Toward a Quantitative Understanding of the Reduction Pathways of a

Dec 13, 2016 - Despite the pivotal role played by the reduction of a salt precursor in the synthesis of metal nanocrystals, it is still unclear how th...
0 downloads 0 Views 4MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Toward a Quantitative Understanding of the Reduction Pathways of a Salt Precursor in the Synthesis of Metal Nanocrystals Tung-Han Yang,†,‡ Hsin-Chieh Peng,† Shan Zhou,§ Chi-Ta Lee,∥ Shixiong Bao,† Yi-Hsien Lee,‡ Jenn-Ming Wu,‡ and Younan Xia*,†,§,∥ †

The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, Georgia 30332, United States ‡ Department of Materials Science and Engineering, National Tsing Hua University, Hsinchu, 30013, Taiwan § School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332, United States ∥ School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States S Supporting Information *

ABSTRACT: Despite the pivotal role played by the reduction of a salt precursor in the synthesis of metal nanocrystals, it is still unclear how the precursor is reduced. The precursor can be reduced to an atom in the solution phase, followed by its deposition onto the surface of a growing nanocrystal. Alternatively, the precursor can adsorb onto the surface of a growing nanocrystal, followed by reduction through an autocatalytic process. With Pd as an example, here we demonstrate that the pathway has a correlation with the reduction kinetics involved. Our quantitative analyses of the reduction kinetics of PdCl42− and PdBr42− by ascorbic acid at room temperature in the absence and presence of Pd nanocubes, respectively, suggest that PdCl42− was reduced in the solution phase while PdBr42− was reduced on the surface of a growing nanocrystal. Our results also demonstrate that the reduction pathway of PdBr42− by ascorbic acid could be switched from surface to solution by raising the reaction temperature. KEYWORDS: Nanocrystal synthesis, precursor reduction, seed-mediated growth, kinetic model

M

etal nanocrystals1−4 have received ever-increasing interest owing to their fascinating properties for a variety of applications, including catalysis,5−7 electronics,8 photonics,9 sensing, 10 and medicine. 11 Most of these applications require the use of metal nanocrystals with a specific shape because this parameter determines the physiochemical properties of a metal nanocrystal.12,13 Thanks to the efforts from many research groups, it is now possible to rationally synthesize metal nanocrystals with diversified shapes.14−16 The synthesis typically involves the reduction of a salt precursor to generate atoms, followed by their nucleation and then growth into nanocrystals.17,18 Such an approach has been used for hundreds of years, with the first documentation made by Michael Faraday for the synthesis of Au colloids in 1856. However, it is still unclear how the salt precursor is reduced to atoms for their evolution into nuclei, seeds, and then nanocrystals. It has been challenging to resolve the details of such a process due to the lack of analytical tools.19−22 Two Different Reduction Pathways. As illustrated in Figure 1a and b, the reduction of a salt precursor can take two completely different pathways during a synthesis of metal nanocrystals. When the ions of a salt precursor are introduced into a solution containing a reductant and growing nanocrystals © XXXX American Chemical Society

(or preformed seeds), they can be directly reduced to atoms in the solution phase through collision and electron transfer with the reductant molecules (Figure 1a, solution pathway).17,18 The resultant atoms then undergo homogeneous or heterogeneous nucleation, with the latter being more favorable due to its lower activation energy barrier. At the beginning of a one-pot synthesis, solution-phase reduction should be the only option for the salt precursor. However, once seeds have been formed (or in the case of seed-mediated growth with the introduction of preformed seeds), the reduction of a salt precursor can undertake an alternative mechanism (Figure 1b, surface pathway). In this case, the precursor ions first adsorb onto the surface of a growing nanocrystal or a preformed seed, followed by their reduction to atoms.23−26 These two reduction pathways can result in completely different types of products, with the second pathway capable of excluding the possible involvement of homogeneous nucleation. With Pd as a typical example, here we design a set of kinetic experiments to shed light on the reduction pathway undertaken Received: October 4, 2016 Revised: December 9, 2016 Published: December 13, 2016 A

