Delivered by Ingenta to:Rensselaer Polytechnic Institute

IP : 128.113.26.88Mon, 15 Feb 2010 08:46:40

COMMUNIC

ATIO

N

Copyright © 2010 American Scientific PublishersAll rights reservedPrinted in the United States of America

Journal ofNanoscience and Nanotechnology

Vol. 10, 2252–2256, 2010

Selective Assembly of Multi-ComponentNanosprings and Nanorods

S. V. Kesapragada1, T.-J. Yim2, J. S. Dordick2, R. S. Kane2, and D. Gall1�∗1Department of Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180, USA

2Department of Chemical and Biological Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180, USA

This communication proposes a new approach to create complex hierarchical nano-to-meso-scalearchitectures based on the use of biological connector molecules to direct the assembly of uniquelyshaped multi-component nanostructures fabricated using glancing angle deposition (GLAD). Mul-tiple sets of 50-nm-wide and 150 to 650-nm-tall Si–Cr/Au multi-stack zigzag nanosprings andnanorods are grown by GLAD on Si substrates. Nanorods, chosen for selective assembly, aredetached from the substrate, suspended in an aqueous solution, and their surfaces are selectivelyfunctionalized by attaching biotin and streptavidin connector-molecules to the Au-regions. Succes-sive mixing of different suspensions leads to the end-to-end assembly of long and short nanorods.This technique provides the path to build hybrid nano-architectures including nano-honeycombs,nanoladders, and 3D nanorod networks, comprised of controlled material combinations.

Keywords: Glancing Angle Deposition (GLAD), Nanorods, Nano-Assemblies, Biotin,Streptavidin.

The controlled assembly of nanoscale units into mesoscaleand ultimately macroscopic materials and devices is thekey to harnessing the unique electronic, optical, and bio-chemical properties of nanounits. An attractive approachfor realizing such a controlled assembly is the use ofbiological connector molecules. In particular, biotin andstreptavidin have been shown to be effective in linkinginorganic nanorods into long chains.1 The magnitude ofthe free energy of association for streptavidin and biotinis among the largest for the non-covalent binding of aprotein to a small ligand in aqueous solution, and eachstreptavidin molecule provides 4 potential sites for biotinattachment.2�3 Site-selective functionalization of goldnanorods with these molecules facilitates controlled end-to-end assembly.1 Similarly, DNA-hybridization-drivenself-assembly of 13-nm-wide gold nanorods has beenreported to lead to 3D bundles,4 while linear nanorodassemblies were achieved using thioalkylcarboxylic acidbased bifunctional molecules that bind to the ends ofnanorods.5

We propose here a particularly promising approach forgenerating macroscopic nanostructured materials: The useof biological connector molecules to direct the arrange-ment of uniquely shaped multi-component nanounitsgrown by glancing angle deposition (GLAD). While

∗Author to whom correspondence should be addressed.

metallic multicomponent nanorods have been fabricatedusing electro-chemical template synthesis,6–8 GLAD hasgained considerable interest in recent years due to itsability to create arrays of uniquely shaped nanostructuresbuilt from a wide range of materials systems.9–11 Itexploits atomic shadowing effects during line-of-sightphysical vapor deposition by controlling azimuthal andpolar deposition angles to nano-engineered columns intovarious shapes such as zigzags,10 pillars,12 chevrons,13

spirals,14 slanted posts,15 multi-stack columns,16 flowers,17

nanotubes,18 and branched rods,19–21 with potential appli-cations as optically active layers,22�23 magnetic storagemedia,13�24 humidity and pressure sensors,25�26 enzymeimmobilization supports,27 emitters,28 actuators,29 radia-tion resistant coatings,30 and fuel-cell electrodes.31�32 Anadditional strength of GLAD is the extreme flexibility bywhich multiple materials can be stacked together in indi-vidual nanostructures,16�33–35 providing a path to createcomplex hybrid architectures and to form surfaces for site-selective functionalization.In this report, we demonstrate that GLAD nano-

structures can be controllably assembled into meso-scalearchitectures using biological connector molecules that areattached at predetermined locations on the nanostructuresurfaces. The processing steps involve (i) the growth ofvarious sets of multi-stack nanorods and nanosprings byGLAD, illustrated in Figures 1(a and b), followed by

