All Rights Reserved.

The widespread perception that the extraction of resources is a fundamentally expensive process is incorrect; the expense results from a clumsy technology that largely relies on the heat-driven partitioning of elements into coexisting phases. The extractive feats of biological systems, which carry out isothermal separation of diverse molecular species by means of molecular mechanisms, furnish a striking contrast. Technological means to carry out similar separations are currently primitive largely due to the absence of atomistic control in their fabrication; however, a strong demand for efficient, selective extraction at low concentrations exists in pollution control and is driving further development. Such applications are obviously a near-term motivation for molecular nanotechnology (MNT), the design and construction of macroscopic objects at molecular scale. As such separation technologies mature and costs fall, they will blur the distinction between a "pollutant" and a "resource", which will make space-based resources unlikely to be attractive for terrestrial use. Although MNT is likely to be vital in space development, it probably will render the structural metals that dominate current practice unimportant, and hence native asteroidal metals may prove irrelevant. Carbon and silicates are instead likely to prove most valuable.

The use of nonterrestrial materials is generally viewed as necessary for the development of space-based infrastructure, because of the cost of importing material out of Earth's gravity well1. In addition, nonterrestrial materials are increasingly proposed for terrestrial use2, a suggestion motivated originally by the "doomsday" scenarios of the early 1970s3, which purported to demonstrate that shortages of energy and materials would shortly cause the collapse of industrial civilization.

Geochemically rare but important metals have been a particular focus of concern4, especially since with present technology they must be extracted from rare, sporadic, and anomalous deposits ("ore bodies") where geologic happenstance has concentrated them. As several such metals are siderophile (e.g., Ni, Co, precious metals), sideritic asteroidal bodies have seemed especially promising as nonterrestrial sources for them, even for terrestrial use5. The fact that these metals are either strategic (e.g., Ni, Co), precious (Au), or both (Pt) has made them particularly attractive.

No shortages of metals have appeared in the intervening 25 years since the Limits to Growth appeared; indeed, the general trend of metals prices has been downward, and this has been reflected in a severe downturn of (for example) the job prospects of geologists specializing in ore deposits. Nonetheless, this is widely perceived as a temporary trend fueled by imports from developing countries, where abundant high-grade ores are still available6. At such point as these ores begin to dwindle, especially as domestic demand in the developing world increases, it is commonly felt that the Limits to Growth scenarios will return with a vengeance.

However, new technological developments are calling the assumptions behind these scenarios into question. In particular, the embryonic field of "molecular nanotechnology", in which macroscopic objects are designed and constructed at atomic scales, promises to invalidate the implicit assumptions on which the traditional paradigms of extractive metallurgy are based.

In this paper, I will first review the traditional paradigms of resource extraction and show that (a) the separation of elements is not intrinsically expensive energetically, and (b) the requirement for a feedstock already anomalously enriched in the desired element (an "ore") is merely a reflection of primitive technology. Biological systems already carry out considerably more spectacular feats of element extraction from dispersed sources, and similar feats seem an obvious near-term application of molecular nanotechnology. Indeed, I will argue that pollution control will be a major economic driver for the development of these technologies, with one result being that the distinction between a "resource" and a "pollutant" is likely to become blurred. Finally, I will examine the implications for space resources from this new perspective.

Separating different kinds of atoms from each other is a fundamental technological problem. From one perspective, it constitutes resource extraction; from another, however, it is the problem of pollution control or purification.

Element separation, particularly with regard to resources, is nearly always viewed as intrinsically energy expensive, and this implicit assumption--if examined at all--is justified with vague appeals to the Second Law of Thermodynamics. However, this is incorrect; the Second Law is a red herring, and the vast costs of resource extraction merely reflect our present clumsy technology. To illustrate, the molar entropy of mixing is given by:

where R is the gas constant and xi is the mole fraction of component i7. For a solid solution, with a component j substituting for a component i, the entropy cost SM per mole of component j is then

SM = Sj/Xj = -R Xj ln Xj / Xj = -R ln Xj,

so that it is proportional to the logarithm of the concentration. This obviously increases much more slowly than the concentration decreases.