DOI: 10.1021/acs.nanolett.6b04151 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 1. Two different reduction pathways possibly involved in a seed-mediated synthesis. (a, b) Schematic illustrations of the reduction pathways (solution reduction vs surface reduction) for a precursor ion in the seed-mediated growth of metal nanocrystals. (c, d) TEM images of products obtained when (c) PdCl42− and (d) PdBr42− were used as precursors, respectively, in the presence of 18 nm Pd cubic seeds. The smaller particles marked in panel c were formed through homogeneous nucleation.

confined to the corners and edges. The site-selected growth can be attributed to the strong passivation of Pd{100} facets by the chemisorbed Br− ions and thereby the retardation of growth at these sites. Apart from the concave cubes grown from the Pd cubic seeds, there were also much smaller particles of 3.4 nm in diameter with a nearly spherical shape (Figure S3), which can be ascribed to homogeneous nucleation. The coexistence of the concave cubes and spherical particles in the final products indicates the involvement of both homogeneous and heterogeneous nucleation and thus the likelihood of solution reduction pathway. In contrast, when PdBr42− was used as the salt precursor, homogeneous nucleation was essentially suppressed for the formation of Pd concave nanocubes only (Figure 1d). To quantitatively understand the effect of precursor type on the reduction pathway in the formation of nanocrystals, we used a UV−vis spectroscopy method to measure the concentrations of Pd(II) ions remaining in the reaction solution at different time points after the introduction of salt precursor (Figures S4 and S5). We conducted two sets of parallel measurements in the absence and presence of Pd cubic seeds, respectively, while all other experimental conditions were kept the same to examine whether the presence of seeds would affect the reduction kinetics. The detailed experimental procedures for kinetic study are provided in the Supporting Information. Figure 2a shows the concentrations of PdCl42− remaining in the reaction solution in the absence and presence of Pd cubic seeds as a function of reaction time. The almost

by a salt precursor during the synthesis of metal nanocrystals. Our quantitative analysis suggests that the pathway is mainly determined by the reduction kinetics involved in the synthesis. Specifically, we measured the reduction kinetics of PdCl42− and PdBr42− in the absence or presence of preformed seeds. A standard experiment involves the use of Pd cubes with an average edge length of 18 nm as the seeds (Figure S1a), along with PdX42− (X = Cl− or Br−), L-ascorbic acid (AA), and poly(vinylpyrrolidone) (PVP) as the metal precursor, reductant, and colloidal stabilizer, respectively. The cubic seeds were slightly truncated at the corner and edge sites, with their {100} side faces passivated by chemisorbed Br− ions (Figure S1b). In the present work, we focus on PdCl42− and PdBr42− to investigate the role of kinetics in determining the reduction pathway. Since the reduction potential of PdCl42−/Pd is more positive than that of PdBr42−/Pd, PdCl42− should be reduced at a faster rate.27,28 When applied to seed-mediated growth under identical conditions, these two precursors led to completely different products. Figure 1c and d shows transmission electron microscopy (TEM) images of the resultant Pd nanocrystals when PdCl42− and PdBr42−, respectively, were mixed with AA and PVP at room temperature in the presence of the Pd cubic seeds. In the case of PdCl42−, the products contained nanocrystals with two distinct sizes and shapes (Figure 1c). The larger nanocrystals exhibited a concave structure on the surface, together with an edge length of 19.9 nm (see Figure S2 for the definition of size), implying that the deposition of Pd did not occur evenly on the entire surface of a cubic seed but B

DOI: 10.1021/acs.nanolett.6b04151 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 2. Quantitative analyses of the kinetic parameters for the reduction of two different Pd(II) precursors by AA. (a) The concentrations of PdCl42− precursor remaining in the reaction solutions as a function of reaction time in the absence or presence of seeds. The experimental data were then fitted using the Finke−Watzky (F−W) model. (b) The rates of solution reduction and surface reduction for the PdCl42− precursor as a function of reaction time in the absence or presence of seeds. (c) The concentrations of PdBr42− precursor remaining in the reaction solutions as a function of reaction time in the absence or presence of seeds. The experimental data were then fitted using the Finke−Watzky (F−W) model. (d) The rates of solution reduction and surface reduction for the PdBr42− precursor as a function of reaction time in the absence or presence of seeds.

