by Ahmed Busnaina, Northeastern University
Much has been said about the great impact and improvements nanotechnology will make in our lives–offering cost-effective green energy, medical treatment and therapy breakthroughs, and new advanced nanoelectronics that are small and ubiquitous, even becoming a part of our future wardrobe. However, one of the biggest impacts of nanotechnology will be on manufacturing. Many of the potential applications and products will be made using a very different technology and processes than what we are used to. We are already seeing early signs of shifting manufacturing of devices and other products from vacuum-based process. For example, some photovoltaics manufacturers use screen printing, and some display applications already commercialized use inkjet printing of circuit patterns. Some of these products are flexible (compliant substrate)
Large-scale assembly of SWNTs on a parylene-polycarbonate wafer.
and some are not–but in both cases they involve room-temperature non-vacuum manufacturing processes to produce electronics (or a major part of an electronic application that in the past was only done using vacuum and high-temperature based processes). Some of these applications utilize a reel-to-reel process in their manufacturing, but these utilize >1µm linewidths, which is comparable to 1980s-era electronics technology.
During the next five years we will see more amazing developments that will allow the fabrication of structures, devices, and circuits with structures and patterns down to 20nm or smaller using room-temperature and non-vacuum processes. These nanomanufacturing processes incorporate nanoelements (such as nanotubes, nanowires, nanoparticles, proteins, DNA, drugs, etc.) to build structures and patterns that can be used for electronics, energy, materials, or biotechnology products; they utilize directed and self-assembly to make devices (electronics or NEMs), sensors, medical devices, energy harvesting, and storage applications.
The details of the nanomanufacturing techniques and examples of application will be covered in this article, examining what is involved in such a nanoscale manufacturing technique and what development is needed to have ready to make products. This involves identifying several fundamental technical barriers to nanomanufacturing:
- How can we assemble different nanoscale elements? What are the mechanisms leading to the assembly and orientation of nanoscale structures? How do we control these mechanisms?
- How can we scale up assembly processes in a continuous or high rate manner? How do we control the interfacial behavior and forces required to assemble, detach, and transfer nanoelements at high rates?
- How can we test for reliability in nanostructures? How can we detect and mitigate defects?
- How can nanomanufacturing processes meet existing and future environmental and health regulations?
The approach to nanomanufacturing processes has to be fully integrated. This nanomanufacturing vision is elucidated by the NSF Nanoscale Science and Engineering Center for High-rate Nanomanufacturing (CHN) and includes: synthesis and/or functionalization of nanoelements; use templates for directed assembly; transfer assembled structures; and integrate assembled structures into applications. This has to also include multiscale modeling to guide processes and also has to consider reliability and defect mitigation.
Nanomanufacturing using directed assembly and transfer
Directed assembly can be used to assemble wide range of nanoelements including nanoparticles, nanotubes, and polymers. Any nanoelements that can address by charge or chemistry can be assembled. However, to have directed assembly, the assembly must be guided to produce the right structure or pattern that will represent the device circuitry. To do that, a template that has the desired pattern and utilizes either electric field and/or chemical functionalization can be utilized to guide the assembly of the nanoelements. The template patterns or structures could include both nanowires and nanotrenches to guide the assembly. Once the assembly is accomplished on the template, the assembled structures need to be transferred to the device substrate of choice (soft or hard) where the device will be housed.
It is also possible to assemble right on the device without using a template if the device circuitry could be used to guide the assembly (template-free directed assembly process). This is often done when a device is made, for example, using a typical semiconductor manufacturing but the actuating, sensing, or switching elements is a nanomaterials such as nanotubes. In this case the circuit can be used to guide and align the assembly of the nanoelements into place.
FIGURE 1: (a) Large-scale assembly of SWNTs (9µm line patterns) and (b) Parallel arrays of aligned SWNT 120nm nanowires on a wafer level.
Electrophoretically-driven directed assembly. Electrophoresis is a promising method for fast, directed assembly of an array of nanoelements. Nanoparticles,1 functionalized colloidal gold films,2 and spherical latex particles3 have been assembled into micro and nanopatterns using electrophoresis. Nanowire templates have been demonstrated to direct the assembly of 800nm to 50nm nanoparticles, conducting polymers,4 and single-wall carbon nanotubes (SWNT) using a DC biased microscale electrode.5 In the case of SWNTs, they assemble onto the wires into non-aligned uniform patterns. This approach is useful for the assembly of SWNTs into patterned structures; however, it does not offer alignment–which is important, especially for electrical and thermal conductivity and structural materials.
Nanotrench templates are fabricated for field-assisted assembly by making PMMA nanotrenches on a conductive film. The trench templates provide superior control over the assembled nanoelements and enable the directed assembly of SWNTs into nanoscale trenches as small as 80 nm in a short time (30-60 seconds) and over a large area (4 inches). The assembly is complete with no gaps demonstrating that an approach to producing nanoscale features at high rate/high volume. Nanoparticles as small as 10nm were assembled to make 10nm nanowires in a short time (30sec) and over a large area (4 inches).