Moreover, this is the worst-case scenario. In most cases a desired element is not present as a solid solution in a single phase but is concentrated into a separate phase; commonly as a sulfide dispersed throughout a silicate mass, as in a porphyry copper. Indeed, the limited solid solution of chalcophile metals such as Cu and Zn in ordinary silicates is of paramount importance for conventional resource extraction. Above a certain limiting concentration such metals form separate sulfide phases, which are much easier to separate from the silicate background8.

S = -R (Xi ln Xi) = R ln 2 ~ 5.8 J/K per mol of grains.

This accounts for the common textbook assertion10 that the entropy of a macroscopic mixture is "zero." Hence, the true thermodynamic entropy of ore minerals dispersed as a separate phase in host rock, or of (say) pieces of scrap metal dispersed in a landfill, is utterly negligible.

One need not rely merely on such calculations, however; the absence of a large energy barrier to the extraction of elements at low concentrations is compellingly demonstrated by biological systems. To name just a few examples: Green plants use a diffuse energy source (sunlight) not only to extract CO2 from ambient air, where it occurs at a concentration of only ~300 ppm, but furthermore to reduce it to build elaborate carbon-based structures. Vertebrate kidneys extract certain solutes out of a background of many other solutes isothermally. Diatoms and similar planktonic Protista extract dissolved H4SiO4 from seawater, where it occurs in concentrations ~2.5 ppm11, to build their shells ("tests").

The enormous energy expense of conventional resource extraction results from what might be termed the "thermal paradigm"; the partitioning of various elements into coexisting thermodynamic phases with the phase changes typically driven by the application and extraction of prodigious amounts of heat. An example from traditional extractive metallurgy is given by the pyrometallurgy of zinc12. Zinc sulfide is first roasted to convert it to oxide:

(sphalerite)

ZnS + 1.5 O2 ==> ZnO + SO2.

(Note, parenthetically, the wastefulness of this initial step; such sulfides contain free energy, as is shown by their use as energy sources by chemolithotrophic bacteria. Thermodynamically it is possible to reduce the ZnS to Zn metal by oxidizing the sulfur alone.) The oxide is then reduced to metal vapor (!) by carbon at elevated temperature:

C + 0.5 O2 ==> CO (g)

The zinc vapor is condensed to separate it from the CO2, while any silicate impurities form a liquid (a "slag") that is left behind.

Fundamentally, such processes exploit the changes in self-organization of vast numbers (~1025-1027) of atoms as a result of changes in state variables. They have the advantage of simplicity and indeed in essence go back to antiquity. However, they are extremely dirty and wasteful for several reasons:

Moreover, the energy requirements increase vastly as the feedstock concentration goes down, both because of all the material that needs to be treated, and because the partitioning becomes relatively less effective. Hence the concentration in the feedstock is of overweening importance, and to be cost-effective anomalous concentrations already enriched in the desired element (an "ore") must be used.

Plants, kidneys, diatoms, and other biosystems are so much more efficient at extraction because they do not rely on phase changes for separation. Instead, they carry out isothermal separations literally at the molecular level by specialized molecular mechanisms. Photosynthesis, for example, begins with the adsorption of individual CO2 molecules by specialized enzymes, followed by the splitting off of the oxygen atoms by electron-transfer mechanisms using energy stored by absorption of solar photons. Kidneys use ion pumps that use chemical energy to move individual solute molecules "uphill" against a concentration gradient.

Obviously, however, all such mechanisms require a structure at molecular scales. Technological approaches to isothermal, atomistic separation, such as semipermeable membranes or highly specific adsorbers, are currently not nearly so effective largely because of the absence of atomistic control in their fabrication.

Molecular nanotechnology (MNT), the design and assembly of macroscopic objects literally atom by atom has become a burgeoning field of study13, and may constitute the "last technological revolution"14. The sweeping changes anticipated from these developments have been treated elsewhere15. A major barrier, however, to nanotechnological development is the vast numbers of atoms in even a minute macroscopic object. Although individual molecules can be manipulated by scanning probe microscopes16, building a macroscopic object in this manner would take a significant fraction of geologic time. Hence, practical assembly must involve vast numbers of parallel operations. Note, for example, that ordinary solution chemistry is massively parallel by its very nature, as it relies on the thermal motion of colliding molecules to test vast numbers of possible rearrangements over short time scales.

To solve this problem, Drexler17 has proposed molecular scale "assemblers", molecular machines capable of building copies of themselves. In this way the huge number of molecular assemblers necessary for parallel construction can themselves be assembled in a reasonable time. Of course, fabricating assemblers capable not just of atomistic control but of self-replication is far beyond current capabilities, though again it is a routine capability of biological systems.