where [Pd0n] is the concentration of Pd nuclei. The rate constants, k1 and k2, for the reduction of PdCl42− in the absence of seeds could be derived from curve fitting according to the Finke−Watzky model (see the Supporting Information for details). We obtained k1 = 6.20 × 10−2 min−1 and k2 = 4.20 × 10−2 min−1 mM−1 (Figures 2a, S7, and eq S5). In comparison, with the introduction of preformed Pd cubic seeds, defined as Pd0n(seed), the seeds would provide extra surfaces for the autocatalytic reduction of Pd(II) (surface

identical data in the absence and presence of seeds indicates that the presence of seeds did not significantly affect the reduction pathway. We further fitted the experimental data with a kinetic model to derive the rate constants of the elementary reactions involved. As shown in Figure 2a, the reduction of PdCl42− by AA in the absence of seeds could be best described using the Finke− Watzky model,23,24 in which Pd(II) ions were first reduced to k1

zerovalent atoms (solution reduction: Pd(II) + 2e− → Pd0) for their aggregation into nuclei through homogeneous nucleation, followed by autocatalytic surface growth enabled by the just-formed nuclei (surface reduction: Pd0n + Pd(II) +

k 2′

reduction: Pd n0(seed) + Pd(II) + 2e− → Pd n0+ 1(seed)), and thus an additional rate constant (as defined as k2′ ) has to be introduced. By taking the values of k1 and k2 from the control experiment in the absence of preformed seeds, we obtained a value of 1.94 × 10−1 min−1 mM−1 for the rate constant (k′2) of surface reduction (Figures 2a, S7, and eq S16). Based on these kinetic parameters, we further derived the reduction rates (Figure 2b) of PdCl42− in the absence and presence of the seeds, respectively. As expected, the rates of solution reduction were much faster than those of surface reduction regardless of the absence/presence of seeds. Furthermore, we calculated the percentages of solution reduction and surface reduction as a function of reaction time by integrating the reduction rate over

k2

2e− → Pd n0+ 1). The resultant particles had an average diameter of 3.7 nm (Figure S6). Note that the amount of reductant (in this case, AA) was in large excess relative to the Pd(II) ions and thus the concentration of the reductant could be assumed as an invariable during the synthesis. Therefore, the rate equation for the reduction of Pd(II) ions in the absence of seeds can be expressed as

rate = k1[Pd(II)] + k 2[Pd(II)][Pd n0]

(1) C

DOI: 10.1021/acs.nanolett.6b04151 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Table 1. Kinetic Parameters for the Reduction of PdCl42− and PdBr42− Precursors in the Absence or Presence of Preformed Seeds at Room Temperature (22 °C)a rate constant

percentage (%)

precursor

seeds (mg)

solution reduction

surface reduction

solution reduction

surface reduction

PdCl42−

0 0.36

k1 = 6.20 × 10−2 min−1 k1 = 6.20 × 10−2 min−1

95.2 89.3

4.80 10.7

PdBr42−

0 0.36

k1 = 9.17 × 10−4 min−1 k1 = 9.17 × 10−4 min−1

k2 = 4.20 × 10−2 mM−1 min−1 k2 = 4.20 × 10−2 mM−1 min−1 k2′ = 1.94 × 10−1 mM−1 min−1 k2 = 3.53 × 10−2 mM−1 min−1 k2′ = 1.72 × 10−1 mM−1 min−1

32.4 19.5

67.6 80.5

a

The percentages of solution reduction and surface reduction are also presented for the Pd(II) precursor at the point when its concentration dropped to 0.17 mM.

