Letter pubs.acs.org/NanoLett
Enzymatically Controlled Vacancies in Nanoparticle Crystals Stacey N. Barnaby,†,‡ Michael B. Ross,†,‡ Ryan V. Thaner,†,‡ Byeongdu Lee,*,§ George C. Schatz,*,†,‡ and Chad A. Mirkin*,†,‡ †
Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States International Institute for Nanotechnology, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States § X-ray Science Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States ‡
S Supporting Information *
ABSTRACT: In atomic systems, the mixing of metals results in distinct phase behavior that depends on the identity and bonding characteristics of the atoms. In nanoscale systems, the use of oligonucleotides as programmable “bonds” that link nanoparticle “atoms” into superlattices allows for the decoupling of atom identity and bonding. While much research in atomic systems is dedicated to understanding different phase behavior of mixed metals, it is not well understood on the nanoscale how changes in the nanoscale “bond” affect the phase behavior of nanoparticle crystals. In this work, the identity of the atom is kept the same, but the chemical nature of the bond is altered, which is not possible in atomic systems, through the use of DNA and RNA bonding elements. These building blocks assemble into single crystal nanoparticle superlattices with mixed DNA and RNA bonding elements throughout. The nanoparticle crystals can be dynamically changed through the selective and enzymatic hydrolysis of the RNA bonding elements, resulting in superlattices that retain their crystalline structure and habit, while incorporating up to 35% random vacancies generated from the nanoparticles removed. Therefore, the bonding elements of nanoparticle crystals can be enzymatically and selectively addressed without affecting the nature of the atom. KEYWORDS: Superlattices, RNA, vacancies, enzymes, nanoparticles, crystals
W
system consisting of three types of gold nanoparticles (A, B, and B′), with two different types of bonding elements. Specifically, particle A has exclusively DNA bonding elements and is capable of recognizing particles with complementary DNA or RNA. Particle B has pure DNA bonding elements, and particle B′ has pure RNA bonding elements, which are complementary to the bonding elements in particle A. Particles A, B, and B′ are assembled to form body-centered cubic (bcc) crystals, and the amount of bonding involving RNA can be controlled by changing the ratio of B to B′ (Tables S1 and S2). We hypothesized that this would allow for the study of how the bonding oligonucleotide directs the crystallization, as there are two different types of bonds that can form: DNA/DNA (from the bonding elements of particles A and B) and DNA/RNA (from the bonding elements of particles A and B′; Figure 1a). Fifteen different aggregates were formed by mixing particle A with different ratios of particles B and B′ and then crystallized using a slow cool annealing process (0.01 °C/min)23 from 5 to 10 °C above the melting temperature down to room temperature.
hen mixing metals in atomic systems, the fundamental limitation for effecting structural control is that the intrinsic bonding capability of the metal cannot be decoupled from the identify of the element. A fundamentally different approach to manipulate bonding without changing the identity of the atom is through nanoparticle superlattice engineering with oligonucleotides, which represents a powerful way to crystallize nanoparticles into superlattices1−7 with independent control over the bond,3,8−14 nanoparticle core,3,15−21 crystalline symmetry,3 and mesoscale shape.22−24 For example, changing the thermodynamic stability of the oligonucleotide bonding element has led to the synthesis of three-component nanoparticle superlattices through topotactic intercalation25 as well as core−shell single crystals of gold nanoparticles and quantum dots.10 The programmable nature of the oligonucleotide is the driving force for crystallization, as opposed to the oligonucleotide identity, which was demonstrated by synthesizing nanoparticle superlattices with RNA−RNA, DNA−RNA, and RNA−DNA bonds, which allowed for specific enzymatic recognition of the bonding elements that were composed of RNA.14 This approach to bonding is chemically driven and is distinct from both atomic and conventional colloidal systems. The programmable nature of oligonucleotides allows for an exploration of how nanoparticle atoms crystallize when they have two different types of nanoscale bonds. We designed a © XXXX American Chemical Society
Received: May 19, 2016 Revised: July 7, 2016
A
DOI: 10.1021/acs.nanolett.6b02042 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
Figure 1. Design of body-centered cubic (bcc) nanoparticle superlattices with different bonding elements. a) Particle A, with DNA bonding elements, can bind to both particles B and B′, with DNA and RNA bonding elements, respectively. b) Four potential crystal phases that could form, based on knowledge of oligonucleotide-directed nanoparticle crystallization. Structure 1 has two separate crystals, one formed from particles A and B, and the second from particles A and B′. Structure 2 has a core−shell of particles A and B′, which is then followed by nucleation driven by the bonding elements of particles A and B. Structures 3 and 4 are random mixtures, with either homogeneous or inhomogeneous mixtures of bonding elements from particles A and B and from particles A and B′, respectively.
