Unravelling Thiol's Role in Directing Asymmetric Growth of Au

Dec 15, 2015 - The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.5b04329. Experiment...
0 downloads 9 Views 2MB Size
Subscriber access provided by The Library | The University of Auckland

Communication

Unravelling Thiol’s Role in Directing Asymmetric Growth of Au Nanorod-Au Nanoparticle Dimers Jianfeng Huang, Yihan Zhu, Changxu Liu, Zhan Shi, Andrea Fratalocchi, and Yu Han Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.5b04329 • Publication Date (Web): 15 Dec 2015 Downloaded from http://pubs.acs.org on December 17, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Unravelling Thiol’s Role in Directing Asymmetric Growth of Au Nanorod-Au Nanoparticle Dimers Jianfeng Huang†,#, Yihan Zhu†,#, Changxu Liu‡, Zhan Shi§, Andrea Fratalocchi‡*, and Yu Han†* †Advanced Membranes and Porous Materials Center, Physical Sciences and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, Saudi Arabia ‡PRIMALIGHT, Faculty of Electrical Engineering; Applied Mathematics and Computational Science, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, Saudi Arabia §State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, College of Chemistry, Jilin University, Changchun 130012, China KEYWORDS: thiol-ligand, asymmetric growth, Au nanorod, dimer, strain

ACS Paragon Plus Environment

1

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 20

ABSTRACT: Asymmetric nanocrystals have practical significance in nanotechnologies but present fundamental synthetic challenges. Thiol ligands have proven effective in breaking the symmetric growth of metallic nanocrystals, but their exact roles in the synthesis remain elusive. Here, we synthesized an unprecedented Au nanorod-Au nanoparticle (AuNR-AuNP) dimer structure with the assistance of a thiol ligand. On the basis of our experimental observations, we unraveled for the first time that the thiol could cause an inhomogeneous distribution of surface strains on the seed crystals as well as a modulated reduction rate of metal precursors, which jointly induced the asymmetric growth of monometallic dimers.

ACS Paragon Plus Environment

2

Page 3 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Colloidal noble metal nanocrystals (NCs) usually exhibit highly symmetric particle morphologies, as dictated by their intrinsic crystallographic symmetries. Breaking this morphological symmetry would bring vast variations to the plasmonic and optical properties of NCs,1 greatly enriching their applications in plasmonics,2, 3 sensing,4 nanophotonics,5 and surfaceenhanced Raman scattering.6-8 One effective approach to breaking the symmetry is to synthesize “dimer” crystals via seeded growth, during which it is possible to play with many factors, such as the size and shape of the seeds, the type of the growth material, the choice of the capping ligand, and the reaction kinetics, to manipulate the dimeric nanostructures.9-11 For example, some bimetallic (e.g., Au-Ag, Pd-Ag, Pt-Ag and Pd-Au) dimers, 12-15 which were also referred to as hybrid dimers or heterodimers, have been fabricated in this way. In our recent work, we fabricated unprecedented monometallic Au nanorod-Au nanoparticle (AuNR-AuNP) dimers that exhibit an extraordinary broadband optical absorption between 400 and 1400 nm and act as an ideal black-body material for generating a new source of light.16 In general, the fabrication of monometallic or bimetallic dimers with their two constituents having nil or little lattice mismatch (e.g., Au-Au, Au-Ag),12, 16, 17 via seed-mediated growth, is rather difficult, because one of the main driving forces for dimerization is the lattice mismatch between two materials. With the growth material the same as or close to the seed in the lattice spacing, epitaxial growth on the entire surface of the seed particle is usually favored over the formation of a dimer.18 Interestingly, in all these successful cases of Au-Au and Au-Ag dimers, thiol ligands were employed in the synthesis to switch the crystal growth modalities from a continuous epitaxial shell to a discrete particle.12, 16, 17 However, the typical utilization of irregular, multiply-twinned AuNP seeds prevents a reliable crystallographic analysis,12, 17 and consequently the role of thiol ligand remains elusive due to the lack of direct experimental evidence.

