No Job Name


1 1171 Chem. Rev. 2005, 105, 1171 − 1196 New Approaches to Nanofabrication: Molding, Printing, and Other Techniques † † † ‡ ,‡ C. Grant Willson,* Byron D. Gates, Declan Ryan, Michael Stewart, Qiaobing Xu, and ,† George M. Whitesides* Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street, Cambridge, Massachusetts 02138 and Department of Chemistry and Biochemistry, The University of Texas at Austin, 1 University Station, Austin, Texas 78712 Received August 4, 2004 6.2. Generating Nanostructures by Exposing the 1187 Contents Edge of a Thin Film 1. Introduction 1171 6.2.1. Edges by Fracturing Thin Films 1187 2. Conventional Techniques for Nanofabrication 1174 1188 6.2.2. Edges by Sectioning Encapsulated Thin 1174 2.1. Overview Films 2.2. Photolithography 1174 6.2.3. Edges by Reorientation of Metal Capped 1188 Posts 2.3. Scanning Beam Lithography 1175 1188 6.3. Summary 3. Nanofabrication by Molding and Embossing 1175 7. Self-Assembly for Nanofabrication 1189 1175 3.1. Hard Pattern Transfer Elements 7.1. Nontemplated Self-Assembly 1189 3.1.1. Step-and-Flash Imprint Lithography (SFIL) 1176 7.2. Templated Self-Assembly 1189 3.1.2. Nanoimprint Lithography (NIL) 1177 7.2.1. Templating from Molecules 1189 3.2. Soft Pattern Transfer Elements 1177 7.2.2. Templating from Particles 1190 3.2.1. Replica Molding (RM) 1178 1190 7.2.3. Templating Using External Forces 1178 3.2.2. Solvent-Assisted Micromolding (SAMIM) 1190 7.3. Summary 1180 3.3. Fundamental Limits of Molding and Embossing 1190 8. Outlook and Conclusions 1180 3.4. Summary 9. Acknowledgments 1191 1181 4. Nanofabrication by Printing 10. References 1192 4.1. Introduction 1181 4.2. Extensions of Microcontact Printing 1181 1. Introduction CP (e- Ì 1181 CP) Ì 4.2.1. Electrical “Nanofabrication” is the process of making func- 4.2.2. Nanotransfer Printing (nTP) 1182 tional structures with arbitrary patterns having 1182 4.3. Fundamental Limits of Printing minimum dimensions currently defined (more-or-less 4.4. Summary 1183 100 nm. Microelectronic devices e arbitrarily) to be 1183 5. Scanning Probe Lithography (SPL) for and information technologies have improved and will Nanofabrication continue to improve as a result of large-scale, com- 1183 5.1. Serial Patterning of Surfaces Using SPL mercial implementation of nanofabrication. The mo- 1184 5.2. Parallel Patterning of Surfaces Using SPL tivation for these improvements is to increase the 5.3. Summary 1185 density of components, lower their cost, and increase 6. Edge Lithography for Nanofabrication 1185 their performance per device and per integrated 1185 6.1. Pattern Generation Directed by Topography circuit. 1185 6.1.1. Material Deposition at Step Edges of The smallest physical gate length of a microproces- 1 Crystalline Lattices sor currently in production is 37 nm, and current 6.1.2. Patterning at Edge-Defined Defects in 1185 half-pitch, or periodicity, of manufactured dynamic SAMs random-access memory (DRAM) is 90 nm. The In- 6.1.3. Controlled Deposition and Undercutting at 1185 ternational Technology Roadmap for Semiconductors Lithographically Defined Step Edges (ITRS), published by the Semiconductor Industry 1186 6.1.4. Phase-Shifting Edge Lithography Association (SIA), projects reaching the 45-nm node in 2010 (corresponding to transistor gate lengths 1 It is down to 18 nm and DRAM spacing of 45 nm). likely that a number of new technologies will evolve * To whom correspondence should be addressed. G.M.W.: Phone with further developments in nanofabrication. (617) 495-9430; Fax (617) 495-9857; E-mail [email protected] C.G.W.: Phone (512) 471-3975. Fax (512) Many materials with minimum dimensions on the 471-7222; E-mail [email protected] nanoscale have properties different than those ob- † Harvard University. ‡ The University of Texas at Austin. served for the bulk material. For example, quantum 10.1021/cr030076o CCC: $53.50 © 2005 American Chemical Society Published on Web 03/04/2005

2 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1172 Byron D. Gates was born in Spokane, WA, in 1975. He received his B.S. Declan Ryan was born in Kerry, Ireland, in 1976. He received his B.S. degree from the University of Limerick in 1998 and his Ph.D. degree with degree from Western Washington University in 1997 and his Ph.D. degree Professor Donald Fitzmaurice from the National University of Ireland with Professor Younan Xia from the University of Washington in 2001. (Dublin) in 2001. From 1998 to 1999 he worked as a research scientist From 2002 to 2004 he was a postdoctoral research fellow at Harvard University for Professor George M. Whitesides. He is now Assistant for NTera, a company focused on the application of nanostructured films in the areas of solar cells, displays, and sensors. From 2002 to 2004 he Professor of Chemistry at Simon Fraser University in Burnaby, BC. His was a postdoctoral fellow at the Department of Chemistry and Chemical research interests include nano- and microfabrication, materials science, Biology, Harvard University, with Professor George Whitesides. He is now surface chemistry, inorganic functional materials, microanalytical systems, a research fellow at the Department of Cell Biology, Harvard Medical and nanostructured materials. School, with Professor Rong Li. His research interests include molecular and colloidal self-assembly, microfluidics, nonlinear dynamics, and cell polarity. Qiaobing Xu was born in 1977 in Jiangsu, China. He received his B.S. degree in 1999 and his M.S. degree in 2002 from Jilin University, P. R. China. He is currently pursuing his Ph.D. degree in Chemistry at Harvard C. Grant Willson received his B.S. and Ph.D. degrees in Organic Chemistry University under the direction of George M. Whitesides. His research from the University of California at Berkeley and his M.S. degree in Organic interests include nanofabrication, material science, microfluidic systems, Chemistry from San Diego State University. He came to the University of and self-assembly. Texas from his position as an IBM Fellow and Manager of the Polymer Science and Technology area at the IBM Almaden Research Center. He is jointly appointed to the Departments of Chemistry and Chemical Engineering, where he holds the Rashid Engineering Regents Chair. Dr. Willson’s research can be characterized as the design and synthesis of functional organic materials with emphasis on materials for microelectronics. George M. Whitesides was born in Louisville, KY, in 1939. He received his A.B. degree from Harvard University in 1960 and his Ph.D. degree with Professor John D. Roberts from the California Institute of Technology in 1964. He was a member of the faculty of the Massachusetts Institute of Technology from 1963 to 1982. He joined the Department of Chemistry of Harvard University in 1982, where he is now a Woodford L. and Ann A. Flowers University Professor. His research interests include physical organ- ic chemistry, materials science, biophysics, complexity, surface science, microfluidics, self-assembly, micro- and nanotechnology, and cell-surface biochemistry. A photograph of George M. Whitesides can be found on p 1105. Michael D. Stewart received is bachelor’s degree from Vanderbilt University 2 - 5 dots can exhibit single-electron tunneling, carbon in Chemical Engineering. In 2003 he earned is Ph.D. degree in Chemical nanotubes can have high electrical conductivity and Engineering at the University of Texas at Austin under his advisor, Dr. 12 6 - mechanical strength, and thin polymer films can Grant Willson. He is currently employed by Molecular Imprints, Inc. as a have glass-transition temperatures higher or lower staff researcher and university liaison. His research interests include 13 - 16 photoresist materials and functional polymers for imprint lithography. than thick films. There is an expectation that

3 1173 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication to photolithography in manufacturing. Tools for new technologies will emerge from the fabrication of nanostructures and nanostructured materials and molding on the nanometer scale are just entering also that nanofabrication will have new applications commercial production. These techniques create op- beyond information processing and storage in areas portunities for fabrication on nonplanar surfaces 17 - 19 such as optics, biomedicine, and materials science. (particularly smooth, curved surfaces) and over large areas and may offer competition in nanofabrication Methods used to generate nanoscale structures and where cost and materials make photolithography nanostructured materials are commonly character- difficult. They are probably the only techniques that ized as “top-down” and “bottom-up”. The top-down are applicable to biological materials and to sensitive approach uses various methods of lithography to organic and organometallic materials. The ability of pattern nanoscale structures. This approach includes s to s conventional or unconventional any technique serial and parallel techniques for patterning features s prototype nanoscale structures rapidly and inexpen- s over length scales typically in two-dimensions (2D) sively will be a factor that influences the acceptance approximately 4 orders of magnitude larger (in linear of that technique. Unconventional techniques have dimension) than an individual structure. The bottom- the potential to be the ultimate, low-cost method for up approach uses interactions between molecules or nanomanufacturing; approaches based on reel-to-reel colloidal particles to assemble discrete nanoscale processing are particularly important for low-cost structures in two and three dimensions. processes. Unconventional approaches are also opera- This manuscript first briefly reviews “conventional” tionally much simpler to use than are conventional techniques for nanofabrication; this review serves as techniques and thus help to open nanoscience and background for discussions of “unconventional” tech- nanotechnology to exploration by a wide range of niques. These top-down techniques include photoli- disciplines, especially those historically weakly con- thography and scanning beam (or maskless) lithog- nected to electrical engineering and applied physics. raphy (e.g., electron beam and focused ion beam lithography). The limitations of these conventional We focus this review on research published in approaches s such as their high capital and operating nanofabrication during the interval from 1999 to mid- costs, the difficulty in accessing the facilities neces- 2004 and to specific techniques demonstrated to be sary to use them, and their restricted applicability capable of patterning a substrate with features e 100 s motivate the to many important classes of problems nm in both lateral and vertical dimensions. In gen- exploration and development of new, or “unconven- eral, this definition excludes techniques with control tional”, nanofabrication techniques. These unconven- over nanoscale features in only one direction, such tional approaches, of course, have limitations of their as the deposition of thin films. Inorganic thin films, 51 own. - or Langmuir self-assembled monolayers (SAMs), Blodgett (LB) films offer precise control over a struc- Unconventional nanofabrication s the focus of the ture only in the vertical direction. These materials can, review s includes both top-down and bottom-up ap- however, sometimes be used to fabricate two-dimen- proaches. We discuss advances in unconventional 20,21 sional structures with nanoscale lateral dimensions. nanofabrication within the areas of molding, 22,23 24 - 27 embossing, scanning probe lithog- printing, The first area of unconventional nanofabrication - 35 43 34 - 28 raphy (SPL), and self- edge lithography, that we review is a set of techniques that uses organic 44 - 47 assembly. The first three techniques are prima- materials to replicate nanoscale patterns or masters. rily top-down approaches to nanofabrication. Scanning These patterns are transferred into the materials by probe lithography and self-assembly, however, bridge molding, embossing, or printing. The second area of top-down and bottom-up strategies for nanofabrica- unconventional nanofabrication that we review (al- tion; these two techniques often use templates fab- beit in less detail) is scanning probe lithography ricated by top-down methods to direct the bottom- (SPL). Techniques based on SPL are serial but can 50 48 - up assembly of components. pattern features on a surface with atomic resolu- 28,52 “Conventional” and “unconventional” techniques tion. We also sketch advances in the development are at different stages of development. Conventional of a parallel approach to SPL. Our review of SPL is techniques for nanofabrication are commercially brief as this technique has been reviewed else- 32,34,53,54 available and widely implemented in manufacturing. where. s The last two areas that we review Conventional techniques have relatively high cost edge lithography and self-assembly s are more limited and low throughput; they are also largely restricted than conventional lithography in generating arbi- to planar fabrication in semiconductor materials and trary patterns but are promising approaches to low- are incompatible with many problems in nonstandard cost, regular arrays of nanostructures. We believe fabrication (e.g., fabrication on nonplanar substrates, that these approaches will be useful in research large area and low-cost fabrication, and three- laboratories wishing to explore ideas in nanoscience. dimensional fabrication). Conventional fabrication There are many forms of edge lithography; generally, techniques also expose substrates to corrosive they are techniques in which the edges of one pattern etchants, high-energy radiation, and relatively high become the features of a second pattern. One ap- temperatures. Alternative techniques are necessary proach to edge lithography converts films that are when patterning relatively fragile materials, such as thin in the vertical direction into structures that are organic materials (especially biological materials) thin in the lateral direction. A second approach to other than photoresist. edge lithography transfers the edges of a patterned thin film into another material. Self-assembly (both Unconventional routes to nanofabrication are often templated and untemplated) offers a final set of new followed in research; they may also offer alternatives

4 Gates et al. 1174 Chemical Reviews, 2005, Vol. 105, No. 4 Table 1. Capabilities of Conventional and Unconventional Nanofabrication Techniques current capabilities (2004) a technique minimum feature resolution pattern 1, b photolithography 37 nm 90 nm parallel generation of arbitrary patterns 88, c scanning beam lithography 20 nm serial writing of arbitrary patterns 5 nm d 116,123,168, molding, embossing, and printing  5 nm 30 nm parallel formation of arbitrary patterns 28,52 scanning probe lithography serial positioning of atoms in arbitrary patterns < 1 nm 1 nm 39, e edge lithography 8 nm 16 nm parallel generation of noncrossing features 353 - 357, f self-assembly 1 nm > 1nm > parallel assembly of regular, repeating structures b a Refers to the minimum demonstrated lateral dimension. A resolution (pitch) of 45 nm is projected for 2010 using 157-nm c light, soft X-rays, or optical “tricks” (e.g., immersion optics). Obtained with a focused ion beam. Limited by photoresist sensitivity e d and beam intensity. Limited by available masters and, ultimately, van der Waals interactions. Potentially smaller sizes could f be obtained using atomic layer deposition. Self-assembly produces structures with critical feature sizes from 1 to 100 nm or larger. approaches to nanofabrication. We believe that tem- plated self-assembly will be very important in nano- science but is early in its development. Table 1 compares the current capabilities of con- ventional and unconventional methods for patterning nanostructures. This table summarizes the current minimum feature size (minimum lateral dimension), the highest resolution (pitch), and the types of patterns that can be generated reproducibly by each technique. The sections of the review that follow also discuss the current applications and limitations of these techniques as well as areas that may lead to further advances in nanofabrication. 2. Conventional Techniques for Nanofabrication 2.1. Overview The microelectronics industry has developed a sophisticated infrastructure for patterning nanoscale features by conventional lithography. There are two Schematic illustration of the fabrication of Figure 1. dominant methods for conventional lithography: pho- topographically patterned surfaces in hard materials by tolithography and particle beam lithography. Photo- conventional photolithography and electroplating. lithography is the method of choice for manufacturing in the microelectronics industry. The most advanced lithography rather than for actual device fabrication. photolithographic systems project collimated light Writing time for scanning beam lithography depends through a quartz plate that supports a patterned on pattern density and feature size. Patterning dense chromium coating. The chromium mask has openings 2 arrays of sub-20-nm features over an area of  1cm with linear dimensions approximately four times requires  24 h; this rate of patterning restricts larger than the final image projected onto a photo- scanning beam lithography techniques to small areas resist located at the focal plane. This projection 2 or low densities of features. lithography can expose an  8cm area of photoresist coated on a planar substrate s typically a semicon- in a few seconds. A photoresist is an ductor wafer s 2.2. Photolithography organic material that cross-links and becomes in- In current semiconductor nanofabrication photoli- soluble or that changes chemically and becomes more thography can pattern 37-nm wide features with 193- soluble in a basic solution upon exposure to high- 55 - 57 nm wavelength light. The microelectronics industry energy short-wavelength light (e.g., UV light). plans to pattern minimum features below 37 nm The exposed photoresist is immersed in solvents 1 using photolithography. Continuing this trend with that dissolve the exposed (positive photoresist) or 193-nm light will require optical proximity correction unexposed (negative photoresist) regions and provide (OPC) or phase-shifting mask technology, which patterned access to the surface of the substrate. The significantly increases the cost of photomasks. patterned photoresist masks the substrate during a subsequent step that chemically modifies the exposed Another potential route to features with sub-50- regions of the substrate (Figure 1). A modern “step- nm resolution using 193-nm light is “immersion 59 64 - and-scan” photolithography system can pattern over lithography”. Immersion lithography is analo- one hundred 300-mm diameter wafers per hour with gous to the better known concept of immersion 65 65-nm resolution; it can also cost tens of millions of microscopy often used with biological specimens. 58 dollars. Imaging resolution for immersion microscopy is Scanning beam lithography is a serial process most improved by increasing the refractive index of the often used to produce photomasks for projection medium between the imaging lens and the imaging

5 1175 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication 65 charge repulsion makes small, high-current beams plane. Increasing the index of the fluid between the unstable). These changes increase the time necessary final lens element and the imaging plane improves depth of focus while also allowing lenses with larger to achieve the same imaging dose. Some improve- capture angles (numerical aperture) to be used in the ments can be realized by using very sensitive resists. imaging system. Switching the immersion fluid from Resists that require a lower dose of electrons or ions, n air ( n ) 1.47 at 193 nm) allows the ) 1) to water ( however, usually have lower resolution than photo- lens systems to be designed with numerical apertures resists that require a higher dose. A chemically approaching 1.3, thus significantly improving ulti- amplified photoresist requires a low dose of electrons 2 66 mate achievable imaging resolution. Ì  10 C/cm ( for an accelerating potential of 100 kV) The technical 81 > 50 nm. challenges of contacting the photoresist-coated sub- to pattern features with a resolution A strate and mask with water (or another solvent) and photoresist requiring a higher dose of electrons, such fabricating complex, aspheric, catadioptric (both re- as poly(methyl methacrylate) (PMMA), however, can flective and refractive) lenses must be solved before have resolution below 20 nm but requires a dose of 82 2 immersion lithography can be implemented, but this 400  Ì C/cm (at 50 kV). A cold developer ( < 10 °C) technology offers a potential route to high-volume may improve the resolution and clarity of features 83 production of devices with sub-50-nm resolution. in both types of resists. To pattern still smaller features, photolithography A focused ion beam can write patterns into a 84 will require further advances, such as decreasing the photoresist or directly onto the substrate. This 67,68 imaging wavelength to 157 nm or to soft X-rays technique can “mill” substrates by selectively remov- (  13.5 nm) s known in the microelectronics industry ing material through ion bombardment or create 69,70 as extreme ultraviolet (EUV) light. The shift to patterns in an additive process by ion deposition or - 85 87 shorter wavelengths of light requires new photore- a localized chemical vapor deposition. This li- sists to alter the wavelength sensitivity and resolu- thography technique can pattern features in a semi- 73 - 70 tion of the resist as well as new light sources and, 20 nm and with  conductor with resolution down to 88 especially, new types of optics based on reflection the smallest lateral dimensions down to  5 nm. - 74 77 rather than transmission to focus the light. Common sources of contamination in FIB lithography are from implanted ions or material displaced from Photolithography has a number of advantages over the substrate after milling. Implantation of ions can, scanning beam lithography in nanofabrication, but however, be useful for transistor fabrication and the time and cost required to fabricate the photo- repair, and the ability to write with different ions is s typically patterned by scanning beam lithog- mask potentially useful in tuning the properties of elec- raphy can be a significant drawback. There is, s tronic nanostructures. however, one photolithographic method that can produce simple patterns (e.g., diffraction gratings) without using a photomask. This process is inter- 3. Nanofabrication by Molding and Embossing 78,79 ferometric lithography, which involves the con- A number of different procedures s molding, em- structive and destructive interference of multiple bossing, and printing have been developed for pat- s laser beams at the surface of a photoresist. This terning nanoscale structures. We divide molding and method does not require a photomask or most of the embossing techniques into two categories: (i) molding expensive projection optics, but the projected patterns and embossing of nanostructures with a hard mold are restricted to regularly spaced arrays of lines or and (ii) molding and embossing of nanostructures dots. Some of the smallest features s patterns of 40- with a soft (elastomeric) mold. Molding involves s nm wide parallel lines separated by 57 nm produced curing a precursor (usually a monomer or a prepoly- by photolithography have, however, been generated 80 mer) against a topographically patterned substrate. using interferometric lithography. This method of pattern transfer is used by techniques 20 such as step-and-flash imprint lithography (SFIL), 2.3. Scanning Beam Lithography 21,89 replica molding (RM) with a soft mask, mi- Scanning beam lithography is a slow process rela- 90 Ì crotransfer molding ( TM), and micromolding in tive to photolithography. This serial technique can, 91 capillaries (MIMIC). Embossing (or imprinting) however, generate high-resolution features with ar- techniques transfer a mold with a structured topog- bitrary patterns. There are three main classes of raphy into an initially flat polymer film. These scanning beam lithography: (i) scanned laser beams 92 techniques include nanoimprint lithography (NIL) with  250-nm resolution are the least expensive; (ii) 93 and solvent-assisted micromolding (SAMIM). focused electron beams with sub-50-nm resolution (depending on tool settings and the choice of photo- 3.1. Hard Pattern Transfer Elements resist) are expensive to purchase and maintain; and (iii) focused ion beam (FIB) systems with sub-50-nm Techniques such as relief printing and injection resolution are primarily (and extensively) used in molding use hard molds or stamps to transfer a research. Typically, high-resolution photomasks are patterned topography into a monomer, prepolymer, patterned using laser writers and electron-beam or polymer substrate. Commercialized processes that tools. use hard molds include patterning of compact discs (CDs), digital versatile discs (DVDs), diffraction There are tradeoffs for high-resolution patterning gratings, holographic gratings (e.g., for identification with an electron or ion beam. Increasing the resolu- markings on credit cards and currencies), and plastic tion requires decreasing the diameter of the particle 94 - 101 parts. beam, which decreases the beam current (charge - Hard molds can also transfer nanoscale