Based on knowledge about how oligonucleotide bonding elements direct nanoparticle crystallization,3,9,23 we hypothesized that four different crystal structures could result (Figure 1b). We hypothesized that more than one crystal could form due to the decreased lattice parameter and increased thermal stability for crystals composed of DNA/RNA bonding elements compared with those of pure DNA bonding elements.14 Structure 1 has two crystals, one formed with particles A and B, and the second with particles A and B′. We hypothesized that phase-separated crystals would form if the difference in lattice parameter and thermal stability for the DNA/DNA and DNA/RNA bonding elements was too great to allow for crystallization into the same crystal.14 Structure 2 has a core of particles A and B′, which would nucleate first due to their higher melting temperature,14 followed by a shell composed of particles A and B, which would nucleate upon further cooling. Structures 3 and 4 are random mixtures. Structure 3 has a homogeneous distribution of bonding elements from particles A and B with those from particles A and B′, which may occur despite the difference in melting temperature between the two bonding elements because of reorganization, which is hypothesized to occur when nanoparticle aggregates are annealed to form crystals.9 Structure 4 has an inhomogeneous distribution of bonding elements, due to incomplete reorganization during thermal annealing. In order to selectively address the bonds without affecting the atoms to generate new structures, the chemical nature for how crystallization proceeds in the presence of two different types of bonding elements must first be elucidated. Melting analyses were conducted, as sharp melting transitions are the result of the cooperative nature of multiple oligonucleotide interactions between each pair of nanoparticles.14,26 Both single sharp melting transitions as well as two superimposed sharp melting transitions were observed when both DNA/DNA and DNA/RNA bonding elements were possible (Figure 2a and Figure S1). A single melting transition is indicative of a single cooperative melting event, whereas a superposition of two
Figure 2. Experimental and theoretical melting curves. a) Experimental results for the melting transition of nanoparticle superlattices with specified ratios of particle B:B′. b) Two-state thermodynamic model for homogeneous (left) and phase separated (right) mixtures of particles with DNA and RNA bonding elements. The traces range from a ratio of 100:0 particle B:B′ (dark purple) to 0:100 particle B:B′ (brown) in increments of 10%.
B
DOI: 10.1021/acs.nanolett.6b02042 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
Figure 3. Characterization of nanoparticle superlattices with controlled ratios of particle B:B′. a) Small-angle X-ray scattering (SAXS) of nanoparticle superlattices with an increasing percent of RNA bonding elements; bcc scattering patterns are observed in all cases. b) Linear decrease in interparticle distance with increasing % particle B′ (error bars are the standard deviation of n = 2 independent experiments). c) Nanoparticle crystals with different ratios of particle B:B′ stabilized in the solid state and imaged with scanning electron microscopy (SEM). Scale bars = 1 μm.
