|Keynote speaker,John Polanyi
1986 Nobel Laureate in Chemistry
The keynote talk by John Polanyi, University of Toronto, 1986 Nobel laureate in chemistry, on "Photon- and electron-induced chemistry as a means to nanolithography" set forth one of the major themes of this conference: nanoscale and atomically precise modification of surfaces. Prof. Polanyi investigated the dynamics of reaction with crystalline surfaces as a result of the photon- or electron induced dissociation of molecules adsorbed to those surfaces. Since in most experiments the photons or electrons are broadcast to the entire surface, the pattern that is written onto the surface will follow the pattern in which the reactive molecules have self-assembled onto the surface.
But the details of which atoms on the surface are modified should follow from the configuration in which each molecule binds to the surface. Indeed, the surface modifications were found on the atoms expected from the configuration of the adsorbed molecules. For example, dissociation of halogenated benzene molecules on the 7x7 reconstructed Si(III) surface produced chlorine atoms on the surface 7 Å apart in the case of 1,2-dichlorobenzene, and 14 Å apart in the case of the 1,4 molecule, in which the Cl atoms are on opposite sides of the benzene ring. In other experiments, it was demonstrated that the number of Si atoms etched from the surface varied with the intensity of the laser used to provide the photons, allowing removal of one Si atom at a time, or "bit etching." In one case, electrons were added to specified points on the surface by pulsing an STM tip at defined intervals as it was dragged across the surface, producing the expected line of Cl atoms written on the Si surface at intervals determined by the pulsing interval.
Prof. Chad A. Mirkin, Northwestern Univ. ("New Chemical, Biochemical, and Physical Methodologies For Preparing Inorganic Nanostructures"), focused on using a "nanopen" to mark surfaces precisely with complex patterns of lines without the need for a resist layer. The nanopen uses the capillary effect in the layer of water that forms (usually as a nuisance) between the tip and surface in scanning force microscopy to transfer molecules dissolved in the water (the "ink") first to the tip and then from the tip to the surface. Factors such as the relative humidity, the relative hydrophilicity of tip and surface, and chemical reaction of the ink with the surface (such as a thiol on a gold surface) control the direction of flow and the width of the lines produced. Typical widths are 15 to 70 nm. A "dark" ink results from using a low friction, hydrophobic molecule, such as a C18 thiol; a contrasting ink from a high friction, hydrophilic molecule, such as a C16 mercaptocarboxylic acid.
A key technical feature was developing a system of registration marks so that a second pattern could be superimposed on an earlier pattern (made with a different ink) to an accuracy of 5 nm, with a drift of only about 1 nm/10 minutes. The system is fast and precise enough that a 115-word paragraph from Feynman's famous 1959 talk was written in only 10 minutes, in lines 60 nm wide. More interesting applications will require a wider variety of inks. Prof. Mirkin proposed using DNA molecules as ink since one set of parameters for using the ink would apply to all DNA molecules, but different sequences of the DNA would give each ink unique molecular recognition properties, facilitating the construction of complex devices for molecular electronics and catalysts. Also, carbon nanotubes used as tips should permit drawing lines much finer than the current 15 nm limit set by the radius of curvature of the silicon nitride tip used for these experiments.
Prof. Joseph W. Lyding, Univ. of Illinois, presented "Silicon-Based Molecular Nanotechnology," and described using high voltage scanning tunneling microscopy to write on the hydrogen passivated Si(100) surface by removing H atoms. Feedback controlled lithography (FCL) uses the STM feedback signal to control the process precisely enough (that is, the tip retracts once the tunneling current indicates that a dangling bond has been formed) to create individual dangling bonds arranged in precise patterns on the Si surface, which can then be used as templates so other molecules spontaneously add to the surface in those (atomically) exact positions. Results were presented with arrays of norbornadiene, copper phthalocyanine, and C60 on the Si surface. For details, see the full paper submitted to the conference.