Figure 3. Quantitative analysis of the reduction of two different Pd(II) precursors by AA in the presence of different amounts of the Pd seeds. (a) A plot showing the concentrations of PdCl42− precursor remaining in the reaction solutions as a function of time when different amounts of seeds were introduced. (b) The rates of solution reduction and surface reduction for the PdCl42− precursor at a concentration of 0.17 mM with the introduction of different amounts of seeds. (c) A plot showing the concentrations of PdBr42− precursor remaining in the reaction solutions as a function of time when different amounts of the seeds were introduced. (d) The rates of solution reduction and surface reduction for the PdBr42− precursor (0.17 mM in concentration) with the introduction of different amounts of seeds.

comparison, the reduction of PdBr42− in the presence of seeds significantly deviated from the concentration profile of PdBr42− in the absence of seeds. Based on the TEM image in Figure 1d, homogeneous nucleation could be excluded so no surface reduction of PdBr42− on the newly formed nuclei would be expected. By setting k1 and k2 to 9.17 × 10−4 min−1 and 0, we obtained a value of k2′ = 1.72 × 10−1 min−1 mM−1 for the surface reduction pathway (Figures 2c, S7, and eq S10). The rates of solution reduction and surface reduction of PdBr42− as a function of reaction time in the absence or presence of seeds are presented in Figure 2d. In the absence of seeds, the surface reduction rate increased rapidly with time and was much faster than the rate of solution reduction after 60 min. In the presence of seeds, the surface reduction always proceeded faster than the solution reduction throughout the synthesis. This result suggests that, once the reaction solution contained a significantly large amount of nuclei or seeds, the reduction of

time (Figure S8). In the absence and presence of seeds, the percentages of solution pathway accounted for over 89% of the reduction (Table 1), suggesting that the PdCl42− ions were reduced to atoms in the solution prior to their deposition onto the surface of nucleus and/or seeds. We further extended the kinetic measurements to the reduction of PdBr42− (Figure 2c and d). Interestingly, the reduction of PdBr42− by AA in the absence and presence of Pd cubic seeds showed two completely different dependences on reaction time. As shown in Figure 2c, the reduction of PdBr42− in the absence of seeds was very slow at the initial stage of a synthesis (0−60 min) and increased drastically later on. By fitting the experimental data to the Finke−Watzky model, we obtained k1 = 9.17 × 10−4 min−1 and k2 = 3.53 × 10−2 min−1 mM−1 for the rate constants (Figures 2c, S7, and eq S5), which were smaller than those of PdCl42− under the same conditions due to the lower reduction potential of PdBr42−.28 In D

DOI: 10.1021/acs.nanolett.6b04151 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 4. Transition of surface reduction to solution reduction for PdBr42− precursor under different reaction temperatures and determination of the corresponding activation energies. (a) Plots showing the concentrations of PdBr42− precursor remaining in the reaction solutions as a function of time at different temperatures in the absence and presence of the seeds. The reduction kinetics for PdBr42− precursor in the absence and presence of seeds are denoted in black and red, respectively. (b) Percentages of solution reduction and surface reduction for PdBr42− precursor at the point when its concentration dropped to 0.17 mM were obtained in the presence of seeds. (c) A plot showing ln k1 and ln k2′ as a function of 1/T for the solution reduction and surface reduction of PdBr42− in the presence of seeds, respectively, where the slope of linear regression line can be used to calculate the corresponding activation energy (Ea) of the reduction pathway using the Arrhenius equation. (d) Potential energy diagrams for the two different reduction pathways. The reduction of the precursor through an autocatalytic mechanism (surface reduction) has a much lower activation energy than the corresponding uncatalyzed reduction mechanism (solution reduction).

