Foresight Update 12

page 3

A publication of the Foresight Institute

Foresight Update 12 - Table of Contents | Page1 | Page2 | Page3 | Page4 | Page5


Recent Progress: Steps Toward Nanotechnology

by Russell Mills

I am on the mailing list of various companies that sell to biotechnologists. The advertising they send provides a way to track technological progress. Could there be any truer indicator of progress than the actual products available for purchase?

Every few weeks a packet of reply cards arrives, each card describing a different company's latest offerings. There's a thrill or two lurking in almost every packet. Here are several recent examples:

Twenty years ago ads like these would have seemed appropriate in science fiction novels about the 21st century. Twenty years from now I expect to see ads with titles like these: "COMPLETE GENOME ANALYSIS FOR ANY ORGANISM, $500" and "CUSTOM ENZYMES--FAST TURNAROUND."

Nanotechnology (i.e., molecular construction with atomic precision) is the expected outcome of our increasing ability to manipulate individual atoms. Manipulating atoms is difficult--our tools tend to be either large and nonspecific, or small and perverse. This difference in tools is what distinguishes the two main approaches to nanotechnology.

The first approach is exemplified by chemistry, a discipline of large tools--flasks, stirrers, heaters, and the like. Atoms are manipulated by ingenious choices of reaction sequence, reagents, catalysts, temperature, and other conditions. Unfortunately the only structures that can be accurately built this way are rather simple ones: molecules of a few hundred atoms, and polymers. To make more complex objects requires more control than can be provided by any choice of reactions or conditions. Chemists are therefore trying to "bootstrap" their way to nanotechnology by using the tools at hand to create molecular-size tools--a challenging task.

The other approach is exemplified by biotechnology, in which many of the tools are microscopic--cells, viruses, enzymes, antibodies, genes, and other objects of evolutionary origin. Since they were not designed by us, they are not easy tools to use. The task here is to modify them, to make them more versatile and obedient.

Let's look at some recent progress along both of these approaches.

Foresight Update 12 - Table of Contents



A major advance in polymer synthesis was announced in February by researchers at Affymax Research Institute (California) [Science 251:767-773, 15Feb91]. Combining methods from the electronics industry with automated peptide and nucleic acid production techniques, the group was able to carry out large numbers of simultaneous syntheses in a small area.

The technique makes use of "protector" molecules that serve as temporary chain terminators for the polymers under construction. The various monomers ("building blocks") are pretreated with protectors such that, of a monomer's two reactive groups, one is capped while the other remains available for chemical reaction. The protectors detach when exposed to light.

The Affymax procedure starts with a flat piece of glass--a stable working surface on which polymer chains will be grown. A layer of linker molecules is chemically bonded to this surface. Protectors are in turn bonded to the linkers. The system with its linkers and protecting groups is now subjected to illumination through a mask of the kind used to define electronic elements in integrated circuits. In illuminated regions, but not elsewhere, the protectors come off, exposing the underlying linkers. The system is next bathed in a solution containing the first monomer, which attaches to the exposed linkers but not to the protected linkers. Since the monomers were pre-equipped with their own protectors, the working surface of the system is once again completely covered with protectors. Illumination through a different mask will now expose a different pattern of attachment sites for a second monomer. In this way, a large number of different polymer molecules can be synthesized simultaneously with high precision.

The technique has been shown to work for both nucleic acids and polypeptides, and it should be applicable to other types of polymers as well. Its inventors envision it as a tool for exploring biological recognition processes and for testing compounds in the pharmaceutical industry. With commercially available photolithography equipment, at least 250,000 different compounds could be synthesized in a 1 cm2 area, then tested in situ for their ability to interact with biological receptors or enzymes. When a promising interaction is detected, the compound responsible can be identified from its location.

This invention should accelerate the development of protein-based nanotechnology by enabling protein engineers to make and test millions of different protein molecules in a matter of days. The time required to home in on a successful design should be considerably reduced. And some of these engineers will be designing molecular machines ...

A class of polymers called "starburst dendrimers" have created a stir among chemists lately. These are molecules that branch repeatedly from a central core until, at a radius of about 5 nm, the density is high enough to form a closed surface. Although starburst dendrimers typically contain tens of thousands of atoms, they can be synthesized with atomic precision. Chemists have discovered a variety of different reaction schemes and initiator cores, each leading to a different end product.

A dendrimer's shape--spherical or cylindrical--is determined by the pattern of reactive sites on the initiator; its surface characteristics depend largely upon the monomer used for the final synthetic step. Some dendrimers are tightly sealed, others are porous. Australian scientists recently found a way to grow dendrimers inward from the periphery back to the core, suggesting that dendrimers with functionally differentiated surfaces may be possible. [Science 251:1562-64,29Mar91]

Physical interfaces--the region of contact between dissimilar materials--are of fundamental importance in many areas of science and technology. The behavior of electronic devices, catalysts, composites and many other things are determined by the structural details of interfaces. Yet, in the absence of suitable manipulators for positioning individual atoms, the preparation of interfaces has been a haphazard affair: a substrate of one material is exposed to a barrage of atoms of a second material, causing an overlayer to accumulate. The arriving atoms are often traveling fast enough to cause damage or undesired chemical bonding, producing an interface different from the one desired.