6 Gates et al. 1176 Chemical Reviews, 2005, Vol. 105, No. 4 viscosity, photocurable liquid or solution fills the void spaces of the quartz mold. The solution consists of a low-molecular-weight monomer and a photoinitiator. Exposing this solution to UV light polymerizes and hardens the precursor while in contact with the mold. Removing the mold leaves a topographically pat- terned (inverse) replica on the substrate. Step-and-flash techniques avoid incomplete mold filling a problem for embossing polymers with a s rigid mold s by using a monomeric fluid with a low viscosity ( 5 cPs). This lithographic technique is also < insensitive to the effects of pattern density reported 114,115 for NIL. Hydrodynamic forces, however, prevent complete displacement of the fluid by the mold, even Figure 2. (A) Schematic illustration of the procedure for for low-viscosity precursors. The incomplete displace- 20 step-and-flash imprint lithography (SFIL). Scanning ment of fluid leaves a residual layer of cured material electron microscopy (SEM) images of (B) 40-nm wide lines between patterned features. The substrate and mold and (C) 20-nm wide lines patterned by step-and-flash must be parallel and both must be rigorously flat to lithography. The mold for the pattern in B was used for more than 1500 previous imprints. (Reprinted with per- ensure that the residual layer is uniform in thickness mission from ref 20. Copyright 1999 SPIE.) over the entire imprinted area. This residual layer can be removed by etching (e.g., RIE). features into polymer films for nanofabrication. Nano- An important consideration in all nanomolding structured hard molds are prepared by transferring techniques is the lifetime of the mold. A release layer a structure patterned in photoresist into a hard reduces the surface free energy of the mold and substrate using reactive ion etching (RIE), wet minimizes adhesion of cross-linked polymer to the chemical etching, or electroplating (Figure 1). Hard mold. If the release layer fails, the cured polymer can molds have been fabricated out of quartz, silicon, and adhere to the mold and foul its surface or break its 106 - 20,92,102 metals. The most commonly used hard features. The first reported release layer was a materials have been quartz and silicon. The smallest 20 fluorinated silane with a lifetime of less than 100  features transferred into a silicon mold are 10-nm patterned substrates. New surface treatments have wide lines written using electron-beam lithogra- been developed (but not described in detail) for 107 phy. The smallest features produced in a quartz patterning that claim to provide more than 1500 - 110 108 mold are 20 nm. 116 consecutive substrates. A hard mold offers a number of advantages for Figure 2B,C shows topographic patterns generated nanofabrication. The rigid mold of silicon or quartz using the SFIL technique. The SFIL process patterns retains nanoscale features with minimal local defor- features down to at least 20 nm across a field size of 2 116 mation (the pressures required for embossing can 6.25 cm This process can pattern per molding step. cause long-range distortions in the substrate). A hard dielectric gates for the fabrication of a metal oxide 104 mold is thermally stable at temperatures used to semiconductor field-effect transistor (MOSFET) cross-link most polymer precursors. Silicon and quartz and is compatible with semiconductor device manu- molds are chemically inert to precursors used to mold facturing. Step-and-flash imprint lithography is also polymers. Surface fouling of the pattern transfer being developed to pattern curved surfaces and element is, however, dependent on the surface free topographies with multiple depths in a single 117,118 energies of the mold and the polymer, and a fluo- step. rosilane [e.g., CF (CF ) SiCl ] is usually co- ) (CH 3 2 2 6 3 2 Step-and-flash imprint lithography uses a rigid, valently linked to the surface of a hard mold to transparent mold to print features at a constant facilitate the release of the mold from the polymer 22 °C) with low applied pressures  temperature ( and reduce surface fouling. One of the main differ- 2 < ( 1 lb/in. ). This combination of factors gives SFIL ences between silicon and quartz substrates is that an inherent advantage in its potential for fine layer- quartz is transparent to ultraviolet and visible wave- to-layer alignment, which is necessary for multilayer lengths of light and silicon is not. It is, therefore, device fabrication. Distortions caused by differential possible to align the quartz mold optically to features thermal expansion are not an issue since the mold on the underlying substrate and initiate photoin- and substrate are not heated. The low printing duced cross-linking of a molded prepolymer by ex- pressure allows imprinting on brittle substrates and posure to UV light through the pattern transfer reduces distortion caused by flexing of the mold or element. Step-and-flash imprint lithography exploits substrate. The alignment accuracy in SFIL has been these properties of a quartz mold for nanofabrica- 119 ( Û reported as high as ). 10 nm (3 20 tion. The development of SFIL techniques has focused primarily on semiconductor nanofabrication. The 3.1.1. Step-and-Flash Imprint Lithography (SFIL) successful implementation of SFIL to other applica- tions will require the development of new photocur- Step-and-flash imprint lithography is a technique able monomers and appropriate economics for the that replicates the topography of a rigid mold using application. The precursors currently available for a photocurable prepolymer solution as the molded 20,104,105,111 - 113 molding do not include the functional materials material. In SFIL (Figure 2A) a low-

7 1177 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication Figure 4. Schematic illustration of the formation of topographically patterned molds (or stamps, depending on the application) and replication of the master into a third functional material. Figure 3. (A) Schematic illustration of nanoimprint 22,131 lithography (NIL). The SEM images (B,C) show typical stress also presents a challenge for alignment during NIL experimental results. (B) Fresnel zone plates with a multilayer fabrication. Thermal cycling of the sub- 125-nm minimum line width. (Reprinted with permission strate also limits throughput to one imprint per 10 from ref 131. Copyright 2000 American Institute of Phys- min. A laser-induced flash heating process may, ics.) (C) Metal dots with a 10-nm diameter and a periodicity 132,133 however, reduce thermal cycling times. To pre- of 40 nm. (Reprinted with permission from ref 22. Copy- right 1997 American Institute of Physics.) vent temperature cycling of the substrate, materials 134 such as poly(dimethylglutarimide) are imprinted required for optoelectronics and ferromagnetic coat- at room temperature at an unstated pressure and ings. polystyrene (PS) printed with pressures above  300 135 atm. Chou et al. imprinted patterns at room 3.1.2. Nanoimprint Lithography (NIL) temperature using a transparent mold and UV- 136 23,103 initiated cross-linking of the molded polymer. The Nanoimprint lithography refers to the pres- difference between UV-based NIL and SFIL is not sure-induced transfer of a topographic pattern from distinct. a rigid mold (typically silicon) into a thermoplastic The high viscosity of the polymer films presents polymer film heated above its glass-transition tem- another challenge for nanofabrication using NIL. perature (Figure 3A). Another term for this method There appears to be an optimal pattern size and is “hot embossing” since the process involves heating 137 feature density for NIL. Embossing micrometer- the molded polymer above its glass-transition tem- scale features can be more challenging than nano- perature. For example, to transfer the pattern from scale features: filling large recesses within the mold a mold into a thin polymer film of PMMA by NIL requires more lateral displacement of the polymer requires heating the polymer film above 110 °C.  than smaller recesses and thus increased process Nanoimprint lithography is a parallel plate print- times (or higher temperatures and pressures). A mold ing process. Entire 100-mm wafers have been pat- containing a range of feature sizes may introduce 102,120 terned in a single imprinting step. A second distortions within the embossed film because of version of NIL uses “rolling molds” to emboss sheets uneven displacement of the polymer and trapping of 121 with a repeating pattern. A third approach uses a air bubbles. The thickness of the residual layer can modified commercial flip-chip bonder to imprint also vary across the imprinted region depending on 122 several fields across a substrate. the pattern density or layout of the pattern. Nonuni- Nanoimprint lithography can mold a variety of formities in the residual layer present a challenge polymeric materials (Figure 3B,C) and pattern fea- for transferring the pattern uniformly into the un- 22,123 5nm  tures as small as and aspect ratios up to derlying substrate by RIE. 124  20 (height-to-width). Materials that have been 125 patterned successfully include biomolecules, block 3.2. Soft Pattern Transfer Elements 126 127,128 copolymers, and fluo- conducting polymers, 129 rescently labeled polymers. This process has been Techniques that prepare a soft mold or stamp by used to pattern components for a range of microelec- casting a liquid polymer precursor against a topo- 130,131 tronic, optical, and optoelectronic devices. For graphically patterned master are commonly referred 24,27 example, gate lengths in a MOSFET have been to as soft lithography (Figure 4). A number of defined by NIL with a minimum feature size as small polymers could be used for molding. Elastomers are 102 as 60 nm. a versatile class of polymers for replication of a topographic master. The most widely implemented Nanoimprint lithography has made great progress and successful elastomer for nanofabrication is poly- in a relatively short period of time. One of the 24,27,54 (dimethylsiloxane) (PDMS). important issues still to be resolved is the useful Other elastomers tested as pattern transfer elements include polyure- lifetime of the mold. Presently, nanoimprint molds thane (PU), polyimide, and cross-linked Novolac require replacement after  50 consecutive im- 138 130 resins (a phenol formaldehyde polymer). prints. Other Heating and cooling cycles and high pres- siloxane elastomers are being developed for soft sures (50 130 bar), applied during embossing, pro- - lithography, such as block copolymer thermoplas- duce stress and wear on nanoimprint molds. This

8 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1178 141 147 139,140 molecular-weight molecules often dissolve in PDMS. tics and fluorocarbon-modified siloxanes. A particularly interesting new class of polymers is Oligomers and prepolymers (and polar or highly 142,143 highly fluorinated elastomers, fluorinated monomers), however, typically do not which show ex- penetrate into PDMS. This low solubility extends the cellent release properties and resistance to swelling by organic solvents and monomers. Molds from other usefulness of each PDMS mold beyond 20 replica- polymers are also finding applications in nanofabri- tions; ultimate limits to the number of cycles over 144 cation, including polyolefins, which a mold can be used are not established. A acrylate-based UV- 145 benefit of molding with PDMS is the ability to mold curable polymers, and an amorphous fluoropoly- 146 mer. against nonplanar, rigid, and soft topographic sur- faces (unlike a hard mold, which requires a planar, Poly(dimethylsiloxane) has a number of useful rigid surface). properties for nanofabrication. This material is du- Replica molding can produce numerous molds, rable, unreactive toward most materials being pat- replicas, and patterned surfaces from each master terned or molded, chemically resistant to many 147 and provide capabilities for nanofabrication not com- solvents, and transparent above a wavelength of monly available in an academic setting. Replication  280 nm. Commercially available kits or precursors 148,149 of the high-cost, high-resolution masters reduces the for this polymer can be obtained inexpensively. financial burden of patterning nanostructures and One of the major advantages of PDMS is that conveniently extends nanofabrication to a range of fabrication of molds or stamps (by replica molding) materials. is so inexpensive that a large number of uses may not be necessary. In fact, sometimes the mold or 30-nm lateral Replica molding has transferred  150,151 stamp becomes a disposable reagent. This mate- features from a diffraction grating, a compressed 21,89 166 rial can be deformed reversibly and repeatedly with- PDMS stamp, and a nanoscale crystal into PU. out permanent distortion or relaxation of the surface  20 nm were also Air bubbles with sizes as small as 21 167 topography. The cured elastomer has a low surface replicated into PDMS. A phase-separated block- 2 152 free energy (21.6 dynes/cm ); this low surface free with one component removed s copolymer film s energy allows PDMS to be easily released after trapped air bubbles in a regular array during molding molding. Fluorosilane chemistry can be used to - with PDMS (Figure 5A C). Features down to at least 2 12 dynes/cm  decrease the surface free energy to s 1.5 nm were also replicated into PDMS from a 163 a value similar to poly(tetrafluoroethylene) or Tef- regular pattern of vertical deflections. Periodic 142,153 lon. Another method to reduce the noncovalent patterns of rings (Figure 5D) or lines (Figure 5G) interactions of the mold substrate interface during - written into a PMMA film using an electron beam release is immersion in a solvent (e.g., methanol or were molded into h-PDMS (Figure 5E,H) and repli- 154,155 hexane) or exposure to an organic vapor. Poly- cated into PU (Figure 5F,I). Surface roughness of the (dimethylsiloxane) is chemically inert, an advantage 0.5 nm over an PMMA, PDMS, and PU are each  2 for patterning many different types of materials. area of at least 1 Ì m . The smallest features repli- 168 These materials include polymers, precursors to  cated using PDMS are 3-nm wide structures and 169 - gel materials, organic and carbons and ceramics, sol  0.5-nm vertical deflections. In the later experi- inorganic salts, colloids, biological macromolecules, ment the surface roughness is 0.2 nm for each  24,27 2 thiols, phosphonic acids, and silanes. An impor- Ì m substrate over an area of 1 . It is still not clear tant limitation of PDMS is that it absorbs many what the ultimate limit to replication using PDMS 147 nonpolar, low-molecular-weight organic compounds. (or other materials) is, but the limit, based on the The swelling of the PDMS both compromises dimen- physics of van der Waals interactions, should be less sional stability and leads to unwanted adhesion after than 0.5 nm. polymerization of monomers. Other materials may Two techniques related to RM are micromolding 142 circumvent this limitation. in capillaries (MIMIC) and microtransfer molding The tensile modulus of 184-PDMS (Dow Corning) ( Ì TM). Micromolding in capillaries can fabricate is relatively low (1.8 MPa) and limits the replication isolated structures by using capillarity to fill channels - 160 156 of nanoscale features. High-resolution masters in a PDMS mold with low-viscosity liquid solutions, 91 are reproduced accurately in composite stamps of such as a photo- or thermally curable prepolymer. 159,161 - 163 170,171 hard PDMS (h-PDMS) or UV-curable PDMS Nanosized channels can be filled with liquids. 164 -PDMS) Ó h ( with tensile moduli of 8.2 and 3.4 MPa, These nanochannels are, however, slow (or impos- respectively. sible) to fill completely since resistance to flow in pressure-driven flow increases rapidly as the size of the channel decreases. The capillary flow can be 3.2.1. Replica Molding (RM) assisted by applying a vacuum to one end of the capillaries, heating the liquid, or applying an electric Replica molding consists of three steps (Figure 4): 172,173 field. Microtransfer molding also fabricates iso- (i) creating a topographically patterned master (usu- lated structures but has been limited to a minimum ally by conventional techniques; see, for example, - 176 90,174 feature size > 100 nm. Figure 1); (ii) transferring the pattern of this master into PDMS by replica molding; and (iii) fabricating a replica of the original master by solidifying a liquid 3.2.2. Solvent-Assisted Micromolding (SAMIM) 27,89,165 precursor against the PDMS mold. The replica can be cast from a photo- or thermally curable Solvent-assisted micromolding uses an elastomeric prepolymer. Nonpolar monomers and other low- mold and an appropriate solvent to emboss polymer

9 1179 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication Figure 6. (A) Schematic illustration of solvent-assisted micromolding (SAMIM). (B) AFM image of 60-nm wide lines in a Novolac photoresist supported on a silicon 93 substrate patterned using SAMIM. Ethanol coated on the PDMS mold softened the photoresist. (Reprinted with permission from ref 93. Copyright 1997 Wiley-VCH Verlag GmbH.) (C) SEM image of square pyramidal structures 20-nm radius of showing a well-defined apex with an  R curvature ( ) molded with a composite PDMS mold using c 159 SAMIM. useful in embossing because the mold conforms to the s s over polymer substrate even when it is not planar 2 1cm > areas and allows uniform pattern transfer over that area. This process has been demonstrated for a number of polymers including Novolac photo- resists, PS, PMMA, cellulose acetate, poly(vinyl chloride) (PVC), and the precursors to conjugated - C) Replica molding of air bubbles trapped Figure 5. (A 179 27,93,177 - 167 The mild processing con- organic polymers. within a porous polystyrene (PS) surface. (Reprinted with permission from ref 167. Copyright 2003 Wiley-VCH Verlag ditions of SAMIM are compatible with patterning 178 GmbH.) (A) Schematic depiction of this replication. (B) polymer-based distributed feedback lasers and Atomic force microscopy (AFM) image of the PS- b -PMMA 177 organic light-emitting diodes (OLEDs). copolymer master after selective removal of the poly(methyl methacrylate) (PMMA), and (C) a PDMS replica of air Although SAMIM makes it possible to mold poly- bubbles trapped within this master of  20-nm wide holes. mers that are difficult to manipulate by SFIL and The inset shows a two-dimensional Fourier transform of NIL, the in-plane dimensional stability of soft masks the PDMS replica. (D - I) Replica molding of nanoscale is believed (or simply assumed ) to be lower than for 1  vertical deflections ( 5 nm) patterned in PMMA by - 163 hard masks. Whether this belief is justified by direct writing with a focused electron beam. The series practice remains to be established. The application of AFM images include the (D,G) PMMA master, (E,H) h-PDMS mold, and (F,I) polyurethane (PU) replica. of SAMIM to patterning nanoscale features has been limited primarily by the lack of appropriate masters 93 films (Figure 6A). This technique processes the rather than by the fundamental characteristics of polymer at ambient conditions with soft molds rather this process. Line widths as small as 60 nm and  than at elevated temperatures with rigid molds. For aspect ratios of at least 1:1 have been patterned in a SAMIM, a solvent swells or dissolves the polymer; Novolac photoresist (Figure 6B) and poly(vinyl pyri- swelling of the elastomeric mold is crucial to the 93 A composite elastomeric mold dine) by SAMIM. process but limited by the amount of solvent con- with a h-PDMS surface has embossed features with tained in the polymer being molded and by the minimum dimensions as small as 20 nm (Figure  solubility and mobility (usually by diffusion) of the 159 6C). solvent in the PDMS. During solvent evaporation the As with other embossing techniques, SAMIM using softened polymer conforms to the surface of the PDMS leaves a residue between isolated features PDMS mold. The gas and solvent permeability of the s with their (molding using fluorinated elastomers mold prevents nonuniform solvent evaporation and may obviate this lower interfacial free energies s trapping of air bubbles at the interface. The precipi- 142 problem ). The processing time for SAMIM depends tated polymer film retains an imprint of the surface on solvent transport through the PDMS. Swelling of topography from the mold. the PDMS by the solvent can cause distortion of the Solvent-assisted micromolding has two character- topographic features. Understanding the interactions istics useful in nanofabrication. It avoids cycling of of solvents with PDMS, such as compatibility and the temperature of the sample and thus limits swelling of the elastomer, will improve the capabili- thermal oxidation or degradation of other system 147,180 components. Elastomeric molds are also especially ties of this technique.

