A Geologic Assessment of Potential Lunar Ores
DAVID S. McKAY and RICHARD J. WILLIAMS
Large amounts of silicon and aluminum and smaller amounts of other elements will be needed to construct large space structures such as solar power satellites. An analysis of transportation costs indicates that it may be much less expensive to use lunar material. Although bulk lunar soil is not a suitable feedstock for extracting metals, certain minerals such as anorthite and ilmenite can be separated and concentrated. These minerals can be considered as potential ores of aluminum, silicon, titanium, and iron. A separation and metal extraction plant could also extract large amounts of oxygen and perhaps hydrogen from these minerals. Anorthite containing 19 percent aluminum and 20 percent silicon can be concentrated from some highland soils where it is present in amounts up to 60 percent. Ilmenite containing 32 percent titanium and 3 7 percent iron can be concentrated from some mare soils where it is present in amounts up to 10 percent. The ideal mining site would be located at the boundary between a high-titanium mare and a high-aluminum highlands. Such areas may exist around the rims of some eastern mania, particularly Tranquilitatis. A location on Earth with raw materials as described above would be considered an economically valuable ore deposit if conventional terrestrial resources were not available.
INTRODUCTION
G. K. O'Neill (ref. 1) points out the potential economic advantage of using nonterrestrial resources, particularly lunar resources, in constructing large space structures in Earth orbit. This advantage accrues because the gravitational well of the Moon is only about 1/22 that of Earth For the very large mass of material necessary to build a large space structure such as a solar power station, this difference in gravitational energy represents a significant savings in launch costs between the Earth and the Moon.
The construction of a large solar power station will require thousands of tons of aluminum and silicon. In addition, a large mass of shielding is necessary to protect the construction crew from exposure to cosmic rays and solar flares. How much of this material can be derived from the Moon? Do economically useful materials exist on the Moon? Can these materials be concentrated and processed to produce the material necessary for constructing large space structures? In the following sections, these questions are considered in the context of the extensive lunar data base resulting from the Apollo program.
BULK LUNAR SOIL AS A RESOURCE
Bulk lunar soil, (The term "soil" is used here to describe the fine-grained debris layer which overlies much of the surface of the Moon; the term "regolith" is also often used for this layer), contains as major constituents aluminum, silicon, iron, titanium, magnesium, calcium, and oxygen. It has been previously proposed that lunar soil could be used directly as an industrial feedstock from which all of these constituents could be extracted in a single plant using a variety of processes (ref. 2) However, nearly all commercial metal-extraction plants are designed to process a single feedstock containing a small number of elements and to extract a single metal or, at most, two or three metals. The plants can then be optimized by a single process designed specifically for the feedstock and the end-product metal. For maximum extraction efficiency, the feedstock is usually concentrated by beneficiation processes before it enters the extraction plant. A plant designed to extract six or seven elements from bulk lunar soil would be extremely complex, it could not be optimized for a single element, and it would have to overcome such complex problems as cross reactions and blocking reactions.
Extensive experience has shown that it is nearly always far cheaper to concentrate an ore by beneficiation technique than to use the unconcentrated ore directly in the energy-intensive chemical processing plant. For example, energy used to concentrate iron ore is only about 1/10 that used in smelting and refining per ton of processed iron (ref. 3) Without concentration, the smelting and refining costs would be much greater and the processes themselves would need to be redesigned.
Based on these considerations and drawing on the extensive experience of commercial metal extraction, we suggest that it is more practical to limit the extraction effort to a small number of elements and to concentrate lunar materials that contain only those elements. The metal-extraction plants can then be designed specifically for those elements using the least complicated processes available
In following this philosophy, we have chosen aluminum as the primary metal to be extracted from lunar material - the mineral anorthite is the primary aluminum resource. In companion papers (refs. 4, 5), the mining and ore concentration operations are discussed as well as the chemical processing necessary to extract aluminum from anorthite. Silicon and oxygen are valuable by-products of this extraction process. Because aluminum and silicon may not be suitable for some applications (e.g., where high structural strength is required), we will also extract, on a smaller scale, iron and titanium from the mineral ilmenite. Again, references. 4, 5, discuss the procedures for mining and concentrating the ilmenite and for chemically extracting its metals and oxygen. A by-product of ilmenite concentration may be recovery of solar-wind hydrogen which can then be used for propulsion fuel or for life-support needs when combined with oxygen.