2252 J. Nanosci. Nanotechnol. 2010, Vol. 10, No. 3 1533-4880/2010/10/2252/005 doi:10.1166/jnn.2010.2059

Delivered by Ingenta to:Rensselaer Polytechnic Institute

IP : 128.113.26.88Mon, 15 Feb 2010 08:46:40

COMMUNIC

ATIO

N

Kesapragada et al. Selective Assembly of Multi-Component Nanosprings and Nanorods

(a)

(c)

(b)

(d)

Nanorod

Nano-spring

Au

Si

Biotin-streptavidinlinker

Fig. 1. Schematic of multi-component (a) nanorods and (b) nanospringsgrown by glancing angle deposition and the hierarchical assemblyof these nanostructures into complex shapes like (c) nanoladders and(d) nanohoneycombs using biotin-streptavidin connectors.

(ii) selective attachment of biological molecules to specificlocations on these nanostructures, and (iii) the directedassembly of the nanostructures into complex meso-scalearchitectures. The uniqueness of this approach stems fromthe extreme flexibility of material combinations and nano-structure shapes that can be achieved by the glancing angledeposition process where any element which exhibits asolid phase at room temperature can be stacked on topof any other one, all within 20–500 nm wide nanorodsor nanosprings and with a vertical resolution of 1–10 nm.In this work, we describe the growth of 25–60 nmwide Cr–Si multi-stack nanosprings and 40–180 nm wideSi–Au multistack nanorods, as well as the directed end-to-end assembly of short and long Si nanorods withAu-caps, using biotin and streptavidin that are selec-tively attached to the Au section of short and long rods,respectively. The procedure presented may be ultimatelyextended to the controlled assembly of nanorods andnanosprings to form interconnected two-dimensional andthree-dimensional architectures like, for example, nanolad-ders and nanohoneycombs shown in Figures 1(c and d),with potential applications in nanorobotics, nanoelectronicdevices, molecular sieving, recognition, and sensing.Prior to deposition by GLAD, the Si wafers were pat-

terned with 20-nm-diameter silica nanoparticles using thefollowing sequential steps: The substrate surfaces wererinsed with isopropanol and acetone, plasma-treated inoxygen plasma to make them hydrophilic, and functional-ized with the cationic polymer polyethyeleneimine (PEI,(Aldrich, St. Louis, MO) by immersing in 10 ml of a0.025 M aqueous PEI solution for 30 min at room tem-perature, followed by a wash with MilliQ water. Silicananoparticles (NYACOL 2040, Eka Chemicals Inc.) were

deposited from a 10 ml MilliQ water suspension con-taining ∼3×1014 particles. The substrates were incubatedfor 30 min and rinsed with MilliQ water to remove anyunbound nanoparticles.Cr–Si nanosprings and Au–Si nanorods were grown in

a load-locked ultra-high vacuum magnetron sputter depo-sition system, described in detail in Ref. [16]. The sub-strates were continuously rotated about the polar axis(perpendicular to the substrate surface) for the growthof nanorods, and rotated sequentially by 180�-steps fornanospring growth. A collimating plate covering the sub-strate was used to prevent any non-directional flux fromstriking the substrate, and to control the azimuthal depo-sition angle to be 84�. Multi-component nanorods wereobtained by depositing different materials in discrete depo-sition steps. The sample heating due to the depositionplasma was monitored by a thermocouple within the sam-ple stage and was below 110 �C for all depositions, whichwere carried out at 0.26 Pa (2.0 mTorr) in 99.999% pureAr that was further purified using a Micro Torr purifier.Sputtering was done at a fixed power of 500 W, yieldingapproximate rod growth-rates of 2.4, 5.3 and 6.3 nm/minfor Cr, Si, and Au respectively. The shape and morphol-ogy of the nanorods and their assemblies were analyzedby scanning and transmission electron microscopy (JEOL6335F Field Emission SEM and Phillips CM12 TEM).Figure 2(a) is a typical scanning electron micrograph of