Thus, although energetically cheap element separation at the molecular level is conceptually possible, it seems at first sight a long way off because of the enormous R&D effort needed to develop a full MNT capability. Drexlerian molecular assemblers are not imminent, and unless developmental pathway(s) can be identified, the relevance of MNT to space development over the near to mid-term is unclear.

It seems such pathways exist, however. First, it appears an interim middle ground exists between present-day "shake and bake" synthesis approaches and full-scale Drexlerian assemblers. One such pathway is molecular self-assembly, which is receiving much study18, and which conceptually provides a way of atomic-scale structuring without assemblers. Another way is to use primitive assemblers, such as gangs of scanning-probe microscopes themselves probably made by conventional microlithography techniques, to "sculpt" molecularly perfect structures on surfaces. Although the number of atoms that can be individually arranged this way remains minuscule, it may be practical for constructing highly selective, tailored catalytic surfaces. In this way the intrinsic parallelism of solution chemistry can be exploited; synthesis yields could be vastly improved through effectively excluding the non-catalyzed reaction pathways19. A major problem in conventional chemical synthesis is the number of unwanted byproducts that form because the synthesis reactions are too unselective; this not only decreases yield but leads to separation and disposal problems. Furthermore, the decrease in yield is substantial for syntheses requiring multiple steps20. (Again, this approach is anticipated by biosystems, which use highly specific catalysts--enzymes--to direct particular synthesis pathways in living cells.)

In addition, such "nanostructured" materials have the advantage of no moving parts19, which eases the engineering difficulties. Nonetheless, the developmental problems remain formidable, and financial incentives must exist if they are to be developed over relevant timescales.

Such an economic driver exists: pollution control. As noted already, it is merely another aspect of separating atoms. Thermal-based approaches, however, as used in traditional pyrometallurgy, are obviously impractical because of the low concentrations involved. Although isothermal phase changes can be used (e.g., precipitation), they still have serious limitations; they require additional reagents (which probably were purified by pyrometallurgy), and there is little control over the precipitates as their nature is set by the laws of chemistry. To wit, there are definite limits (set by the solubility products) to the concentrations that can be treated; the nature of the precipitated phase may be inconvenient (e.g., through being vulnerable to oxidation); and finally, changes in solution composition can cause unwanted phases to form, depending on the species present, their concentrations, and the stability fields of possible solid phases.

Selectivity is also an issue; it is common to have low levels of a toxic ion (e.g., Pb) among a much larger concentration of an innocuous ion (e.g., Ca), and a practical extractive process thus must strongly discriminate in favor of the rare ion21.

More promising approaches to separation involve nanostructured materials, such as highly selective semipermeable membranes, which could filter out and concentrate particular solutes (e.g., heavy metals). Already, molecular sieves such as zeolites are used to separate gaseous N2 from O222, but wider use of such separation is hindered by the expense of crystallizing the sieves. Specific adsorbers, with molecular binding sites highly specific for certain ligands, furnish another example. Note also that such devices do not involve nanotechnological machines--i.e., devices with molecular-scale moving parts; they instead operate passively.

Both selectivity and extraction of solutes at low concentration requires precision at atomic scales23, indeed, current limitations in the applications of membrane technology largely result from fabrication difficulties. The materials are expensive, rather delicate, and molecularly imprecise. Hence the desirability of atomically precise assembly provides incentives for near-term nanofabrication techniques 24.

In addition, a vast and growing literature exists on highly specific complexing agents (typically macrocyclic compounds such crown ethers and calixarenes) for various metal ions, both for potential therapeutic uses as well as for environmental and hydrometallurgical applications25. However, many such compounds are not currently economic due to their costly syntheses. Hence, directed catalysis, as by nanotailored catalytic surfaces as described above, may make such compounds economic and provide another economic motivation for "interim" nanotechnology.

Initially, pollution control will drive these technologies because it is the high value application; the value of the extracted material itself will be insufficient to pay for the technology. Applications initially will lie in such areas as the clean-up of industrial wastewater streams, which is required before their discharge into surface waters. Environmental remediation, such as the clean-up of dump and mining sites (in particular, the amelioration of acid drainage resulting from the oxidation of reactive tailings), are also obvious near-term applications.