ANORTHITE AS AN ALUMINUM RESOURCE
Aluminum may be used extensively to construct a large solar power station. Fortunately, the Moon contains abundant aluminum as an essential constituent of the mineral anorthite, the aluminum-rich end member of the plagioclase series. Pure anorthite - CaAl2 Si2O8 - contains (by weight) 19.4 percent aluminum metal, 20.2 percent silicon metal, 14.4 percent calcium metal, and 46.0 percent oxygen. Anorthite can be considered to be a potential aluminum ore in the sense that it is a naturally occurring concentration of aluminum from which it may be economically feasible to extract the metal. Bauxite, which contains about 25 percent aluminum, is currently the major terrestrial aluminum ore. However, terrestrial anorthite has been used in some countries as a commercial aluminum ore (ref. 6) The United States Bureau of Mines (ref. 7) recently studied the economics of extracting aluminum from anorthite using a lime-soda sinter process. They concluded that the cost of extracting aluminum from anorthite was within a factor of 2 of the cost of extracting aluminum from bauxite and would become even more competitive as the cost of bauxite increased. The Bureau of Mines is currently planning to build a pilot plant to extract aluminum from anorthite (ref. 8). Alcoa Corp., which recently purchased a large area of land in Wyoming estimated to contain as much as 30 billion tons of recoverable anorthosite (ref. 9), is developing plans to recover aluminum from this source.
If anorthite is becoming attractive as a terrestrial aluminum resource, it is even more attractive as a lunar aluminum resource. The lunar crust contains a much higher proportion of anorthite than does the Earth's crust and the lunar highlands are particularly rich in anorthite.
Table 1 presents Al2O3 data on lunar soils sampled by the Apollo and Luna missions (refs. 10- 12). Normative anorthite is calculated using a value of 36.6 percent Al2 O3 for pure anorthite and assuming that all the Al3O3 in a regolith will go into anorthite. Normative anorthite is a good measure of the theoretical maximum amount of anorthite that can be present. The true or modal anorthite content will always be somewhat less because some aluminum will be present in solution in pyroxenes and glasses. Table 1 shows clearly that the highlands contain notably more normative anorthite than the mare regions. It is also apparent that the highlands show considerable variation in normative anorthite content from site to site and that the Apollo 16 site is the region richest in anorthite.
Mission | Stations | No. of Soils | Al2O3 percent | Normative anorthite, percent | |
Highland soils | Ap 14 | LM,G | 6 | 17.5 | 48 |
Ap 15 | 2,6,7 | 11 | 16.6 | 45 | |
Ap 16 | 1,2,3,4,8 | 10 | 27.5 | 75 | |
Ap 16 | 11 | 4 | 29.0 | 79 | |
Ap 17 | 2,3 | 11 | 20.7 | 56 | |
Ap 17 | 6,7 | 7 | 17.8 | 49 | |
Luna 20 | - | 1 | 22.8 | 62 | |
Mare soils | Ap 11 | LM | 1 | 13.6 | 37 |
Ap 12 | L,M | 11 | 14.0 | 38 | |
Ap 15 | 1,4,8,9 | 7 | 13.1 | 36 | |
Ap 15 | 9A | 2 | 10.0 | 27 | |
Ap 17 | 0,1.1A,5 | 7 | 11.2 | 31 | |
Luna 16 | - | 1 | 15.3 | 42 |
ANORTHITE IN APOLLO 16 ROCKS
The rocks at the Apollo 16 site consist of a variety of cataclastic breccias, porous breccias, metamorphosed breccias, and crystalline melt rocks. The rocks display a considerable range of anorthite content. Anorthosites are richest in anorthite, containing more than 95 percent modal plagioclase that is more than 95 percent of the anorthite molecule (An 95 ). If we require an aluminum ore that is more than 90 percent anorthite, then anorthosite rocks are an ideal ore. These rocks are represented in the Apollo collection by such rocks as 60015 (mass, 5547 g), which is more than 98 percent plagioclase of An96-97 ; 60025 (mass, 1836 g), which is more than 98 percent plagioclase of An 95-97 ; 60215 (mass, 385 g),' which is more than 99 percent plagioclase of An 96-97; and 65315 (mass, 300 g), which is more than 98 percent plagioclase of An97-98 (ref. 13). These rocks contain about 35 percent Al 2O3 or 18 percent aluminum metal.
