Application of Giant Unilamellar Vesicles as Templates for Conducting Nanostructures
J. G. Linhardt*, a, H. Bowmana, E. Evansb, D. A. Tirrellc
aPolymer Science and Engineering Department, University of Massachusetts, Amherst, MA 01003 bDepartment of Physics and Pathology, University of British Columbia, Vancouver, BC Canada V6T 1Z1 c Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA 91125
Over the last decade, there has been a great deal of development in the production of meso- and nanoscale structures. Structures with reduced dimensions often exhibit properties that are quite distinct from those of the bulk material, particularly when the transverse dimension approaches the electronic wavelength (Fermi wavelength) of the material. The wavelike nature of the electron becomes evident with the appearance of quantized conductance, where conduction occurs through distinct electronic states or energy levels. In addition, these small structures frequently show charging effects characterized by the presence of a Coulomb blockade. Production of small structures of this type will continue to impact electronics development.
The self-assembly of carbon nanotubes has been extensively studied, and a variety of techniques have been developed for their production. Researchers are now gaining control over the diameter, number of concentric shells, and chirality of the formed tubes [1, 2]. Recently, evidence of both single electron charging and resonant tunneling through quantized energy levels was shown in carbon nanotubes [3, 4].
In addition, a number of templating approaches to produce mesoscopic structures have been developed. The spontaneous self-aggregation of certain chiral lipids into submicron tubules has been used to produce templates for the deposition of metals , silica , and metal oxides . Gin and coworkers have used lyotropic liquid-crystalline monomers that form hexagonally packed aqueous channels in which poly(p-phenylene vinylene) (PPV) can be formed from its water soluble precursor . Martin and coworkers developed a templating procedure where nanofibrils and nanotubes are formed inside the pores of track-etch membranes. These polycarbonate or polyester membranes have been used to template metals, semiconductors, and organic conductors such as poly(pyrrole), poly(3-methylthiophene), and poly(aniline) .
Despite the advances made in these studies, their limitations include modest control of the final dimensions of the structure, and difficulty in integrating these structures into circuits with macroscopic electrodes. We believe that the method of using bilayer membrane templates for producing nanoscale conduits and networks will circumvent the above problems. This technique was recently demonstrated , and we anticipate that point to point attachments can be easily made by utilizing the fluid properties of a bilayer membrane.
Giant unilamellar vesicles are used as a template to produce flexible polymeric wires with diameters in the range from 20 to 200 nm, controllable by adjusting membrane tension. A lipid tubule can be pulled from a feeder vesicle when the vesicle is attached to a substrate and retracted using micromanipulators. In order for a tube to be pulled, there must be a reservoir of excess lipid bilayer. This is accomplished by placing the hydrated vesicles in a hyperosmotic solution causing slight dehydration. At this point, the excess bilayer is drawn into the tip of the micropipette, forming a projection as shown in Figure 1A. Then, the vesicle is brought into adhesive contact with a substrate. The substrate is decorated with a ligand specific to the head group of a lipid present in the bilayer. In previous work, the bacterial adhesion complex biotin-avidin was chosen as a prototype to demonstrate attachment to a surface. An avidin coated microsphere held by a second pipette was brought into contact with a vesicle containing a small amount of biotinylated lipid. When the feeder vesicle was retracted, a nanometer size tubule was formed. It was visualized with an optical microscope using epi-illumination by excitation of a fluorescently labeled lipid, as shown in Figure 1B.
Figure 1. Video microscope images of a lipid bilayer vesicle (~20 µm diameter) held by micropipette suction and tethered to a solid microsphere (~4 µm diameter) by an invisible nanotube of bilayer (~40 nm diameter) pulled from a vesicle surface. (A) A bright field image shows only the vesicle and microsphere. (B) Epi-illumination is used to excite fluorescence from labeled lipids doped in the bilayer emanating from the vesicles. [Note: The tube appears thick in the fluorescent image due to optical diffraction (~0.5 µm) but its actual diameter is much smaller (~40 nm)]
The resulting nanotube can be stabilized by photopolymerization of the lipid sheath or the lumenally confined prepolymer. If the structure is not stabilized, membrane tension will simply pull the tubule back into the feeder vesicle upon release. In the described work, poly(ethylene glycol)1000 dimethacrylate (PEGDMA 1000) was photo-crosslinked using a standard light-sensitive polymerization initiator. Examples of structures formed from the polymerization of PEGDMA encapsulated in lipid templates are shown in Figure 2.
