-period comet to their credit. The comet 1978a was also discovered by West, for example, while 1973e (1973 VII) and 1973f (1973 XII) were both discovered by Kohoutek, as for that matter were 1970 III, P/Kohoutek (1975 III), and P/West-Kohoutek-Ikemura (1975 IV).

Recent review papers on comets can be found in Comets Asteroids Meteorites Interrelations, Evolution and Origins, A. H. Delsemme (ed.), University of Toledo, 1977 and in The Study of Comets, Proceedings of IAU Colloquium No. 25, B. Donn, M. Mumma, W. Jackson, M. A. Hern and R. Harrington (eds.), NASA SP-393, 1976.

$5$From Marsden "Catalog of Cometary Orbits, 2nd Edition." This is the primary source of orbital data for all comets.

$4$Magnitude at 1 AU from Earth and from Sun (Extrapolated to that distance, if the comet doesn't actually achieve it, using a 4th-power law with heliocentric distance).

(Lunar, 1978)

The solar wind is the outward flow of fully ionized gases (at least 10$3$ t/yr) from the Sun. Presumably, all elements present in the Sun are represented in the solar wind. Table 4.11 gives estimates of the annual output tonnages of the elements, assuming each is present with the same distribution as the cosmic abundance (Allen, 1976). (It is assumed that the hydrogen flux (2&times;10$3$ ions/cm$4$-sec) is omnidirectional from the Sun.) Some type of magnetodynamic systems would clearly be required to collect the solar wind. Perhaps flux-braking by the solar gravitational force is possible, via many convergent magnetic nozzles. Significant fractions of the solar wind might be condensed into grains in convergent regions of such magnetic loops. Though the team can offer no conceptual designs for such systems, it is intriguing that the solar wind output of most elements rivals or far exceeds their corresponding current annual terrestrial production rates. In particular, this source could provide enormous masses of hydrogen throughout the Solar System. Given a means of collecting large fractions of the solar wind, eventually it may be possible to tap tiny portions of the 2&times;10$8$ ton mass of the Sun itself. Such a "star-centered" resource technology capability could decouple the extrasolar spread of humanity and its artifacts from the need for detailed knowledge of the star system of destination.

$22$Freitas, R., 1980a.