Unfortunately, anorthosites are not particularly common among the Apollo 16 rocks. A survey of 342 rock fragments 2-4 mm in size showed that only 9 percent were anorthosites (ref. 13). A more abundant rock type of a slightly lower grade is the light matrix breccia. Light matrix breccias contain about 34 percent Al 2O3, or about 93 percent normative anorthite (ref. 14). A study of 31 1-2-mm fragments of light matrix breccias showed a mean modal plagioclase content of 79 percent (ref. 15), or somewhat less than the normative anorthite calculated from the A1203 content. Studies of 2-4-mm fragments at each Apollo 16 station showed a range in the total anorthosite plus light matrix breccia fragments from a low of 25 percent at stations 9 and 10 to a high of 63 percent at station 11 (ref. 14). The other rock types present at the Apollo 16 site generally contain less anorthite plagioclase than either anorthosite or light matrix breccia. These rock types include feldspathic melt rocks with 62-84 percent plagioclase (ref. 16), metamorphosed breccia, and poikilitic rocks with 50-55 percent plagioclase (ref. 17), Consequently, they are lower grade potential ores of aluminum.
ANORTHITE IN APOLLO 16 SOILS
Lunar soil consists mainly of fine particles formed by meteorite impact communition of lunar rocks (for a detailed discussion of the lunar soil or regolith, see refs. 2, 18). Lunar soil has the advantage of being already pulverized with a mean grain size generally between 40 and 100 m. From an economic point of view, this means that lunar soil is very easy to mine by simple surface mining techniques and furthermore need
not be pulverized before being fed into ore concentration devices. This is in contrast to the much coarser rock or megaregolith that lies beneath the fine-grained regolith. The megaregolith would require much more elaborate mining processes and would also require energy intensive grinding equipment. Therefore, we conclude that lunar soil is the best source of anorthositic aluminum ore.
A soil made up of meteorite -pulverized anorthosite rock would be the best ore. However, as shown in table 1, such soils have not been found and the richest soils contain no more than 79 percent normative anorthite. This normative anorthite in the soil is not all available for the extraction processing plant, however. Some of it is dissolved in impact-produced glasses (mainly agglutinates) and some of it is intimately inter-grown with pyroxene, ofivine, and ilmenite and cannot be liberated without pulverizing. The amount of available anorthite that can be physically separated can be approximated on the basis of particle counts obtained from the 90-150-/m grain size fractions of lunar soils. This grain size is close to the mean grain size of immature lunar soils and may therefore be a good average of
the coarser and finer grain size populations. Table 2 presents particle counts on 16 Apollo 16 soils from references 19 and 20. Only the anorthite-rich particles and agglutinates are shown. For comparison, the normative anorthite of each soil from data in references 10 and 21 is also shown.
Sample | Plagioclase | Anorthosite | Light matrix breccia | Totalb | Normative anorthitec | Aggultinates | |
Core | 60009,455 | 47.0 | 6.8 | 3.8 | 57.6 | 82 | 12.5 |
60009,457 | 76.2 | 3.8 | 0.8 | 80.8 | - | 12.5 | |
61161 | 14.7 | 4.7 | 13.6 | 33.0 | 72 | 37.0 | |
61181 | 5.3 | 4.3 | 7.3 | 16.9 | 74 | 59.6 | |
STA 1 | 61221 | 17.0 | 13.6 | 10.0 | 40.6 | 77 | 6.3 |
61241 | 12.3 | 5.0 | 18.3 | 35.6 | 74 | 27.1 | |
STA 2 | 61281 | 16.0 | 5.6 | 11.3 | 32.9 | 73 | 40.0 |
63321 | 9.6 | 11.2 | 14.0 | 34.8 | 79 | 32.6 | |
STA 3 | 63341 | 12.6 | 5.9 | 14.9 | 33.4 | 79 | 40.0 |
63501 | 10.3 | 3.0 | 16.7 | 30.0 | 76 | 44.6 | |
STA 4 | 64501 | 20.3 | 3.0 | 6.6 | 29.9 | 75 | 51.6 |
STA 8 | 64501 | 12.3 | 1.9 | 29.3 | 43.5 | 73 | 38.6 |
67481 | 15.0 | 10.9 | 20.3 | 45.2 | 79 | 23.0 | |
67601 | 14.0 | 3.6 | 24.0 | 41.6 | 79 | 36.0 | |
STA 11 | 67701 | 21.0 | 3.3 | 34.0 | 58.3 | 78 | 15.6 |
67711 | 41.0 | 4.6 | 36.6 | 82.2 | 80 | 1.6 |
Column 1 in table 2 shows the percent of individual plagioclase grains in each soil. Based on analysis of plagioclase grains in the rocks, these plagioclase grains can be expected to be about An95-98. Column 2 shows the anorthosite rock fragments, which are generally more than 98 percent plagioclase. Column 3 shows light matrix breccia which, based on analyses of larger rock fragments discussed previously, can be expected to be about 93 percent anorthite. Column 4 represents the total available anorthite concentrate that can be physically separated from the soils without additional grinding. The mean value for the 16 numbers in column 4 is 43.5 percent available anorthite concentrate. The purity of the anorthite concentrate depends on the relative amounts of plagioclase and anorthosite at about 97 percent purity and fight matrix breccias at about 93 percent purity. For the soils in table 2, the concentrate is more than 95 percent pure anorthite.