10 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1180 polymer due to solvent evaporation and/or cross- 3.3. Fundamental Limits of Molding and linking during polymerization. A third cause for Embossing distortion is collapse or deformation of the molded features (or features on the mold) due to mechanical The resolution of nanofabrication by molding and instability of the polymer (e.g., flexibility and polymer embossing is primarily limited by five factors: (i) the chain mobility): capillary and van der Waals interac- practicality of fabricating masters with small fea- tions can be very important at small scales. A fourth tures; (ii) the ability of a material to mold, with high cause for distortion is the forces required to overcome fidelity, the features of the master; (iii) the distortion 182 adhesion between the master and mold. One ap- of features in the transferred pattern; (iv) the swell- ing of the master by the monomers used or the proach to reducing the adhesion between a polymer solvent used to dissolve polymers; and (v) the ability mold and a silicon or quartz master is to reduce the of the molded material to fill the mold completely, surface free energy of the master by modifying its the tendency of the system to trap bubbles of gas, surface with a fluorosilane. A second approach is to the kinetics of filling of the mold, the thickness of carry out the separation between mold and molded the residue or “scum” layer (if any) between isolated material in the presence of a vapor or liquid with very 154,155 features, and related issues concerning the mold, low viscosity, such as methanol. Further work substrate, and polymer as a system. The practicality is necessary to develop surface release layers for of nanofabrication by molding and embossing is high-resolution masters. limited by release of the mold from the polymer, cycle Elastomeric molds are susceptible to distortion time, mold maintenance, and registration (or exact because of their low elastic modulus and high ther- alignment) of features on the mold with structures mal expansion coefficient. Hard molds can also on a substrate. experience distortion even with relatively minor Advances in the capabilities of conventional litho- 115,183 pressure differentials. The alignment of patterns graphic tools will continue to decrease the smallest with nanoscale precision is challenging for two lateral dimensions in the masters. For example, reasons: (i) distortion of the mold and (ii) the low direct-write electron-beam lithography can pattern contrast in refractive index between the master (e.g., 82 a minimum feature size down to 5 nm. The ability PDMS or quartz) and polymer precursor. These to arrange and assemble small molecules or particles limitations often restrict molding and embossing could also present a new strategy for fabricating techniques to fabricating structures with one or at masters with nanoscale features (further details are most two layers. Quartz molds have been aligned provided in section 7). The minimum feature size 119 ( with an accuracy as small as Û ); 10 nm (3 similar replicated by molding and embossing has, so far, values will ultimately be achievable with other hard decreased steadily with improvements in fabricating (and perhaps soft) molds. masters. A number of factors limit the fidelity of replicating nanostructures by molding and embossing. The fidel- 3.4. Summary ity is, in principle, limited by the size of the molecular precursors of the polymer and the separation between Molding and embossing are the most widely pur-  the mold and master. A minimum separation of 0.1 sued and successful techniques for unconventional 0.5 < nm is predicted by van der Waals contacts, and nanofabrication. There are a number of molding and 163,169,181 nm is suggested by current experimental data. embossing techniques that can pattern, in parallel, There is, thus, no fundamental physics-based limita- nanometer-scale features over large areas (i.e., entire tion to molding in the nanometer range, although the silicon wafers). These techniques have been used to actual limit to replica molding remains to be deter- pattern functional structures for inorganic- and mined experimentally. The size and shape of the 25,104,184 organic-based microelectronics and optics. The that is, molecules making up the molded species s formation of these structures requires a high-resolu- s material used in the polymer and the polymer (and tion master, typically generated by conventional filler) making up the mold also limits the fidelity of nanofabrication techniques. These masters can, how- pattern transfer. The granularity of matter at the ever, be inexpensively replicated by molding or 1 nm) thus provides atomic and molecular scale (0.1 - embossing. The replication process can be repeated a chemical limit to resolution. As a consequence, a number of times for a range of organic materials. small molecules such as monomers and prepolymers The number of replications is limited by surface and amorphous materials such as silica gels may give fouling, which is related to interfacial free energies. better resolution for molding than higher molecular Another limitation is the variation in thickness of the weight precursors (i.e., polymers). residual layer (the so-called “scum layer”) between The distortion of nanoscale features in the master, nanoscale features. It is challenging to transfer the the mold, or the replica product limits the number pattern uniformly into an underlying substrate by of materials that can be used in molding and emboss- RIE because of this variation in thickness. The ing for nanofabrication. One cause for distortion of number of replicas, surface area of each replica, and replicated features is a difference in thermal expan- range of materials that can be patterned by molding sion between the materials used to fabricate the and embossing have steadily increased, while the master, mold, and replica. A second cause for distor- resolution and minimum feature size have steadily tion, which influences the fidelity of replication, is decreased. polymer shrinkage upon curing a precursor to a

11 1181 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication 4. Nanofabrication by Printing 4.1. Introduction Transferring a material onto a substrate by print- ing from a topographically patterned stamp is also useful in nanofabrication. A process known as mi- crocontact printing ( CP) transfers molecules from Ì a patterned PDMS stamp to a substrate by the - 191 138,185 formation of covalent bonds. In this process a solution of molecules (e.g., 5 mM alkanethiol in ethanol) is inked onto the surface of the PDMS, and this stamp is brought into contact with an appropri- ate substrate (e.g., a thin gold, silver, or palladium (A) Schematic illustration depicting the proce- Figure 7. film). The molecules are transferred in a pattern dure for electrical microcontact printing (e- Ì CP). Kelvin defined by the topography of the stamp (i.e., the probe force microscopy (KFM) measurements of a thin film regions of atomic-level contact between the stamp CP show (B) no of PMMA after patterning charge by e- Ì and substrate) with a minimum feature size as small change in the surface topography and (C) a surface 138,192 196 - as  30 nm. The flexibility of the PDMS potential with positive features for a test pattern of 620- 221 nm diameter rings. The full-width half-maximum (fwhm) stamp and the ability to achieve conformal, atomic- 135 nm for the positive ring of charge in the plot of is  level contact between the stamp and the substrate surface potential. (Reprinted with permission from ref 221. are both advantageous for printing over large areas Copyright 2001 American Association for the Advancement 2 201 197,198 - 199 ( 50 cm > and on curved surfaces. ) This of Science.) patterning technique has been developed primarily for PDMS stamps, although recent demonstrations removing this electrode the changes in surface po- have explored surface modifications of these stamps tential across the thin dielectric were measured by 202 - 205 and other soft materials. Further details con- 225 Kelvin probe force microscopy (KFM). Figure 7B CP of SAMs on metal films are covered in Ì cerning and C shows the measured topography and corre- 51 another article of this issue. sponding surface potential, respectively, for an 80- nm thick PMMA surface patterned using a current 2 221 4.2. Extensions of Microcontact Printing density of 20 mA/cm ( + 18 V). A number of methods exist for charging electrets, Microcontact printing has patterned a number of 226 - 228 such as corona discharge and tribocharging. materials other than self-assembled monolayers Typically, these methods pattern charge in electrets 206 - 213 (SAMs), such as biomolecules, colloidal par- with micrometer-scale lateral resolution. Scanning 218 - 220 - 202,214 217 ticles, and polymers. In this section we probe techniques can pattern charge in electrets over focus on more recent extensions of Ì CP for patterning multiple length scales and have demonstrated fea- nanostructures. These extensions include patterning 2 ture densities of 7 Gbits/cm for 120-nm wide features of surfaces by electrical contact (electrical CP, or Ì 229 - 231 at a resolution of 100 nm. The writing speed 221 e- CP) Ì and release of material coatings from a 2 1cm  is, however, slow: a pattern covering requires 222 patterned stamp (nanotransfer printing, or nTP). 24 h to write. Alternatively, e- CP can pattern  Ì Both techniques use a PDMS stamp or a PDMS 2 1cm > charge over areas in less than 20 s with a substrate to achieve compliant, conformal contact and 221 100 nm. resolution of  The minimum feature size, uniform pattern transfer. the maximum area that can be patterned in one impression, and the maximum charge density that Ì Ì 4.2.1. Electrical CP) CP (e- CP remains to be established. can be patterned by e- Ì The stability of the patterned charge (for periods Electrical microcontact printing uses a flexible 221 greater than months) is unexpectedly high. electrode to pattern a thin film of a material that is The patterned, flexible electrode used in e- Ì CP is an electret (i.e., that accepts and maintains an reusable and serves as a tool for fundamental studies electrostatic potential), probably by injecting and 26,221,223,224 of thin film electrets with high lateral resolution. trapping charges. The electrode was a Recent studies have explored the application of e- CP Ì 5 nm of chromium (adhe- PDMS stamp coated with  221 2 in high-density data storage ( > 5 Gbits/cm and ) 80 nm of gold (electrode material)  sion layer) and electrostatic printing of particles, such as graphitized (Figure 7A). This flexible electrode was brought into 221,224,232,233 carbon, carbon toner, and iron oxide. Simi- contact with a thin dielectric film (the electret, a lar procedures can pattern thin-film waveguides in material such as PMMA) supported on a second 2 > 90 s over large areas ( < 1cm ) by changing the local 〈 silicon). Electrical 〉 electrode (typically n-doped 100 223 index of refraction of a doped polymer. microcontact printing uses the flexibility of the PDMS stamp to allow conformal contact between the Ì CP include buckling of Current limitations of e- top, patterned electrode and the dielectric film. A the PDMS stamp after metal deposition from thermal 221 - 30 V) was applied between the two voltage pulse (10 expansion and contraction of the surface. Deposit- ing metal onto small features or cooling the PDMS 10 s with current densities of  10 electrodes for  223 2 prevents buckling. mA/cm Other studies have explored . Charge remained in the electret where the alternative designs for the electrodes using a pat- flexible electrode contacted the dielectric film. After

12 Gates et al. 1182 Chemical Reviews, 2005, Vol. 105, No. 4 222,235 tions and transfers nanostructures in one step. Nanotransfer printing can pattern features with a lateral resolution of at least 70 nm and an edge 222,246 roughness down to 10 nm. Nanotransfer printing is well suited for transfer- ring electrodes to fragile surfaces. For example, this contact printing technique can pattern parallel lines 246 and circular dots as electrical contacts on SAMs. These discontinuous structures adhere to the sub- 235 strate under Scotch tape adhesive tests. The components of devices fabricated directly on plastic substrates include complementary inverter circuits, 235 234 organic thin-film transistors, and capacitors, 222 electrostatic lenses. This patterning technique can 236 also transfer arrays of sacrificial etch masks and Figure 8. (A) Schematic illustration of one approach to 237 stable ferromagnetic stacks of cobalt. nanotransfer printing (nTP) s transferring thin films from The morphology and continuity of the transferred a topographically patterned PDMS stamp by forming bonds metal structure is important for functional devices. between the thin film and the surface chemistry of another Uniformity of the metal film is dependent on the substrate. SEM images of (B) a 20-nm gold layer trans- ferred onto a GaAs substrate functionalized with 1,8- wetting and grain size of the metal on the stamp. A octanedithiol and (C) a multilayered stack of 20-nm thick 2-nm thick) improves the thin adhesion layer ( < layers of gold containing parallel grooves with each layer uniformity of a gold layer on the PDMS stamp and 236 oriented perpendicular to the one below. After printing 236,237,242 in the transferred layer. A metal film on an the initial gold layer each subsequent layer is transferred elastomeric stamp can crack from thermal expansion by cold welding. during metal deposition. These cracks can be pre- vented by rapid deposition of metal (rates g 0.3 nm/s terned array of nanometer-scale, in-plane edges to minimize thermal stress on the PDMS) and by 41 (further details are provided in section 6.2). 223,237 cooling the stamp. The stress in the metal film from thermal expansion is also avoided by depositing 4.2.2. Nanotransfer Printing (nTP) the metal onto a stamp with a higher thermal The process of transferring a thin, solid film from conductivity than PDMS (e.g., silicon or gallium a stamp with nanoscale-patterned features to a arsenide). The surface of these stamps must, how- substrate is referred to as nanotransfer print- ever, be modified with a release layer. Mechanical - 39,222,234 242 ing. The stamp can be either a soft or a stress during printing can also introduce cracks into hard material, such as PDMS or silicon. Figure 8A the metal structure as indicated by the arrows in 236 illustrates a typical procedure for this contact print- Figure 8B. The consequences of mechanical defor- ing technique using a PDMS stamp. The stamp was mations on nanofabrication using nTP remain to be 20-nm thick)  coated with a continuous layer of gold ( established. without an adhesion layer between the gold and the An alternative approach to printing structured 150,151 PDMS. This stamp was brought into contact with a materials is decal transfer printing. This process substrate coated with a dithiol (e.g., 1,8-octanedi- transfers a structure (e.g., PDMS membrane or 222,237 thiol). The dithiol formed a SAM on the sub- isolated PDMS features) from one planar surface to strate (GaAs in this case) and the exposed thiol group another. The PDMS decals can be made to adhere 247,248 covalently bound to the gold layer in the regions of reversibly to the first substrate (i.e., a PDMS slab) contact. Removing the elastomeric stamp from the while forming covalent bonds with the second sub- substrate left the gold layer bound to the SAM and strate. The PDMS slab serves as a handle for pat- the underlying substrate (Figure 8B). Alternatively, terning continuous or discontinuous features that are 243,244 cold welding between two metal surfaces could otherwise difficult to manipulate. Decal transfer also transfer the structured metal film. Three- printing can transfer submicrometer features, but dimensional structures can be fabricated by repeating extending this technique to nanoscale features will 236 this procedure (Figure 8C). Another method of require further investigation of the interfacial adhe- releasing the structured film relies on condensation sion between the PDMS (or other) substrate and the OH) - reactions between surface-bound silanols (Si decal. 222,234 and/or titanols (Ti - OH). Techniques relying on 4.3. Fundamental Limits of Printing noncovalent interactions between the metal film and the substrate have also been explored, although the A number of factors determine the smallest fea- minimum dimensions of transferred features are tures that can be printed by soft lithography. The 239,245 100 nm. currently > fundamental limits to printing are determined by A number of methods can pattern nanoscale metal- three main constraints: (i) minimum size of features lic structures such as narrow, periodic lines. Con- in the stamp; (ii) lateral dimensions and resolution ventional methods include lift-off, wet chemical etch- of the transferred material; and (iii) preferential ing, RIE, and shadow evaporation. These patterning adhesion of the printed material to the second techniques require exposure to high temperatures, surface. basic or acidic solutions, and/or organic solvents. The smallest feature in the stamp depends on the Nanotransfer printing avoids harsh processing condi- size of features within the master, the fidelity of the

13 1183 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication of the surface in well-defined, nanometer-scale re- molding process, and the ability of the mold to retain gions. High-resolution printing normally requires a nanoscale features. Further details on these limita- mechanically compliant stamp. The surface can be tions are provided in section 3.3. Distortion of the modified with covalently bound molecules (e.g., SAMs) stamp while in contact with the printed surface also or thin films (e.g., gold or silica). The printed material limits the minimum size of the transferred feature. can have applications as etch resists, reversible or The elastic deformation of a PDMS stamp can col- 156,158 permanent electrical contacts, patterns for use in lapse high-aspect ratio features. A composite biotechnology, and bits of information in high-density stamp of h-PDMS with a 184-PDMS backing can, data storage. Resolution is, however, limited by however, retain nanoscale features without col- 159,161 lapse. surface diffusion of printed molecules and distortion of the features within the stamp during printing. Resolution and lateral dimensions of nanoscale Different materials are being developed to improve features transferred from these stamps depends, the fidelity of pattern transfer from nanometer-scale among other things, on the interaction of the “ink” features. These printing techniques are easily imple- with the printed surface. Lateral spreading of the mented and attractive because the stamps are gener- pattern (either by transport across the surface or by ated inexpensively from readily available commercial transport through the vapor) can result from diffu- precursors. sion of molecular ink on the printed surface. For Ì CP example, alkanethiols patterned on gold by while immersing the substrate in water illustrates 5. Scanning Probe Lithography (SPL) for the effect of lateral surface diffusion of alkanethiol Nanofabrication 194 across gold on broadening of the printed pattern. In this example the lateral dimensions of the exposed Scanning probe lithography provides a versatile set gold decreased from 500 to 35 nm by a time-  of tools for both manipulating and imaging the dependent surface-mediated diffusion as the stamp topography of a surface with atomic-scale resolu- 34 - 31 was left in contact with the substrate for longer tion. At present, these tools seem well suited for intervals of time. Diffusion of the ink and blurring applications in research but will require substantial of the features can be minimized by using high- development before they can be used for patterning molecular-weight inks. Patterns of dendrimers printed large areas in manufacturing. The most important 40 by Ì CP can have lateral dimensions as small as  SPL techniques include scanning tunneling micros- 220,249 nm. The outline of the edge of features printed copy (STM), atomic force microscopy (AFM), and using CP often follows the edges of grains in the Ì near-field scanning optical microscopy (NSOM). A metal film used as a substrate; grain size is, thus, striking example of the potential of these techniques also an important determinant of resolution. for nanoscale fabrication is the precise positioning of individual Fe atoms with an STM tip (Figure The resolution and minimum feature size of thin 28,52 9A,B). This atomic-scale manipulation is interest- films transferred by nTP also depends on the integ- ing scientifically but is not yet a practical technology. rity of the transferred material. The wetting of the Our review of the current capabilities of SPL for stamp by a metal determines the minimum thickness nanofabrication is brief as this topic is extensively and lateral dimension of isolated features. The grain 32,34,53 reviewed elsewhere. size of the metal determines the thickness required to produce a continuous film. For example, the minimum thickness of a continuous film of Pd (grain 5.1. Serial Patterning of Surfaces Using SPL size of  20 nm) is less than that required for Au Scanning probe lithography is a versatile method 50 nm) when deposited by electron-  (grain size of for depositing clusters of atoms or molecules onto a 190 beam evaporation onto titanium-coated silicon. 30 surface in a well-defined pattern. One approach to Cracks in the metal film can also form as a result of deposit nanoparticles or molecules selectively onto a 236 mechanical deformation during printing. Other 252 - 30,31,250 surface is dip-pen nanolithography (DPN). sources of defects in the printed pattern include An AFM tip is “inked” with a solution of the material disparities in the stamp or on the printed surface. to be transferred to the surface. The material ad- The preferential adhesion of the material to be sorbed onto the AFM tip transfers to the surface in printed onto a second surface is a third limiting factor an arbitrary pattern “written” with the scanning in printing nanoscale features. The stamp can be probe (Figure 9C). This technique can, with care, coated with a release layer to assist the removal of reproducibly pattern lateral features as small as 50 this material (e.g., a metal film) from the stamp 32 nm. Molecules patterned by this technique include by decreasing the surface free energy of the stamp. SAMs for binding oligonucelotides, proteins, and This release layer can, however, also decrease the 256 - 253 viruses. Similar SAMs can also mask the 237 wetting of metal deposited onto the stamp. The substrate during wet etching to pattern nanostruc- formation of covalent bonds between the transferred 257 tures of metals such as Au, Pd, and Ag (Figure 9D). material (e.g., molecular ink or metal film) and the This process can fabricate trenches with lateral printed surface also improves release from the 258 dimensions from 12 to 100 nm. The mechanism of 222,234,236,237 stamp. material transfer by DPN is not yet clear. One possibility is that water between the tip and surface 4.4. Summary 259 mediates the process; another is the transfer of solid material as a result of tip - surface interactions. Printing materials onto a surface using a topo- The spreading of the ink on the substrate depends graphically patterned stamp changes the properties