Cr–Si zigzag nanosprings, illustrating the uniquely shapednanostructures that can be grown by GLAD. This cross-sectional image, obtained by cleaving the Si substrate,shows the as-grown springs which are still attached tothe substrate surface. The structures were grown in foursequential steps with alternating deposition of Cr fromthe right and Si from the left, yielding two-componentzigzag nanosprings with four arms consisting of the respec-tive elements. The width of the springs near the substrateranges from 25 to 30 nm. However, the width increasesto 45± 5 nm at the end of the first (Cr) arm and to65± 15 nm for the fourth (Si) arm, that is, the top ofthe spring. This increase in width is also illustrated bythe dotted line highlighting the contour of a nanospringin Figure 2(a). The broadening as a function of height isattributed to strong intercolumnar competition which hasbeen reported to increase with homologous temperature36

and to lead to column extinction during GLAD.37–39 Theincrease in the width in each consecutive arm of the zigzagsis compensated by a fraction of the springs terminatingtheir growth prematurely, leading to a reduction in theirnumber density and resulting in an increase in the averagespacing between the springs, from 25± 5 nm for the firstarm to 135±10 nm for the fourth arm of the zigzags.A corresponding TEM micrograph of Cr–Si nanosprings

is shown in Figure 2(b). The TEM specimen was preparedby detaching the springs from the substrate in a 30-minuteultrasonic bath of 99.9% acetone, followed by drying of a

J. Nanosci. Nanotechnol. 10, 2252–2256, 2010 2253

Delivered by Ingenta to:Rensselaer Polytechnic Institute

IP : 128.113.26.88Mon, 15 Feb 2010 08:46:40

COMMUNIC

ATIO

N

Selective Assembly of Multi-Component Nanosprings and Nanorods Kesapragada et al.

(a)

(b)

(c)

(d)

Fig. 2. Multi-component nanostructures grown by GLAD. (a) Scanningand (b) transmission electron micrographs of Cr–Si nanosprings withthe four arm-levels labeled (1)–(4), (c) secondary and (d) back-scatteredelectron micrographs of the same Si–Au nanorods.

15±5-�l-droplet of the resulting suspension on a carbon-foil of a Cu-TEM grid (Ted Pella, Inc.). The bright fieldimage shows strong diffraction contrast between the Crand the Si arms, that is, the crystalline Cr appears darkas most of the electron beam intensity is diffracted, whilethe Si is brighter due to its amorphous structure and lower

atomic mass. The nanosprings in the micrograph consist offour arms, Cr–Si–Cr–Si, as labeled (1)–(4) in Figure 2(b).The first Cr arm (1) is less pronounced in the image, whichis attributed to its smaller width, as discussed above, andthe related detachment or mechanical degradation duringTEM specimen preparation. In addition, the micrographshows 5–7 second-level Si arms (2) which merge duringCr deposition (arm 3) into only two fourth-level Si arms(4). That is, the image illustrates the reduction of the num-ber of springs as a function of height, in this case by afactor of ∼3, which is attributed to a competitive growthmode which is also responsible for the observed broaden-ing discussed above. Such column competition and relatedbroadening can be delayed or completely suppressed bycontrolling the initial column separation during GLADusing appropriate seed pattern dimensions.40–42

Figures 2(c) and (d) are cross-sectional SEM micro-graphs of Si–Au multi-stack nanorods, showing the sameportion of the sample, imaged using secondary electronsand back-scattered electrons, respectively. The rods weregrown with continuous rotation of the substrate about thepolar axis using two alternating sequences of Si and Audeposition steps, with a nominal Si:Au atomic ratio of2:1 The nanorods are 600± 25-nm tall and their widthincreases with height from 50± 15 nm near the substrateto 170± 10-nm on the top, as determined by plan-viewmicrographs (not shown). The increase in width is dueto a competitive growth mode and is accompanied by adecrease in the rod number density, similar to the case forthe nanosprings discussed above. This leads to rods thatterminate their growth prematurely, as those shown in theforeground in Figure 2(c), with a height of 150±35 nmand a width of 45±15 nm. The larger rods exhibit increas-ingly rough surfaces with protrusions that elongate alongthe growth direction. These protrusions can be attributedto the same growth instability that causes the formationof the entire nanorod. That is, small surface irregulari-ties are exacerbated during the growth process since anyroughness on the rod surface results in an enhanced cap-turing rate of the flux, yielding locally increased growthrates and, in turn, surface protrusions. Similar GLADnanostructure surface morphologies have been reported forcobalt and silicon films.43�44 The rods in Figure 2(c) areslightly tilted with respect to the substrate surface nor-mal due to an anisotropy in the deposition flux causedby unintentional off-centering of the substrate with respectto the collimating plate. The backscatter electron image(BSEI) in Figure 2(d), from the same portion of a Si–Aunanorod layer shown in Figure 2(c), shows clear compo-sitional contrast between the Au and Si portions of therods. Au appears brighter in the micrograph due to itshigher atomic number compared to Si (ZAu = 79 vs. ZSi =14). This image demonstrates the controlled positioning ofAu-sections within Si nanorods, where the position andsize of the Au is directly determined by the subsequentdeposition-sequence times.