The Apollo 16 samples are the most anorthite-rich soils of the Apollo missions. They contain 75 percent normative anorthite. An anorthite concentrate making up 43.5 percent of the average soil can be recovered without additional grinding. This concentrate is 95 percent pure anorthite. If soil from only station 11 is considered, a 58 percent anorthosite concentrate can be recovered. This concentrate is more than 92 percent pure anorthite. Thus it would appear feasible to obtain concentrates that are more than 90 percent anorthite and comprise from about 40-60 percent of the soils.
EFFECT OF SOIL MATURITY ON AVAILABLE ANORTHITE
The amount of available anorthite must always be somewhat less than the normative anorthite. However, for a constant normative anorthite content, the amount of available anorthite depends strongly on soil maturity. This relationship is shown in figure 1, where the available anorthite concentrate is plotted against the agglutinate content for the soils in table 1. Agglutinates are impact produced glassy aggregates of glass, mineral, and lithic grains. The agglutinate abundance has been shown to be a useful index of surface exposure age of soil maturity (ref. 22). As agglutinate content increases, more of the aluminum in a soil is incorporated into the agglutinates, either in the glass itself or in feldspar grains engulfed and buried by the agglutinate glass. This aluminum cannot be concentrated by physical mineral separation techniques, so it is not available as an ore. Equally important is the dilution effect of the agglutinates on the recoverable anorthite concentrate. As shown in figure 1, , a mature soil has less available anorthite concentrate by a factor of 2 to 4 compared to an immature soil. The soils richest in available anorthite concentrate are the extremely immature soils with agglutinate contents less than 3 percent. Such soils are created by moderately large impact craters and are found at or near the surface close to fresh large craters.
DO HIGH-GRADE ANORTHITE DEPOSITS EXIST ON THE MOON?
At all Apollo and Luna sites, it is clear that ore beneficiation by mineral separation techniques is necessary to provide anorthite feedstock that is more than 90 percent pure for chemical processing. Do areas exist where the regolith is already more than 90 percent anorthite? If such areas were available as mining sites, there might be no need for ore beneficiation.
As discussed above, at the Apollo 16 site, two rock types have greater than 90 percent normative anorthite: anorthosite and light matrix breccia. No large body of anorthosite has been found, but a 30-m-thick layer of light matrix breccia is postulated to underlie the region around N. Ray crater (refs. 14, 15). If this layer were at the surface, the overlying regolith might be expected to contain 90 percent pure anorthite. However, no surface regolith of this grade has been identified in the Apollo 16 area. Furthermore, no large body of anorthosite has been found at Apollo 16 or any of the other Apollo or Luna sites.
The Apollo 15 and 16 orbital x-ray experiments mapped Al/Si ratios over the Command Service Module ground tract for these missions. Al/Si ratios in highland regions ranged from 0.52 for the region west of Fecunditatis to a high of 0.71 for the farside highlands west of Mendeleev (ref. 23).
We conclude that no large anorthite-rich region is now known. Additional mapping by remote techniques would be desirable to try to locate such areas. Earth-based telescopic spectra sensing has been used to determine iron content in lunar regolith (ref. 24) Since iron content varies inversely with anorthite content, areas extremely low in iron would be rich in anorthite. Additional Earth-based telescopic observations might discover such areas. Surveys from lunar orbiting spacecraft would be highly desirable.