ACS Paragon Plus Environment

3

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 20

In this work, we endeavored to unveil the mechanism of thiol ligand-induced asymmetric growth, on the basis of experimental observations during the growth of a single AuNP on singlecrystalline AuNRs to form AuNP-AuNR dimers. The choice of single-crystalline AuNRs as seeds is beneficial to our study, because their anisotropic yet well-defined morphology (with known surface facet indices) allows one to easily identify the formation of a dimer interface and to track the subsequent evolution of the AuNPs. This is crucial for understanding why one site on the AuNR, at which the AuNP is grown, can be differentiated from many other symmetryequivalent sites by thiol ligands. The important roles of the thiol ligand, in both thermodynamic and kinetic aspects, have been carefully investigated and identified.

Figure 1. TEM images of (a) the AuNR seeds and (b-d) various Au nanostructures after seeded growth with different concentrations of 4-MP: (b) peanut-shaped AuNRs (4-MP: 0 mM), (c) irregular rugged AuNRs (4-MP: 1 mM), and (d) AuNR-AuNP dimers (4-MP: 11 mM). Single-crystalline AuNRs were synthesized following our previous procedures (Figure 1a).19, 20

To grow AuNR-AuNP dimers, the as-synthesized AuNRs were first incubated with a thiol-

ACS Paragon Plus Environment

4

Page 5 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

ligand 4-mercaptophenol (4-MP) for 1 hour to allow for a complete exchange of the original surface ligand hexadecyltrimethylammonium bromide (CTAB) by 4-MP. Then, a Au precursor HAuCl4 and a reducing agent ascorbic acid was added. After 5 minutes, the product was washed and collected via centrifugation (see Supporting Information). We found that the concentration of 4-MP in the solution determines the final Au nanostructure. In the absence of 4-MP, a layer of Au was continuously grown around the AuNR with a preference for the rod ends, leading to the “peanut” shape of the final structure (Figure 1b). The formation of such structure was attributed to the site-dependence of the CTAB binding strength on the surface of the nanorods. CTAB molecules were believed to have a larger packing density at the lateral sides, while the deposition of Au atoms would preferentially take place at the ends, resulting in the production of peanutshaped nanorods.21 As a matter of fact, the stabilizing power of CTAB can be adjusted by varying its concentration.22 Peanut-shaped nanorods were obtained using a relatively small concentration of CTAB in the overgrowth process, as is the case for our system.23 When the concentration was significantly increased, the Au nanorods would undergo an isotropic overgrowth, leading to the formation of cuboidal Au nanocrystals with uniform Au shells.22, 23 Selected-area electron diffraction (SAED) demonstrated that each “peanut” remains a single crystal, while the high-resolution transmission electron microscopy (HRTEM) image accordingly showed a clear epitaxial interface between the newly grown layer and the original NR (Figure S1). When the AuNRs were incubated with 1 mM of 4-MP, the growth of Au on the seed crystals became irregular, resulting in NRs with rugged surfaces (Figure 1c). A careful TEM investigation revealed that the newly grown Au layer was mainly composed of multiple-twinned crystals, while epitaxial growth was also observed at the rod ends (Figure S2). These observations suggest that two different mechanisms of secondary growth coexist under these

ACS Paragon Plus Environment

5

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 20

conditions. Interestingly, when the concentration of 4-MP was increased to 11 mM, the desired AuNR-AuNP dimer structure was successfully produced in a high yield, in which a single AuNP (~ 30 nm) was mainly located at the “neck” (the transition region between the rod end and the rod body) of each individual AuNR that was otherwise nearly unchanged (Figure 1d, and a lowmagnification TEM image Figure S3). As illustrated by HRTEM, the AuNR in the dimer remains single crystalline with a smooth surface (Figure 2a, b), whereas the AuNP is a multipletwinned crystal (Figure 2a, S4). Successive nanotwins at the interface between the NP and NR were observed in the HRTEM image (Figure 2a) and confirmed by the diffuse streaks along the [111] direction in the fast Fourier transform (FFT) diffractogram (Figure 2c). The fact that the AuNR and AuNP are intergrown through a metallic lattice (metallic bonding) distinguishes the present AuNR-AuNP dimer from many previous dimers formed by the ligand-assisted assembly of two separate Au particles.24-26 These results indicate that with a high concentration of 4-MP, the continuous epitaxial growth of the Au layer was essentially inhibited, whereas the asymmetric growth of the twinned particle became dominant. In this sense, the irregular rugged AuNRs synthesized with a medium concentration of 4-MP (i.e., 1 mM) can be considered as an intermediate between the “peanut” and the “dimer”.