2254 J. Nanosci. Nanotechnol. 10, 2252–2256, 2010

Delivered by Ingenta to:Rensselaer Polytechnic Institute

IP : 128.113.26.88Mon, 15 Feb 2010 08:46:40

COMMUNIC

ATIO

N

Kesapragada et al. Selective Assembly of Multi-Component Nanosprings and Nanorods

In the following, we present initial data demonstratingbiomolecule-directed assembly of GLAD nanostructuresinto more complex architectures. For this purpose, twosets of Au capped Si nanorods were grown by GLAD.The long rods, 665± 10 nm tall and 185± 10 nm wide,are ∼four times longer than the short rods, 150± 15 nmtall and 50± 10 nm wide. Biotin was attached to the Auportion of the short nanorods by first immersing the Siwafer containing the rods into a 5 ml solution of 2 mM2-aminoethanethiol dissolved in 100% ethanol for 4 hours.The substrate (and rods) were then rinsed with 5 ml ofethanol, immersed in 5 ml 10 mM EZ-Link Sulfo-NHS-LC-Biotin in 0.1 M sodium phosphate buffered saline(0.15 M NaCl, pH 7.2) for 30 min at room temperature,and washed with MilliQ water to remove any unboundlinker. Streptavidin was attached to the Au portions ofthe long rods by first immersing the substrate containingthe rods in 5 ml of 2 mM mercapto-undecanoic acid dis-solved in 100% ethanol for 4 hours followed by a rinse in5 ml solution of ethanol, an immersion in 5 ml of aque-ous solution containing 0.05 M N -Hydroxysuccinimide(NHS) and 0.2 M 1-ethyl-3-(3-dimethylaminopropyl) car-bodiimide (EDC) for 7 min, and washing with 5 ml of50 mM MES buffer (pH 6.2). The two silicon supportswere then sonicated (separately) for 30 min in MilliQwater, resulting in the formation of aqueous suspensionsof functionalized nanorods. The two suspensions, contain-ing short and long nanorods, were mixed together andwere allowed to settle for 30 minutes A 5 �L drop ofthe resulting mixture was dispersed and dried on top ofa plasma-cleaned Cu grid with a carbon membrane (TedPella, Inc.) for imaging by TEM. In addition, three controlsamples containing a mixture of (i) biotin-functionalizedshort nanorods and non-functionalized long nanorods,(ii) streptavidin-functionalized long nanorods and non-functionalized short nanorods and (iii) non-functionalizedlong and short nanorods, were prepared to demonstratethat biotin-streptavidin linkage is necessary for nanorodassembly.Figure 3 is a representative bright-field TEM micro-

graph of a Si–Au nanorod assembly. The image shows 3long nanorods, with a measured length of 674± 18 nm.The rods are capped with Au, which appears dark inFigure 3 due to strong scattering of electrons from crys-talline (high Z) Au grains in comparison to the brighteramorphous (low Z) Si rods. The dark region in the centerof the left long rod is due to an overlapping 4th long rodwhich extends out of the image at the left bottom corner.The Si portion of the nanorods exhibits strong porosity,associated with atomic shadowing caused by the intention-ally high deposition angle in combination with low adatommobility due to a low homologous growth temperature ofTs/Tm(Si� = 300 K/1687 K = 0�18. The short nanorodsappear in Figure 3 as bundles of 1–3 parallel rods. Suchbundles form during the nanorod growth, when neighbor-ing rods touch and develop a sufficiently strong bond so

Fig. 3. TEM micrograph of short and long Au-capped Si nanorodsassembled using biotin-streptavidin connectors.