ILMENITE AS A RESOURCE FOR TITANIUM AND IRON
Although rutile is more desirable, ilmenite is also considered to be a commercial ore for producing titanium. Dupont Corp., for example, has used ilmenite ores on a commercial basis and the United States Bureau of Mines has recently reviewed the feasibility of replacing imported rutile with domestic ilmenite as a source for U.S. titanium (ref. 25) Ilmenite - FeTiO3 - contains 31.6 percent titanium and 36.8 percent iron. In a companion paper (ref. 5), an extraction process is discussed that recovers titanium, iron, and oxygen from ilmenite.
LUNAR ILMENITE
Ilmenite is a constituent of most lunar rocks but occurs only in small amounts in highland rocks. Ilmenite is most abundant in mare basalts. Table 3 presents the normative ilmenite identified in each of the large returned mare basalts (ref. 26) and also shows the normative ilmenite calculated from the TiO2 content by assuming that all TiO2 is present as ilmenite. The Apollo 11 and 17 basalts are clearly richest in ilmenite. The chemical composition of lunar ilmenite is presented in table 4, (ref. 27)
ILMENITE IN LUNAR REGOLITH
As with anorthite, we consider soil as a source for ilmenite rather than the actual rocks. Again, regolith has the advantage of being easy to mine and of being already pulverized. The disadvantage is that some of the ilmenite it incorporated into glassy agglutinates and cannot be recovered by mineral separation techniques. Additionally, some of the ilmenite is present as fine-grained intergrowths or rock fragments and cannot be separated
from pyroxene and plagioclase without additional grinding, which we hope to avoid. Consequently, the amount of ilmenite available without additional grinding is considerably less than the normative ilmenite. Table 5 presents the modal ilmenite in the 90-150-
The normative ilmenite calculated from the TiO2 content of the bulk soil is also given. An important point to note in table 5 is that the available ilmenite in the average Apollo 17 mare soil is only 1/3 of the normative ilmenite. All modal data in table 5 are for the 90-150-
SOIL MATURITY AND AVAILABLE ILMENITE
As with anorthite, the amount of available ilmenite is influenced by the maturity of the soil. As soils mature, they generally become finer grained, which should have the effect of liberating more mineral grains. However, agglutinates are also formed, which tends to reduce the abundance of mineral grains. For Apollo 17 mare soils, the variation in the ratio of available ilmenite to normative ilmenite with agglutinate content is shown in
figure 2. No trend is apparent in this figure and, at least for the 90-150-
DO HIGH-GRADE ILMENITE RESOURCES EXIST ON THE MOON?
The most ilmenite-rich mare basalt from the Apollo samples contains about 25 percent ilmenite (table 3), and the most ilmenite-rich soil contains about 9 percent available ilmenite (table 5). Is it likely that more ilmenite-rich rocks and soils exist on the Moon? TiO 2 content can be approximated from optical spectral data using Earth-based telescopic observations (ref. 29). From observations made by this technique, the high titanium basalts of Apollos 11 and 17 appear to represent maximum TiO2 contents for lunar materials. It is possible that some mare lava flows could have undergone near-surface fractionation and settling of ilmenite to form near-surface ilmenite cumulates. However, little support for this process can be found from the chemistry of the basalt samples (refs. 30, 31). Most researchers favor the theory that the ilmenite-rich basalts originated when a cumulate melted at some depth within the Moon (refs. 30, 31).Therefore, near-surface cumulates of ilmenite may not exist. If they do exist, they have not yet been sampled nor have they been detected by remote-sensing techniques.
BLACK GLASS AS AN ILMENITE RESOURCE
The orange glass and its partly crystallized equivalent, black glass, are apparently abundant in numerous locations, mainly around the rims of circular maria (ref. 32). This material contains 8.8 percent TiO2 and, in its crystallized form, contains olivine, ilmenite, and other minerals. It has been suggested that this material might be a useful titanium ore (ref. 33). However, the crystallized ilmenite is always very fine-grained and tightly intergrown with olivine. It is therefore not possible to separate and concentrate the ilmenite unless the material were ground to a grain size approaching 1
EXTRACTION OF HYDROGEN AND WATER
DURING ILMENITE PROCESSING
It is well known that lunar soils contain essentially no water, except possibly in permanently shadowed polar cold traps. However, more mature lunar soils contain some hydrogen implanted from the solar wind. How much hydrogen is present, where is it located, and is it feasible to extract it from the lunar soils? Data on hydrogen content of lunar materials are relatively sparse and mostly limited to bulk soil analyses. Data from various investigators vary by about a factor of 2 for the same soil sample. Typical analyses by Epstein and Taylor (refs. 34-36) and Merlivat et al. (ref. 37). are in the range 30 to 60 ppm with a maximum of 61 ppm. Friedman et al. (ref. 38). DesMarais et al, (ref. 39). and Stoenner et al. (ref. 40) found generally higher values ranging up to 90 ppm for 10084 (ref. 38), 92 ppm for part of the Apollo 15 core (ref. 39), and 145 ppm for part of the Apollo 17 core (ref. 40), Eberhardt (ref. 41) considers 70 ppm to be a typical concentration.