ACS Paragon Plus Environment

6

Page 7 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2. (a) HRTEM image of a AuNR-AuNP dimer taken along the [110] axis of the AuNR, showing single-crystalline AuNR, multiple twinned AuNP, and a stacking fault-rich interface. (b, c) FFT diffractograms of (b) region I and (c) region II marked in (a). When the concentration of 4-MP was further increased (> 15 mM), no obvious secondary growth on the AuNR seeds was observed, possibly because the reduction of Au ions was inhibited by highly concentrated 4-MP ligands. We also found that the reaction for the peanutshaped AuNRs was completed within ~ 5 s, while the time required for the formation of AuNRAuNP dimers was > 60 s (vide infra). These observations suggest a correlation between the reduction rate of the Au precursor and the final nanostructure, where the dimer structure is favored by relatively slow reaction kinetics. To verify this, we modified the synthetic system for the dimer structure to increase the reduction rate by either elevating the concentration of ascorbic acid, replacing ascorbic acid with sodium ascorbate, or heating the reaction solution. In all cases, irregular NRs instead of dimers were produced. These results demonstrate the important influence of the reduction rate on the formation of the dimer structure. There are two possible

ACS Paragon Plus Environment

7

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 20

reasons for the slow reduction rate associated with the high concentration of 4-MP: (i) thiolligands and AuCl4− form more stable thiolate-Au(I) complexes with lower reduction potentials;27 (ii) thiol-ligands occupy the surface sites of AuNRs through strong chemical adsorption, inhibiting the reduction of Au(I) ions by ascorbic acid for which the exposed Au surface acts as a catalyst.28 However, controlled reduction kinetics of Au ions is a necessary but insufficient condition for the formation of the dimer, because when the concentration of 4-MP was low, dimers could not be obtained no matter how the reduction rate was reduced by varying other conditions, e.g., using lower temperatures or a smaller amount of ascorbic acid. In another control experiment, equal amounts of as-synthesized AuNRs (without incubation with 4-MP) and 4-MP incubated AuNRs were mixed and then used as seeds immediately after the mixing (within 10 seconds) for the synthesis of the dimer, keeping other conditions identical. Despite the presence of free thiol molecules in the mixed solution, the as-synthesized AuNRs remained CTAB-capped during the secondary growth, because the ligand exchange between CTAB and the free thiol-molecules proved to be a relatively slow process (see Figure S5). This ensured that there were two types of AuNRs in the system. We supposed the two types of AuNRs to experience the same reaction kinetics in the same reaction system. However, we found that the product was a roughly 1:1 mixture of irregular crystals and dimer crystals (Figure S6). Therefore, in addition to controlling the reaction rate, the 4-MP ligand must play other role(s), most likely to do with the thermodynamics, in directing the formation of the dimer structure. Understanding the roles of 4MP would be helpful to answering some as yet unanswered questions relevant to the dimer structure: (i) why are AuNPs preferentially grown in the neck region of the AuNR? (ii) why are the AuNPs all multiple-twinned crystals? (iii) why is there only one AuNP grown at one out of

ACS Paragon Plus Environment

8

Page 9 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

many symmetry-equivalent sites for each AuNR? (iv) how does the concentration of 4-MP influence the mechanism of secondary growth?