The collective melting temperature, Tm, takes the basic form Tm = Tm,i + (i − 1)α, where Tm,i is the melting temperature of an individual sticky end (determined from the nearest neighbor model), i is the number of connections, and α is a fitting parameter (SI Materials and Methods).26,28,29 To understand the melting behavior of the two distinct types of curves (one melting transition vs two distinct transitions), two adaptations of the two-state model (eq 1) were used. In a superlattice where the DNA and RNA bonding elements are randomly mixed, the Tm is assumed to scale with the ratio of DNA−DNA/DNA−RNA nearest neighbor connections; here a single cooperative transition was observed where Tm increases linearly as a function of RNA composition from 0% to 100% RNA (dark purple to brown, Figure 2b, left). By comparison, when the DNA and RNA components are either locally phase separated or entirely distinct, the Tm was assumed to scale as the linear weight of the 100% DNA−DNA transition with the 100% DNA−RNA transition; here two distinct superimposed melting transitions were observed that exchange intensity as the
distinct melting transitions suggests the presence of separate DNA−DNA connections (Tm = 51−52 °C) as well as DNA− RNA connections (Tm = 55 °C). To understand the cooperative nature of such mixed-oligonucleotide melting transitions, a two-state model was used that has accurately described the thermodynamics of melting transitions in both DNA-linked gold nanoparticles with long DNA overlap regions and many DNA connections26 as well as molecular DNA− hybrid systems with only a few DNA connections.27 Such a two-state melting model can be represented as f=
1 ⎡ ΔH 1 + exp⎢ Rtotal ⎣
(
1 T
−
1 Tm
)⎤⎥⎦
(1)
where f is the fraction melted, ΔHtotal is the total enthalpy of melting, R is the gas constant, and Tm is the melting transition temperature. From this model, it can be inferred that ΔHtotal is responsible for the melting transition width, while Tm is the weighted average of the individual strand melting temperatures. C
DOI: 10.1021/acs.nanolett.6b02042 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
Figure 4. Nanoparticle superlattices after enzymatic removal of particle B′. a) Upon incubation of nanoparticle crystals with RNase A, the RNA bonding elements of particle B′ are hydrolyzed, releasing the B′ particles and leaving behind empty sites in the positions previously occupied by particle B′. b) Photograph of nanoparticle crystals after incubation with 1 μg of RNase A.
Figure 5. Measurement of the number of vacancies in nanoparticle superlattices after incubation with 1 μg of RNase A under ambient conditions for 12 h. a) SAXS patterns, showing loss of bcc scattering pattern at >80% particle B′. b) Scattering pattern for 50% particle B′, showing the emergence of the (100) reflection, indicating the removal of the B′ particles and the remaining nanoparticles in a simple cubic arrangement. c) Illustrations of particles in a bcc (blue) and simple cubic (black) arrangement. d) SEM images. Scale bars = 1 μm. e) A figure of merit, site occupancy factor (SOF), used to measure the number of vacancies generated after particle B′ was removed. Plot of one minus SOF (1 − SOF) versus fraction of particle B′, fit to a linear regression. These data allow for a quantitative measure of vacancies in the nanoparticle crystals.
DNA:RNA ratio varies (Figure 2b, right). This is the first example of utilizing the two-state model to examine melting transitions resulting from short sticky ends (i.e., 8-base pair vs previous examples with an 18-base pair overlap), 26,27
demonstrating the versatility and potential of this model to examine different types of DNA and RNA melting transitions in nanoparticle superlattices.27 These data suggest that there are two different regions in the superlattice, those with bonding D
DOI: 10.1021/acs.nanolett.6b02042 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