Prof. Hongjie Dai, Stanford Univ. ("Integrated Nanotubes"), described progress in chemical control of the growth of carbon nanotubes on silicon substrates, thus integrating nanotubes with conventional microstructures. First, lithographic techniques are used to pattern silicon substrates with micrometer-scale islands of catalytic metals, and then nanotubes are grown from those islands by chemical vapor deposition. Depending on substrate and CVD conditions, the products are either single walled nanotubes (SWNT) or multi-walled nanotubes (MWNT) held tightly by van der Waals forces in self-assembled rectangular blocks (for example, 50 microns square and 250 microns high). Current research includes growing nanotubes as AFM tips, molecular wires, and as very sensitive sensors in which electrical conductivity changes strongly upon exposure to, for example, NH3 or NO2.
|1999 Feynman Prize winner Bill Goddard and 1998 Feynman Prize Winner Ralph Merkle discuss their theories about Molecular Nanotechnology|
As usual, computational simulation was applied to a wide range of problems in nanotechnology. Prof. K. J. Cho, Stanford Univ. ("Multiscale Simulations for Computational Nanomechanics"), discussed integrated quantum calculations, atomistic simulation, and continuum modeling methods to model the mechanical and electronic properties of carbon nanotubes. His findings included the demonstration that local areas of sp3-hybridized carbon atoms (diamond) formed in buckling SWNTs. Since diamond is an electrical insulator, this provides a way to introduce insulators into electronic devices made from SWNTs. Other calculations show the effects of small amounts of NH3 or NO2 on the conductivity of 10,0 SWNTs, explaining the experimental observations of Prof. Dai above and providing a way to dope nanotubes, with NH3 reducing conductivity and NO2 increasing it. Taken together, these simulations suggest that mechanical and chemical control could be combined to make complete devices from nanotubes.
Dr. Susan Sinnott, Univ. of Kentucky ("Computational Design of New Nano Composite Materials"), used molecular dynamics simulation with the reactive empirical bond-order potential for hydrocarbons to investigate the effect of chemical functionalization of the walls of carbon nanotubes. Chemical functionalization of the nanotube wall would greatly facilitate using the nanotubes in a polymer matrix to form nanocomposites, but what would the effect be on the properties of the nanotube? As a first step, Dr. Sinnott simulated ion bombardment of 10,10-SWNT with methyl radicals at incident energies from 10 to 80 eV. At 10 eV CH3- functionalization without the creation of defects predominates. At 45 eV the methyl groups tend to decompose to form CH and CH2, which both adsorb to and create defects within the wall. At 80 eV, there is more decomposition of the methyl radicals and more defect formation, including the formation of 7- and 8-carbon rings within the nanotube wall and cross-linking between nanotubes. Interestingly, impacts of nanotubes that already carried pentagon-heptagon defects tended to heal the defects.
Another major theme of conference presentations was nanotechnological uses of biological molecular machines. Addressing the problem of powering nanotechnology devices with biomolecular motors, Dr. Carlo D. Montemagno, Cornell Univ. ("Integrating Life Processes into Engineered Nanofabricated Systems"), turned to genetically engineering the F1-ATPase from the bacterium Bacillus PS3 (for background information, see his full paper from the 1998 Sixth Foresight Conference and his web site). For example, by adding metal binding amino acid residues to the ATPase, the motor molecule was made to bind tightly to nickel nanodot arrays prepared by electron beam lithography, a single, properly oriented motor molecule per nanodot. This year Dr. Montemagno reported attaching a silicon nitride bar several hundred nm long to the rotor subunit of the motor molecule. He also engineered bacteriorhodopsin into liposomes to use light to produce ATP molecules to power his ATPase bio-motor. Working towards the goal of having an engineering device fully integrated into a living system within 3-5 years, he envisions a 25-nm diameter molecule sorting rotor working in vivo and powered by the physiological environment.