Assembly and transfer of conducting polymers. The potential application of polyaniline (PANI) in nanoelectronic or photonic devices and sensors has spurred research in patterning conducting polymers onto substrates. End uses include flexible electronics, paper-like displays, fuel cells, light-emitting diodes, circuits, and biosensors.6 Electrophoretic assembly using nanowires was successfully used to assemble PANI on templates. The assembled PANI wires need to be transferred intact to another substrate. The CHN also demonstrated successful transfer using thermoset elastomers; this method of transfer allows for the template to be reused, unlike many conventional lithographic technologies. The same approach has demonstrated successful transfer of CNTs a polymer substrate.
Chemically driven directed assembly. In addition to electrophoretically-driven assembly, it is also possible to have an assembly that does not require a conductive film in the template. Chemically driven assembly techniques could be done on a variety of substrates conductive or nonconductive which increase manufacturing flexibility.
Assembly of organized SWNT networks on silicon and flexible substrates
Electrophoresis to conduct assembly is useful, but requires supplying an electric-field to the template patterns. Capillary force can also be used to produce ordered nanoelements in trenches,7,8 but the assembly of SWNTs in relief structures is more difficult due to the high aspect ratio of individual tubes or bundles. It has been demonstrated (Small, JACS) that a surface-controlled microfluidic approach can be used to assemble highly organized SWNT networks in various dimensions and geometries using a template assisted dip-coating method, as shown in Figure 1. The approach also allowed the alignment and orientation of assembled SWNTs, which produced SWNT networks that can be directly used as interconnect wires, diverse sensing elements, and other active components in electronic devices. Moreover, this approach is compatible to CMOS processes.
The assembly of ordered arrays of carbon nanotubes onto flexible substrates offers many opportunities for realizing novel functional devices.9,10,11 The CHN has developed a method for directed assembly of SWNT microstructures on soft polymer substrates by using a surface-controlled microfluidic assembly technique. A treatment method to adjust the hydrophilic/hydrophobic properties of the surface allowed direct assembly of carbon nanotubes on Parylene-C. The approach has also been scaled to assembly on a 4-in. polycarbonate wafer.
FIGURE 2: Non-uniform geometries by direct assembly of PS/PAA blends.
Directed self-assembly of polymer blends. Block copolymers and polymer blends have attracted interest because of their nanoscale two-phase morphology,12 which is useful for a number of applications.13,14,15 Defect-free morphologies (for block copolymers) have been obtained using nanopatterned surfaces16,17,18 and electric fields.19,20 However, these techniques are not suitable for scale-up because of long annealing times. Other limitations of block copolymers are an inability to form complex and nonuniform patterns, and a restriction of the inherent size of the blocks. Using patterned polymer blends can overcome this restriction and offers more flexibility in terms of pattern size and shape.
FIGURE 3: FESEM image of SWNTs assembled between the top and bottom metal.
We have used chemically functionalized templates prepared by patterning of alkanethiols with different chemical functionality on our nanotrench templates to direct the assembly of polymer blends. With this approach, the selective assembly process can be finished in 30 seconds and does not require such long annealing times (3-7 days) as are often required in the conventional assembly of block copolymers. We have demonstrated this method for a variety of complex geometries including 90° bends, T-junctions, and square and circle arrays (Figure 2), which have many application in nanoelectronics. In addition, the resultant polymeric structures can be patterned over a very large area and with high resolution, overcoming the constraint of limited areas for patterning with dip-pen nanolithography21 and low resolution by microcontact printing.22,23
Three-dimensional assembly of SWNTs
Techniques developed to selectively place or assemble nanometer-sized materials24,25,26,27,28,29 are restricted to planar substrates. Currently the only 3D approach is based on chemical vapor deposition (CVD)30,31 but is limited by high processing temperatures (>500°C) and nonselectivity to nanotube types.32 The CHN has developed a new hybrid technique combining both bottom-up dielectrophoresis and top-down microfabrication techniques to enable low-temperature integration of SWNTs and gold nanoparticles into 3D architectures. SEM images of the assembled SWNTs are shown in Figure 3. This approach was also demonstrated for nanoparticles.
Application in nanotube memory devices
The SWNT switch is a non-volatile memory element with potentially orders-of-magnitude higher density than silicon memories. Such a switch could support an estimated market value on the order of $100 billion.33 The CHN is fabricating bistable nanoelectromechanical (NEMS) switches employing SWNTs as the actuation element.
FIGURE 4: A schematic and a FESEM image of the assembled SWNTs on the nanoswitch.
A schematic diagram of the proposed bistable nanoswitch and high-angle and top-viewed SEM micrographs for one such final device is shown in Figure 4. The SEM image shows that only an individual SWNT or a SWNT bundle is assembled. This illustrates that the degree of controlled assembly is achieved with a high yield (95%). A cross-sectional view of the bi-stable switch is shown in Figure 5. The geometry is scalable such that many pairs of electrodes (similar to A and C) and trenches (similar to B and D) can be fabricated resulting in large memory arrays. The principle of operation of the device also is shown schematically.