In order to quantify the number of vacancies present after enzymatic removal of particle B′, the SAXS data from Figure 5a were analyzed to generate a figure of merit called the site occupancy factor (SOF), which compares the intensity of the (100) reflection for simple cubic to that of the (110) reflection for a bcc lattice (Figure 5c and S6; SI Materials and Methods), such that SOF = 1 is the case where there are no vacancies. For particle A, lattice sites (within error) are expected to be filled at positions (0, 0, 0), and thus one minus SOF (1 − SOF) for particle A is 0 (meaning zero empty sites). For particles B and B′, the number of sites at positions (1/2, 1/2, 1/2) can be varied from 0 to 1 by RNase A, and thus 1 − SOF for particle B, B′ will range from 0 (no vacancies) to 1 (100% vacancies; no crystal). A plot of 1 − SOF for particles B, B′ versus the fraction of particle B′ fit to a linear regression has a slope of 0.389 (Figure 5e). These data reveal that 30−40% of the B′ particles that are added to the system are ultimately observed as vacancies. The equation of the line allows one to input a desired number of vacancies and calculate the ratio of particle B:B′ required at synthesis. Taken together, these data suggest that RNA bonding elements can be hydrolyzed up to a certain point (80% particle B′), after which the larger crystal can no longer stay intact. We note that these values are calculated based on the model for Structure 3. They may account for the fractions of homogeneously distributed vacancies and underestimate the fractions of overall vacancies including inhomogeneously distributed ones. When a B′ particle rich region is hydrolyzed, there may not be an ordered array of particles (as expected from S(q) of samples with high percentages of B′ particles; Figure 5a), and thus remaining particles in the region do not contribute to the diffraction peaks. These data all confirm the presence of Structure 4 and also provide insight into the nucleation and growth processes of nanoparticle superlattices. Polycrystalline nanoparticle superlattices crystallize through rapid hybridization and dehybridization of sticky ends between adjacent nanoparticles.9 Here, when the melting temperature of DNA and RNA bonding elements is well-matched (i.e., ≈3 °C difference), such that reorganization of both particle types can occur simultaneously, and only one crystal forms (evident by SAXS) even though nucleation between the bonding elements of particles A and B′ happens first. Future work investigating greater melting temperature differentials between the bonding elements may lead to different types of structures, such as core−shell and phaseseparated crystals. In addition, we also demonstrate a strategy to manipulate nanoscale bonds without changing the identity of the nanoparticle atom, something not observed in atomic systems. These data lay the groundwork for new types of structures such as those with programmable amounts of defects and dopants. All atomic crystalline materials contain some type of intrinsic defect structure that is often responsible for tuning and controlling the material properties beyond what is already present due to the atomic bonding and lattice structure.32−34 We believe that methodically combining different oligonucleotide bonds can lead to defect structures in nanoparticle superlattices with enhanced properties, which would be of interest for catalytic and optical materials.
elements of particles A and B and those with bonding elements of particles A and B′. This can be most easily seen in the melting transitions for the superlattices with 20−50% particle B′, where two distinct transitions were observed both before and after thermal annealing (Figure 2a and S1). Therefore, Structure 3 can be eliminated because a single cooperative transition would result if all the bonding elements were randomly distributed as the melting temperature is an average of all the bonds in the superlattices.26 In order to distinguish between Structures 1, 2, and 4 (Figure 1b), small-angle X-ray scattering (SAXS) was used to characterize the superlattice symmetry and interparticle distance. In all cases, we observed scattering patterns indicative of bcc symmetry, where a shift to larger values of the scattering vector q (units of Å−1, therefore smaller interparticle distance) was observed as the percent of particle B′ increased (Figure 3a). The peak position of the initial scattering peak (q0) was used to calculate the interparticle distance, where a linear decrease in interparticle distance was observed with increasing percent of particle B′ (Figure 3b, SI Materials and Methods), which is consistent with previous results14 as well as with the structural differences between DNA/DNA and DNA/RNA molecular duplexes.28 These data are significant because the decreasing interparticle distance with increasing percent particle B′ coupled with the presence of only a single scattering pattern confirms the incorporation of both particles B and B′ within a single superlattice, eliminating Structures 1 and 2 because they would both exhibit two distinct X-ray scattering patterns, and thus confirming the presence of Structure 4 (Figure 1b, Figure S3). Upon transferring the nanoparticle superlattices to the