14 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1184 (Figure 9F). Irregularities in these patterns can result from variations in surface topography and, therefore, a nonuniform contact between the tip and substrate. A second material (e.g., SAMs or nano- 265,268 particles) can replace the removed film. This substitution lithography is a convenient method for patterning multiple types of SAMs on a surface. Mechanical displacement of a thin polymer film with 269 an AFM tip, or “nanoplowing”, can also generate nanoscale holes or trenches. These nanoscale pits are 270 templates for the growth of nanowires or the formation of nanosized electrical contacts with the 271 supporting substrate. Scanning probe lithography is also used to modify 272 a surface chemically. One example is the localized oxidation of a surface (metal, semiconductor, or SAM) in a pattern scanned by a conductive AFM or STM 277 - 267,272 tip (Figure 9G). A local electric field between the conductive tip and the surface induces oxidation of the surface. Typically, this method can generate 50-nm wide features. A carbon nanotube-modified  AFM probe can, however, pattern 10-nm wide lines  of silicon oxide on a silicon hydride surface (Figure 278 9H). A conductive AFM tip can also locally modify organosilane SAMs to direct the deposition of Au 55 279,280 nanoclusters. Another SPL method for chemi- cally modifying a surface is photochemical oxidation 281 - 283 by NSOM. For example, the photochemical oxidation of SAMs of mercaptoundecanoic acid (with 244-nm light), followed by selective wet chemical  etching of the unmasked substrate, can pattern 55- 281 nm wide trenches in gold. The commercial availability of AFM, STM, and NSOM make these tools convenient for nanofabrica- tion. These instruments are also capable of nanoscale registration. This approach to writing nanoscale patterns with a single tip is, however, fundamentally slow. The serial nature of SPL results in a low sample Figure 9. Schematic representations of four approaches throughput. Single-probe methods are probably re- to scanning probe lithography, and patterns produced using stricted to research applications and possibly to them: (A) Scanning tunneling microscopy (STM) can fabrication of customized patterns or mask repair. position atoms on a surface with high precision to generate patterns, such as (B) a quantum corral of a 48-atom Fe 52 ring formed on Cu enclosing a defect-free region. (Re- 5.2. Parallel Patterning of Surfaces Using SPL printed with permission from ref 52. Copyright 1993 American Association for the Advancement of Science.) (C) A practical approach to SPL for large-volume, Dip-pen nanolithography can direct the deposition of SAMs parallel production may emerge by simultaneously (e.g., 16-mercaptohexadecanoic acid) on Ag as an etch resist 53,250,251,284 - 288 writing patterns with multiple probes. 257 to pattern (D) 70-nm wide features. (E) Nanoshaving can An array of cantilevers scanned in parallel may allow remove regions of SAMs to pattern features such as (F) a 266 square hole within octadecanethiolate SAMs on Au. (G) higher sample throughput. The concept of the “Mil- Scanning electrochemical oxidation with a carbon nano- lipede” was developed as a 2D array of independently tube-modified AFM tip can selectively oxidize a surface to addressable AFM probes for high-density data stor- pattern (H) 10-nm wide (2-nm tall) silicon oxide lines 53,284 - 286,289,290 age. Each probe in this array can be 278 spaced by 100 nm. (Reprinted with permission from ref mechanically deflected in the vertical direction and 278. Copyright 1999 American Institute of Physics.) resistively heated. These arrays of cantilevers can  locally heat a thin polymer film to pattern 40-nm on the humidity, the reactivity of the ink with the 284,285 2 11 10 > wide holes with feature densities /in. A . substrate, the radius of curvature of the probe, and - 259 264 parallel approach to patterning surfaces by DPN is the linear velocity of the probe. 250,251,287,288 also being explored. The current capabili- Another approach to nanofabrication by SPL is the ties of parallel DPN include writing multiple copies selective removal of material from a surface by force- 265 - 267 of a pattern using a linear array of passive (i.e., induced patterning. An AFM tip in contact with nonactuated) probes or writing a series of different the surface can displace SAMs in a process referred 266 patterns (e.g., alkanethiols in the pattern of numerals to as nanoshaving (Figure 9E). This process re- 287 9) using thermally actuated probes. 0 - moves SAMs from a surface in a well-defined pattern

15 1185 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication generate nanowires of metals (e.g., Ag, Pd, Cu, and 5.3. Summary Au), oxides (e.g., MoO and Cu O), and semiconduc- x 2 313 - 43,308 Scanning probe lithography can precisely position tors (e.g., MoS These nanowires Te ). and Bi 2 3 2 43 atoms on a surface and selectively deposit or remove can have lateral dimensions down to  15 nm and 310,314 regions of etch resist to pattern surfaces. These can be used in the fabrication of gas sensors. techniques may find applications in mask or device For example, palladium nanowires, transferred to a repair and information storage. Parallel approaches cyanoacrylate polymer film, can be used to detect the 310,315,316 in SPL are being developed to overcome the serial presence of hydrogen gas. Nanowires grown limitations of standard SPL technologies. Surface by this method are, however, randomly positioned on diffusion of molecular inks and colloidal suspensions the substrate in a pattern determined by the orienta- broadens the features patterned by SPL. It is also tion and spacing of the step edges in the HOPG challenging to generate reproducible structures be- substrate. Their grain structures and edge roughness tween scans because of variations in the surface have not been characterized. The diameter of these topography of the substrate and differences in the nanowires can also vary across the substrate, and shape of the tip (and variations in this shape with nanoparticles can nucleate on defects in the sub- time and use). strate. 6.1.2. Patterning at Edge-Defined Defects in SAMs 6. Edge Lithography for Nanofabrication Another strategy for patterning nanostructures by We define edge lithography as either pattern 36,42,291,292 edge lithography is the selective deposition or re- transfer directed by the edge of a feature or moval of material in regions defined by defects at the the process of transforming a feature that is thin in 36,291,317,318 edges of topographic features. For example, the vertical direction into a feature that is thin in 37,39,40,293 SAMs form polycrystalline lattices on a planar me- the lateral direction. General interest, as tallic surface but remain disordered at the edges of measured by volume of publications, in these two 35,36 this surface. Sharp metal corners within a topo- areas has increased markedly in the past few 43 - 35 graphically patterned metal substrate prevent the years. These methods are, currently, limited in formation of well-ordered SAMs and expose the the types of patterns they can form but can be used 36,291,317 underlying metal at these edges. Selectively < 100-nm structures in parallel to pattern arrays of etching the exposed metal transfers the outline of the for a range of materials. patterned metallic topography into the underlying 50 nm. film with line widths as small as  6.1. Pattern Generation Directed by Topography A modification of this technique included a thin In this section we discuss varieties of edge lithog-  5-nm thick) between a patterned titanium layer ( 318 raphy that use the edge (e.g., the perimeter or the silver film and a planar supporting layer of silver. outline) of a topographic feature to generate nano- Immersing this patterned substrate in a solution of scale structures. We divide this area into four strate- alkanethiols formed SAMs on the silver but not on gies for generating nanostructures: (i) depositing the titanium. The exposed edge of titanium formed 297 - 43,294 material at step edges of crystalline lattices;  5-nm wide) in the SAM-coated substrate. a gap ( (ii) adding or removing material at edge-defined This edge served as a well-defined nanoelectrode for 36 318 defects in SAMs; (iii) depositing or undercutting at the electrodeposition of copper. Electrodeposition 38,298 lithographically defined step edges; and (iv) on edges made by engineering defects in SAMs has patterning photoresists at regions defined by vertical generated nanowires with lateral dimensions as edges in a soft stamp using phase-shifting photo- small as 70 nm. 42,292 lithography. These edge lithographic techniques generate nano- structures in a pattern defined by the outline of a 6.1.1. Material Deposition at Step Edges of Crystalline topographic template. Photolithography can pattern Lattices regular arrays of topographic features, and elec- trodeposition on edges within this well-defined sur- The exposed edges of steps within a crystalline 318 face can generate aligned nanowires. Removing an lattice can have properties that are different from array of parallel nanowires with Scotch tape trans- those of the bulk material. For example, gas-phase fers these nanostructures to a transparent, flexible catalytic studies have shown that the exposed substrate and generates an efficient optical polar- edges of the catalyst can promote a number of 318 izer. Nanowires supported on this adhesive sub- reactions and are often the sites of highest reac- - 301 295,299 strate can also conform to a curved or flat substrate. tivity. The step edges on single-crystalline surfaces can also direct the growth of metal nano- 6.1.3. Controlled Deposition and Undercutting at 43,294,296,297,302 - 306 structures. Particles of Cu, Co, and Lithographically Defined Step Edges Ag have been deposited at the step edges of crystal- line metallic substrates (e.g., Mo (110), Ag (111), and Thin-film deposition onto a topographic template 303 - 305,307 Cu (111)). These metals are often deposited and selective etching of the substrate can also gener- by physical vapor deposition. ate nanoscale features. For example, shadow evapo- Material deposition at atomic step edges has been ration of metal onto the side walls of topographic extended to the growth of continuous nanowires by features, followed by the selective etching of the 319 electrodeposition on highly oriented pyrolytic graph- substrate, generates narrow, vertical structures. 43,308 White et al. used this method to fabricate an array - C). ite (HOPG) (Figure 10A This approach can

16 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1186 Figure 11. (A) Schematic illustration of phase-shifting edge lithography using a topographically patterned com- 161 s “hard” PDMS (h-PDMS) posite PDMS stamp with a soft 159 PDMS backing s in contact with a photoresist. The schematic depicts the intensity profile for a destructive modulation of the phase-shifted light s a constructive modu- lation is also possible (see text for further details). The phase shift is optimal when the thickness of the air gap ) of light Ï (h) is equivalent to the ratio of the wavelength ( to twice the change in refractive index ( ¢ ) between PDMS n 162 and air. The SEM images show (B) 30-nm wide rings 159 and (C) 50-nm wide lines patterned in a positive-tone photoresist using phase-shifting edge lithography with a PDMS stamp and destructive modulation of the phase- shifted light. Controlled undercutting of photoresist has also pat-  terned 50-nm wide islands of gold for growing ZnO 322 nanowires. The controlled undercutting of thin (A) Schematic illustration of the process used Figure 10. to generate nanowires at step edges of a graphite surface films is not limited to semiconductor substrates. For by electrodeposition of a material such as molybdenum example, patterning an aluminum film supported on 43 oxide (MoO ). SEM images (B,C) show dense arrays of x a calcium fluoride (CaF ) substrate generated a 2 2 - MoO nanowires deposited from 1.0 mM MoO . (Re- x 4 298 frequency-selective surface (Figure 10F). printed with permission from ref 43. Copyright 2000 Each of these techniques is limited in its ability to American Association for the Advancement of Science.) (D) pattern arbitrary features by the features whose Schematic illustration of the use of controlled undercutting  to pattern trenches with lateral dimensions as small as 50 edges are being used and by the characteristics of nm. (E) SEM images for a cross-section of  75-nm wide light. Photolithography and electron-beam lithogra- trenches transferred into a silicon substrate through a phy can pattern regular arrays of topographic fea- patterned chrome mask. (F) An infrared transmission plot tures, and nanoscale structures generated at the step showing the frequency selectivity of 100-nm wide trenches edge of these features are regularly spaced. Pattern- in Al on a CaF substrate, prepared by selective undercut- 2 298 ing intersecting (crossing) lines of metal using these ting. (Reprinted with permission from ref 298. Copyright 2001 Wiley-VCH Verlag GmbH.) methods is, however, not straightforward. 6.1.4. Phase-Shifting Edge Lithography of 30-nm wide lines of silica with heights of  300 38 nm. Another approach to nanofabrication directed by Depositing a low-temperature oxide uniformly over topographic features and etching the substrate the edges of a topographic feature is near-field phase- 42,162,241,292,323 - 325 by RIE can also pattern nanoscale features at the shifting photolithography (Figure 11A). 320 outline of each feature. In this technique the vertical edges of a transparent, Somorjai et al. used this 10-nm wide vertical struc-  method to generate topographically patterned substrate induce abrupt 321 tures. changes in the phase of incident, collimated light over short distances (an edge in a conformal, transparent, Patterned arrays of nanostructured trenches can PDMS stamp). This shift in phase of the incident be fabricated by the controlled undercutting of topo- light creates narrow regions of constructive and graphic features using isotropic wet etching, followed 298 destructive interference. Phase-shifting photolithog- by deposition of a thin film (Figure 10D). In this raphy uses this interference to project “dark” or approach to nanofabrication the initial step was to “bright” regions of incident light onto the surface of pattern a photoresist supported on a metal-coated a photoresist. The smallest lateral dimensions are substrate (e.g., chromium on silicon). The exposed  produced when the light has a phase shift of metal film was isotropically wet etched with con- radians at the photoresist mask interface (Figure - trolled undercutting of the photoresist. Coating this 11A). substrate with a second metal film, followed by lift- off of the photoresist, produced nanostructured Masks for phase-shifting edge lithography must trenches at the edges of the photoresist (Figure 10E). be transparent and situated as close as possible to This method has patterned well-defined trenches the films of photoresist. (Ideally the mask and 298 with lateral dimensions as small as  50 nm. photoresist should be in conformal contact.) These

17 1187 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication requirements limit the choice of materials. A hard mask for phase-shifting edge lithography (e.g., quartz) is expensive to design and requires accurate control of the distance between the phase mask and the photoresist; accidental contact between the two easily damages or contaminates the mask. An elastomeric mask is much less expensive to fabricate and use. A PDMS mask spontaneously and nondestructively achieves conformal contact with the photoresist; conformal contact eliminates any gap between the mask and the resist and places the resist directly in the optical near-field of the mask. Phase-shifting edge lithography using an elastomeric phase mask has  30 nm (Figure patterned features as small as 162,292,323,324,326,327 11B,C). Soft masks for phase shift- ing can also conform reversibly to nonplanar sur- 162,292,323,328 faces. Near-field phase-shifting photolithography has been used to fabricate a number of simple patterns, such as rings and lines of photoresist (Figure 11B,C). These patterns can be transferred into a metal film 162,292,323,327 by lift-off or selective wet etching. The narrow photoresist features can also mask a sub- strate during RIE. This method produced uniform, single-crystalline silicon nanostructures with well- 1 > defined features as small as 40 nm and lengths 326 cm. Components of devices fabricated by phase- (A) Schematic illustration of a method to create Figure 12. 329 shifting edge lithography include optical polarizers nanowires by physical vapor deposition on a structured 327 and gates for organic transistors having dimen- edge, which was fabricated by molecular beam epitaxy 39 100 nm. An array of patterned  sions as small as (MBE) and etching in buffered oxide etch (BOE). (Re- printed with permission from ref 39. Copyright 2003 features in Al can generate a frequency-selective 328 American Association for the Advancement of Science.) optical filter, and nanostructured holes patterned (B,C) SEM images showing the edge of the MBE-grown in photoresist can direct the crystallization of various substrate with an array of deposited Pt nanowires. The 162 330 salts or the deposition of nanoparticles. higher resolution micrograph (C) shows 10-nm diameter Phase-shifting lithography using a soft, conformal Pt nanowires with a pitch of 60 and 30 nm. (D) SEM image mask requires that the mask have vertical, straight of a cross-bar array of Pt nanowires fabricated from two sets of nanowires transferred to adhesive substrates. (E) sidewalls. Distortions in the mask broaden the Schematic diagram showing the fabrication of a patterned features in the photoresist. A composite stamp (Fig- 40 array of epoxy-embedded conducting metal edges. (F) ure 11A) of h-PDMS with a “soft” PDMS backing such SEM image of 50-nm thick Au lines exposed by sectioning as Sylgard 184 improves the fidelity of the sidewalls with the glass knife of a microtome. (G) Electrochemical in comparison to a stamp of only Sylgard 184-based deposition of a metal onto the exposed metal edge identified PDMS. These composite stamps can pattern  30-nm the conductive regions. 162 features by phase-shifting edge lithography. this approach to nanofabrication. For example, 6.2. Generating Nanostructures by Exposing the Pfieffer et al. fractured multilayered MBE-grown Edge of a Thin Film substrates to direct the growth of quantum wires and 37,336,337 quantum dots. An MBE-grown substrate con- A second type of edge lithography takes advantage sisting of alternating layers of AlGaAs and GaAs was 331,332 of the numerous methods that can grow thin also used to fabricate an array of field-effect transis- films over large areas with a thickness between 1 and 336  20-nm gate lengths. tors (FETs) with which are thin in the 50 nm. Converting these films s The edge of a multilayered substrate that has been s into structures that are thin in the vertical direction lateral direction is an approach to fabricating nano- fractured can also template the formation of parallel 37,39,293,333,334 39 structures. There are now three demon- nanowires by physical vapor deposition (Figure 12A). strated approaches to exposing a nanostructured Selectively etching one of the components (e.g., 37 edge: (i) fracturing a thin film; (ii) sectioning an AlGaAs in a buffered oxide etch) reveals an array of 40,335 encapsulated thin film; and (iii) reorienting posts narrowly spaced grooves. These grooves are partially 41 capped with a thin film. coated with a metal by selective angle deposition to generate parallel nanowires supported on the frac- 6.2.1. Edges by Fracturing Thin Films tured edge (Figure 12B,C). The width of the groove, corresponding to the original thickness of the etched Thin films deposited onto a crystalline semiconduc- film, determines the spacing between the nanowires. tor substrate, such as silicon, can be exposed as a Modifying the thickness of each layer of the substrate uniform narrow edge by fracturing the substrate. will change the spacing between the nanowires and Multilayered structures grown by molecular beam epitaxy (MBE) are the most common substrates for the width of the nanowires.

18 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1188 An adhesive tape can remove the deposited lines of metal from the edge of the substrate. Free-standing Pt nanowires have been fabricated with diameters 39  16 nm. down to 8 nm at a pitch of Parallel nanowires supported on adhesive substrates can be overlapped perpendicular to each other to create a cross-bar array (Figure 12D). A narrow edge can also 338 direct the assembly of nanorods. For example, nanorods of cadmium selenide (7-nm diameter, 35- nm long) can align along an exposed edge of ZnS (5- nm wide) functionalized with SAMs of hexanedithiol. 6.2.2. Edges by Sectioning Encapsulated Thin Films Figure 13. (A) Schematic illustration of an array of A nanostructured edge can also be fabricated by microdominos (posts of an epoxy-based photoresist) uni- embedding a thin film in a soft matrix and cutting formly collapsed by applying a horizontal shear using a slab or sectioning the matrix to expose its cross-sec- of PDMS. Selectively depositing a Pd film to coat the top 40,335 tion. An early approach to exposing the edge of and part of one side of each post prior to collapse generated an embedded thin film was to polish the substrate 41 an array of nanometer-scale Pd edges after collapse. (B) 335 with an abrasive film (e.g., silica-coated paper). In SEM images of the array of collapsed microdominos adher- ing to the PDMS slab, and (C) a  15-nm wide edge of Pd this approach a Pt film deposited on mica was on a collapsed post. (D) Topography and (E) surface encapsulated in an epoxy matrix. Sanding this soft potential of a 100-nm thick PMMA surface after patterning matrix exposed the edge of the Pt. Another approach regions of charge (  400 mV peak surface potential). The to expose the edge of the encapsulated film is to inset in E shows the side profile for one region of charge. section a polymer-encased metal film with the glass (Reprinted with permission from ref 41. Copyright 2004 40 knife of a microtome. This approach to edge lithog- Wiley-VCH Verlag GmbH.) raphy can be combined with soft lithography to 2 areas ( > 1cm ). The reorientation of the microdomi- generate a patterned edge structure (Figure 12E). nos also generates an array of metal features (edges) The surface roughness of the sectioned epoxy was with nanometer dimensions in the plane of the array.  10 nm. This smooth surface encapsulated arrays The collapsed microdominos adhere to the PDMS of isolated, narrow gold lines with line widths as slab that is used to apply the shear force. If a small as 50 nm (Figure 12F). conducting polymer [e.g., polyaniline (PANI)] had Exposed edges can direct the deposition of material been grafted onto the surface of the PDMS before into a well-defined pattern. For example, these edges using it to shear the microdominos, this layer estab- can be electrically addressed from the opposite side lished an electrical connection to each of the arrayed of the encapsulating matrix for the electrodeposition 41 40 edges. This array of narrow electrodes supported of metal (Figure 12G). This backside electrical on a conducting, elastomeric substrate can be used connectivity cannot be easily achieved for structures as a flexible electrical contact for printing charge into patterned by conventional techniques. We believe the electrets (Figure 13D,E). Multiple regions of charge combination of edge lithography with molding and can be patterned in parallel by simultaneously ad- embossing will be a useful method for generating dressing all of the nanoscale edges. substantially more complex features than those cur- A nonuniformly applied shear force can result from rently demonstrated. uneven contact between the microdominos and the A major challenge in fabricating encapsulated horizontally translated substrate. An elastomeric nanostructures by sectioning a soft matrix is mini- slab, such as PDMS, applies a shear force evenly metal interface. mizing delamination at the matrix - across the substrate by conforming to asperities in Sectioning the matrix at temperatures below - 120 the surface. A limitation of the current approach is °C significantly minimized the delamination between 40 the inability to address each nanoscale edge indi- epoxy and an embedded metal film. Exposing the vidually. Patterning the conductive substrate in surfaces to an oxygen plasma before embedding also contact with these edges may overcome this limita- improved adhesion between the interfaces. tion. 6.2.3. Edges by Reorientation of Metal Capped Posts 6.3. Summary Ordered arrays of nanoscale edges have been patterned by capping an array of posts with a thin Edge lithographic techniques are, currently, re- film and tipping each post onto one side (Figure stricted to generating certain, limited types of line 41 13A). Arrays of epoxy posts were patterned by structures (e.g., noncrossing lines) in one step of photolithography on a silicon substrate. A thin metal fabrication. Crossed lines can sometimes be gener- 39,40 film, deposited by selective angle deposition, coated ated by stacking features. These techniques are the top and part of one side of each post. These still being developed as tools for research. Recent structures (“microdominos”) fractured from the sup- developments include “wiring-up” nanostructures to 40,41 porting substrate under a horizontally applied shear external magnetic or electric fields and directing force. They collapsed in a uniform pattern determined the formation of parallel nanowires with applications 318 39 by the direction of the applied shear (Figure 13B,C). in nanoelectronics and tunable optical polarizers. These patterned, collapsed arrays can cover large We believe the formation of more complex nanostruc-