As these technologies mature and their costs fall, however, they will ultimately blur the distinction between a "pollutant" and a "resource"; that is, the value of the extracted material will become important in itself. Indeed, as demand increases many sources containing metals in aqueous solution may become attractive. (Note also that the byproduct of such extraction processes would be pure water, which hardly poses a disposal problem.) Seawater is an obvious possibility, but highly saline natural brines may be more attractive. Indeed, deep, saline groundwaters such as those associated with oilfields currently pose a disposal problem.

Because these technologies involve extraction from solutions, it may also prove economic to leach materials containing useful elements, and recover metals from the leachates. This could be looked on as an extension of present hydrometallurgy, as with the present-day cyanide-based solution extraction of Au, or the leaching of Cu with dilute H2SO4. For a further example, during World War II an experimental process for magnesium production involved the dissolution of olivine ((Mg,Fe)SiO4) by a strong mineral acid, such as HCl26. Obviously Fe could be a byproduct (unwanted at the time); these authors also noted that Ni and Co, which commonly substitute for Fe and Mg at concentrations up to ~2000 and ~130 ppm27, respectively, could be recovered. Finally, another unwanted byproduct was silica gel formed from the disaggregated mineral, but this may itself prove useful in a silicate-based nanotechnology, as discussed below. Olivine is a ubiquitous mineral; it makes up most of planetary mantles, and is locally abundant at the surface of both the Earth and Moon.

One hindrance to wider application of such solution-based extractive processes has been the necessity for selective extraction of solutes from dilute solutions. This is the very problem of pollution control again, and underscores the fundamental fact that what's a "resource" and what's a "pollutant" is merely a matter of perspective. Note also that biosystems have anticipated a solution-based approach to extracting raw materials; consider digestion.

In addition, over the longer term MNT is likely to change substantially what elements are desired. In particular, current technology relies heavily on metals for structural members. It is well-known, however28, that ordinary macroscopic materials are a couple of orders of magnitude weaker than the ultimate strength limits set by chemical bonds because of their extremely high densities of defects, such that the strengths are determined instead by such things as grain boundaries and dislocations. Under such circumstances metals are useful because they are highly tolerant of microflaws, even at extreme densities; incipient cracks tend to "heal" via plastic deformation rather than propagate. However, metals are intrinsically weak because of this readiness to deform; brittle materials are potentially far stronger, but liable to catastrophic failure via propagation of Griffith cracks unless they are essentially defect-free at a molecular level. MNT should allow fabricating such defect-free materials, with profound potential consequences29.

Most theoretical studies have focused on tetrahedral (sp3) carbon frameworks ("diamondoid") as the structural basis of MNT30. This is partly motivated by the enormous strength/weight ratio theoretically possible with such networks, but the familiarity and vast knowledge base of organic chemistry also provide a motivation. However, silicates, compounds of Si and O, are a potentially valuable alternative31. Silicates are based on an SiO4 tetrahedron that easily enters 3-D coordination; that is, each vertex can be shared with an adjacent tetrahedron such that all oxygens "bridge" between two silicon atoms. Furthermore, the Si-O bond is strong and directional, due to its partial covalent character. Moreover, in contrast to "diamondoid" carbon, silicates can polymerize at STP, even from aqueous solution; hence a silicate-based MNT may well be nearer term.

Finally, the crust of the Earth is largely made of silicates; oxygen and silicon, respectively, make up 60.4 and 20.5 atom percent of the crust32, and thus raw materials are literally everywhere. However, conventional ores are seldom silicates, simply because of the difficulty of breaking up the Si-O bonds with current pyrometallurgy. Indeed, the waste from conventional mining largely consists of comminuted silicates; ore minerals are typically sulfides, and must be separated from the silicate "gangue" by grinding and flotation. The left-over silicate debris ("tailings") currently constitutes an environmental problem; it is unesthetic, commonly constitutes a dust hazard, and the oxidation of residual sulfides commonly leads to acidic drainage. Its very comminution, however, suggests that tailings might be ideal feedstock for a silicate-based MNT, and certainly there would be no environmental objection to its reprocessing.

Ironically, therefore, the silicates that make up the bulk of the Earth, and that have been ignored in traditional resource scenarios, may yet prove to be among the most valuable raw materials for a truly mature technology. Indeed, the metals such as Fe, Al, Mg, and so on that make up a large percentage of common rocks may ultimately become a (largely) unwanted byproduct of a silicate-based nanotechnology.