solid state30 and examining their morphology using scanning electron microscopy (SEM), single crystal rhombic dodecahedra were observed in all cases (Figure 3c).14,23 We hypothesized that the crystal lattices synthesized with random DNA and RNA bonding elements throughout could be dynamically addressed with ribonuclease A (RNase A),31 which would selectively and enzymatically hydrolyze the RNA bonding elements and remove the B′ particles from the lattice structure while leaving particles A and B intact (Figure 4a). Upon addition of RNase A under ambient conditions, the supernatants changed from clear to red over the course of 12 h, depending on the ratio of particle B:B′ incorporated into each crystal (Figure 4b). Melting analyses were conducted on the pellet, and UV−visible spectroscopy of the supernatant was used to confirm the complete removal of the B′ particles from the crystals (Figures S2 and S4a and SI Discussion). SAXS analysis revealed that well-ordered bcc scattering patterns were maintained for structures that contained up to 50% particle B′, after which broadening of the q0 peak coupled with a decrease in the number of higher order scattering peaks was observed (Figures 5a and S4b). We also observed a (110) reflection, the initial scattering peak for a simple cubic lattice, thus further confirming the removal of B′ particles from the crystal (Figure 5b,c). To determine the effect that enzymatic removal of particle B′ had on the crystal morphology, we turned to SEM, where it was observed that, at low percentages of particle B′ removed (50%, faceted structures were no longer observed after treatment with RNase A; by >80% particle B′, all mesoscale shape was lost, and only free particles and silica were observed.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.6b02042. E
DOI: 10.1021/acs.nanolett.6b02042 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
■
(11) Pal, S.; Zhang, Y.; Kumar, S. K.; Gang, O. J. Am. Chem. Soc. 2015, 137 (12), 4030−4033. (12) Maye, M. M.; Kumara, M. T.; Nykypanchuk, D.; Sherman, W. B.; Gang, O. Nat. Nanotechnol. 2010, 5 (2), 116−120. (13) Zhang, Y.; Pal, S.; Srinivasan, B.; Vo, T.; Kumar, S.; Gang, O. Nat. Mater. 2015, 14 (8), 840−847. (14) Barnaby, S. N.; Thaner, R. V.; Ross, M. B.; Brown, K. A.; Schatz, G. C.; Mirkin, C. A. J. Am. Chem. Soc. 2015, 137 (42), 13566−13571. (15) Zhang, C.; Macfarlane, R. J.; Young, K. L.; Choi, C. H. J.; Hao, L.; Auyeung, E.; Liu, G.; Zhou, X.; Mirkin, C. A. Nat. Mater. 2013, 12 (8), 741−746. (16) Sun, D.; Gang, O. J. Am. Chem. Soc. 2011, 133 (14), 5252− 5254. (17) Young, K. L.; Ross, M. B.; Blaber, M. G.; Rycenga, M.; Jones, M. R.; Zhang, C.; Senesi, A. J.; Lee, B.; Schatz, G. C.; Mirkin, C. A. Adv. Mater. 2014, 26 (4), 653−9. (18) Jones, M. R.; Macfarlane, R. J.; Lee, B.; Zhang, J.; Young, K. L.; Senesi, A. J.; Mirkin, C. A. Nat. Mater. 2010, 9 (11), 913−917. (19) O’Brien, M. N.; Jones, M. R.; Lee, B.; Mirkin, C. A. Nat. Mater. 2015, 14 (8), 833−9. (20) Lu, F.; Yager, K. G.; Zhang, Y.; Xin, H.; Gang, O. Nat. Commun. 2015, 6, 6912. (21) Ross, M. B.; Ku, J. C.; Lee, B.; Mirkin, C. A.; Schatz, G. C. Adv. Mater. 2016, 28 (14), 2790−4. (22) Senesi, A. J.; Eichelsdoerfer, D. J.; Macfarlane, R. J.; Jones, M. R.; Auyeung, E.; Lee, B.; Mirkin, C. A. Angew. Chem., Int. Ed. 2013, 52 (26), 6624−6628. (23) Auyeung, E.; Li, T. I. N. G.; Senesi, A. J.; Schmucker, A. L.; Pals, B. C.; de la Cruz, M. O.; Mirkin, C. A. Nature 2013, 505 (7481), 73− 77. (24) Ross, M. B.; Ku, J. C.; Vaccarezza, V. M.; Schatz, G. C.; Mirkin, C. A. Nat. Nanotechnol. 2015, 10 (5), 453−8. (25) Macfarlane, R. J.; Jones, M. R.; Lee, B.; Auyeung, E.; Mirkin, C. A. Science 2013, 341 (6151), 1222−1225. (26) Jin, R.; Wu, G.; Li, Z.; Mirkin, C. A.; Schatz, G. C. J. Am. Chem. Soc. 2003, 125 (6), 1643−1654. (27) Prytkova, T. R.; Eryazici, I.; Stepp, B.; Nguyen, S. T.; Schatz, G. C. J. Phys. Chem. B 2010, 114 (8), 2627−2634. (28) Sugimoto, N.; Nakano, S.-i.; Katoh, M.; Matsumura, A.; Nakamuta, H.; Ohmichi, T.; Yoneyama, M.; Sasaki, M. Biochemistry 1995, 34 (35), 11211−11216. (29) Sugimoto, N.; Nakano, S.; Yoneyama, M.; Honda, K. Nucleic Acids Res. 1996, 24 (22), 4501−4505. (30) Auyeung, E.; Macfarlane, R. J.; Choi, C. H.; Cutler, J. I.; Mirkin, C. A. Adv. Mater. 2012, 24 (38), 5181−6. (31) Raines, R. T. Chem. Rev. 1998, 98 (3), 1045−1066. (32) Banhart, F.; Kotakoski, J.; Krasheninnikov, A. V. ACS Nano 2011, 5 (1), 26−41. (33) Watkins, G. D. Mater. Sci. Semicond. Process. 2000, 3 (4), 227− 235. (34) Tang, M.; Colombo, L.; Zhu, J.; Diaz de la Rubia, T. Phys. Rev. B: Condens. Matter Mater. Phys. 1997, 55 (21), 14279−14289.