Materials chemists at the University of Minnesota have now developed a method for gently lowering an overlayer onto a substrate. The trick is to create a protective cushion on the substrate surface by letting a noble gas (like xenon) condense on it at low temperature. An overlayer is then deposited on this cushion in the usual manner--by evaporating it from a hot filament, for example. Increasing the temperature evaporates the cushion, causing the overlayer to settle upon the substrate. [Science 251:1444-51,22Mar91]

What we have here is a simple nanometer-scale manipulator controlled by temperature and pressure adjustments. True, the manipulator only works in one dimension and one direction. And the cushion is so "slippery" that the overlayer atoms slide around on it uncontrollably and form clusters. On the other hand, the concept may be generalizable: a substance more dynamic than xenon might be used--one having responses specific to different overlayer atoms or to stimuli other than temperature and pressure.

Foresight Update 12 - Table of Contents



At New York University Junghuei Chen and Nadrian Seeman report that they have built a cube-like object out of DNA. Each face of the cube is defined by a closed strand of DNA; the object therefore consists of six strands. Each cube's edges are helical segments of double-stranded DNA composed of the DNA strands of the adjacent faces wrapped around each other. The whole structure contains 480 nucleotides and is built in several stages from ten carefully designed DNA strands.

The researchers suggest that DNA might be used to make larger frameworks to which proteins or other molecules could be attached. A framework would regularize the positions of such substituents, allowing X-ray diffraction mapping to be performed even on molecules that resist crystallization. The study of interactions between molecules occupying adjacent lattice sites would be another possibility. [Nature 350:631-33,18Apr91]

Ferritin is an iron-storage protein, variants of which are found throughout much of the animal kingdom. The ferritin molecule, roughly spherical with a central cavity 8 or 9 nanometers across, is composed of 24 subunits. Channels in the molecule's walls allow the passage of iron atoms; an apparatus in the channel oxidizes the iron and transfers it to crystallization sites inside the cavity. When filled, each ferritin molecule holds about 2000 molecules of iron oxide, forming what might well be an atomically precise object.

Ferritin subunits are of two common types: L and H. Since L ferritin crystallizes easily it was mapped in detail by X-ray diffraction several years ago. H ferritin, however, resists crystallization. European researchers decided that the reason lay in the presence or absence of a metal binding site on the molecular surface--crystals of L ferritin are stabilized by cadmium atoms at the points of intermolecular contact. A metal-binding site was therefore introduced into H ferritin by changing a single amino acid. The resulting molecules crystallized easily and have now been mapped. [Nature 349:541-544,3Feb91]

That a single amino acid substitution can make a drastic difference in how proteins interact is no surprise; what is remarkable is to see the interaction so effectively redesigned. Similar techniques might permit other proteins to be assembled into larger structures or "programmed" to rearrange themselves in response to changes in their chemical environment.

In other ferritin news, researchers at the University of Bath (England) used ferritin molecules as reaction vessels for producing beads of several kinds of materials, including manganese oxide, uranyl oxyhydroxide, and iron sulfide. The first two products were made by allowing the reactants to penetrate into empty ferritin spheres; the metals underwent oxidation and precipitated in the cavity. The iron sulfide was made by exposing iron-filled ferritin to hydrogen sulfide; the reaction took place inside the ferritin cavity. Electron microscope analysis showed all the products to be beads 7-8 nm in diameter. [Nature 349:684-687,21Feb91]

How might ferritin be adapted for nanotechnology? Let's indulge in some speculation. Genetic engineering applied to ferritin's L and H subunits would let us create new subunits of our own design. But whereas native ferritin is assembled out of semi-arbitrary combinations of 24 L and H subunits, the assembly of engineered ferritin could be controlled by designing the interfaces between different subunits. The shape of the interior ferritin cavity would be determined by the design of the corresponding surfaces of the subunits. The specificities of the catalytic sites might be modified so that different subunits would sequester different elements or small molecules. The result: a set of molds for making nanometer-scale machine parts with controlled shapes and compositions. (Possible drawbacks: crystals of this size may not be very stable; and small machine parts may be a lot easier to make than to assemble into a machine.)

Russell Mills directs a small research company in California.

Foresight Update 12 - Table of Contents


Molecular Manufacturing for Space

by Chris Peterson

For decades the space development effort has been plagued by expensive and unreliable hardware, so it isn't surprising that this community has shown a strong interest in the molecular manufacturing concept since its earliest days. To follow up on this interest, the Foresight Institute and the Institute for Molecular Manufacturing sent representatives to the International Space Development Conference in San Antonio in late May. Attending were Jim Bennett, IMM Executive Director Lynne Morrill, Eric Drexler, and myself. The results were all that we had hoped, and more.