Within a soil, hydrogen concentration varies inversely with grain size. This variation results from the surface correlation of all solar-wind species; these atoms are implanted and trapped near the surfaces of individual grains. For hydrogen, the only published data on this variation are by DesMarais et al. (ref. 39), who found an increase by a factor of about 2 in going from bulk soil to soil finer than 20
Maturity also influences hydrogen content. Immature soil 62221 contains only 8 ppm hydrogen and the extremely immature orange glass sample 74220 contains only 0.2 ppm hydrogen (ref. 36),
In summary, solar-wind hydrogen is present in some mature lunar soils in amount exceeding 70 to 100 ppm. The hydrogen is not homogeneously distributed but is concentrated in the finer grain sizes, by a factor of about 2 for the grains of less than 20
SELECTION OF A MINING SITE From the previous discussions, a set of criteria can be constructed for selecting the optimum lunar mining site for Al, Si, Ti, Fe, and H2 ores:
However, mass-driver siting seems to require a location farther east, preferably at long. 33o40' E and lat. 1o44'N in southeastern Tranquilitatis (ref. 50).This region appears to be lower in TiO2 and the closest highland region appears somewhat lower in Al/Si ratio (ref. 23). More detailed remote data are needed to choose the optimum site for high TiO
The requirement that the highland site be relatively immature is an important one, but one which can fairly easily be met by location near a relatively large fresh crater that has ejected a large quantity of fresh, immature soil. Mature mare regions should also be easy to locate as they predominate in areas away from fresh craters.
Most mare sites should be flat enough for easy surface mining. A problem might arise with the highland area because highlands at the boundary of mare basins often have a mountain front. However, experience from Apollos 15 and 17 has shown that highland soils are easily accessible in talus slopes or debris aprons at the base of highland mountains. At the Apollo 17 site, relatively pure mare soil (80 percent mare, 20 percent highland material) was located at station 5 within 7 kin of highland soil (92 percent highland, 8 percent mare material) at station 2 (ref. 51). Even over the 4 km between station 6 and the LM, the character of the soil changed from 70 percent highland to 70 percent mare. At the Apollo 15 site, mare soil at station 9 was about 85 percent mare material although it was only 4 km from the highland front (ref. 51). It would appear that a separation of about 5 km between a highland and mare mining pit would be sufficient to provide a reasonably clean separation of soil types. Even less distance might be appropriate for properly chosen sites. A debris apron, typically about 300 m wide, usually separates highland fronts from mare areas (ref. 52). Mining pits separated by only 300 in might be sufficient if located on either side of a debris apron.
SUMMARY AND CONCLUSIONS
We have discussed reasons why bulk lunar soil is not a suitable raw material for chemically extracting metals. On the basis of considerable terrestrial experience, it is much more desirable to concentrate minerals containing the desired metals and then to use extraction techniques specifically designed for these minerals. A suitable lunar ore would be anorthite feldspar for aluminum and silicon.
Anorthite contains 19.4 percent aluminum and 20.2 percent silicon. It has been used commercially as an alurninum ore and will be used more and more in the future as the supply of bauxite declines. For titanium and iron, lunar ilmenite would be an excellent ore. Ilmenite contains 31.6 percent titanium and 36.8 percent iron. Chemical processing of both anorthite and ilmenite can produce abundant oxygen as a by-product.
Mineral separation techniques should be used on the lunar surface to concentrate the anorthite and ilmenite ores before shipping. Mining and ore beneficiation are greatly assisted by the fine-grained nature of the lunar soil, which eliminates the need for blasting, crushing, and grinding. An anorthite concentration mill located at the Apollo 16 site could recover 40 to 60 percent of the soil as a 90 percent pure anorthite concentrate. An ilmenite concentration mill on a high TiO2 mare soil could concentrate 5-10 percent of the soils as 90 percent pure ilmenite concentrate. These numbers are based on the available soil data. However, more accurate figures are necessary and can be determined only if lunar soil is analyzed specifically for extractable anorthite and ilmenite content.