Environmental demands are only going to increase in the coming years, and although this has been viewed as increasing the potential demand for space-based resources, it also increases the demand for technologies to ameliorate environmental problems. As described above, this may have the paradoxical effect of quelling demand for off-earth resources, at least for use on Earth. When (say) Ni and Co can be extracted from a wastewater stream at parts-per-million levels, there seems little incentive to mine them from a sideritic asteroid. Indeed, as argued above metals may become a largely unwanted byproduct of a maturing nanotechnology, as the greater intrinsic strengths of brittle materials can be exploited.

Another growing issue is the pressure for total product life-cycle closure, such that the disposal and byproduct costs of a commodity are initially factored into its price. Even if metal is imported for terrestrial use, the ultimate disposal of that metal, due to wear and replacement, represents a cost that must be dealt with.

Energy.

The above considerations, however, do not apply to energy derived from space, as with the oft-repeated proposals for solar power satellites33. Nonetheless, it should be noted that the vast energy demand of present technology is largely because energy is used as heat, the most disorganized and wasteful form of energy. Not only does the application and extraction of heat pervade contemporary processing, as in the traditional pyrometallurgy sketched above, but mechanical motion is nearly always ultimately fueled by a Carot-limited heat engine. Even immature MNT should yield much more efficient energy usage, because little will be used directly as heat; fuel cells, for example, which are not Carnot-limited, are another obvious application of the cheap fabrication of nanostructured materials. Again, a salutary example is provided by biological systems; consider the capabilities of photosynthesizing plants, which, moreover, use only ~1% of the incident sunlight.

Calling MNT "convenient" for space applications is likely to be a major understatement; it may indeed be vital for a viable off-Earth civilization. The value of the extreme strengths of MNT-based materials is only one aspect29; as described above for terrestrial uses, MNT also makes practical a wide variety of raw materials. I had previously argued, based on several millennia of terrestrial experience, that anomalous concentrations of elements--"ores"--would be necessary for space-based resource extraction, just as they have been on Earth34. With the advent of MNT, this seems merely another naive extrapolation of current technology. In addition, because structural metals are likely to be unimportant even with a relatively immature MNT, the ready availability of even high-quality Ni-alloy steels on sideritic asteroids may prove irrelevant.

Of course, C is also abundant in carbonaceous chondrite-like bodies, and thus asteroidal bodies may still prove to be extremely attractive sources of raw materials, quite apart from their low gravity wells. Conversely, C is nearly absent from many rocky Solar System bodies, the Moon in particular, so a diamondoid-based MNT seems unattractive there. (Parenthetically, however, it might be noted that the largest off-Earth reservoir of C in the inner Solar System is the CO2 atmosphere of Venus, which thus may have unexpected long-term value.)

However, a silicate-based nanotechnology is likely to find many applications in space, as silicates dominate rocky bodies such as the Moon just as they do the Earth. Indeed, the regoliths mantling bodies like the Moon, which consist of silicate debris comminuted by eons of meteoritic impact, may prove to be ideal feedstocks. A silicate MNT devised to handle terrestrial mining debris should be readily adaptable to such regoliths.

The separation of elements at an atomic level is an obvious near-term application of molecular nanotechnology. Viewed in one way, this is the problem of resource extraction; but viewed in another, it is the problem of pollution control. Indeed, pollution control is likely to be an economic driver for molecularly precise fabrication, because of the ongoing financial incentives involved.

This has two major implications for space development. First, materials from lunar or asteroidal mines are unlikely to be significant for terrestrial use; when desired elements can be recovered at ppm levels from aqueous solutions, whether wastewater streams, leachates, or natural brines, bringing them in from space is unlikely to make economic sense. Moreover, under a "total product lifetime closure" approach, even space-derived material will have hidden environmental costs due to its ultimate costs of disposal, and such costs will have to be addressed in any case.

Second, by the same token such technologies vastly broaden the potential sources of raw material in space for development in space. When even low concentrations of a desired element can be exploited, "ores" in the traditional sense become unnecessary. However, MNT is also likely to change considerably the desired elements; in particular, structural metal is likely to become unimportant, whereas carbon will become highly sought after. More unexpectedly, the silicates that make up the bulk of the rocky bodies in the inner Solar System may also prove extremely valuable for MNT applications. Hence, the comminuted, rocky regoliths of bodies such as the Moon may prove to be ideal feedstocks, especially as a silicate nanotechnology is likely to be developed in any case for terrestrial applications.