Switching from rotary to linear motors, Dr. Loren Limberis, Univ. of Utah ("Kinesin-powered MicroChemoMechanical Systems (MCMS)" ), demonstrated that kinesin motor proteins could be attached to 10-micron square silicon chips and act as motors when the chips were placed on a glass surface covered with microtubules, the 24-nm diameter protein tracks along which kinesin molecules move in the intact cell. In the cell, kinesins are known to take an 8 nm step along a microtubule for every ATP molecule hydrolyzed, generating a maximum force of 6 pN and a top speed of about 1 micron per second. Since kinesins can be bound to silicon at up to 105 molecules per square micron, it should be possible to generate forces on the silicon chips of 100's of nN per square micron. The kinesins were observed to move on the glass slides at up to 8 microns per second, rotate at 20° per second (presumably because the chip contacted two microtubules of opposite polarity), or flip from one side to the other. It is hoped this technology will be able to power more complex microdevices, such as gears, levers, etc.
Despite the impressive array of molecular machines that have evolved in biological systems, there is not one available for every function that one would like to have in nanotechnology. Dr. Andrew Pohorille, NASA Ames Research Center ("Towards the Creation of Simple, Functional, Cell-like Structures"), tackled the problem of creating a "general strategy for building simple, cell-like systems capable of performing functions that may or may not exist in a cell." They have developed liposomes encapsulating bacteriorhodopsin and ATP synthase to use light to generate energy in the form of ATP molecules. To make these little, artificial cells useful for nanotechnology, they needed a general method to develop protein catalysts that do not exist in nature. In a major advance, Dr. Pohorille provided "a proof of concept" for evolving proteins to perform any function for which a suitable selection exists.
The method is based upon technology published two years ago by one of Dr. Pohorille's collaborators, Jack Szostak ("RNA-peptide fusions for the in vitro selection of peptides and proteins," by RW Roberts and JW Szostak, Proc. Natl. Acad. Sci. USA94: 12297-12302), in which a mRNA molecule is covalently fused to the protein molecule produced when the RNA is translated in vitro. The method is efficient enough so that if a pool of different (randomly mutated) RNA molecules is used, libraries larger than 1013 independent fusion molecules can be screened for proteins with the desired characteristics. The messenger RNAs encoding the selected proteins are automatically selected with the proteins and can then be put through another round of amplification and variation to look for proteins with even better characteristics. Applying this technique to the directed evolution of a protein toward a target function, Dr. Pohorille reported evolving a 78-amino acid protein that binds ATP, much smaller than natural proteins with that function. Future goals include evolving proteins that act as kinases (add phosphates to proteins or other molecules) and as amino acid ligases (form bonds between two amino acids).
Other approaches were also presented for adding to the molecular tool kit. Dr. Jan H. Hoh, Johns Hopkins Univ. School of Medicine ("Controlling DNA Folding in Vitro"), looked at the collapse of DNA from a stiff rod-like structure with a persistence length of 50 nm to compact rod or toroid structures that occurred in the presence of certain polycations, reminiscent of the biological problem of how a meter-scale DNA molecule fits into a micron-scale cell. Investigating the mechanism of this collapse in the presence of a novel class of silane cations suggests a new class of DNA-DNA contacts stabilized by the silane cations. Taken together with the cross-linking chemistry that exists with silanes, this observation suggests the possibility of a new class of DNA-silica particles.
|Highlights of the conference included the high-quality poster session and the Nanomanipulator demo.|
Molecular construction kits were presented by Prof. Joseph Michl, Univ. of Colorado ("Molecular Grids and Rotors in the Computer and in Practice"). Prof. Michl focused on various designs for molecular rotors mounted on molecular grids, that would rotate due either to a flow of inert gas past the rotor blades or to the rotation of an external electric field. The strategy was to use molecular dynamics to simulate rotors that could actually be synthesized in the laboratory if the simulation showed the design to be promising. An early design uses a rectangular molecular grid built around a dirhodium tetracarboxylate moiety to which the molecular rotor is attached. The rotor is a complex of rhenium with two substituted o phenanthroline ligands, one positively and one negatively charged so that the molecule has a electric dipole moment of 42 debyes.