Figure 3. TEM images of various intermediates of the AuNR-AuNP dimer at different growth stages, which were obtained by quenching the reaction at (a) 3, (b) 10, (c) 20, (d) 60, and (e) 180 s. The circles in (a) encircle the tiny budding particles. The arrows in (b and c) indicate wormlike (red arrows) and cauliflower-like agglomerates (green arrows). (f) HRTEM image of the intermediate at 3 s, showing a small particle protruding from the AuNR along the [111] direction. Insets I and II are FFT diffractograms of region I and II marked in (f). The diffuse reflections in (II) indicate the presence of stacking faults. (g) HRTEM image of the intermediate at 20 s. The inset is an enlarged view of the interface, in which two arrows point to nanotwin boundaries. The relatively slow kinetics of the formation of the dimers allows us to track the evolution of the AuNP on the AuNR by quenching the reaction to get intermediates. Figure 3 shows TEM

ACS Paragon Plus Environment

9

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 20

images of the products at different reaction stages (3, 10, 20, 60, and 180 s). In the 3 s sample, the initial emergence of AuNP on the AuNR is observed. As shown in Figure 3a, most AuNRs have a small Au particle (~3 nm) protruding from the neck region. The HRTEM image clearly shows that the small particle is grown from the NR along the [111] direction (Figure 3f). Different from the defect-free body of the AuNR, the budding AuNP contains stacking faults, as indicated in the FFTs (Insets of Figure 3f). In the next stage (3 to 20 s), the tiny Au buds gradually grow into larger irregular agglomerates of small grains with two typical overall morphologies: cauliflower-like and worm-like (Figure 3b, c). HRTEM images indicate that regardless of the morphology, these agglomerates are full of randomly grown nanotwins (large stacking faults) (Figure 3g, S7). As in the case of the final dimer structure, twin boundaries are clearly observed at the interface between the AuNR and the growing agglomerate (inset of Figure 3g). Interestingly, the obvious difference in morphology vanishes as the reaction time is prolonged. In the 60 s sample, most AuNRs have a nearly round AuNP (Figure 3d), which resemble the NPs in the final dimers (i.e., the 180 s sample, Figure 3e) but have rougher surfaces and smaller sizes. It has been documented that strong ligands induce remarkable strains on metal surfaces,29 which in turn modulate the stacking-fault energy (SFE) of the metal.30, 31 Such surface strains vary in type and magnitude depending on the exposed facets and the local surface curvature. Different surface strains influence the SFE differently. When a strain lowers the SFE, the formation of stacking faults or nanotwins in the crystal is promoted. 30, 31 In the present synthetic system, anisotropic AuNRs were incubated with 4-MP (a strong thiol-based ligand) before they were used for growing the dimer structure. The 4-MP ligand would give rise to a non-uniform surface strain field on the AuNR, which helps to differentiate different crystal facets and

ACS Paragon Plus Environment

10

Page 11 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

“magnify” the tiny structural difference between symmetry-related sites (the strain is highly sensitive to local surface curvature). This might account for the observed preferential growth of a single AuNP on the AuNR in the dimer. Meanwhile, the 4-MP-induced surface strain reduces the SFE, facilitating the growth of AuNP in the form of successive nanotwins. On the basis of these observations and analysis, we propose a mechanism that describes the growth of AuNP in the dimer structure as follows. The chemisorption of 4-MP ligands induces inhomogeneous strains on the surface of AuNR. Among different surface facets of the AuNR, the (111) bridging facets at the rod neck are the most strained and are thus preferred locations for Au deposition to form a twinning structure via stacking faults that can largely relieve the surface strain.32 Once the first twin structure is formed on a (111) facet, this facet is differentiated from other symmetry-equivalent facets, becoming an “active” site for the further growth of Au through successive twinning, as driven by the same strain-relieving mechanism (Figure 3a). New twins can be formed randomly in different directions, resulting in various agglomerates of small grains. Worm-like structures are formed when most new grains grow laterally along the rod, while radiative twinning leads to cauliflower-like structures (Figure 3b, c). At the final stage, the agglomerates undergo a recrystallization process with small grains fused into large ones, accompanied by the disappearance of grain boundaries. As a consequence, differently-shaped agglomerates gradually develop into single AuNPs with a similar shape. If the reaction is prolonged, the NPs have fewer grains and smoother surfaces (Figure 3d, e). The key hypothesis of our proposed mechanism is that 4-MP ligands induce significant and inhomogeneous surface strain on the AuNR. In order to verify this hypothesis, we used HRTEM to characterize the 4-MP-incubated AuNR, and then visualized its strain distributions using geometric phase analysis (GPA).33 An atomic-resolution HRTEM image was taken along the