PdBr42− mainly underwent the faster, autocatalytic surface pathway and thus diminished the probability of homogeneous nucleation. The percentage of surface reduction reached up to a level of 80% in the presence of seeds (Figure S8 and Table 1). The difference in reduction kinetics between PdCl42− and PdBr42− in the absence/presence of seeds can be attributed to the involvement of different reduction pathways. In the case of PdCl42−, regardless of the absence/presence of seeds in the reaction solution, the reduction kinetics was more or less the same (Figure 2a and b). This observation implies that the

presence of Pd seeds did not affect the reduction of PdCl42− by AA under the experimental conditions used. In other words, the precursor ions are likely reduced to atoms in the solution phase, followed by their deposition onto the surface of seeds. In contrast, the reduction of PdBr42− by AA showed a completely different behavior depending on the absence/presence of seeds. In contrast to the sluggish kinetics in the absence of seeds, the reduction rate was significantly accelerated in the presence of seeds (Figure 2c and d), suggesting that the seeds indeed participated in the reduction of the precursor. From the kinetic E

DOI: 10.1021/acs.nanolett.6b04151 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters data, we believe that the reduction of PdBr42− by AA under the experimental conditions used in this work underwent a different mechanism where the precursor was reduced on the surface of a seed rather than in the solution phase. Validation of the Proposed Mechanism. According to eqs S9 and S15, the rate of autocatalytic reduction is directly proportional to both the concentration of precursor and the number of seeds. To further validate this correlation, we conducted a set of experiments with the introduction of different amounts of Pd cubic seeds to alter their concentration. The corresponding reduction rates and rate constants were again obtained from the kinetic data acquired using the spectroscopy method (Figures S9−S14 and Table S1). As shown in Figure 3, the reduction of PdCl42− and PdBr42− shows markedly distinctive dependence on the amount of Pd cubic seeds. Specifically, in the case of PdCl42−, the total reduction rate (solution plus surface) showed a negligible dependence on the amount of seeds involved (Figure 3b). This is in agreement with the proposed reduction pathway for PdCl42−, where solution reduction dominates over surface reduction. In comparison, there was clearly a positive correlation between the total reduction rate and the amount of seeds for the syntheses involving PdBr42− (Figure 3d). As the amount of seeds was increased, the total area of active surface that could catalyze the reduction of PdBr42− would increase proportionally, leading to a greater reduction rate. These results suggest that the reduction rate of a precursor would be highly dependent on the area of active surface present in the reaction solution if the reduction is an autocatalytic process, and vice versa. Taken together, it is clear that the difference in reduction kinetics can lead to different pathways for the reduction of a precursor during the formation of metal nanocrystals. Transition from Surface Reduction to Solution Reduction. We also conducted a set of experiments to further demonstrate how the reduction pathway of PdBr42− could be altered by manipulating the reaction temperature and thus reduction kinetics. Specifically, we measured the reduction kinetics of PdBr42− in the absence/presence of Pd cubic seeds at different temperatures. As shown in Figure 4a, the reduction kinetics shows a strong dependence on temperature. The reduction of PdBr42− was greatly accelerated after the introduction of seeds at temperatures ranging from 0 to 30 °C. Besides this, the reduction kinetics of PdBr42− in the presence of seeds was surprisingly consistent with those for the syntheses conducted in the absence of seeds at 40, 50, and 60 °C, suggesting that the reduction of PdBr42− might follow the same pathway (i.e., solution reduction) at elevated temperatures regardless of the absence/presence of seeds. Again, the corresponding kinetic parameters for the reduction of PdBr42− under different conditions could be derived from the Finke− Watzky model (Figures S15−S20 and Table S2). Figure 4b shows the percentages of solution reduction and surface reduction of PdBr42− in the presence of seeds as a function of reaction temperature. We observed a transition from surface reduction to solution reduction as the temperature was raised, which was supported by the observation of tiny, homogeneously nucleated nanoparticles in the product if the synthesis was conducted at an elevated temperature (Figure S21). Taken together, these results indicate that the two pathways involved in the reduction of PdBr42− could be switched from one to another by varying the experimental conditions such as reaction temperature.