19 1189 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication tures will result from a combination of edge lithog- conventional and unconventional s raphy with other s nanofabrication techniques. 7. Self-Assembly for Nanofabrication Self-assembly is defined as the spontaneous orga- nization of two (or more) components into larger 339 aggregates using covalent and/or noncovalent bonds. We described a number of techniques for nanofabri- cation that use top-down approaches to patterning Figure 14. (A) AFM image of a 2D array of DNA strands nanostructures. Self-assembly, a bottom-up approach that templates the electroless deposition of palladium 364 to nanostructures or nanostructured materials, is a nanowires. (B) SEM image of poly(styrene- -methyl block methacrylate) self-assembled onto a photopatterned tem- second strategy for nanofabrication. This approach 366 plate of SAMs with a periodicity of 48 nm. (Reprinted relies on cooperative interactions of small components with permission from ref 366. Copyright 2003 Macmillan that assemble spontaneously in a predefined way to Magazines Ltd.) produce a larger structure in two or three dimen- sions. structure and sometimes increase the order of the We will focus on two types of self-assembly: (i) self-assembled structure. Self-assembly can be di- nontemplated self-assembly, where individual com- rected using surface topography, electric and mag- without ponents interact to produce a larger structure netic fields, or shear forces. Templated self-assem- the assistance of external forces or spatial con- 352 bly often uses top-down strategies to fabricate straints, and (ii) templated self-assembly, where components that direct the bottom-up assembly of and individual components interact with each other molecules, macromolecules, or colloidal particles. an external force or spatial constraint. In this section Templated self-assembly is an alternative to non- we review recent advances in both nontemplated and templated self-assembly for the controlled fabrication templated self-assembly and evaluate the current of patterned structures with nanometer-scale local status of self-assembly for nanofabrication. For other order and for the generation of micrometer-size, or aspects of self-assembly we direct the interested larger, domains of defect-free patterns. 46,340 344 - reader to recent reviews. 7.1. Nontemplated Self-Assembly 7.2.1. Templating from Molecules One of the most appealing aspects of self-assembly 45,339,353,354 Molecular and supramolecular chemistry spontaneous assembly of components into a is the can produce structures that range in size from 1 to desired structure. The notion that a fabrication - 46,174,353,355 359 100 nm and beyond. Generation of these strategy requires only the mixing of components to nanostructures has, largely, remained an academic achieve an ordered structure is appealing both for exercise in design and synthesis because they are not its simplicity and its potential efficiency. We refer to functional s they are not, for example, electrical con- this type of self-assembly as “nontemplated self- ductors, transistors, or motors. These organic nano- assembly.” Examples of materials fabricated using 51 structures have, however, been used as templates to this approach include SAMs and structures that 360 345,346 mask the deposition of metal or guide the growth self-assemble from block copolymers and nano- 363 - 361 343,347 of metal nanoparticles and nanowires (Figure particles. Recent advances in the fabrication of 364 14A). The inorganic nanoparticles or nanowires are functional nanostructures using self-assembly in- often fabricated at electrodes and are electrically clude self-assembled arrays of magnetic nanopar- 365 conductive. ticles that can be used as magnetic data storage 348 The combined use of a top-down approach, such as devices and self-assembled arrays of nanorods that 349 photolithography, and bottom-up self-assembly has display birefringence. been used to pattern block copolymers (Figure Nontemplated self-assembly, while attractive for 369 - 366 14B). Patterns of oxidized SAMs (or random its minimalist use of materials and energy (in con- 370 copolymers) on silica can be produced using ex- trast to conventional lithography), is not widely used treme ultraviolet interferometric lithography that for nanofabrication. Examples where this approach exhibit a periodicity on the order of the lamellar is used for nanofabrication include patterning arrays 350 -methyl methacrylate) block spacing of poly(styrene- of nanoelectrodes and generating arrays of metal 351 copolymer (that is, in this instance, 48 nm). When  nanoparticles for sensing biomolecules. Self-as- this block copolymer is allowed to self-assemble on sembly is, however, prone to producing defects, and the patterned SAM (or random copolymer), this the perfect periodicity of self-assembled structures pattern acts as a template that guides the phase from nanoscale components is generally limited to separation of the polymer. Annealing a block copoly- micrometer-sized areas. The components also have mer film confined by physical boundaries can also a limited number of different ordered arrangements direct the assembly of the copolymer into a regular and generate a limited number of functional struc- 371,372 structure. These methods illustrate the use of tures. templating to overcome the major disadvantages of 7.2. Templated Self-Assembly nontemplated phase separation as a method of gen- erating regular structures: that is, the high level of By templating self-assembly it is possible to intro- defects, the inability to control or pattern the phase- duce an element of pattern into the self-assembled

20 Gates et al. 1190 Chemical Reviews, 2005, Vol. 105, No. 4 wire self-assembly results in partially ordered, small superlattices. This issue has been addressed by several new methods that direct the assembly of nanowires, including the use of microfluidic chan- 402 403 nels and electric fields. Another review in this issue contains a full account of methods available to 404 assemble nanowires. 7.3. Summary s Self-assembly as a stand-alone method for nano- Figure 15. (A) Transmission electron microscopy (TEM) fabrication s is presently unable to produce structures image of a polymer shell fabricated using layer-by-layer assembly of alternating layers of poly(styrenesulfonate) and with precise spatial positioning and arbitrary shapes poly(allylamine hydrochloride) assembled onto a spherical with a low concentration of defects and functionality 380 template. a melamine In this example the template s - that can be achieved using conventional nanofabri- has been selectively removed. (B) SEM formaldehyde bead s cation. It is also unable to generate the range of image of 50-nm diameter gold nanoparticles assembled into 400 patterns required for even simple electronic func- wells defined by electron-beam lithography. Capillary tionality. Nontemplated self-assembly may represent forces generated by an evaporating solvent directed the particles into these wells. a useful method of generating materials for informa- tion technology. For example, crystalline arrays of magnetic nanocrystals can store large amounts of segregated regions, and the uncontrolled drift in the 348 information. pattern over nonlocal dimensions. Neither nontemplated nor templated self-assembly 7.2.2. Templating from Particles strategies have yet demonstrated a route to the level of functionality necessary to contribute to micro- Charged polymers, or polyelectrolytes, can be used 375 - 373 electronics other than by generating materials or to modify the surfaces of colloidal particles. The  positioning objects with large ( m) dimen- Ì 100 electrostatic attraction between a charged surface - 197,405 409 sions. We are, however, optimistic that self- and a charged macromolecule is sufficient for adsorp- assembly will play a significant role in nanofabrica- tion. Often the charged polymer has excess charge 376 tion in the future, particularly when we consider the and reverses the charge on the surface. This potential for fabrication in three dimensions, the reversal of charge permits a second polymer, with 410 opportunity for reversible and reconfigurable self- charge opposite to the first, to assemble on the first 411 377,378 assembly, and the implication that self-assembled layer of polymer. The layer-by-layer assembly structures can undergo self-repair or self-replica- of polyelectrolytes can be repeated to produce robust 412 tion. The cell is the ultimate demonstration and multilayers of self-assembled polymers. inspiration for continuing work in nanometer-scale Layer-by-layer assembly has also been used to 379 - 383 sophisticated than more self-assembly, and it is much fabricate hollow colloidal particles (Figure 15A). current microelectronic systems. A demonstration of A multilayer polyelectrolyte shell around a colloidal the principle for very complex and functional forms particle is robust enough to survive the removal of of self-assembly thus already exists. the core, generally achieved by dissolving the core template in a solvent. The hollow and permeable 8. Outlook and Conclusions shell that remains encapsulates an attoliter vol- 377,384 ume. Colloidal particles have also been used to The expectations surrounding nanotechnology and 388 - 385 template the self-assembly of nanoparticles. nanofabrication are high. Governments and compa- Typically, the diameter of these particles is larger nies around the world are spending billions of dollars than 100 nm, but the assembly of thin, metallic shells on research related to nanotechnology. The U.S. may be important for nanofabrication. government promoted an initiative announced in optimistic) very 2000 based on optimistic (probably 7.2.3. Templating Using External Forces predictions that it would eventually lead to “materi- Nanospheres that are monodisperse in size and als with ten times the strength of steel and only a shape self-assemble to produce thin films of close- fraction of the weight, the ability to shrink the 389 packed, ordered lattices. Electric and magnetic information housed in the Library of Congress into 391 390 fields and spatial con- as well as shear forces a volume the size of a sugar cube and the ability to 392,393 straints have been used to direct the assembly detect dangerous cancerous tumors when they are 17 of nanoparticles and nanorods into different configu- only a few cells in size”. This initiative predicts rations. For example, template-assisted self-assem- nanotechnology will add a trillion dollars to the gross 394 bly has been used to direct the assembly of national product and add 2 million new jobs to the spherical and tetrapod-shaped nanoparticles where U.S. economy by 2013. Whether nanotechnology will the capillary force exerted by the edge of a drop of or will not have a revolutionary economic impact is evaporating solvent confines particles against an not certain but also not highly relevant at this early, 395 396 399 - edge, or in a well in a narrow channel, exploratory stage in its development. It is unques- 400 (Figure 15B). tionably fascinating science, and it will certainly lead to significant technologies. The assembly of nanowire arrays is more challeng- Nanoscience and nanotechnology cover many dif- ing than the assembly of nanoparticles and nanorods 401 ferent areas, but one key set of methods for both is due to the anisotropic shape of the object. Nano-

21 1191 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication The intrinsic limits of photolithography are ines- nanofabrication. The development of microelectronic circuits with 100-nm-scale features is proceeding < capably related to the diffraction of light. The attrac- rapidly by extensions of existing, conventional pho- tion of many unconventional nanofabrication tech- tolithographic techniques. Unconventional tech- set niques is that their intrinsic limitations are not niques and new materials will be required, if at all, by optics and diffraction but by van der Waals for structures with dimensions below 20 nm and  interactions and the granularity of matter at the to change the cost structure of this very capital- molecular or atomic scale. The resolution of molding 163,169 intensive industry. New products and technologies methods in the laboratory is currently  1 nm, outside the field of microelectronics, but still requir- but extensive development will be required to achieve ing nanoscale fabrication, are being developed in < 10-nm feature sizes reliably in manufacturing. widely different areas such as biology, materials Unconventional nanofabrication methods may, how- science, and optics. Our intuition is that there will ever, offer higher resolution patterning at a lower cost be many opportunities for the application of uncon- than photolithography. Conventional scanning beam ventional nanofabrication in these areas. Some pos- lithography also has high initial equipment costs. sible areas of application include the following: (i) Scanning beam lithography is, however, required by printed, low-cost organic microelectronics; (ii) sub- many nanofabrication techniques to pattern the wavelength optics; (iii) tools for biology for investigat- master or mask. The major advantage most uncon- ing individual cells and cell cell interactions; (iv) - ventional methods have over conventional scanning nanofluidics; (v) nanoelectrical mechanical systems beam techniques is the high throughput in replicat- (NEMS); and (vi) single-molecule studies. ing the original pattern. The high capital and operating cost of conventional Each technique for nanofabrication has certain equipment for nanofabrication (photolithographic characteristic advantages, and it is unlikely that a steppers and scanning beam tools) has given an single technique will dominate all areas of applica- opening for unconventional techniques in nanofab- tion. For example, molding, embossing, or printing rication, initially in the areas of research and product 414,415 121,184 from a cylindrical drum or by wave printing development. Molding, embossing, and printing tech- or perhaps self-assembly have the potential to gener- niques have demonstrated usefulness, at least at the ate regular patterns over very large areas inexpen- level of generating prototypes. A new generation of sively. Photolithography will, however, probably re- s nanofabrication tools is emerging mainly from small main the exclusive method for producing high- businesses located in the United States, Europe, and performance integrated circuits for years to come. Asia that enable various forms of stamping, mold- s Regardless of their ultimate commercial applica- ing, and embossing. These products vary in sophis- tions, unconventional techniques for nanofabrication tication and capability. Most of the tools that are provide uncomplicated methods for the research being sold are going to research laboratories or to communities in a broad range of disciplines to explore fabrication facilities for products other than semi- nanoscience and begin to develop nanotechnology. conductor-based devices. The nanofabrication equip- Their variety, low cost, operational simplicity, and ment manufacturers are selling tools to companies broad collective applicability open the door to survey- that are reportedly developing a variety of single- ing and exploring many areas to which high-resolu- layer products such as surface acoustic wave devices, tion photolithography and particle-beam writing photonic crystals, microfluidics, and biosensors. We simply are not applicable. Unconventional techniques expect that unconventional nanofabrication will play for nanofabrication are being widely accepted and an important role in the commercialization of nano- developed in this exploratory spirit and are thus a technology-based devices that require low-cost manu- key part of the development of nanoscale research facturing of nanoscale structures over large areas, on nonplanar surfaces, and with new materials. and development. From them may come unexpected inventions and technologies that will allow nanotech- Historically, the motivation for developing uncon- nology to live up to the optimistic expectations for ventional nanofabrication techniques was to continue its future. patterning semiconductor devices beyond the pre- dicted resolution limit of photolithography to fore- s 413 Photolithography stall an end to “Moore’s Law”. 9. Acknowledgments continues to overcome obstacles to achieve new 1 resolution requirements. These improvements have This research was supported in part by the Defense not always been simple. This challenge is reflected Advanced Research Projects Agency (DARPA) and in the increasing costs of conventional photolithog- the Harvard NSEC under NSF Award No. PHY- raphy processes when shifting to shorter imaging 01,17795. This work made use of MRSEC shared wavelengths, improved imaging optics, phase-shifting facilities supported by the NSF under Award No. masks, and optical proximity correction. Complex DMR-98,09363. The authors thank Dr. Doug Resnick engineering challenges remain. It is, however, now of Motorola Laboratories and Dr. S. V. Sreenivasan clear that it is not feature sizes (or even registration) of Molecular Imprints, Inc. for discussions on SFIL that is the limitation in microelectronic device per- and NIL. The authors also thank Dr. Emmanuel formance. The major challenges are connections Delamarche and Dr. Heinz Schmid of the IBM Zurich within or from the chip, power distribution, and heat research laboratory for useful discussions on evapo- dissipation. These areas may offer opportunities for unconventional nanofabrication. ration of metals onto PDMS.