Simulations showed the rotor to turn, both from gas flow and from an external electric field, but it also moves like a pendulum, probably because it is attached to the grid at only one end of the axle. Movies are available at the Web site of Prof. Michl's collaborator, Jaroslav Vacek. A more recent model is a five-bladed carborane propeller attached to the grid at both ends of the axle. Simulations show that this solves the pendulum problem, but the blades are not stiff, and friction is an issue. While working to improve designs for large rotors attached to nanostructured rectangular grids, Prof. Michl is also proceeding in the laboratory to attach a substituted benzene with a dipole moment of 5.5 D to a quartz substrate via an acetylene bond about which rotation can occur.
|Marcia Seidler,conference planner,still has a smile at the end of four long days.|
Looking to biology for a molecular construction kit, Prof. Dietmar Pum, Center for Ultrastructure Research, Austria ("S-Layer Proteins as Basic Building Blocks in a Biomolecular Construction Kit"), focused on proteins isolated from bacterial cell surfaces that reassemble on a variety of surfaces, including metals, silicon wafers, and lipid monolayers, into crystalline two-dimensional arrays that have either oblique, square or hexagonal lattice symmetry and unit cell dimensions in the range of 3 to 30nm. Since these proteins provide a high density of targets for chemical functionalization (about 1.6 x 106 carboxyl groups per µm2), a wide variety of macromolecules have been covalently immobilized on surfaces of S-layer proteins. Further, the proteins contain pores of 2 8 nm (depending upon bacterial species) which can serve as templates for precipitation of inorganic salts to form quantum dot arrays of, for example, CdS or Au nanoparticles. For details, check the full paper.
Directly addressing the question of what types of molecular building blocks would work with positional assembly to build molecular machine systems, Dr. Ralph C. Merkle, Zyvex ("Molecular Building Blocks and Development Strategies for Molecular Nanotechnology"), proposed adamantane as a possible molecular building block. The C10H16 molecule is tetrahedrally symmetric, resembles a small piece of diamond, and its chemistry has been well studied, with over 20,000 derivatives known. One approach to linking together such building blocks, Dr. Merkle suggested, could be dipolar bonds between N atoms substituted for the corner C atoms of one adamantane and B atoms substituted at the corners of another. For details, check the draft of the full paper at Dr. Merkle's Web site.
Still the favorite building block by far for current research is the carbon nanotube. Prof. Rodney S. Ruoff, Washington Univ. ("Tensile Strengths of Carbon Nanotubes and Mechanochemistry of Carbon Nanotubes"), reported further characterization of nanotubes using the 4-degree-of-freedom nanomanipulator announced at last year's Conference. Looking at the breaking of MWNTs under tension, the most common failure mode was pulling the inner tube out of the outer tube, like pulling a sword out of a sheath. The few cases in which the entire force curve was analyzed implied a Young's modulus of from 250-950 GPa for the outermost shell of the MWNT. Analysis of a rope of 217 SWNT's gave only 12 GPa per SWNT, but it is likely that only some of the SWNT are intact for the whole length of the rope so that the true strength of individual SWNTs could be much higher.
Scanning force microscopy was used for positional assembly of nanoparticles by Roland Resch, Univ. of Southern California ("Building Nanostructures from the Bottom Up by Manipulation and Self Assembly"). Special software was written to control the probe to facilitate maneuvering the particles. Gold nanoparticles (5-30 nm in diameter) were coated with a bifunctional reagent, an alkanedithiol, so that particles brought into contact would stay together. Dr. Resch was able to manipulate these particles into structures like 3-particle straight lines and rings of 5 particles that could be translated along the surface and rotated without breaking the structure. Four-particle straight lines, however, broke when moved. Furthermore,these 3-5 particle structures could be immobilized on the surface by laying a siloxane layer over them, permitting the construction of new sets of particles on top of the embedded ones.