that they do not break apart during the subsequent soni-cation. As reported previously, isolated nanorods can begrown by choosing pattern dimensions that prevent col-umn competition.16 The short rods have a measured lengthof 165± 10 nm, in reasonable agreement with the lengthof 150±15 nm, measured by SEM. All short rods arrangethemselves so that their Au-portions are in close proxim-ity (≤5 nm) to the Au-portion of the large rods. This isan indication that the biotin at the short rods connectsto the streptavidin at the long rods, during the mixing ofthe two suspensions. A statistical analysis from multipleTEM micrographs, corresponding to a total sample area of104 �m2, shows that 30±10% of long rods are linked toshort rods. In contrast, two of the three control samplesshow no nanorod assembly while the third exhibits a rodagglomerate, corresponding to ∼5% of the observed rodsin this sample. That is, samples with biotin and strepta-vidin molecules are at least six times more likely (30%versus <5%) to exhibit linking of long and short nanorods,indicating that the biotin-streptavidin linkage is the pri-mary cause for nanorod assembly. We attribute the rod-assembly in the absence of bio-linkers to hydrophobicinteractions45 or capillary, electrostatic, and/or Van-der-Waals forces during the drying process of the rod suspen-sion on the TEM grid. We expect that these secondarydriving forces can likely be further suppressed by an opti-mized assembly procedure. Overall, the results demon-strate the feasibility of the envisioned two-componentnanostructure assembly by biological connector molecules.In conclusion, we have demonstrated that uniquely-

shaped multi-component nanostructures can be grown byGLAD, including Si–Au nanorods and Cr–Si nanosprings.

J. Nanosci. Nanotechnol. 10, 2252–2256, 2010 2255

Delivered by Ingenta to:Rensselaer Polytechnic Institute

IP : 128.113.26.88Mon, 15 Feb 2010 08:46:40

COMMUNIC

ATIO

N

Selective Assembly of Multi-Component Nanosprings and Nanorods Kesapragada et al.

We have also shown that Si–Au nanorods can be con-trollably assembled (end-to-end) using biological con-nector molecules, in particular biotin and streptavidin,that are selectively attached to the Au portion at theend of the nanorods. This new hybrid physical-vapor-deposition/wet-chemistry approach will be useful forassembling complex hierarchical nanoarchitectures includ-ing nano-honeycombs, nanoladders, and 3D nanorod net-works, comprised of controlled materials combinations.We acknowledge the financial support for this

exploratory research from the Donors of the AmericanChemical Society Petroleum Research Fund under grantNo. 44226-G10 and from the National Science Foundationunder grant No. CMMI-0727413.

References and Notes

1. K. K. Caswell, J. N. Wilson, U. H. F. Bunz, and C. J. Murphy, J. Am.Chem. Soc. 125, 12914 (2003).

2. N. M. Green, Adv. Protein Chem. 29, 85 (1975).3. T. Sano, S. Vajda, and C. R. Cantor, J. Chromatogr. 715, 85

(1998).4. E. Dujardin, L.-B. Hsin, C. R. C. Wang, and S. Mann, Chem.

Commun. 1264 (2001).5. K. G. Thomas, S. Barazzouk, B. I. Ipe, S. T. S. Joseph, and P. V.

Kamat, J. Phys. Chem. B 108, 13066 (2004).6. S. R. Nicewarner-Pena, R. G. Freeman, B. D. Reiss, L. He, D. J.

Pena, I. D. Walton, R. Cromer, C. D. Keating, and M. J. Natan,Science 294, 137 (2001).

7. A. K. Salem, M. Chen, J. Hayden, K. W. Leon, and P. C. Searson,Nano Lett. 4, 1163 (2004).

8. B. D. Reiss, R. G. Freeman, I. D. Walton, S. M. Norton, P. C. Smith,W. G. Stonas, C. D. Keating, and M. J. Natan, J. Electroanal. Chem.522, 95 (2002).

9. K. Robbie, M. J. Brett, and A. Lakhatia, Nature 384, 616 (1996).10. K. Robbie and M. J. Brett, J. Vac. Sci. Technol. A 15, 1460 (1997).11. R. M. A. Azzam, Appl. Phys. Lett. 61, 3118 (1992).12. B. Dick, M. J. Brett, and T. Smy, J. Vac. Sci. Technol. B 21, 2569

(2003).13. B. Dick, M. J. Brett, T. J. Smy, M. R. Freeman, M. Malac, and R. F.

Egerton, J. Vac. Sci. Technol. A 18, 1838 (2000).14. S. R. Kennedy and M. J. Brett, J. Vac. Sci. Technol. B 22, 1184

(2004).