22 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1192 Pure Appl. (47) Boncheva, M.; Bruzewicz, D. A.; Whitesides, G. M. 10. References , 621. 2003 75 , Chem. (48) Yang, P.; Deng, T.; Zhao, D.; Feng, P.; Pine, D.; Chmelka, B. F.; (1) Anon ( Science , 2244. 282 , 1998 Whitesides, G. M.; Stucky, G. D. (2) Brus, L. E. J. Chem. Phys. , 4403. 80 1984 , 2001 , J. Am. Chem. Soc. (49) Yin, Y. D.; Lu, Y.; Gates, B.; Xia, Y. N. ,1. , 27 (3) Feldheim, D. L.; Keating, C. D. Chem. Soc. Rev. 1998 , 8718. 123 , 24. (4) Kastner, M. A. Phys. Today 1993 , 46 Adv. Funct. Mater. , 95. 11 , 2001 (50) Ozin, G. A.; Yang, S. M. (5) Klein, D. L.; Roth, R.; Lim, A. K. L.; Alivisatos, A. P.; McEuen, (51) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, 389 , 699. P. L. Nature 1997 , 105 , 1103. G. M. Chem. Rev. 2005 , (6) Venema, L. C.; Wildoer, J. W. G.; Janssen, J. W.; Tans, S. J.; 262 1993 Science (52) Crommie, M. F.; Lutz, C. P.; Eigler, D. M. , , Tuinstra, H. L. J. T.; Kouwenhoven, L. P.; Dekker, C. Science 218. , 52. 283 1999 , Rev. Sci. Instrum. (53) Wilder, K.; Soh, H. T.; Atalar, A.; Quate, C. F. , 35 , 1026. 2002 Acc. Chem. Res. (7) Avouris, P. 70 1999 , , 2822. , 1552. 290 , 2000 Science (8) Zhou, C.; Kong, J.; Yenilmez, E.; Dai, H. (54) Geissler, M.; Xia, Y. , 1249. 16 , 2004 Adv. Mater. (9) Liang, W.; Bockrath, M.; Bozovic, D.; Hafner, J. H.; Tinkham, (55) Stewart, M. D.; Patterson, K.; Somervell, M. H.; Willson, C. G. Nature M.; Park, H. 2001 , 411 , 665. J. Phys. Org. Chem. , 767. 13 , 2000 (10) Bockrath, M.; Liang, W.; Bozovic, D.; Hafner, J. H.; Lieber, C. , J. Photopolym. Sci. Technol. (56) Willson, C. G.; Trinque, B. C. 2003 291 , 283. , M.; Tinkham, M.; Park, H. Science 2001 16 , 621. (11) Bockrath, M.; Cobden, D. H.; McEuen, P. L.; Chopra, N. G.; Zettl, , (57) Ito, H. J. Polym. Sci., Part A: Polym. Chem. 2003 41 , 3863. , 1922. 275 , 1997 Science A.; Thess, A.; Smalley, R. E. (58) ASML ( (12) Baughman, R. H.; Cui, C.; Zakhidov, A. A.; Iqbal, Z.; Barisci, J. (59) Switkes, M.; Kunz, R. R.; Rothschild, M.; Sinta, R. F.; Yeung, N.; Spinks, G. M.; Wallace, G. G.; Mazzoldi, A.; De Rossi, D.; , 2794. M.; Baek, S. Y. J. Vac. Sci. Technol. B 2003 , 21 , Science 1999 Rinzler, A. G.; Jaschinski, O.; Roth, S.; Kertesz, M. (60) Mulkens, J.; Flagello, D. G.; Streefkerk, B.; Graeupner, P. J. 284 , 1340. , 3 2004 Microlith. Microfab. Microsys. , 104. (13) Torres, J. A.; Nealey, P. F.; de Pablo, J. J. Phys. Rev. Lett. 2000 , , 2004 J. Microlith. Microfab. Microsys. (61) Owa, S.; Nagasaka, H. 85 , 3221. 3 , 97. 34 , 9139. (14) Tsui, O. K. C.; Zhang, H. F. Macromolecules 2001 , (62) Hoffnagle, J. A.; Hinsberg, W. D.; Sanchez, M.; Houle, F. A. J. (15) Fryer, D. S.; Peters, R. D.; Kim, E. J.; Tomaszewski, J. E.; de 1999 Vac. Sci. Technol. B , 17 , 3306. Macromol- Pablo, J. J.; Nealey, P. F.; White, C. C.; Wu, W. L. (63) Owa, S.; Nagasaka, H.; Ishii, Y.; Hirakawa, O.; Yamamoto, T. ecules , 5627. 2001 , 34 Solid State Technol. 2004 , 47 , 43. (16) Singh, L.; Ludovice, P. J.; Henderson, C. L. Thin Solid Films , (64) Switkes, M.; Rothschild, M. J. Vac. Sci. Technol. B 2001 19 , , 231. 2004 , 449 2353. (17) US-NNI US National Nanotechnology Initiative (http:// (65) Savile Bradbury, B. B. ; BIOS Introduction to Light Microscopy Scientific Publishers Ltd: Oxford, U.K., 1998. , 1161. Nat. Biotech. (18) Whitesides, G. M. , 21 2003 (66) Burnett, J. H.; Kaplan, S. G. J. Microlith. Microfab. Microsys. (19) Maier, S. A.; Brongersma, M. L.; Kik, P. G.; Meltzer, S.; , 68. 3 , 2004 13 Adv. Mater. Requicha, A. A. G.; Atwater, H. A. , 1501. 2001 , (67) Goethals, A. M.; Bisschop, P. D.; Hermans, J.; Jonckheere, R.; J. Van Roey, F.; Van den Heuvel, D.; Eliat, A.; Ronse, K. (20) Colburn, M.; Johnson, S.; Stewart, M.; Damle, S.; Bailey, T. C.; , 549. Photopolym. Sci. Technol. 2003 , 16 Choi, B.; Wedlake, M.; Michaelson, T.; Sreenivasan, S. V.; (68) Mulkens, J.; McClay, J.; Tirri, B.; Brunotte, M.; Mecking, B.; , 1999 Ekerdt, J.; Willson, C. G. Proc. SPIE-Int. Soc. Opt. Eng. 2003 , 753. Jasper, H. 5040 , Proc. SPIE Int. Soc. Opt. Micro. - , 379. 3676 (69) Cerrina, F.; Bollepalli, S.; Khan, M.; Solak, H.; Li, W.; He, D. (21) Xia, Y.; Kim, E.; Zhao, X.-M.; Rogers, J. A.; Prentiss, M.; , 13. 53 , 2000 Microelectron. Eng. 1996 , 273 , 347. Science Whitesides, G. M. J. Photopolym. Sci. (70) Brainard, R. L.; Cobb, J.; Cutler, C. A. J. (22) Chou, S. Y.; Krauss, P. R.; Zhang, W.; Guo, L.; Zhuang, L. 16 , 401. Technol. , 2003 15 , 1997 Vac. Sci. Technol. B , 2897. (71) Golovkina, V. N.; Nealey, P. F.; Cerrina, F.; Taylor, J. W.; Solak, (23) Chou, S. Y.; Krauss, P. R.; Renstrom, P. J. Science 1996 , 272 , , 22 , H. H.; David, C.; Gobrecht, J. J. Vac. Sci. Technol. B 2004 85. 99. (24) Xia, Y.; Whitesides, G. M. Angew. Chem., Int. Ed. Engl. 1998 , (72) Talin, A. A.; Cardinale, G. F.; Wallow, T. I.; Dentinger, P.; , 550. 37 2004 , Pathak, S.; Chinn, D.; Folk, D. R. J. Vac. Sci. Technol. B Chem. Rev. (25) Xia, Y.; Rogers, J. A.; Paul, K. E.; Whitesides, G. M. 22 , 781. 99 , , 1823. 1999 (73) Toriumi, M.; Ishikawa, T.; Kodani, T.; Koh, M.; Moriya, T.; (26) Gates, B. D.; Xu, Q.; Love, J. C.; Wolfe, D. B.; Whitesides, G. M. Yamashita, T.; Araki, T.; Aoyama, H.; Yamazaki, T.; Furukawa, , Annu. Rev. Mater. Res. 2004 34 , 339. T.; Itani, T. , 27. 22 J. Vac. Sci. Technol. B , 2004 , 153. 28 , 1998 Annu. Rev. Mater. Sci. (27) Xia, Y.; Whitesides, G. M. , 21 J. Vac. Sci. Technol. B 2003 (74) Li, Y.; Ota, K.; Murakami, K. , 344 , , 524. Nature (28) Eigler, D. M.; Schweizer, E. K. 1990 127. 386 1997 , 259. (29) Quate, C. F. Surf. Sci. , 1997 , (75) Stulen, R. 97 - 3 , 515. Proc. Electrochem. Soc. , (30) Piner, R. D.; Zhu, J.; Xu, F.; Hong, S.; Mirkin, C. A. 1999 Science (76) Larruquert, J. I.; Keski-Kuha, R. A. M. Appl. Opt. 2002 , 41 , 5398. 283 , 661. (77) Montcalm, C.; Grabner, R. F.; Hudyma, R. M.; Schmidt, M. A.; Angew. Chem., Int. Ed. (31) Ginger, D. S.; Zhang, H.; Mirkin, C. A. Spiller, E.; Walton, C. C.; Wedowski, M.; Folta, J. A. Appl. Opt. 43 2004 , , 30. , 3262. 41 , 2002 Chem. Rev. (32) Kraemer, S.; Fuierer, R. R.; Gorman, C. B. 2003 , 11 J. Vac. Sci. Technol. B , (78) Zaidi, S. H.; Brueck, S. R. J. 1993 , , 4367. 103 658. , 1997 Chem. Rev. (33) Nyffenegger, R. M.; Penner, R. M. 97 , 1195. (79) Bozler, C. O.; Harris, C. T.; Rabe, S.; Rathman, D. D.; Hollis, , 43 , (34) Wouters, D.; Schubert, U. S. 2004 Angew. Chem., Int. Ed. 1994 J. Vac. Sci. Technol. B , 629. M. A.; Smith, H. I. 12 , 2480. (80) Hoffnagle, J. A.; Hinsberg, W. D.; Houle, F. A.; Sanchez, M. I. , Nature (35) Aizenberg, J.; Black, A. J.; Whitesides, G. M. 1999 , 398 16 , 2003 J. Photopolym. Sci. Technol. , 373. 495. (81) Huang, W.-S.; He, W.; Li, W.; Moreau, W. M.; Lang, R.; Medeiros, Nature (36) Aizenberg, J.; Black, A. J.; Whitesides, G. M. , 1998 394 , D. R.; Petrillo, K. E.; Mahorowala, A. P.; Angelopoulos, M.; 868. Deverich, C.; Huang, C.; Rabidoux, P. A. Proc. SPIE-Int. Soc. (37) Pfeiffer, L.; West, K. W.; Stormer, H. L.; Eisenstein, J. P.; Opt. Eng. 2003 , 5130 , 58. 1990 , Appl. Phys. Lett. Baldwin, K. W.; Gershoni, D.; Spector, J. (82) Yasin, S.; Hasko, D. G.; Ahmed, H. Appl. Phys. Lett. 2001 , 78 , 56 , 1697. 2760. (38) Flanders, D. C.; White, A. E. J. Vac. Sci. Technol. 1981 , 19 , 892. (83) Hu, W.; Sarveswaran, K.; Lieberman, M.; Bernstein, G. H. J. (39) Melosh, N. A.; Boukai, A.; Diana, F.; Gerardot, B.; Badolato, A.; Vac. Sci. Technol. B 22 2004 , , 1711. Petroff, P. M.; Heath, J. R. , 112. 300 , 2003 Science Microelectron. Eng. 32 , 1996 , 159. (84) Gamo, K. , (40) Xu, Q.; Gates, B.; Whitesides, G. M. J. Am. Chem. Soc. 2004 (85) Morita, T.; Kometani, R.; Watanabe, K.; Kanda, K.; Haruyama, , 1332. 126 Y.; Hoshino, T.; Kondo, K.; Kaito, T.; Ichihashi, T.; Fujita, J.-i.; (41) Gates, B. D.; Xu, Q.; Thalladi, V. R.; Cao, T.; Knickerbocker, T.; Ishida, M.; Ochiai, Y.; Tajima, T.; Matsui, S. J. Vac. Sci. Technol. , 2004 43 Angew. Chem., Int. Ed. Whitesides, G. M. , 2780. B 2003 , 21 , 2737. Proc. SPIE-Int. Soc. (42) Toh, K. K. H.; Dao, G.; Singh, R.; Gaw, H. (86) Prewett, P. D.; Gentili, M.; Maggiora, R.; Mastrogiacomo, L.; Opt. Eng. 1991 , 1496 , 27. Watson, J. G.; Turner, G. S.; Brown, G. W.; Plumb, D.; Leonard, 290 , 2120. , (43) Zach, M. P.; Ng, K. H.; Penner, R. M. Science 2000 Q.; Cerrina, F. , , 249. 1992 Microelectron. Eng. 17 , , 1991 Science (44) Whitesides, G. M.; Mathias, J. P.; Seto, C. T. 254 (87) Blauner, P. G.; Ro, J. S.; Butt, Y.; Thompson, C. V.; Melngailis, 1312. J. Mater. Res. Soc. Symp. Proc. 1989 , 129 , 483. Supramolecular Chemistry: Concepts and Perspec- (45) Lehn, J. M. 1993 , 567. (88) Kubena, R. L. Mater. Res. Soc. Symp. Proc. 279 , ; John Wiley & Sons: New York, 1995. tives (89) Xia, Y.; McClelland, J. J.; Gupta, R.; Qin, D.; Zhao, X. M.; Sohn, Angew. (46) Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F. L. L.; Celotta, R. J.; Whitesides, G. M. Adv. Mater. , 9 , 147. 1997 Chem., Int. Ed. , 3349. 39 , 2000 1996 , 8 , 837. Adv. Mater. (90) Zhao, X. M.; Xia, Y.; Whitesides, G. M.

23 1193 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication , 376 , 581. (91) Kim, E.; Xia, Y.; Whitesides, G. M. 1995 Nature Pfeiffer, K.; Bleidiessel, G.; Gruetzner, G.; Maximov, M. V.; , 1995 Appl. Phys. Lett. (92) Chou, S. Y.; Krauss, P. R.; Renstrom, P. J. , Heidari, B. Mater. Sci. Eng., C 2003 , 23. 23 , 3114. 67 2000 (131) Li, M.; Wang, J.; Zhuang, L.; Chou, S. Y. , Appl. Phys. Lett. , 1997 (93) Kim, E.; Xia, Y.; Zhao, X. M.; Whitesides, G. M. Adv. Mater. , 673. 76 , 651. 9 Appl. (132) Xia, Q.; Keimel, C.; Ge, H.; Yu, Z.; Wu, W.; Chou, S. Y. Angew. Chem., (94) Emmelius, M.; Pawlowski, G.; Vollmann, H. W. , 2003 Phys. Lett. , 4417. 83 , 1475. 101 , 1989 Int. Ed. (133) Grigaliunas, V.; Tamulevicius, S.; Tomasiunas, R.; Kopustinskas, J. Inject. Mold. (95) Kikuchi, A.; Coulter, J. P.; Angstadt, D. C. 2004 - 453 , , Thin Solid Films V.; Guobiene, A.; Jucius, D. 454 2002 , 91. 6 , Technol. 13. (96) Schift, H.; David, C.; Gobrecht, J.; D’Amore, A.; Simoneta, D.; (134) McAlpine, M. C.; Friedman, R. S.; Lieber, C. M. Nano Lett. 2003 , 2000 Kaiser, W.; Gabriel, M. J. Vac. Sci. Technol. B , 18 , 3564. , 443. 3 (97) Schift, H.; David, C.; Gabriel, M.; Gobrecht, J.; Heyderman, L. (135) Khang, D.-Y.; Yoon, H.; Lee, H. H. , Adv. Mater. 13 , 749. 2001 J.; Kaiser, W.; Koppel, S.; Scandella, L. Microelectron. Eng. 2000 , (136) Otto, M.; Bender, M.; Hadam, B.; Spangenberg, B.; Kurz, H. 53 , 171. 2001 Microelectron. Eng. , 57 - 58 , 361. (98) Schift, H.; Heyderman, L. J.; Gobrecht, J. Chimia 2003 , 56 , 543. 2003 , Microelectron. Eng. (137) Schulz, H.; Wissen, M.; Scheer, H. C. Macromol. Mater. Eng. (99) Gadegaard, N.; Mosler, S.; Larsen, N. B. 67 - 68 , 657. 2003 , 76. 288 , Appl. Phys. Lett. (138) Kumar, A.; Whitesides, G. M. 1993 63 , 2002. , , (100) Janucki, J.; Owsik, J. , 63. 2003 228 Opt. Commun. (139) Fichet, G.; Stutzmann, N.; Muir, B. V. O.; Huck, W. T. S. Adv. (101) Colburn, W. S. , 41 , 443. 1997 J. Imag. Sci. Technol. , , 47. 14 2002 Mater. , 1632. Appl. Phys. Lett. 2003 , (102) Zhang, W.; Chou, S. Y. 83 (140) Trimbach, D.; Feldman, K.; Spencer, N. D.; Broer, D. J.; J. Vac. Sci. Technol. (103) Chou, S. Y.; Krauss, P. R.; Renstrom, P. J. Bastiaansen, C. W. M. , 10957. 19 , 2003 Langmuir 1996 B , 14 , 4129. 2003 , 15 , 647. (141) Tsibouklis, J.; Nevell, T. G. Adv. Mater. (104) Smith, B. J.; Stacey, N. A.; Donnelly, J. P.; Onsongo, D. M.; (142) Rolland, J. P.; Van Dam, R. M.; Schorzman, D. A.; Quake, S. R.; Bailey, T. C.; Mackay, C. J.; Resnick, D. J.; Dauksher, W. J.; DeSimone, J. M. J. Am. Chem. Soc. 2004 , 126 , 2322. Mancini, D. P.; Nordquist, K. J.; Sreenivasan, S. V.; Banerjee, (143) Rolland, J. P.; Hagberg, E. C.; Denison, G. M.; Carter, K. R.; De S. K.; Ekerdt, J. G.; Willson, C. G. Proc. SPIE-Int. Soc. Opt. Eng. , 43 , 5796. Angew. Chem., Int. Ed. Simone, J. M. 2004 2003 , 5037 , 1029. (144) Csucs, G.; Kuenzler, T.; Feldman, K.; Robin, F.; Spencer, N. D. (105) Colburn, M.; Grot, A.; Amistoso, M. N.; Choi, B. J.; Bailey, T. 2003 Langmuir 19 , 6104. , C.; Ekerdt, J. G.; Sreenivasan, S. V.; Hollenhorst, J.; Willson, (145) Choi, S. J.; Yook, P. J.; Baek, S. J.; Kim, T. W.; Lee, H. H. J. C. G. , 453. 3997 , 2000 Proc. SPIE-Int. Soc. Opt. Eng. , Am. Chem. Soc. , 7744. 126 2004 417 , 2002 Nature (106) Chou, S. Y.; Keimel, C.; Gu, J. , 835. 20 , 2004 Langmuir (146) Khang, D.-Y.; Lee, H. H. , 2445. , 237. 35 , (107) Chou, S. Y.; Krauss, P. R. Microelectron. Eng. 1997 , , 75 2003 Anal. Chem. (147) Lee, J. N.; Park, C.; Whitesides, G. M. (108) Resnick, D. J.; Mancini, D.; Dauksher, W. J.; Nordquist, K.; 6544. Bailey, T. C.; Johnson, S.; Sreenivasan, S. V.; Ekerdt, J. G.; (148) DowCorning ( 2003 Willson, C. G. 69 Microelectron. Eng. , 412. , (149) Gelest ( (109) Resnick, D. J.; Dauksher, W. J.; Mancini, D. P.; Nordquist, K. (150) Childs, W. R.; Nuzzo, R. G. J. Am. Chem. Soc. 2002 , 124 , 13583. J.; Ainley, E. S.; Gehoski, K. A.; Baker, J. H.; Bailey, T. C.; Choi, (151) Childs, W. R.; Nuzzo, R. G. Adv. Mater. 2004 , 16 , 1323. B. J.; Johnson, S.; Sreenivasan, S. V.; Ekerdt, J. G.; Willson, C. Siloxane polymers (152) Clarson, S. J.; Semlyen, J. A. ; Prentice Hall: 4688 , , 205. 2002 Proc. SPIE-Int. Soc. Opt. Eng. G. Englewood Cliffs, NJ, 1993. (110) Resnick, D. J.; Dauksher, W. J.; Mancini, D.; Nordquist, K. J.; (153) Perutz, S.; Wang, J.; Kramer, E. J.; Ober, C. K.; Ellis, K. Ainley, E.; Gehoski, K.; Baker, J. H.; Bailey, T. C.; Choi, B. J.; Macromolecules , 31 , 4272. 1998 Johnson, S.; Sreenivasan, S. V.; Ekerdt, J. G.; Willson, C. G. J. 1991 , 7 , 1013. (154) Chaudhury, M. K.; Whitesides, G. M. Langmuir Microlith. Microfab. Microsys. 2002 , 1 , 284. (155) Kim, C.-J.; Kim, J. Y.; Sridharan, B. 1998 , Sens. Actuators A (111) Johnson, S. C.; Bailey, T. C.; Dickey, M. D.; Smith, B. J.; Kim, , 17. 64 E. K.; Jamieson, A. T.; Stacey, N. A.; Ekerdt, J. G.; Willson, C. (156) Delamarche, E.; Schmid, H.; Michel, B.; Biebuyck, H. Adv. Mater. G.; Mancini, D. P.; Dauksher, W. J.; Nordquist, K. J.; Resnick, 9 , 1997 , 741. D. J. Proc. SPIE-Int. Soc. Opt. Eng. , 197. 5037 , 2003 (157) Rogers, J. A.; Paul, K. E.; Whitesides, G. M. J. Vac. Sci. Technol. (112) Resnick, D. J.; Dauksher, W. J.; Mancini, D. P.; Nordquist, K. 16 B 1998 , 88. , J.; Bailey, T. C.; Johnson, S. C.; Stacey, N. A.; Ekerdt, J. G.; , (158) Hui, C. Y.; Jagota, A.; Lin, Y. Y.; Kramer, E. J. 2002 Langmuir Willson, C. G.; Sreenivasan, S. V.; Schumaker, N. E. Proc. SPIE- , 1394. 18 , 12. Int. Soc. Opt. Eng. 5037 2003 , (159) Odom, T. W.; Love, J. C.; Wolfe, D. B.; Paul, K. E.; Whitesides, (113) Jung, G. Y.; Ganapathiappan, S.; Ohlberg, D. A. A.; Olynick, D. Langmuir G. M. 2002 , 18 , 5314. , , 4 Nano Lett. L.; Chen, Y.; Tong, W. M.; Williams, R. S. 2004 (160) Bietsch, A.; Michel, B. 2000 , 88 , 4310. J. Appl. Phys. 1225. Macromolecules 2000 , 33 , 3042. (161) Schmid, H.; Michel, B. (114) Scheer, H. C.; Schulz, H.; Hoffmann, T.; Sotomayor Torres, C. J. (162) Odom, T. W.; Thalladi, V. R.; Love, J. C.; Whitesides, G. M. , M. J. Vac. Sci. Technol. B 1998 16 , 3917. Am. Chem. Soc. , 12112. 124 , 2002 (115) Scheer, H. C.; Schulz, H. Microelectron. Eng. , 56 , 311. 2001 , (163) Gates, B. D.; Whitesides, G. M. J. Am. Chem. Soc. 2003 , 125 5347 (116) McMackin, I. 2004 Proc. SPIE-Int. Soc. Opt. Eng. , 232. , 14986. (117) Ruchhoeft, P.; Colburn, M.; Choi, B.; Nounu, H.; Johnson, S.; (164) Choi, K. M.; Rogers, J. A. J. Am. Chem. Soc. 2003 , 125 , 4060. Bailey, T.; Darmle, S.; Stewart, M.; Ekerdt, J.; Sreenivasan, S. (165) Xia, Y.; Whitesides, G. M. 77 , 1997 Polym. Mater. Sci. Eng. , 596. , 17 , 1999 J. Vac. Sci. Technol. B V.; Wolfe, J. C.; Willson, C. G. , 8078. 20 , 2004 Langmuir (166) Deng, Z.; Mao, C. 2965. (167) Kim, D. H.; Lin, Z.; Kim, H.-C.; Jeong, U.; Russell, T. P. Adv. (118) Colburn, M.; Grot, A.; Choi, B. J.; Amistoso, M.; Bailey, T.; , 15 , 811. Mater. 2003 Sreenivasan, S. V.; Ekerdt, J. G.; Grant Willson, C. J. Vac. Sci. (168) Hua, F.; Sun, Y.; Gaur, A.; Meitl, M.; Bilhaut, L.; Rotkina, L.; , 2162. 19 , 2001 Technol. B Nano Wang, J. F.; Geil, P.; Shim, M.; Rogers, J. A.; Shim, A. (119) Stewart, M. D.; Johnson, S. C.; Sreenivasan, S. V.; Willson, C. Lett. , 2467. 4 2004 , G. 4 2005 , in press. J. Microlithogr., Microfabr., Microsyst. , (169) Xu, Q.; Mayers, B.; Lahav, M.; Vezenov, D. V.; Whitesides, G. (120) Roos, N.; Luxbacher, T.; Glinsner, T.; Pfeiffer, K.; Schulz, H.; J. Am. Chem. Soc. 127 M. 2005 , , in press. Scheer, H.-C. , 427. 4343 , 2001 Proc. SPIE-Int. Soc. Opt. Eng. (170) Dujardin, E.; Ebbesen, T. W.; Hiura, H.; Tanigaki, K. Science , J. Vac. Sci. Technol. B 1998 (121) Tan, H.; Gilbertson, A.; Chou, S. Y. , 1850. 1994 , 265 , 3926. 16 , (171) Martin, C. R. 266 1994 , 1961. Science (122) Haatainen, T.; Ahopelto, J. , 357. 67 , 2003 Phys. Scr. (172) Jeon, N. L.; Choi, I. S.; Xu, B.; Whitesides, G. M. Adv. Mater. (123) Austin, M. D.; Ge, H.; Wu, W.; Li, M.; Yu, Z.; Wasserman, D.; , , 946. 1999 11 2004 , 5299. Appl. Phys. Lett. Lyon, S. A.; Chou, S. Y. 84 , (173) Pisignano, D.; Sariconi, E.; Mazzeo, M.; Gigli, G.; Cingolani, R. (124) Ansari, K.; van Kan, J. A.; Bettiol, A. A.; Watt, F. Appl. Phys. , 1565. 14 , 2002 Adv. Mater. Lett. , 476. 85 , 2004 1997 J. Mater. Chem. (174) Zhao, X. M.; Xia, Y. N.; Whitesides, G. M. , (125) Hoff, J. D.; Cheng, L.-J.; Meyhoefer, E.; Guo, L. J.; Hunt, A. J. , 1069. 7 Nano Lett. 4 , 2004 , 853. (175) LaFratta, C. N.; Baldacchini, T.; Farrer, R. A.; Fourkas, J. T.; (126) Li, H.-W.; Huck, W. T. S. , 1633. 4 , 2004 Nano Lett. J. Phys. Chem. B Teich, M. C.; Saleh, B. E. A.; Naughton, M. J. (127) Behl, M.; Seekamp, J.; Zankovych, S.; Torres, C. M. S.; Zentel, 108 2004 , , 11256. , R.; Ahopelto, J. 2002 Adv. Mater. , 588. 14 (176) Leung, W. Y.; Kang, H.; Constant, K.; Cann, D.; Kim, C. H.; J. Vac. Sci. (128) Makela, T.; Haatainen, T.; Ahopelto, J.; Isotalo, H. 2003 J. Appl. Phys. Biswas, R.; Sigalas, M. M.; Ho, K. M. , , 93 2001 19 , 487. , Technol. B 5866. (129) Finder, C.; Beck, M.; Seekamp, J.; Pfeiffer, K.; Carlberg, P.; Appl. Phys. Lett. 1998 , 73 , 294. (177) Rogers, J. A.; Bao, Z.; Dhar, L. Maximov, I.; Reuther, F.; Sarwe, E. L.; Zankovych, S.; Ahopelto, Appl. Phys. (178) Lawrence, J. R.; Turnbull, G. A.; Samuel, I. D. W. J.; Montelius, L.; Mayer, C.; Sotomayor Torres, C. M. Microelec- Lett. 2003 , 82 , 4023. 67 68 tron. Eng. 2003 , , 623. - (179) Kim, Y. S.; Park, J.; Lee, H. H. Appl. Phys. Lett. , 81 , 1011. 2002 (130) Sotomayor Torres, C. M.; Zankovych, S.; Seekamp, J.; Kam, A. Langmuir (180) Duineveld, P. C.; Lilja, M.; Johansson, T.; Inganaes, O. P.; Clavijo Cedeno, C.; Hoffmann, T.; Ahopelto, J.; Reuther, F.; , 18 , 9554. 2002