The ilmenite concentrate may be rich in solar-wind-derived hydrogen, perhaps containing as much as 1000-2000 ppm hydrogen in the fine-grained ilmenite. This quantity of hydrogen is equivalent to nearly 1-2 percent water. This water could be extracted from the ilmenite concentrate with little additional effort beyond that needed to mine and concentrate the ilmenite. However, actual analyses of hydrogen in ilmenite are necessary before the hydrogen concentration in ilmenite can be known with any certainty. In any case, hydrogen extraction would probably only be worthwhile as a by-product of ilmenite concentration processes.
The ideal mining site must be near both an immature highland area and a more mature TiO2 mare area. Such sites exist around the margins of some of the eastern maria including Tranquilitatis. The southeastern margin of Tranquilitatis is also ideally suited for a mass-driver location. Additional chemical and soil maturity data are needed to determine whether acceptable mass-driver sites and acceptable geological sites coincide.
After candidate sites are located by Earth-based and lunar-orbiter remote techniques, detailed geologic prospecting, mapping, surveying, and sample analysis would be necessary before a site could be certified for mining operations. Such a site, if transported to Earth, would have such abundant and easily accessible anorthite and ilmenite as to make it a valuable terrestrial ore body, if conventional terrestrial ores did not exist.
Sample Modal ilmenite
content, percentNormative ilmenite
content, percent
1003 14-18 21.5
10017 14-24 22.4
1044 6-12 17.1
10045 7-11 21.3
10049 16-17 21.5
10072 13-22 22.8
AP 11 Mean 14.5 21.1
12202 8-11 4.9
12021 5-12 6.6
12022 9-23 9.3
12039 8-10 5.7
12051 8-11 8.9
12063 8-10 9.5
AP 12 Mean 10 7.5
15016 6 4.4
15076 0.5 3.6
15475 1.0 3.4
15555 1-5 4.0
15556 2 5.1
AP 15 Mean 2.6 4.1
75055 12-20 20.5
70215 13-37 24.7
70035 15-24 24.7
70017 19-23 24.7
AP 17 Mean 20.4 23.7
aReference 26
TiO2 53.1
FeO 43.7
Al2 O3 0.26
Cr2 O3 0.72
MnO 0.33
MgO 2.0
CaO 0.18
100.29 m size fraction for the Apollo 11 soil and for nine Apollo 17 mare soils (A. Basu, unpublished data on 10084, 1977) (ref. 28).
m GRAIN SIZE FOR THE APOLLO 11 SOILa AND FOR NINE APOLLO 17 MARE SOILSb, NORMATIVE ILMENITE CALCULATED FROM THE TiO2 CONTENT IS ALSO SHOWNc
Sample Modal
ilmenite, percentNormative
ilmenite, percent
10084 14.7
79221 1.3 12.4
78501 3.7 10.4
75081 7.6 17.9
75061 5.3 19.6
71501 9.0 18.1
71061 4.6 17.7
71041 5.6 18.2
70181 2.3 8.1
70161 5.0 17.1
AP17 Mean 4.9 15.5
aReference 28
bReference 29
cReference 10
m grain size fraction. How does available ilmenite vary with grain size? Table 6 presents the modal data for one Apollo 17 mare soil which shows the variation in ilmenite content over five grain size fractions (reference 18
). The overall trend is for more ilmenite to be available at finer grain sizes. This trend is to be expected because a larger proportion of ilmenite grains are liberated from lithic fragments as the overall grain size approaches the average grain size of the mineral grains in the lithic fragments. For sizes below 45 pin, no data are available, but more than 6 percent ilmenite grains should be present if the trend continues. Consequently, the 90-150-
m abundance of 4.6 percent may be a representative average for the whole soil. It appears that 90 percent ilmenite concentrates from mare soils is feasible. On average such concentrates would comprise about 5 percent of the soil; possibly as much as 10 percent could be recovered in favorable areas.