Foresight and IMM shared an information table, and many of those stopping by reported that they came to the meeting specifically for the nanotechnology coverage. A well-attended lecture on that topic was given by Eric Drexler. That evening IMM held a small gathering of supporters and garnered its first $1000 donation, from John Baccellieri of Arlington, Texas. (Mr. Baccellieri told us that he had been unable to attend the main talk himself, since he was serving as coordinator of volunteers at the conference: his volunteers had deserted him in order to attend that talk, leaving him on duty alone.)

The next day included another talk relating the old goals of space settlement--familiar to many in the audience who had been members of the L5 Society--to the new means of molecular manufacturing. Running concurrently was the National Space Society Board of Directors meeting, at which a motion in support of molecular manufacturing (see text below) was unanimously approved. Then that evening at the Awards Banquet, a Space Pioneer award in the Scientist/Engineer category was presented to Eric Drexler for his work on nanotechnology.

In parallel, a new special interest group was formed within the National Space Society: the Molecular Manufacturing Shortcut Group. They plan to get the needed technology developed sooner by diverting more government research funds to the relevant enabling science and technologies. The group is open to any member of the National Space Society. For more information, write to the president of the new group: Stewart Cobb, Molecular Manufacturing Shortcut Group, 555 Bryant St. Suite 253, Palo Alto, CA 94301.

[Editor's note: The current mailing address is: Molecular Manufacturing Shortcut Group, 8381 Castilian Drive, Huntington Beach, CA 92646.]

As an incentive to new members, IMM distributed souvenir T-shirts with the IMM logo on the front and the slogan "Ad Astra Per Nanotechnologia" (To the Stars through Nanotechnology) on the back. A few of these collector's items are still available; contact IMM.

The following is an abridged version of the text distributed to the NSS Board:

Improvements in manufacturing capabilities have played an essential role in opening the space frontier and will continue to do so. The mechanical manipulation of matter at the molecular level, proposed by Richard Feynman[1], has now been demonstrated[2]. Molecular manipulation can be used to construct general structures[3], ultimately including high-performance, low-cost aerospace systems[4,5]. Concrete examples include the manufacture of structural components with order-of-magnitude improvements in strength and stiffness, and order-of-magnitude reductions in cost. This by itself would suffice to make single-stage-to-orbit vehicles efficient and inexpensive.

Major steps toward this objective are now feasible[6], requiring perhaps 3 to 5 years with a budget of perhaps $1-2 million per year. Large-scale space applications will await the emergence of advanced molecular manufacturing and related nanotechnologies, requiring many years of goal-oriented research--a long time by some standards, but well within the 10- to 25-year time horizon that has been traditional in the space movement.

Molecular nanotechnology promises to play a central role in 21st century industry, both on Earth and in space, but the U.S. is falling behind. The journal Nature writes[7] of "blossoming interest in Japan in nanotechnology, a field which with the backing of the powerful Ministry of International Trade and Industry (MITI) seems destined to become Japan's next priority target for industrial research."

Proposed resolution

Substantial sentiment on the Board and within the Society favors the promotion of nanotechnology development as one of several approaches that can accelerate the achievement of long-term space goals. Because initial results can apparently be had for a cost that is insignificant relative to that of any major space project, a small effort today can speed developments with enormous future payoffs for space technology. Further developments will be more expensive, but will proceed on the basis of concrete technological achievements, many with immediate applications. An NSS role in these further developments can guide them more swiftly toward space applications. To help the Society take advantage of this opportunity, the following resolution is proposed to facilitate activities by NSS members:

Whereas, advances in materials, computers, and manufacturing have in the past led to advances in space technology, and

Whereas, advances in molecular systems engineering, leading to molecular-precision fabrication and related molecular nanotechnologies, promise to yield dramatic advances in materials, computers, and manufacturing, be it therefore

Resolved, that the National Space Society encourages the pursuit of research leading toward molecular nanotechnology as a means of accelerating space development, and encourages the formation of a group within the Society to promote efforts in this direction.


  1. Richard Feynman, "There's Plenty of Room at the Bottom," in Miniaturization, H. Gilbert, ed., (New York: Reinhold, 1961).
  2. D. M. Eigler and E. K. Schweizer, "Positioning single atoms with a scanning tunnelling microscope," Nature 344: 524-526 (1990).
  3. K. Eric Drexler, "Molecular engineering: An approach to the development of general capabilities for molecular manipulation," Proceedings of the National Academy of Sciences (USA) 78: 5275-5278 (1981).
  4. K. Eric Drexler, "Molecular Engineering: Assemblers and Future Space Hardware," American Astronautical Society, AAS-86-415 (1986).
  5. Stewart Cobb, "Molecular Magic," Ad Astra 2,6:22-25 (June 1990).
  6. K. Eric Drexler, "Molecular Tip Arrays for AFM Imaging and Nanofabrication," Journal of Vacuum Science and Technology B 9(2): 1394-1397 (March/April 1991).
  7. David Swinbanks, "Nanotechnology: MITI aims at small research," Nature 349:449 (1991).

Foresight Update 12 - Table of Contents | Page1 | Page2 | Page3 | Page4 | Page5

From Foresight Update 12, originally published 1 August 1991.

Foresight thanks Dave Kilbridge for converting Update 12 to html for this web page.