24 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1194 (181) Israelachvili, J. N. ; McGraw- Intermolecular and Surface Forces (226) Wiles, J. A.; Grzybowski, B. A.; Winkleman, A.; Whitesides, G. Hill Publishing Co.: Tokyo, Japan, 1991. M. Anal. Chem. 2003 , 75 , 4859. (227) Grzybowski, B. A.; Winkleman, A.; Wiles, J. A.; Brumer, Y.; Polymer Interface and Adhesion ; Marcel Dekker: New (182) Wu, S. 2003 Whitesides, G. M. Nat. Mater. , 2 , 241. York, 1982. (228) Ferreira, G. F. L.; Figueiredo, M. T. IEEE Trans. Electron. Insul. (183) Abdo, A. Y.; Zheng, L.; Wei, A.; Mikkelson, A.; Nellis, G.; , 1992 27 , 719. , - 73 , 2004 Microelectron. Eng. Engelstad, R. L.; Lovell, E. G. 74 13 , 2001 Adv. Mater. (229) Mesquida, P.; Stemmer, A. , 1395. 161. (230) Stemmer, A.; Mesquida, P.; Naujoks, N. Chimia 2003 , 56 , 573. (184) Rogers, J. A.; Bao, Z.; Meier, M.; Dodabalapur, A.; Schueller, O. Surf. Interface Anal. (231) Mesquida, P.; Knapp, H. F.; Stemmer, A. J. A.; Whitesides, G. M. Synth. Met. 115 , 2000 ,5. , 159. 2002 , 33 Langmuir (185) Kumar, A.; Biebuyck, H. A.; Whitesides, G. M. , 1994 Appl. Phys. (232) Barry, C. R.; Lwin, N. Z.; Zheng, W.; Jacobs, H. O. 10 , 1498. , 5527. 83 , 2003 Lett. (186) Xia, Y.; Zhao, X.-M.; Kim, E.; Whitesides, G. M. Chem. Mater. (233) Barry, C. R.; Steward, M. G.; Lwin, N. Z.; Jacobs, H. O. , 2332. 7 , 1995 , 1057. 2003 Nanotechnology , 14 (187) Kumar, A.; Abbott, N. L.; Biebuyck, H. A.; Kim, E.; Whitesides, J. Am. (234) Loo, Y.-L.; Willett, R. L.; Baldwin, K. W.; Rogers, J. A. G. M. Acc. Chem. Res. 1995 28 , 219. , 124 , 7654. 2002 , Chem. Soc. Adv. Mater. (188) Wilbur, J. L.; Kumar, A.; Kim, E.; Whitesides, G. M. (235) Loo, Y.-L.; Willett, R. L.; Baldwin, K. W.; Rogers, J. A. Appl. , 600. 1994 , 6 2002 Phys. Lett. 81 , 562. , (189) Delamarche, E.; Geissler, M.; Wolf, H.; Michel, B. J. Am. Chem. (236) Zaumseil, J.; Meitl, M. A.; Hsu, J. W. P.; Acharya, B. R.; Baldwin, , 3834. 124 , 2002 Soc. , 1223. K. W.; Loo, Y.-L.; Rogers, J. A. Nano Lett. 2003 , 3 (190) Love, J. C.; Wolfe, D. B.; Chabinyc, M. L.; Paul, K. E.; Whitesides, (237) Schmid, H.; Wolf, H.; Allenspach, R.; Riel, H.; Karg, S.; Michel, G. M. J. Am. Chem. Soc. 2002 , 124 , 1576. , 2003 Adv. Funct. Mater. B.; Delamarche, E. , 145. 13 (191) Koide, Y.; Such, M. W.; Basu, R.; Evmenenko, G.; Cui, J.; Dutta, , Appl. Phys. Lett. (238) Kim, C.; Shtein, M.; Forrest, S. R. 2002 , 80 P.; Hersam, M. C.; Marks, T. J. , 86. 19 , 2003 Langmuir 4051. (192) Geissler, M.; Wolf, H.; Stutz, R.; Delamarche, E.; Grummt, U.- Adv. (239) Wang, Z.; Yuan, J.; Zhang, J.; Xing, R.; Yan, D.; Han, Y. W.; Michel, B.; Bietsch, A. 2003 Langmuir , 6301. 19 , , 1009. 15 , 2003 Mater. (193) Geissler, M.; Schmid, H.; Bietsch, A.; Michel, B.; Delamarche, 3 , 2003 Nano Lett. (240) Schaper, C. D. , 1305. 18 , 2374. , E. Langmuir 2002 (241) Jeon, S.; Menard, E.; Park, J.-U.; Maria, J.; Meitl, M.; Zaumseil, J. Am. Chem. Soc. (194) Xia, Y.; Whitesides, G. M. 1995 , , 3274. 117 16 , 1369. 2004 Adv. Mater. J.; Rogers, J. A. , (195) Biebuyck, H. A.; Larsen, N. B.; Delamarche, E.; Michel, B. IBM Langmuir (242) Menard, E.; Bilhaut, L.; Zaumseil, J.; Rogers, J. A. J. Res. Dev. , 41 , 159. 1997 2004 20 , 6871. , J. (196) Li, S. P.; Natali, M.; Lebib, A.; Pepin, A.; Chen, Y.; Xu, Y. B. 1986 Appl. Phys. Lett. (243) Lasky, J. B. 48 , 78. , Magn. Magn. Mater. , 447. 241 , 2002 , 323. (244) Tong, Q. Y. 2001 , B87 Mater. Sci. Eng., B (197) Rogers, J. A.; Baldwin, K.; Bao, Z.; Dodabalapur, A.; Raju, V. (245) Helt, J. M.; Drain, C. M.; Batteas, J. D. , 2004 J. Am. Chem. Soc. , 2001 Mater. Res. Soc. Symp. Proc. R.; Ewing, J.; Amundson, K. 126 , 628. , JJ7 1/1. 660 Nano Lett. (246) Loo, Y.-L.; Lang, D. V.; Rogers, J. A.; Hsu, J. W. P. 1996 , 1015. 8 , (198) Xia, Y.; Qin, D.; Whitesides, G. M. Adv. Mater. 2003 , 3 , 913. , 13 , 5349. Langmuir 1997 (199) Tien, J.; Terfort, A.; Whitesides, G. M. (247) Jackman, R. J.; Duffy, D. C.; Cherniavskaya, O.; Whitesides, G. Adv. Mater. (200) Rogers, J. A.; Jackman, R. J.; Whitesides, G. M. , 15 , 2973. M. 1999 Langmuir , 1997 9 , 475. (248) Jackman, R. J.; Duffy, D. C.; Ostuni, E.; Willmore, N. D.; (201) Jackman, R. J.; Wilbur, J. L.; Whitesides, G. M. Science 1995 , , 2280. Anal. Chem. Whitesides, G. M. 1998 70 , 269 , 664. (249) Li, H.; Kang, D.-J.; Blamire, M. G.; Huck, W. T. S. Nano Lett. (202) Delamarche, E.; Donzel, C.; Kamounah, F. S.; Wolf, H.; Geissler, 2002 , , 347. 2 M.; Stutz, R.; Schmidt-Winkel, P.; Michel, B.; Mathieu, H. J.; , 1808. 2000 Science (250) Hong, S.; Mirkin, C. A. , 288 19 , 8749. Schaumburg, K. Langmuir 2003 , (251) Hong, S.; Zhu, J.; Mirkin, C. A. 286 , , 523. Science 1999 (203) Donzel, C.; Geissler, M.; Bernard, A.; Wolf, H.; Michel, B.; (252) Ryu, K. S.; Wang, X.; Shaikh, K.; Bullen, D.; Goluch, E.; Zou, , 13 , 1164. Hilborn, J.; Delamarche, E. Adv. Mater. 2001 J.; Liu, C.; Mirkin, C. A. Appl. Phys. Lett. 2004 , 85 , 136. Adv. (204) Hyun, J.; Ma, H.; Zhang, Z.; Beebe, T. P., Jr.; Chilkoti, A. , (253) Lee, K.-B.; Lim, J.-H.; Mirkin, C. A. J. Am. Chem. Soc. 2003 , 15 , 576. Mater. 2003 , 5588. 125 (205) Kunzler, J. F.; Salamone, J. C. Polym. Prepr. (Am. Chem. Soc., (254) Smith, J. C.; Lee, K.-B.; Wang, Q.; Finn, M. G.; Johnson, J. E.; 2003 44 , 215. , Div. Polym. Chem.) 2003 Nano Lett. , 3 Mrksich, M.; Mirkin, C. A. , 883. (206) Lange, S. A.; Benes, V.; Kern, D. P.; Hoerber, J. K. H.; Bernard, Nano Lett. (255) Lee, K.-B.; Kim, E.-Y.; Mirkin, C. A.; Wolinsky, S. M. , 2004 Anal. Chem. , 1641. 76 A. , 4 , 1869. 2004 (207) Bernard, A.; Renault, J. P.; Michel, B.; Bosshard, H. R.; (256) Nam, J.-M.; Han, S. W.; Lee, K.-B.; Liu, X.; Ratner, M. A.; Adv. Mater. 2000 Delamarche, E. , 12 , 1067. Angew. Chem., Int. Ed. 2004 , 43 Mirkin, C. A. , 1246. (208) Bernard, A.; Delamarche, E.; Schmid, H.; Michel, B.; Bosshard, , 1480. (257) Zhang, H.; Mirkin, C. A. Chem. Mater. 2004 , 16 Langmuir 1998 , 14 H. R.; Biebuyck, H. , 2225. , 43. Nano Lett. 2003 , 3 (258) Zhang, H.; Chung, S.-W.; Mirkin, C. A. (209) James, C. D.; Davis, R. C.; Kam, L.; Craighead, H. G.; Isaacson, (259) Rozhok, S.; Piner, R.; Mirkin, C. A. J. Phys. Chem. B 2003 , 107 , 1998 , Langmuir M.; Turner, J. N.; Shain, W. 14 , 741. 751. (210) Branch, D. W.; Corey, J. M.; Weyhenmeyer, J. A.; Brewer, G. Langmuir 2001 , 17 , 5971. (260) Schwartz, P. V. 1998 Med. Biol. Eng. Comput. J.; Wheeler, B. C. 36 , 135. , 18 (261) Schwartz, P. V. Langmuir 2002 , , 4041. (211) Csucs, G.; Michel, R.; Lussi Jost, W.; Textor, M.; Danuser, G. , Phys. Rev. Lett. 2004 (262) Jang, J.; Schatz, G. C.; Ratner, M. A. 92 , , 1713. 24 Biomaterials 2003 , 08550401. (212) Renault, J. P.; Bernard, A.; Juncker, D.; Michel, B.; Bosshard, 90 , (263) Jang, J.; Schatz, G. C.; Ratner, M. A. , 2003 Phys. Rev. Lett. H. R.; Delamarche, E. , 2320. 41 , 2002 Angew. Chem., Int. Ed. 15610401. (213) Bernard, A.; Fitzli, D.; Sonderegger, P.; Delamarche, E.; Michel, (264) Jang, J.; Hong, S.; Schatz, G. C.; Ratner, M. A. J. Chem. Phys. Nat. Biotech. B.; Bosshard, H. R.; Biebuyck, H. 2001 , 19 , 866. , 115 , 2721. 2001 Langmuir (214) Hidber, P. C.; Helbig, W.; Kim, E.; Whitesides, G. M. (265) Liu, J.-F.; Cruchon-Dupeyrat, S.; Garno, J. C.; Frommer, J.; Liu, , 1375. 1996 , 12 , 937. G.-Y. Nano Lett. 2002 , 2 (215) Santhanam, V.; Liu, J.; Agarwal, R.; Andres, R. P. Langmuir (266) Liu, G.-Y.; Xu, S.; Qian, Y. , 457. 33 , 2000 Acc. Chem. Res. 2003 , 7881. 19 , (267) Rolandi, M.; Suez, I.; Dai, H.; Frechet, J. M. J. 2004 , Nano Lett. (216) Shin, H. S.; Yang, H. J.; Jung, Y. M.; Kim, S. B. Vib. Spectrosc. 4 , 889. 29 , 2002 , 79. (268) Garno, J. C.; Yang, Y.; Amro, N. A.; Cruchon-Dupeyrat, S.; Chen, (217) Cherniavskaya, O.; Adzic, A.; Knutson, C.; Gross, B. J.; Zang, , 2003 3 S.; Liu, G.-Y. , 389. Nano Lett. , 7029. 18 , 2002 Langmuir L.; Liu, R.; Adams, D. M. , (269) Carpick, R. W.; Salmeron, M. 1997 Chem. Rev. , 1163. 97 2004 , (218) Choi, D.-G.; Yu, H. K.; Yang, S.-M. Mater. Sci. Eng., C (270) Porter, L. A., Jr.; Ribbe, A. E.; Buriak, J. M. 3 , 2003 Nano Lett. , C24 , 213. 1043. Adv. Mater. (219) Wang, M.; Braun, H.-G.; Kratzmuller, T.; Meyer, E. (271) Bouzehouane, K.; Fusil, S.; Bibes, M.; Carrey, J.; Blon, T.; Du, 13 , 2001 , 1312. M. L.; Seneor, P.; Cros, V.; Vila, L. Nano Lett. 2003 , 3 , 1599. Langmuir (220) Li, H.-W.; Muir, B. V. O.; Fichet, G.; Huck, W. T. S. (272) Dagata, J. A.; Schneir, J.; Harary, H. H.; Evans, C. J.; Postek, 2003 19 , 1963. , Appl. Phys. Lett. 1990 , 56 , 2001. M. T.; Bennett, J. , , 1763. 291 Science 2001 (221) Jacobs, H. O.; Whitesides, G. M. Appl. Phys. Lett. (273) Snow, E. S.; Campbell, P. M.; McMarr, P. J. (222) Loo, Y.-L.; Hsu, J. W. P.; Willett, R. L.; Baldwin, K. W.; West, , 749. 63 , 1993 K. W.; Rogers, J. A. , , 2853. 20 J. Vac. Sci. Technol. B 2002 Appl. Phys. Lett. , 1932. 64 , 1994 (274) Snow, E. S.; Campbell, P. M. (223) Wolfe, D. B.; Love, J. C.; Gates, B. D.; Whitesides, G. M.; Conroy, (275) Day, H. C.; Allee, D. R. , 2691. 62 , 1993 Appl. Phys. Lett. 2004 R. S.; Prentiss, M. , 84 , 1623. Appl. Phys. Lett. (276) Song, H. J.; Rack, M. J.; Abugharbieh, K.; Lee, S. Y.; Khan, V.; (224) Jacobs, H. O.; Campbell, S. A.; Steward, M. G. 2002 Adv. Mater. , , 3720. Ferry, D. K.; Allee, D. R. 12 , 1994 J. Vac. Sci. Technol. B , 1553. 14 (277) Gwo, S.; Yeh, C. L.; Chen, P. F.; Chou, Y. C.; Chen, T. T.; Chao, (225) Jacobs, H. O.; Stemmer, A. , Surf. Interface Anal. , 361. 1999 27 , 74 , 1090. 1999 T. S.; Hu, S. F.; Huang, T. Y. Appl. Phys. Lett.