1. e.g., Gillett, S.L., Extraterrestrial resources, Northwest Mining Association, Spokane, WA (Paper presented at the 90th Annual Meeting of the Northwest Mining Association, December 8, 1984); Lewis, John S., & Ruth A. Lewis, Space Resources : Breaking the Bonds of Earth, New York: Columbia University Press, 1987; Lewis, J. S., M.S. Matthews, & M.L. Guerrieri,, eds., Resources of Near-Earth Space, University of Arizona Press, 1993.

2. e.g., Gillett, 1984, op. cit.; Lewis, J. S., Mining the Sky: Untold Riches from the Asteroids, Comets, and Planets, Addison-Wesley, 1996, 274 pp.

3. Meadows, D.H., D.L. Meadows, J. Randers, & W.W. Behrens, III, The Limits to Growth, New American Library, 1972.

4. e.g., Brown, H., The challenge of man's future; an inquiry concerning the condition of man during the years that lie ahead, New York, Viking Press, 1954., p. 187-219; Lovering, T.S., Resources and Man, Report by the Committee on Resources and Man, National Academy of Sciences/National Research Council, Freeman, 1969; Skinner, B.J., Earth resources, Proc. Natl. Acad. Sci. USA, 76, 4212-4217, 1979.

5. e.g., Kuck, D.L., Near-Earth extraterrestrial resources, Fourth Princeton/AIAA Conference on Space Manufacturing Facilities, AIAA #79-1377, Princeton, NJ, May 14-17, 1979; Lewis, John S., 1996, op. cit.

6. Hatfield, C. B., Opinion: Is the greenhouse effect the foe?, Geology , 21, 3, 1993.

7. e.g., Moore, W. J., Physical Chemistry, 4th edition, Prentice-Hall, 1972, p. 239.

8. e.g., Gordon, R. B., Tjalling C. K., W. D. Nordhaus, & B. J. Skinner, Toward a New Iron Age?, Harvard University Press, 1987, p. 24-28.

9. My previous calculation (Gillett, S.L., Nanotechnology, resources, and pollution control, Nanotechnology, 7, 177-182, 1996a) is slightly in error; that value should be multiplied by the mean molecular weight of ~60.

10. e.g., Broecker, W. S., & V. M. Oversby, Chemical Equilibria in the Earth, McGraw-Hill, 1971, p. 238-239.

11. e.g., Krauskopf, K.B., & D.K. Bird, Introduction to Geochemistry, McGraw-Hill, 1995, p. 589.

12. Rosenqvist, T., Principles of Extractive Metallurgy, 2nd ed., McGraw-Hill, 1983, p. 277 ff.

13. e.g., Drexler, K.E., Nanosystems, Wiley, 1992; Crandall, BC, and J. Lewis, eds., Nanotechnology: research and perspectives: papers from the First Foresight Conference on Nanotechnology, Cambridge, Mass, MIT Press, 1992; Krummenacker, M. & J. Lewis, Prospects in Nanotechnology: Toward Molecular Manufacturing (Proceedings of the Foresight Conference, 1992), Wiley, 1995.

14. Drexler, K. E.., Engines of Creation, New York : Anchor Books/Doubleday, 1990, c1986.

15. e.g., Drexler, 1986, op. cit.; Crandall, BC, ed., Nanotechnology: Molecular Speculations on Global Abundance, MIT Press, 1996

16. e.g., Weisenhorn, A.L., J.E. MacDougall, S.A.C. Gould, S.D. Cox, W.S. Wise, J. Massie, P. Maivald, V.B. Elings, G.D. Stucky, & P.K. Hansma, Imaging and manipulating molecules on a zeolite surface with an atomic force microscope, Science, 247, 1330, 1990; Cuberes, M.T., R.R. Schlittler, & J.K. Gimzewski, Room-temperature repositioning of individual C60 molecules at Cu steps: operation of a molecular counting device, Appl. Phys. Lett., 69, 3016-3018, 1996.