ACS Paragon Plus Environment

11

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 20

[110] axis, as the displacement of the densely packed (111) layers could be identified in this direction

Figure 4. (a) Atomic-resolution HRTEM image of the 4-MP-incubated AuNR taken along the [110] axis. (b) Corresponding strain distributions of the shear component (εxy, the magnitude cutoff is ± 8%) determined by geometric phase analysis. (c) HRTEM image of a AuNR taken at the earliest stage of the seeded growth along the [110] axis, in which the formation of the first stacking fault at one bridging facet is captured, as indicated by the arrow. (d) Enlarged image of the highlighted region in (c) with the stacking manner of (111) planes specified. (Figure 4a). Interestingly, the GPA analysis showed that the 4-MP-incubated AuNR exhibited large shear deformation on the rod surface that was particularly localized on the (111) bridging facets in the neck regions (Figure 4b). Aligning the y-axis of the GPA map with the growth direction (i.e., [001]) of the AuNR, the shear strain fields (εxy) in left and right bridging facets have opposite signs, indicating the same type of shear deformation caused by gliding of the {111} planes (Figure 4b). The shear strain was determined to be from the {111} slip

ACS Paragon Plus Environment

12

Page 13 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(Figure S8).31 It is interesting to note that the right bridging facet has markedly greater strain than does its left counterpart (10% vs. 5%). This means that the intrinsic symmetry of the AuNR can be broken upon the exertion of uneven ligand-induced surface strains. Consequently, symmetry-equivalent surface facets are distinguished by their different SFEs, and the nucleation of stacking faults would preferentially take place at the one with the lowest SFE. This explains the fact that AuNP grows from only one of eight symmetry-related {111} bridging facets on the AuNR. We examined a few more randomly selected Au NRs, and the results were consistent (Figure S9). We performed the same HRTEM-based GPA analysis for the as-synthesized AuNR (with weaker ligand CTAB capped). The results showed only little strain fluctuation over the entire crystal (Figure S10), confirming the essential role of 4-MP in generating surface strains on the AuNR. We fortunately captured the occurrence of the first stacking fault at one bridging facet of a AuNR, which is likely at the earliest stage of seeded growth (Figure 4c, d). This observation strongly supports our hypothesis that the nucleation of the AuNP starts from a stacking fault formed on the most strained bridging facet. The appearance of the stacking fault and of the subsequent twinning structures at only one bridging facet breaks the symmetry of the AuNR. The mis-stacking of atoms produces re-entrant edges, accelerating the crystal growth along the boundary.34 Meanwhile, the vicinity of the twin boundary has lower SFEs, where the nucleation of new stacking faults is greatly facilitated.35 In this sense, the growth of AuNP from the first stacking fault through continuous twinning is a self-accelerating and highly favorable process, which largely inhibits the nucleation/growth of new AuNPs on other sites of the AuNR. With the gradual consumption of the HAuCl4, the growth of AuNP reaches a plateau, while a slower “recrystallization” process becomes dominant. This is because irregular agglomerates of