25 1195 Chemical Reviews, 2005, Vol. 105, No. 4 New Approaches to Nanofabrication , 73 , 1508. (278) Dai, H.; Franklin, N.; Han, J. 1998 Appl. Phys. Lett. (325) Maria, J.; Jeon, S.; Rogers, J. A. J. Photochem. Photobiol., A , 149. 2004 , 166 , 1055. 2002 2 Nano Lett. (279) Liu, S.; Maoz, R.; Schmid, G.; Sagiv, J. , , 1426. 12 , 2000 Adv. Mater. (326) Yin, Y.; Gates, B.; Xia, Y. 2000 Adv. Mater. , (280) Maoz, R.; Frydman, E.; Cohen, S. R.; Sagiv, J. (327) Rogers, J. A.; Dodabalapur, A.; Bao, Z.; Katz, H. E. Appl. Phys. 12 , 725. 1999 Lett. , 1010. 75 , (281) Sun, S.; Leggett, G. J. Nano Lett. 2002 , 2 , 1223. Appl. Opt. (328) Paul, K. E.; Zhu, C.; Love, J. C.; Whitesides, G. M. , (282) Sun, S.; Chong, K. S. L.; Leggett, G. J. J. Am. Chem. Soc. 2002 , 2001 40 , 4557. 124 , 2414. (329) Aizenberg, J.; Rogers, J. A.; Paul, K. E.; Whitesides, G. M. Appl. (283) Sun, S.; Leggett, G. J. , 1381. 4 Nano Lett. 2004 , , Opt. 37 , 2145. 1998 (284) Vettiger, P.; Brugger, J.; Despont, M.; Drechsler, U.; Durig, U.; Adv. (330) Babayan, Y.; Barton, J. E.; Greyson, E. C.; Odom, T. W. Haberle, W.; Lutwyche, M.; Rothuizen, H.; Stutz, R.; Widmer, 2004 , 1341. 16 , Mater. Microelectron. Eng. , 11. 46 , 1999 R.; Binnig, G. (331) Crowell, J. E. 2003 , S88. 21 , J. Vac. Sci. Technol. A (285) Vettiger, P.; Despont, M.; Drechsler, U.; Durig, U.; Haberle, W.; (332) Rossnagel, S. M. 21 , S74. , 2003 J. Vac. Sci. Technol. A Lutwyche, M. I.; Rothuizen, H. E.; Stutz, R.; Widmer, R.; Binnig, (333) Fasol, G.; Runge, K. 1997 , 70 , 2467. Appl. Phys. Lett. , 323. G. K. IBM J. Res. Dev. 2000 , 44 Nano (334) Chopra, N.; Kichambare, P. D.; Andrews, R.; Hinds, B. J. (286) Vettiger, P.; Binnig, G. 1 , 47. Sci. Am. 2003 , Lett. , 1177. 2 , 2002 (287) Bullen, D.; Chung, S.-W.; Wang, X.; Zou, J.; Mirkin, C. A.; Liu, , (335) Morris, R. B.; Franta, D. J.; White, H. S. J. Phys. Chem. 1987 C. 84 Appl. Phys. Lett. , 789. 2004 , , 3559. 91 (288) Zhang, M.; Bullen, D.; Chung, S.-W.; Hong, S.; Ryu, K. S.; Fan, Appl. (336) Stormer, H. L.; Baldwin, K. W.; Pfeiffer, L. N.; West, K. W. , Z.; Mirkin, C. A.; Liu, C. , 212. 13 Nanotechnology 2002 1991 , 1111. 59 , Phys. Lett. (289) Minne, S. C.; Flueckiger, P.; Soh, H. T.; Quate, C. F. J. Vac. (337) Natelson, D.; Willett, R. L.; West, K. W.; Pfeiffer, L. N. Appl. , 13 , 1380. Sci. Technol. B 1995 Phys. Lett. , 77 , 1991. 2000 (290) Despont, M.; Brugger, J.; Drechsler, U.; Durig, U.; Haberle, W.; 3 , Nano Lett. (338) Artemyev, M.; Moeller, B.; Woggon, U. , 509. 2003 Lutwyche, M.; Rothuizen, H.; Stutz, R.; Widmer, R.; Rohrer, H.; , (339) Lehn, J. M. Angew. Chem., Int. Ed. Engl. 1990 29 , 1304. Binnig, G.; Vettiger, P. Technical Digest 12th IEEE Int’l Micro (340) Chakraborty, A. K.; Golumbfskie, A. J. Annu. Rev. Phys. Chem. Electro Mechanical Systems Conf. “MEMS ’99” , Orlando, FL, , 2001 52 , 537. 1999, p 564. (341) De Wild, M.; Berner, S.; Suzuki, H.; Ramoino, L.; Baratoff, A.; J. Am. (291) Black, A. J.; Paul, K. E.; Aizenberg, J.; Whitesides, G. M. ; New York Academy of Molecular Electronics III Jung, T. A. In , 8356. Chem. Soc. , 121 1999 Sciences: New York, 2003; Vol. 1006. (292) Rogers, J. A.; Paul, K. E.; Jackman, R. J.; Whitesides, G. M. (342) Muchado, V. G.; Baxter, P. N. W.; Lehn, J. M. J. Braz. Chem. 70 1997 , 2658. Appl. Phys. Lett. , 12 , , 431. 2001 Soc. (293) Fasol, G. 280 , 1998 Science , 545. (343) Pileni, M. P.; Lalatonne, Y.; Ingert, D.; Lisiecki, I.; Courty, A. (294) Nichols, R. J.; Kolb, D. M.; Behm, R. J. J. Electroanal. Chem. , 251. Faraday Discuss. 2004 , 125 1991 Interface Electrochem. , 109. 313 , Proc. Natl. Acad. Sci. U.S.A. (344) Whitesides, G. M.; Boncheva, M. (295) Himpsel, F. J.; Ortega, J. E. 1994 , 50 , 4992. Phys. Rev. B , 2002 , 4769. 99 (296) Himpsel, F. J.; Jung, T.; Kirakosian, A.; Lin, J. L.; Petrovykh, , 31 , 2001 Annu. Rev. Mater. Res. (345) Fasolka, M. J.; Mayes, A. M. , 24 , 20. 1999 MRS Bull. D. Y.; Rauscher, H.; Viernow, J. 323. (297) Dekoster, J.; Degroote, B.; Pattyn, H.; Langouche, G.; Vantomme, 14 , 2002 Adv. Mater. (346) Krausch, G.; Magerle, R. , 1579. , 938. A.; Degroote, S. Appl. Phys. Lett. 1999 , 75 Annu. Rev. Mater. (347) Murray, C. B.; Kagan, C. R.; Bawendi, M. G. (298) Love, J. C.; Paul, K. E.; Whitesides, G. M. Adv. Mater. 2001 , 2000 Sci. , 30 , 545. 13 , 604. (348) Sun, S. H.; Murray, C. B.; Weller, D.; Folks, L.; Moser, A. Science J. Appl. Phys. (299) Mundschau, M.; Bauer, E.; Telieps, W.; Swiech, W. , 1989. 287 , 2000 1989 , 65 , 4747. 2002 Nano Lett. (349) Li, L. S.; Walda, J.; Manna, L.; Alivisatos, A. P. , , Phys. Rev. Lett. 74 1995 (300) Jung, T.; Mo, Y. W.; Himpsel, F. J. , 2 , 557. 1641. (350) Jeoung, E.; Galow, T. H.; Schotter, J.; Bal, M.; Ursache, A.; (301) de la Figuera, J.; Huerta-Garnica, M. A.; Prieto, J. E.; Ocal, C.; Tuominen, M. T.; Stafford, C. M.; Russell, T. P.; Rotello, V. M. , Miranda, R. Appl. Phys. Lett. 1995 66 , 1006. , 2001 , 6396. 17 Langmuir (302) Jung, T.; Schlittler, R.; Gimzewski, J. K.; Himpsel, F. J. Appl. , 1057. , 2003 Nano Lett. (351) McFarland, A. D.; Van Duyne, R. P. 3 Phys. A 1995 , 61 , 467. (352) Lindsey, J. S. , 153. 1991 , 15 New J. Chem. (303) Gambardella, P.; Blanc, M.; Brune, H.; Kuhnke, K.; Kern, K. 27 Angew. Chem., Int. Ed. Engl. , 89. (353) Lehn, J. M. 1988 , 2000 , 2254. 61 , Phys. Rev. B (354) Gale, P. A. , , 431. 2000 Philos. Trans. R. Soc. A 358 Phys. (304) Morin, S.; Lachenwitzer, A.; Magnussen, O. M.; Behm, R. J. , (355) Caulder, D. L.; Raymond, K. N. Acc. Chem. Res. 1999 , 975. 32 83 , 5066. , 1999 Rev. Lett. 2003 (356) Seeman, N. C. Chem. Biol. , 10 , 1151. (305) Petrovykh, D. Y.; Himpsel, F. J.; Jung, T. Surf. Sci. , 407 , 1998 (357) Zimmerman, S. C.; Lawless, L. J. In ; Springer- Dendrimers IV 189. Verlag: Berlin, 2001; Vol. 217. Phys. Rev. (306) Hahn, E.; Schief, H.; Marsico, V.; Fricke, A.; Kern, K. Nano Lett. 2004 , (358) Shu, D.; Moll, W.-D.; Deng, Z.; Mao, C.; Guo, P. 72 , 3378. Lett. 1994 , 4 , 1717. (307) Otero, R.; Rosei, F.; Naitoh, Y.; Jiang, P.; Thostrup, P.; Gourdon, J. Am. Chem. (359) Liu, D.; Wang, M.; Deng, Z.; Walulu, R.; Mao, C. A.; Laegsgaard, E.; Stensgaard, I.; Joachim, C.; Besenbacher, Soc. 2004 , 126 , 2324. F. , 75. 4 , 2004 Nano Lett. Angew. Chem., Int. Ed. , 2004 43 (360) Deng, Z.; Mao, C. , 4068. J. Phys. Chem. B (308) Penner, R. M. , 3339. 106 , 2002 (361) Yang, J.; Mayer, M.; Kriebel, J. K.; Garstecki, P.; Whitesides, (309) Walter, E. C.; Murray, B. J.; Favier, F.; Kaltenpoth, G.; Grunze, 43 , 2004 Angew. Chem., Int. Ed. G. M. , 1555. , 11407. 106 , 2002 J. Phys. Chem. B M.; Penner, R. M. (362) Knez, M.; Sumser, M.; Bittner, A. M.; Wege, C.; Jeske, H.; (310) Favier, F.; Walter, E. C.; Zach, M. P.; Benter, T.; Penner, R. M. Martin, T. P.; Kern, K. 14 , 2004 Adv. Funct. Mater. , 116. , 2227. 293 2001 Science , (363) Knez, M.; Bittner, A. M.; Boes, F.; Wege, C.; Jeske, H.; Maiss, , 4 2004 , 2009. Nano Lett. (311) Menke, E. J.; Li, Q.; Penner, R. M. E.; Kern, K. , Nano Lett. 2003 3 , 1079. 2004 , 3402. 16 , Chem. Mater. (312) Li, Q.; Olson, J. B.; Penner, R. M. 3 , 1545. Nano Lett. 2003 , (364) Deng, Z. X.; Mao, C. D. (313) Li, Q.; Newberg, J. T.; Walter, E. C.; Hemminger, J. C.; Penner, (365) Yan, H.; Park, S. H.; Finkelstein, G.; Reif, J. H.; LaBean, T. H. 2004 , 4 , 277. R. M. Nano Lett. Science 301 , 1882. 2003 , , 4 2004 , (314) Murray, B. J.; Walter, E. C.; Penner, R. M. Nano Lett. (366) Kim, S. O.; Solak, H. H.; Stoykovich, M. P.; Ferrier, N. J.; de 665. 2003 Nature Pablo, J. J.; Nealey, P. F. 424 , , 411. , 74 , 2002 Anal. Chem. (315) Walter, E. C.; Favier, F.; Penner, R. M. (367) Shin, K.; Leach, K.; Goldbach, J.; Kim, D.; Jho, J.; Tuominen, 1546. 2 , 933. 2002 Nano Lett. M.; Hawker, C.; Russell, T. , (316) Kaltenpoth, G.; Schnabel, P.; Menke, E.; Walter, E. C.; Grunze, (368) Thurn-Albrecht, T.; Steiner, R.; DeRouchey, J.; Stafford, C.; M.; Penner, R. M. 2003 , 75 , 4756. Anal. Chem. Adv. Huang, E.; Bal, M.; Tuominen, M.; Hawker, C.; Russell, T. , 2259. (317) Sundar, V. C.; Aizenberg, J. , 83 2003 Appl. Phys. Lett. , Mater. 2000 12 , 787. (318) Yang, H.; Love, J. C.; Arias, F.; Whitesides, G. M. Chem. Mater. (369) von Werne, T. A.; Germack, D. S.; Hagberg, E. C.; Sheares, V. 14 , 2002 , 1385. J. Am. Chem. Soc. V.; Hawker, C. J.; Carter, K. R. , 125 , 2003 , 1980 Appl. Phys. Lett. (319) Prober, D. E.; Feuer, M. D.; Giordano, N. 3831. , 94. 37 (370) Edwards, E. W.; Montague, M. F.; Solak, H. H.; Hawker, C. J.; (320) Tas, N. R.; Berenschot, J. W.; Mela, P.; Jansen, H. V.; Elwens- , 1315. Nealey, P. F. Adv. Mater. 2004 , 16 2002 , , 1031. Nano Lett. poek, M.; van den Berg, A. 2 2004 , 4 , Nano Lett. (371) Sundrani, D.; Darling, S. B.; Sibener, S. J. (321) Choi, Y.-K.; Zhu, J.; Grunes, J.; Bokor, J.; Somorjai, G. A. J. 273. , 3340. 2003 Phys. Chem. B , 107 , (372) Sundrani, D.; Darling, S. B.; Sibener, S. J. Langmuir 2004 , 20 16 Adv. Mater. 2004 , (322) Greyson, E. C.; Babayan, Y.; Odom, T. W. , 5091. 1348. 277 , 1232. (373) Decher, G. Science 1997 , (323) Rogers, J. A.; Paul, K. E.; Jackman, R. J.; Whitesides, G. M. J. 4 , 1999 (374) Hammond, P. T. Curr. Opin. Colloid Interface Sci. , 430. 1998 16 Vac. Sci. Technol. B , , 59. (375) Clark, S.; Handy, E.; Rubner, M.; Hammond, P. Adv. Mater. 78 , 2431. Appl. Phys. Lett. , (324) Li, Z.-Y.; Yin, Y.; Xia, Y. 2001 , 1031. 11 1999 ,

26 Gates et al. Chemical Reviews, 2005, Vol. 105, No. 4 1196 (376) Afsharrad, T.; Bailey, A. I.; Luckham, P. F.; Macnaughtan, W.; (399) Yang, X.; Liu, C.; Ahner, J.; Yu, J.; Klemmer, T.; Johns, E.; , Colloids Surf. 25 Chapman, D. 1987 , 263. , Weller, D. J. Vac. Sci. Technol. B 2004 22 , 31. 36 , 2551. , Anal. Lett. (377) Campas, M.; O’Sullivan, C. 2003 (400) Cui, Y.; Bjork, M. T.; Liddle, J. A.; Sonnichsen, C.; Boussert, B.; Adv. Mater. 2004 , 1271. 16 , (378) Hammond, P. T. Alivisatos, A. P. Nano Lett. , 4 , 1093. 2004 , 11. 13 , 2001 Adv. Mater. (379) Caruso, F. (401) Xia, Y. N.; Yang, P. D.; Sun, Y. G.; Wu, Y. Y.; Mayers, B.; Gates, , Chem. Mater. 11 , 1999 (380) Caruso, F.; Schuler, C.; Kurth, D. G. , 353. 2003 , 15 Adv. Mater. B.; Yin, Y. D.; Kim, F.; Yan, Y. Q. 3394. Science , 2001 (402) Huang, Y.; Duan, X. F.; Wei, Q. Q.; Lieber, C. M. 2004 Nano Lett. (381) Schneider, G.; Decher, G. , , 1833. 4 291 , 630. Adv. Mater. , 671. 2004 , (382) Shchukin, D. G.; Sukhorukov, G. B. 16 (403) Smith, P. A.; Nordquist, C. D.; Jackson, T. N.; Mayer, T. S.; (383) Shchukin, D. G.; Shutava, T.; Shchukina, E.; Sukhorukov, G. Martin, B. R.; Mbindyo, J.; Mallouk, T. E. 2000 , Appl. Phys. Lett. , 16 , 3446. Chem. Mater. 2004 B.; Lvov, Y. M. , 1399. 77 ; Springer-Verlag: Berlin, (384) Caruso, F. In Colloid Chemistry II (404) Lieber, C. M. Personal communication. 2003; Vol. 227. (405) Srinivasan, U.; Liepmann, D.; Howe, R. J. Microelectromech. Sys. J. Mater. Chem. , 459. 14 , 2004 (385) Wang, D. Y.; Mohwald, H. , 2001 , 17. 10 (386) Pham, T.; Jackson, J. B.; Halas, N. J.; Lee, T. R. Langmuir 2002 , (406) Srinivasan, U.; Helmbrecht, M.; Rembe, C.; Muller, R.; Howe, , 4915. 18 R. 8 , 2002 IEEE J. Sel. Top. Quantum ,4. , 1146. Chem. Mater. 2001 , (387) Yin, Y.; Lu, Y.; Gates, B.; Xia, Y. 13 (407) Snyder, E.; Chideme, J.; Craig, G. Jpn. J. Appl. Phys. 2002 , , 41 4 , 1525. , 2004 Nano Lett. (388) Barton, J. E.; Odom, T. W. 4366. (389) Harnack, O.; Pacholski, C.; Weller, H.; Yasuda, A.; Wessels, J. (408) Rogers, J. A.; Bao, Z.; Baldwin, K.; Dodabalapur, A.; Crone, B.; M. , 1097. , 2003 Nano Lett. 3 Raju, V. R.; Kuck, V.; Katz, H.; Amundson, K.; Ewing, J.; Drzaic, (390) Love, J. C.; Urbach, A. R.; Prentiss, M. G.; Whitesides, G. M. J. , P. 98 Proc. Natl. Acad. Sci. U.S.A. , 4835. 2001 Am. Chem. Soc. , 12696. 125 , 2003 , J. Polym. Sci., Part A: Polym. Chem. 2002 (409) Rogers, J. A.; Bao, Z. (391) Gourdon, D.; Yasa, M.; Alig, A. R. G.; Li, Y.; Safinya, C. R.; , 3327. 40 2004 , Adv. Funct. Mater. Israelachvili, J. N. , 238. 14 181 (410) Anfinsen, C. B. Science , 223. 1973 , Nano Lett. , 2003 (392) Nagle, L.; Ryan, D.; Cobbe, S.; Fitzmaurice, D. (411) Grzybowski, B. A.; Radkowski, M.; Campbell, C. J.; Lee, J. N.; 3 , 51. Whitesides, G. M. 2004 , 84 , 1798. Appl. Phys. Lett. (393) Nagle, L.; Fitzmaurice, D. Adv. Mater. 2003 15 , 933. , , 629. (412) Conn, M. M.; Rebek, J. 4 , 1994 Curr. Opin. Struct. Biol. , 2001 J. Am. Chem. Soc. (394) Yin, Y.; Lu, Y.; Gates, B.; Xia, Y. 123 , 8718. , April 19. 38 , 1965 Electron. Lett. (413) Moore, G. E. (395) Hutchinson, T. O.; Liu, Y.-P.; Kiely, C.; Kiely, C. J.; Brust, M. (414) Schellekens, J.; DBurdinski, D.; Saalmink, M.; Beenhakkers, M.; Adv. Mater. 2001 , , 1800. 13 Mater. Res. Soc. Symp. Proc. , Gelinck, G.; Decre, M. M. J. 2004 (396) Lu, N.; Chen, X.; Molenda, D.; Naber, A.; Fuchs, H.; Talapin, D. , M2.9. EXS-2 V.; Weller, H.; Mueller, J.; Lupton, J. M.; Feldmann, J.; Rogach, (415) Schellekens, J.; DBurdinski, D.; Saalmink, M.; Beenhakkers, M.; , 885. Nano Lett. 2004 A. L.; Chi, L. 4 , Gelinck, G.; Decre, M. M. J. , 2004 Mater. Res. Soc. Symp. Proc. 4 , 2004 (397) Xia, D.; Brueck, S. R. J. , 1295. Nano Lett. EXS-2 , M4.9. 2004 Adv. Mater. (398) Xia, D.; Biswas, A.; Li, D.; Brueck, S.-R. J. , CR030076O 16 , 1427.

Related documents