Most importantly, from the rate constants at different temperatures (Table S2), the activation energy (Ea) of the reduction reaction involving PdBr42− and preformed seeds can be derived using the Arrhenius equation. Figure 4c shows a plot of ln k1 and ln k′2 as a function of 1/T for the solution reduction and surface reduction of PdBr42− in the presence of seeds, respectively. Based on the Arrhenius equation, the activation energies of the solution reduction and surface reduction were determined to be 131.3 and 43.4 kJ/mol, respectively. In general, the precursor should follow the surface reduction pathway due to a small activation energy of 43.4 kJ/mol. The precursor could also be reduced in the solution phase, but with a much greater activation energy of 131.3 kJ/mol. Figure 4d shows an illustration of the potential energy diagrams for these two parallel reduction pathways. Since the precursor ions collide more frequently with the reducing agents than with the seeds, solution reduction is always more favorable than surface reduction as long as the temperature is high enough to overcome the activation energy barrier. In summary, we have quantitatively analyzed the reaction kinetics involved in the seed-mediated growth of Pd nanocrystals using a spectroscopy method and further had it correlated to the reduction pathway of a salt precursor. Based on the quantitative data, it can be concluded that the precursor would be reduced on the surface of a seed through an autocatalytic process under slow reduction kinetics, whereas it would be reduced in the reaction solution when the kinetics was fast. Most significantly, we further demonstrated that the reduction pathway could be switched from one to another by manipulating the experimental parameters such as the type of precursor and the reaction temperature, to modulate the reduction kinetics. We believe that the quantitative understanding of reduction pathway reported in this work would provide a guideline for achieving rational design and synthesis of nanocrystals.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.6b04151. Experimental details, detailed methods for analyzing the reaction kinetics, TEM images, size distributions of the particles formed through homogeneous nucleation, UV− vis spectra, curve fittings based on the Finke−Watzky model, the rates for solution reduction and surface reduction of Pd(II) precursors, two tables of kinetic parameters (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Tung-Han Yang: 0000-0001-6161-4397 Younan Xia: 0000-0003-2431-7048 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported in part by a research grant from the NSF (DMR 1506018) and start-up funds from the Georgia F

DOI: 10.1021/acs.nanolett.6b04151 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

(28) Zhang, H.; Jin, M.; Wang, J.; Li, W.; Camargo, P. H. C.; Kim, M. J.; Yang, D.; Xie, Z.; Xia, Y. J. Am. Chem. Soc. 2011, 133, 6078−6089.

Institute of Technology. As a visiting Ph.D. student from National Tsing Hua University, T.H.Y. was also partially supported by a fellowship from the Ministry of Science and Technology, Taiwan.