Grain size, m
Ilmenite, percent
45-75 6.0
75-90 3.3
90-150 4.6
150-250 3.3
250-500 2.3
aReference 18
m fraction, the degree of soil maturity appears to have little effect on the proportion of normative ilmenite available for mineral separation. This contrasts with anorthite, for which soil maturity strongly influenced the amount of available anorthite.
m, an extremely impractical undertaking. The bulk orange and black. glass would not make a good titanium ore as it contains slightly less titanium than soils from ilmenite-rich basalt regions (such as soil 75080) and contains only about 1/6 as much titanium as ilmenite.
m. Data on the variation with grain size of solar-wind helium are more plentiful. The behavior of hydrogen should be roughly comparable to helium, and helium generally shows an increase by about a factor of 2 in going from bulk soil to soil finer than about 20
m. (refs. 42, 43). The concentration of hydrogen has been measured on individual grains as a function
of distance inward from the surface. The outermost zone ay contain 400 to 600 ppm (refs. 44, 45).
Hydrogen content also varies with the type of particle. Agglutinates in a single -grain-size fraction may contain more than twice the hydrogen concentration of the bulk-size fraction and more than 10 times the hydrogen concentration of plagioclase (ref. 40), Although no data exist for hydrogen, the mineral ilmenite is known to retain helium much more readily than other minerals. Hintenberger et al. (refs. 46) found for some Apollo 17 soils that, in the 35-54-
m grain size, the ilmenite grains were enriched in 4He by a factor of about 3 to 6 over he bulk material in that size range. Eberhardt et al. ref. 47) found that for soil 12001, the 11-
m ilmenite concentrate was enriched in 4He by a factor of nearly 22 over the bulk 4He content of the soil. If hydrogen shows the same enrichment factor as helium in ilmenite, the 11-
m grain size fraction of ilmenite in 12001 might be expected to contain 22 times the bulk hydrogen content of 44 ppm (refs. 48)The fine-grained ilmenite would then contain about 1000 ppm hydrogen. If the bulk soil contained 100 ppm hydrogen and if the 22x enrichment factor occurred, the fine-grained ilmenite would contain over 2000 ppm hydrogen. Whether hydrogen actually follows these trends of extreme enrichment in ilmenite remains to be shown. Only the lighter rare gases show his enrichment; actually, argon, krypton, and xenon how a depletion in ilmenite compared to bulk soil. Because of the enrichment of helium in ilmenite, soils how a correlation between TiO2 content and helium content. Mare soils are up to an order of magnitude richer in helium than highland soils. The available hydrogen data do not show such a great difference. However, relatively few mare soils have been analyzed for hydrogen.
m. Hydrogen is also concentrated by a factor of about 2 in agglutinates. The largest concentration, however, may be in ilmenite grains. For solar-wind helium, finer-grained ilmenite may contain more than 20 times as much helium per gram of ilmenite compared to a gram of bulk soil. Whether hydrogen follows this trend is yet to be determined. If it does follow the trend, however, the finer-grained
ilmenite or mature mare soils may contain as much as 1000-2000 ppm hydrogen. Concentrates of this ilmenite could conceivably constitute a hydrogen ore. If this hydrogen were converted to water during the extraction process, the ilmenite concentrate would yield 1 to 2 percent water. Particularly as a by-product of the concentration of ilmenite, hydrogen might be extracted from ilmenite on the Moon with little additional effort. It should be emphasized however that the amount of hydrogen extracted from any concentrate must always be less than the hydrogen present in the bulk soil from which the concentrate came. A thermodynamic analysis of the amount of hydrogen that could be extracted and a discussion of the method and conditions for extraction are found in reference 4.
The requirement for an area rich in both anorthite and ilmenite can only be met at a boundary between a mare region and a highland region. In addition, the mare region must contain a high TiO2 basalt. These requirements are met at many places on the front side of the Moon, namely, around the margins of the high TiO2 eastern maria. The requirement that the highland area be very rich in anorthite rather than, for example, KREEP or highland basalt, places an additional restriction on the site. The Descartes highlands area near the Apollo 16 site meets this requirement as determined from the returned samples and from the relatively high Al/Si and low Fe of the orbital x-ray experiment (refs. ref.23, 49). Whether this high Al region continues northward to Mare Tranquilitatis is not clear from the orbital data. Mare Tranquilitatis is relatively high in its southwestem quadrant near
the lobe of highlands extending north from the Apollo 16 region. Consequently, this margin of Tranquilitatis might be an excellent site.
Curators:
Al Globus and
Bryan Yager
NASA Responsible Official:
Dr. Ruth Globus
Last Updated: July 10, 2002
If you have problems accessing the information
on this site, please contact access@mail.arc.nasa.gov. If you find any errors on this page contact Bryan Yager.