17. Drexler, K.E., Nanosystems, Wiley, 1992

18. e.g., Lehn, J.M., Supramolecular chemistry--Scope and perspectives: Molecules, supermolecules, & molecular devices, Angew. Chem. Int. Ed. Engl., 27, 89-112, 1988; Lehn, J.M., Supramolecular chemistry, Science, 260, 1762-1763, 1993; Voegtle, Fritz, Supramolecular Chemistry: An Introduction, Wiley, 1991; Whitesides, G. M., E.E. Simanek, J.P. Mathias, & C.T. Seto, Noncovalent synthesis: Using physical-organic chemistry to make aggregates, Acc. Chem. Res., 28, 37, 1995

19. Gillett, S.L., Near-term nanotechnology: the molecular fabrication of nanostructured materials, Nanotechnology, 7, 168-176, 1996b.

20. Hudlicky, Tomas, Design constraints in practical syntheses of complex molecules: Current status, case studies with carbohydrates and alkaloids, and future perspectives, Chem. Rev., 96, 1, 3-30, 1996.

21. e.g., Ramana, A., & A.K. Sengupta, Removing selenium(IV) and arsenic(V) oxyanions with tailored chelating polymers, J. Environ. Eng., 118, 755-775, 1992.

22. Szostak, R., Molecular Sieves: Principles of Synthesis and Identification, Van Nostrand Reinhold, 1989, p. 17-18.

23. e.g., Martin, C.R., W. Liang, V. Menon, R. Parthasarathy, & A. Parthasarathy, Electronically conductive polymers as chemically-selective layers for membrane-based separations, Synth. Met., 57, 3766, 1993.

24. Gillett, S.L., 1996a, op. cit.

25. e.g., Ramana & Sengupta, op. cit.; Abu-Dari, K., T.B. Karpishin, & K.N. Raymond, Lead sequestering agents. 2. Synthesis of mono- and bis(hydroxypyridinethione) ligands and their lead complexes. Structure of bis(6-(diethylcarbamoyl)-1-hydroxy-2(1H)-pyridine-2-thionato-O,S)lead(II), Inorganic Chemistry, 32, 3052-5, 1993; Durbin, P. W., B. Kullgren, & K.N. Raymond, Octadentate catecholamide ligands for Pu(IV) based on linear or preorganized molecular backbones, Human and Experimental Toxicology, 15, 352, 1996; McMurray, T. J., K.N. Raymond, & P.H. Smith, Molecular recognition and metal ion template synthesis, Science, 244, 938, 1989; Raymond, K.N., Biomimetic metal encapsulation, Coordination Chemistry Reviews, 105, 135, 1990; Xu, J., B. Kullgren, & K.N. Raymond, Specific Sequestering Agents for the Actinides. 28. Synthesis and initial evaluation of multidentate 4-carbamoyl-3-hydroxy-1-methyl-2(1h)-pyridinone ligands for in vivo plutonium (iv) chelation, Journal of Medicinal Chemistry, 38, 2606, 1995; also recent issues of Separation Science and Technology.

26.Houston, E.D., Magnesium from olivine, Trans. AIME, 182, 113-126, 1949.

27. Turekian, K.K., & K.H. Wedepohl, Distribution of the elements in some major units of the Earth's crust, Geol. Soc. Amer. Bull., 72, 175-192, 1961.

28. e.g., Kelly, A., & N.H. MacMillan, Strong Solids, 3rd. ed., Clarendon, 1986.

29. e.g., McKendree, T., Planning scenarios for space development, in Space Manufacturing 10, B. Faughnan, ed., pp. 254-264, 1995.

30. Drexler, 1992, op. cit.; Merkle, Ralph C., Design-ahead for nanotechnology, in Krummenacker & Lewis, op. cit., pp. 23-52.

31. Gillett, S.L., Toward a nanotechnology of silicates, in preparation for CRC volume on nanotechnology, L. Chambers, ed.; Gillett, S.L., to be submitted to the Fifth Foresight Conference on Molecular Nanotechnology, November, 1997.

32. Mason, B., Principles of Geochemistry, 3rd ed., Wiley, 1966.

33. e.g., Space Studies Institute, Solar Power Satellite Built of Lunar Materials, Final Report of a study conducted by Space Research Associates, Inc., for SSI, 1985; previous volumes of this Conference; and previous issues of Space Power.