■

REFERENCES

(1) Ahmadi, T. S.; Wang, Z. L.; Green, T. C.; Henglein, A.; El-Sayed, M. A. Science 1996, 272, 1924−1925. (2) Sun, Y.; Xia, Y. Science 2002, 298, 2176−2179. (3) Chiu, C. Y.; Li, Y.; Ruan, L.; Ye, X.; Murray, C. B.; Huang, Y. Nat. Chem. 2011, 3, 393−399. (4) Ye, X.; Chen, J.; Engel, M.; Millan, J. A.; Li, W.; Qi, L.; Xing, G.; Collins, J. E.; Kagan, C. R.; Li, J.; Glotzer, S. C.; Murray, C. B. Nat. Chem. 2013, 5, 466−473. (5) Yamada, Y.; Tsung, C. K.; Huang, W.; Huo, Z.; Habas, S. E.; Soejima, T.; Aliaga, C. E.; Somorjai, G. A.; Yang, P. Nat. Chem. 2011, 3, 372−376. (6) Wang, X.; Choi, S.; Roling, L. T.; Luo, M.; Ma, C.; Zhang, L.; Chi, M.; Liu, J.; Xie, Z.; Herron, J. A.; Mavrikakis, M.; Xia, Y. Nat. Commun. 2015, 6, 7594. (7) Zhang, L.; Roling, L. T.; Wang, X.; Vara, M.; Chi, M.; Liu, J.; Choi, S.; Park, J.; Herron, J. A.; Xie, Z.; Mavrikakis, M.; Xia, Y. Science 2015, 349, 412−416. (8) Choi, J. H.; Wang, H.; Oh, S. J.; Paik, T.; Jo, P. S.; Sung, J.; Ye, X.; Zhao, T.; Diroll, B. T.; Murray, B. T.; Kagan, C. Science 2016, 352, 205−208. (9) Henzie, J.; Andrews, S. C.; Ling, X. Y.; Li, Z.; Yang, P. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 6640−6645. (10) Liu, N.; Tang, M. L.; Hentschel, M.; Giessen, H.; Alivisatos, A. P. Nat. Mater. 2011, 10, 631−636. (11) Yavuz, M. S.; Cheng, Y.; Chen, J.; Cobley, C. M.; Zhang, Q.; Rycenga, M.; Xie, J.; Kim, C.; Song, K. H.; Schwartz, A. G.; Wang, L. V.; Xia, Y. Nat. Mater. 2009, 8, 935−939. (12) Xia, Y.; Xiong, Y.; Lim, B.; Skrabalak, S. E. Angew. Chem., Int. Ed. 2009, 48, 60−103. (13) Xia, Y.; Xia, X.; Peng, H. C. J. Am. Chem. Soc. 2015, 137, 7947− 7966. (14) Buck, M. R.; Bondi, J. F.; Schaak, R. E. Nat. Chem. 2012, 4, 37− 44. (15) Auyeung, E.; Li, T. N. G.; Senesi, A. J.; Schmucker, A. L.; Pals, B. C.; de la Cruz, M. O.; Mirkin, C. A. Nature 2014, 505, 73−77. (16) Wu, X. J.; Chen, J.; Tan, C.; Zhu, Y.; Han, Y.; Zhang, H. Nat. Chem. 2016, 8, 470−475. (17) Lamer, V. K.; Dinegar, R. H. J. Am. Chem. Soc. 1950, 72, 4847− 4854. (18) Thanh, N. T. K.; Maclean, N.; Mahiddine, S. Chem. Rev. 2014, 114, 7610−7630. (19) Cacciuto, A.; Frenkel, D. Nature 2004, 428, 404−406. (20) Yao, T.; Liu, S.; Sun, Z.; Li, Y.; He, S.; Cheng, H.; Xie, Y.; Liu, Q.; Jiang, Y.; Wu, Z.; Pan, Z.; Yan, Y.; Wei, S. J. Am. Chem. Soc. 2012, 134, 9410−9416. (21) Liao, H. G.; Zherebetskyy, D.; Xin, H.; Czarnik, C.; Ercius, P.; Elmlund, H.; Pan, M.; Wang, L. W.; Zheng, H. Science 2014, 345, 916−919. (22) Peng, Y.; Wang, F.; Wang, Z.; Alsayed, A. M.; Zhang, Z.; Yodh, A. G.; Han, Y. Nat. Mater. 2015, 14, 101−108. (23) Watzky, M. A.; Finke, R. G. J. Am. Chem. Soc. 1997, 119, 10382−10400. (24) Watzky, M. A.; Finney, E. E.; Finke, R. G. J. Am. Chem. Soc. 2008, 130, 11959−11969. (25) Liu, Q.; Gao, M. R.; Liu, Y.; Okasinski, J. S.; Ren, Y.; Sun, Y. Nano Lett. 2016, 16, 715−720. (26) van den Berg, R.; Elkjaer, C. F.; Gommes, C. J.; Chorkendorff, I.; Sehested, J.; de Jongh, P. E.; de Jong, K. P.; Helveg, S. J. Am. Chem. Soc. 2016, 138, 3433−3442. (27) Timoshkin, A. Y.; Kudrev, A. G. Russ. J. Inorg. Chem. 2012, 57, 1362−1370. G

DOI: 10.1021/acs.nanolett.6b04151 Nano Lett. XXXX, XXX, XXX−XXX