FIGURE 5: Schematic diagrams of the two states of the bistable switch.
The CHN is a team of three universities: Northeastern University (lead university) in Boston, MA; University of Massachusetts/Lowell; and the University of New Hampshire/Durham. This partnership combines the skills of 39 scientists in the fields of engineering, physics, chemistry, material science, business, and humanities in addition to more than 100 students and 15 post-docs working on developing technology to overcome critical barriers.
Ahmed A. Busnaina is William Lincoln Smith Professor and director of the NSF Nanoscale Science and Engineering Center for High-rate Nanomanufacturing at Northeastern University, Boston, MA. Contact: firstname.lastname@example.org, www.nano.neu.edu.
- Y.Fukada, N. Nagarajan, W. Mekky, Y. Bao, H.S Kim,P.S. Nicholson, Journal of Material Science, 39, 787 (2004).
- Ryan C. Baily, Keith J. Stevenson, and Joseph T. Hupp, Advanced Materials, 12, No 24 (2000).
- Eugenia Kumacheva, Robert Kori Golding, Mathieu Allard and H. Sargent, Adv. Mater., 14, No 3 (2002).
- M. Wei, Z. Tao, X. Xiong, M. Kim, J. Lee, S. Somu, S. Sengupta, A. Busnaina, C. Barry, J. Mead, “Fabrication of Patterned Conducting Polymers on Insulating Polymeric Substrates by Electric-Field-Assisted Assembly and Pattern Transfer,” Macromolecular Rapid Communications, 27, 1826 (2006)
- Xiong, X.; Makaram, P.; Bakhtari, K.; Somu1, S.; Busnaina, A.; Small, J.; McGruer N.; Park, J.; Proceedings of the Materials Research Society Symposium 2005, 901E, 0901-Ra04-01.
- J. A. Rogers, Z. Bao, Journal of Polymer Science: Part A: Polymer Chemistry, 2002, 40, 3327.
- F. Juillerat, H. H. Solak, P. Bowen, and H. Hofmann, Nanotechnology 16, 1311 (2005).
- Y. Yin, Y. Lu, B. Gates, and Y. Xia, J. American Chemical Society 123, 8718 (2001).
- H. Ko, V. V. Tsukruk, Nano Letters 2006, 6, 1443.
- S-H. Hur, O. O. Park, J. A. Rogers, Applied Physics Letters 2005, 86, 243502.
- S. J. Kang, C. Kocabas, T. Ozel, M. Shim, N. Pimparkar, M. A. Alam, S. V. Rotkin, J. A. Rogers, Nature Nanotechnology 2007, 2, 230.
- N. Hadjichristidis, S. Pispas, G. Floudas, Block Copolymers : Synthetic Strategies, Physical Properties, and Applications, John Wiley and Sons, Hoboken, NJ, 2003.
- G.M. McClelland, M.W. Hart, C.T. Rettner, M.E. Best, K.R. Carter, and B.D. Terris, “Nanoscale patterning of magnetic islands by imprint lithography using a flexible mold,” Applied Physics Letters, 81, 1483 (2002).
- D.H. Kim, Z. Lin, H.-C. Kim, U. Jeong, and T.P. Russell, “On the replication of block copolymer templates by poly(dimethylsiloxane) elastomers,” Advanced Materials, 15, 811 (2003).
- Y.S. Kim, H.H. Lee, and P.T. Hammond, “High density nanostructure transfer in soft molding using polyurethane acrylate molds and polyelectrolyte multilayers,” Nanotechnology, 14, 1140 (2003).
- L. Rockford, S.G. J. Mochrie, and T. P. Russell, “Propagation of Nanopatterned Sustrate Templated Ordering of Block Copolymers in Thick Films,” Macromolecules, 34, 1487 (2001).
- X. M. Yang, R. D. Peters, T. K. Kim, P. F. Nealy, S. L. Brandow, M-S. Chen, L. M. Shirey, and W. J. Dressick, “Proximity X-ray Lithography Using Self-Assembled Alkylsilixone Films: Resolution and Pattern Transfer,” Langmuir, 17, 228 (2001).
- S. O. Kim, H. H. Solak, M. P. Stoykovich, N. J. Ferrier, J. J. DePablo, and P. F. Nealy, “Epitaxial Self-Assembly of Block Copolymers on Lithographically Defined Nanopatterned Substrates,” Nature, 424, 411, (2003).
- E. Schaffer, T. Thurn-Albrecht, T. P. Russell, and U. Steiner, “Electrically Induced Structure Formation and Pattern Transfer,” Nature, 403(6772) 874 (2000).
- T. Thurn-Albrecht, J. DeRouchy, T. P. Russell, and H. M. Jaeger, “Overcoming Interfacial Interactions with Electric Fields,” Macromolecules, 33, 3250 (2000).
The full list of references is viewable online.