34. Gillett, S.L., Lunar resources: Thoughts of an economic geologist, Space Power, 10, 3-17, 1991.

Dr. Hugh Ross predicted five years ago that such a discovery was inevitable because of the meteoric debris that floats between planets. Robert Zubrin, Chairman of the National Space Society, has pointed out that Martian rocks have been landing on Earth at a rate of about 500 kilograms a year, and that Martian bacterial spores could have survived the trip. The reverse trip is also possible for over a hundred Terran bacteria, so why the big fuss?

We should send more probes to Mars to determine the issue completely, but I predict that any life we find will have DNA that looks terrestrial. This would only prove that we humans are not the only space-traveling species in this solar system, and that life is a bit tougher than we thought. Of course, such a discovery wouldn’t solve the issue of where life really started: Earth, Mars, both independently in parallel paths of biochemical evolution, or some other common source, such as interstellar Space.

What bothers me is that an enormous unasked question underlies all this hoopla, and nobody in the popular or academic press is asking it, although physicist Enrico Fermi raised it almost a generation ago. He pointed out that at our current pace of technological growth, humans will soon be building starships and visiting neighboring stars. Like the American explorers, some of them will settle where they land, while others (or their children) will move on to the next star for elbow room, religious freedom, or new opportunities. A conservative estimate of this process predicts that humans or their self-replicating robots will explore, settle, and develop every planet around every star in our Milky Way within 250,000 years, a mere cosmic eye blink in the lifetime of the Universe.

We live on an ordinary planet that orbits an ordinary sun at the edge of an ordinary galaxy. Our "average-ness" makes it extremely unlikely that we are the first intelligent species to leave our planet. Our best scientific estimates show that our galaxy contains up to a billion stars that could have developed intelligent life—so at least half should have done it before us. Now it is conceivable that many of these alien species live on water worlds (and therefore never discovered fire), or blew themselves up in nuclear wars (because their technological power outstripped their moral character), or preferred poetry to engineering (inadvertently leading to their extinction when the next big meteor hit), so they never built their first starship. But why has NONE of these possible aliens done what life has ALWAYS done - expand into EVERY niche available to it? And why have they left no trace of their existence? An advanced civilization could leave at least three traces:

First, radio: Every viewer with a standard radio telescope within 35 light-years can watch "I Love Lucy", and Earth would be the brightest radio source in the sky. In fact, an Aricebo-sized radio telescope could pick it up from across our galaxy. But SETI has found no trace of anything.

Second, starship tracks: Physical objects traveling at significant fractions of lightspeed leave trails of Cerenkov radiation that would crisscross the sky for our detectors.

Third, Dyson Spheres: Surrounding a star with an enormous sphere would essentially turn an entire solar system into a giant space ship, and would multiply the living area and/or standard of living by a factor of a billion. Such a macro-engineered structure would block the visible light of a sun from our sight, but it would radiate heat as inferred light.

But physical presence and planetary development would be the most obvious. Everything we know predicts that aliens should have been here a long time ago, and they should have developed our planet out from under us. Out of a billion species that could have gotten here, there are many reasons that would keep many of them away. But there is no reason that adequately explains why NONE of them are here. That fact indicates that there is a gaping hole in our knowledge about the universe, and our place in it. As Sherlock Holmes said, when the obvious is ruled out, then only the alternatives, no matter how fantastic, MUST be considered.

Are we in a cosmic wildlife preserve? Then where are the tourists and the poachers? Is Earth under some sort of universal interdict by an organization whose border guards cannot be tempted, reprogrammed, or overpowered? Is the emergence of intelligence an automatic death sentence to a biosphere because of the technological power it gives for self-destruction? Or does advanced technology mean that the race "moves on" into a Singularity? Do automated war machines from forgotten interplanetary wars roam between the stars? Does the Anthropic Principle, which shows that we are in a unique epoch since the Big Bang, predict that we must choose the final destiny of the Universe? Is our concept of "little green men in flying saucers from Alpha Centuri" adequate? Are our preconceptions of angels and demons too limited?

As we move off this planet to insure the survival of our biosphere, I’m scared of hard vacuum, meteorites, and fatal radiation, and I’m really scared of the dangers we don’t even know about.

But what scares me the most today is that scientists and the media are refusing to address the real implications of life on Mars. By ignoring an enormous paradox in our knowledge, we will remain ignorant—and our ignorance may be fatal.