Discussions, materials and methods, additional melting curves, details on the theoretical models used, as well as additional characterization data of nanoparticle superlattices after RNase A (PDF)
AUTHOR INFORMATION
Corresponding Authors
*Byeongdu Lee:
[email protected]. *George Schatz:
[email protected]. *Chad Mirkin:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This material is based upon work supported by the following awards: Air Force Office of Scientific Research FA9550-11-10275; Department of Defense National Security Science and Engineering Faculty Fellowship N00014-15-1-0043; Department of the Navy, Office of Naval Research N00014-11-1-0729, and the National Science Foundation’s MRSEC program (DMR-1121262) at the Materials Research Center of Northwestern University. S.N.B. and R.V.T. acknowledge National Science Foundation Graduate Research Fellowships. S.N.B. also acknowledges a P.E.O. Scholar Award. M.B.R. acknowledges a National Defense and Science Engineering Graduate Research Fellowship. This work made use of the EPIC facility of the NUANCE Center at Northwestern University, which has received support from the Soft and Hybrid Nanotechnology Experimental (SHyNE) Resource (NSF NNCI-1542205); the MRSEC program (NSF DMR-1121262) at the Materials Research Center; the International Institute for Nanotechnology (IIN); the Keck Foundation; and the State of Illinois, through the IIN. Portions of this work were carried out at the DuPont-Northwestern-Dow Collaborative Access Team (DND-CAT) beamline located at Sector 5 of the Advanced Photon Source (APS). DND-CAT is supported by E. I. DuPont de Nemours & Co., Dow Chemical Company, and the state of Illinois. Use of the APS was supported by the U.S. DOE, Office of Science, Office of Basic Energy Sciences, under contract DEAC02-06CH11357.
■
REFERENCES
(1) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382 (15), 607−609. (2) Park, S. Y.; Lytton-Jean, A. K. R.; Lee, B.; Weigand, S.; Schatz, G. C.; Mirkin, C. A. Nature 2008, 451 (7178), 553−556. (3) Macfarlane, R. J.; Lee, B.; Jones, M. R.; Harris, N.; Schatz, G. C.; Mirkin, C. A. Science 2011, 334 (6053), 204−208. (4) Alivisatos, A. P.; Johnsson, K. P.; Peng, X.; Wilson, T. E.; Loweth, C. J.; Bruchez, M. P.; Schultz, P. G. Nature 1996, 382 (6592), 609− 611. (5) Tan, S. J.; Campolongo, M. J.; Luo, D.; Cheng, W. Nat. Nanotechnol. 2011, 6 (5), 268−276. (6) Nykypanchuk, D.; Maye, M. M.; van der Lelie, D.; Gang, O. Nature 2008, 451 (7178), 549−552. (7) Jones, M. R.; Seeman, N. C.; Mirkin, C. A. Science 2015, 347 (6224), 1260901. (8) Hill, H. D.; Macfarlane, R. J.; Senesi, A. J.; Lee, B.; Park, S. Y.; Mirkin, C. A. Nano Lett. 2008, 8 (8), 2341−2344. (9) Macfarlane, R. J.; Thaner, R. V.; Brown, K. A.; Zhang, J.; Lee, B.; Nguyen, S. T.; Mirkin, C. A. Proc. Natl. Acad. Sci. U. S. A. 2014, 111 (42), 14995−15000. (10) Seo, S. E.; Wang, M. X.; Shade, C. M.; Rouge, J. L.; Brown, K. A.; Mirkin, C. A. ACS Nano 2016, 10 (2), 1771−1779. F
DOI: 10.1021/acs.nanolett.6b02042 Nano Lett. XXXX, XXX, XXX−XXX