Figure 2. Bright-field video images of rigid casts of vesicle and tube shapes after photopolymerization and detergent removal of the lipid bilayer template. (a) The cross-linked PEGDMA replica of a bilayer vesicle (diameter ~20 µm) aspirated initially into a micropipette. (b) The cross-linked PEGDMA core of a large bilayer tube (diameter ~0.5-1.0 µm) connected to the polymerized feeder vesicle. (c) A small-bore pipette pulls on the polymerized cylinder to demonstrate its mechanical strength. (d) After release, the tube relaxes and coils loosely like a flexible "rope."
This demonstration of the use of bilayers as templates to create stable tubes opens the door to a variety of processing techniques to create functionalized nanostructures. The versatility of this system allows alteration of the lipid sheath or encapsulation of a wide range of water soluble components in order to obtain desired properties. Our current efforts focus on stabilizing tubes between gold electrodes and making them electrically conductive. We have recently demonstrated attachment of lipid nanotubes to gold pads utilizing the biotin-avidin adhesion complex. This has involved a number of processing steps including: deposition of an amine functionalized self-assembled monolayer (SAM) on the gold surface, reaction of the amines with succinimidyl-biotin, and reaction of the biotin with avidin. This functionalized surface can be used to attach a giant vesicle that contains a small amount of biotinylated lipid and allows the pulling of nanotubes between gold pads (Figure 3A and B). We would like to develop single step attachment to the gold surface using the interaction between gold and sulfur. The adsorption of thiol-containing compounds onto gold has been widely studied [11, 12]. Our initial attempts using phosphatidylthioethanol as a component in the membrane failed to demonstrate adhesion of the bilayer. We thought that in this case, the thiol group did not extend far enough away from the membrane, remaining buried in the bilayer. Experiments are currently underway in which we are using a thiolipid where the thiol group is connected to the head group of the phospholipid via a short hydrophilic spacer. Preliminary results show better adhesion to gold. After the lipid tubule is attached to gold pads, and stabilized by photopolymerization of encapsulated material, we would like to visualize structures using electron or atomic force microscopy. Electron microscopy on polymerized PEGDMA structures is currently being carried out. Attempts at making the nanotubes conducting will include the encapsulation and crosslinking of water soluble conductors, metallization via electroless deposition, and the reduction of encapsulated metal salts to form continuous wires. Recently, we have shown that a water soluble PPV (1) can be encapsulated inside membranes.
Figure 3. (a) Bright-field image of the micropipette held feeder vesicle being brought into adhesive contact with one of two lithographically defined gold pads (5 µm X 5 µm) that are 60 µm apart. A nanotube can be drawn between gold pads, with the use of micromanipulators, that is invisible in bright-field microscopy, but becomes visible in fluorescence microscopy (b) by excitation of a fluorescent lipid doped in the bilayer. (Micrographs taken by Howard K Bowman)
We would like to thank Prof. Mark Tuominen for help in preparing lithographic substrates and for many helpful discussions and Prof. Robert H. Grubbs for providing us with a sample of water soluble
PPV (1). We would also like to thank Prof. Horst Vogel for providing a sample of a thiolipid synthesized in his group.
Iijima, S. Nature 1991, 354, 56-58.
Thess, A.; Lee, R.; Nikolaev, P.; Dai, H.; Petit, P.; Robert, J.; Xu, C.; Lee, Y. H.; Kim, S. G.; Rinzler, A. G.; Colbert, D. T.; Scuseria, G. E.; Tomanek, D.; Fischer, J. E.; Smalley, R. E. Science 1996, 273, 483-487.
Bockrath, M.; Cobden, D. H.; McEuen, P. L.; Chopra, N. G.; Zettl, A.; Thess, A.; Smalley, R. E. Science 1997, 275, 1922-1925.
Tans, S. J.; Devoret, M. H.; Dai, H.; Thess, A.; Smalley, R. A.; Geeriigs, L. J.; Dekker, C. Nature 1997, 386, 474-477.
Schnur, J. M. Science 1993, 262, 1669-1676.
Baral, S.; Schoen, P. Chem. Mater. 1993, 5, 145-147.
Archibald, D. D.; Mann, S. Nature 1993, 364, 430-433.
Smith, R. C.; Fischer, W. M.; Gin, D. L. JACS 1997.