A novel form of nanomanipulation was presented by Dr. Michael P. Hughes, Univ. of Surrey, United Kingdom ("AC Electrokinetics: Applications for Nanotechnology"). AC electrokinetic techniques, in which dynamic electric fields induce dipoles that can be used to manipulate polarizable particles, have been used for years to manipulate µm-scale particles, such as yeast cells. Dr. Hughes explained how the advent of micromachined electrodes, with µm-scale separation between electrodes, opens the possibility of applying these techniques to nm-scale particles. Dr. Hughes cited work in which virus particles and latex spheres as small as 14 nm had been trapped by AC electrokinetic methods. For details, check the full paper.
|Co-chairs Jan Hoh and Deepak Srivastava congratulate each other at the end of the conference|
|Foresight Update 39 - Table of Contents|
The Tutorial on Foundations of Nanotechnology was held on October 14, 1999, immediately preceding the Seventh Foresight Conference on Molecular Nanotechnology, to give participants in the Tutorial an overview of the technologies underlying molecular nanotechnology. The Tutorial Chair this year was Prof. Donald W. Brenner, Materials Science and Engineering, North Carolina State University.
Phillip Russell, Professor of Materials Science and Engineering and Director of the Analytical Instrumentation Facility at North Carolina State University, gave a presentation on scanning probe microscopy (SPM), an extraordinarily useful tool for imaging and manipulating atoms and nanometer-scale objects. Part of his tutorial is available on the web. Prof. Russell reviewed the history of SPM and explained the working principles of the two major forms of SPM: scanning tunneling microscopy (STM) and scanning force microscopy (SFM, originally designated AFM for atomic force microscopy). Reviewing the different limitations and strengths of STM and SFM and the different modes of doing SFM led to the take-home conclusion: "There is no universal approach to achieving atomic resolution on arbitrary materials!" But with a wide variety of techniques from which to choose, it is possible to come close.
Computational Methods in nanotechnology were covered by Prof. Brenner. Beginning with the exact solution that quantum mechanics provides for the hydrogen atom, he provided a detailed tour of the successive approximations and alternatives needed to model increasingly complex systems, culminating with current research on systems of interest for nanotechnology, such as devices made with fullerene components and diamond. One example: computational modeling predicts increased chemical reactivity of kinked carbon nanotubes at the point of the kink, and such reactivity has been observed experimentally by Rod Ruoff's group. One caution: commercial and free software for modeling is available, but is easy to misuse without a clear understanding of the limitations at each level of approximation.
Nanostructures defined as structures large enough to be controlled in interesting ways and small enough to exhibit quantum properties were discussed by Prof. Paul McEuen, University of California at Berkeley and Lawrence Berkeley National Laboratory. Nanostructures discussed included quantum dots and similar structures fabricated by lithography or molecular beam epitaxy, various nanocrystals in the range of 10 to 100 atoms in diameter, and monolayers and other self-assembled structures, but most attention was given to carbon nanotubes, particularly to their interesting electronic properties. Those properties present great opportunities, but major problems include the inability to tell by casual inspection, with AFM for example, whether a given nanotube is metallic or semiconducting, or to purify or prepare nanotubes with uniform properties, such as semiconducting tubes of a given length. Prof. McEuen concluded that the best way to solve problems in nanotechnology is to use all the various technologies available, and that the greatest obstacle to success is that scientific training is too fragmented.
Bio-nanotechnology: Lessons from Nature, presented by Dr. David S. Goodsell, Scripps Research Institute, surveyed the only way complex molecular machinery can currently be fabricated using biology. The mechanisms of evolution place strict limitations on the types of molecular machinery biology can create, but modern cells have had over three billion years to exploit the possibilities within those limitations. Dr. Goodsell demonstrated through an overview of the properties of biological molecules how evolution has produced machines that would be very difficult to design. For example, self-assembly of biological machines requires large complementary surfaces with many weak interactions, yielding flexible machines in which the resulting motion is essential for the function of the machine. One trick that biology has exploited is symmetry, producing complex associations of proteins having functions the individual proteins lack.
From Foresight Update 39, originally published 30 December 1999.
Foresight materials on the Web are ©1986–2018 Foresight Institute. All rights reserved. Legal Notices.