15. D.-X, Ye, T. Karabacak, R. C. Picu, G.-C, Wang, and T.-M. Lu,Nanotechnology 16, 1717 (2005).

16. S. V. Kesapragada and D. Gall, Thin Solid Films 494, 234 (2006).17. Y.-P. Zhao, D.-X. Ye, G.-C. Wang, and T.-M. Lu, Nano Lett. 2, 351

(2002).18. S. V. Kesapragada, P. R. Sotherland, and D. Gall, J. Vac. Sci. Technol.

B 26, 678 (2008).19. J. Wang, H. Huang, S. V. Kesapragada, and D. Gall, Nano Lett.

5, 2505 (2005).20. C. M. Zhou and D. Gall, Appl. Phys. Lett. 88, 203117 (2006).21. S. V. Kesapragada and D. Gall, Appl. Phys. Lett. 89, 203121 (2006).22. M. Suzuki and Y. Taga, J. Appl. Phys. 71, 2848 (1992).23. I. J. Hodgkinson and Q. H. Wu, Opt. Eng. 37, 2630 (1998).24. A. Lisfi and J. C. Lodder, Phys. Rev. B 63, 174441 (2001).25. P. C. P. Hrudey, A. C. Van Popta, J. C. Sit, and M. J. Brett, Proc.

SPIE Int. Soc. Opt. Eng. 5931, 593113 (2005).26. S. V. Kesapragada, P. Victor, O. Nalamasu, and D. Gall, Nano Lett.

6, 854 (2006).27. T.-J. Yim, D.-Y. Kim, S. S. Karajanagi, T.-M. Lu, R. Kane, and J. S.

Dordick, J. Nanosci. Nanotechnol. 3, 479 (2003).28. J. P. Singh, F. Tang, T. Karabacak, T.-M. Lu, and G.-C. Wang, J. Vac.

Sci. Technol. B 22, 1048 (2004).29. J. P. Singh, D.-L. Liu, D.-X. Ye, R. C. Picu, T.-M. Lu, and G.-C.

Wang, Appl. Phys. Lett. 84, 3657 (2004).30. R. Nagar, B. R. Mehta, J. P. Singh, D. Jain, V. Ganesan, S. V.

Kesapragada, and D. Gall, J. Vac. Sci. Technol. A 26, 887 (2008).31. M. D. Gasda, R. Teki, T.-M. Lu, N. Koratkar, G. A. Eisman, and

D. Gall, J. Electrochem. Soc. 156, B614 (2009).32. C. M. Zhou and D. Gall, J. Appl. Phys. 103, 014307 (2008).33. Y. P. He, J. S. Wu, and Y. P. Zhao, Nano Lett. 7, 1369 (2007).34. C. M. Zhou and D. Gall, Small 4, 1351 (2008).35. C. M. Zhou, H. F. Li, and D. Gall, Thin Solid Films 517, 1214

(2008).36. S. Mukherjee, C. M. Zhou, and D. Gall, J. Appl. Phys. 105, 094318

(2009).37. C. M. Zhou and D. Gall, Thin Solid Films 515, 1223 (2006).38. C. M. Zhou and D. Gall, Appl. Phys. Lett. 90, 093103 (2007).39. C. M. Zhou and D. Gall, J. Vac. Sci. Technol. A 25, 312 (2007).40. Y.-P. Zhao, D.-X. Ye, P.-I. Wang, G.-C. Wang, and T.-M. Lu, Int. J.

Nanosci. 1 87 (2002).41. M. Malac and R. F. Egerton, Nanotechnology 12 11 (2001).42. C. M. Zhou and D. Gall, Thin Solid Films 516, 433 (2007).43. B. Dick, M. J. Brett, and T. Smy, J. Vac. Sci. Technol. B 21, 23

(2003).44. Y.-P. Zhao, D.-X. Ye, P.-I Wang, G.-C. Wang, and T.-M. Lu, Intl. J.

Nanosci. 1, 87 (2002).45. S. Park, J.-H. Lim, S.-W. Chung, and C. A. Mirkin, Science 303, 348

(2004).

Received: 9 June 2008. Accepted: 3 December 2008.

2256 J. Nanosci. Nanotechnol. 10, 2252–2256, 2010