ACS Paragon Plus Environment

13

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 20

small grains are thermodynamically unstable and they tend to ripen into larger crystals with fewer grain boundaries and smaller surface-to-volume ratios to decrease the surface energy. Eventually, single large AuNPs (~ 30 nm) are evolved in the neck regions. These single particles have multiple-twinned structures that appear to be close to decahedra or icosahedra. During TEM characterization of the 20 s sample, we indeed observed the evolution of irregular agglomerates into a single particle that was induced by prolonged electron beam irradiation (Movie S1). Based on this observation and the well-recognized high mobility of metal atoms in nano-sized crystals,36 it is plausible to speculate that the recrystallization takes place through the migration of Au atoms in the solid phase, without dissolution and re-deposition processes. We note that unlike the initial stages of the NP growth when the growing NPs are almost exclusively attached to the rod necks (Figure 3a-c), a fraction of dimers at the final stage (after recrystallization) have AuNPs located at the rod body (Figure 1d, 3e). This observation suggests that the recrystallization may occasionally cause the final AuNP deviating from the original nucleation sites. We speculate that the NPs located at the necks and bodies evolve from the cauliflower-like and worm-like agglomerates, respectively. An alternative possibility is that there exists a small quantity of flawed AuNRs in the seed solution, which favor the growth of AuNP from the defects in the rod body, while we did not observe this process due to the limited sampling ability of TEM. A schematic illustration of the growth pathway of the AuNR-AuNP dimer structure is shown in Figure 5. The above-discussed mechanism is mainly based on thermodynamics. The question about the influence of the concentration of 4-MP, which determines not only the surface strains on AuNR seeds but also the reduction rate of the Au precursor, on the secondary growth of Au has yet to be fully addressed (refer to Figure 1). To achieve a thermodynamically controlled process

ACS Paragon Plus Environment

14

Page 15 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

requires sufficiently slow reaction kinetics. As discussed earlier, a higher 4-MP concentration leads to a slower reduction rate in this system. In the absence of 4-MP, the AuNR has trivial surface strain and epitaxial growth is therefore highly favored even with a fast reduction rate (Figure 1b). At the other extreme, an optimally high concentration of 4-MP induces strong inhomogeneous strains on the AuNR surface to distinguish a single bridging facet for nucleation and, at the same time, results in a slow reduction rate to ensure thermodynamically controlled growth of AuNP at the sole nucleation site (Figure 1d). With a medium concentration of 4-MP, despite the existence of local surface strains, the relatively fast reduction rate leads to indiscriminate crystal growth on the AuNR in both twinning and epitaxial manners. Consequently, irregular NRs with different secondary growth mechanisms are obtained (Figure 1c).

Figure 5. Schematic illustration of the growth pathway of a AuNR-AuNP dimer. (a) AuNR seed; (b) Two orthogonal views of initial nucleation of a stacking fault (in silver color) on one (111) bridging facet; (c) Subsequent growth of the Au agglomerate of small grains by random twinning; and (d) The final formation of a multi-twinned particle (an icosahedron was taken as an example) at the neck (left) and body (right) of the AuNR after recrystallization. In addition, we notice an interesting phenomenon in the current synthetic system, which is a strong hint that the growth mechanism discussed here is extendable to the fabrication of dimers from nanocrystals of other shapes. Specifically, we found that in the AuNR seed solution, there

ACS Paragon Plus Environment

15

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 20

existed Au nanocubes and other-shaped Au nanocrystals (octahedron, quasi-sphere, and irregular particle) as impurities, and that these non-rod seeds could form dimers as well at the end of the synthesis (Figure S11). This phenomenon also confirms that the asymmetric growth of Au nanocrystals in our system is not driven by the site-blocking effect described in a recent study, where S2O32− ions exhibited influences on the morphology of seed-mediated Au nanocrystals but could not induce asymmetric growth.37 In conclusion, we investigated asymmetric monometallic AuNR-AuNP dimer nanocrystals that were fabricated through a selective growth of single AuNPs on single-crystalline AuNRs, and identified the growth pathway with TEM. We unveiled crucial roles the thiol-ligand 4-MP plays in directing the formation of the dimer. Thermodynamically, it induces significant and inhomogeneous surface strain on the AuNR to initiate the growth of AuNP from stacking faults; kinetically, it modulates the reduction rate of the Au precursor to prevent uncontrolled deposition. The growth mechanism revealed here represents the first example of identification with direct and solid evidences of thiol’s role in directing asymmetric growth, and provides a promising guideline for developing asymmetric nanocrystals with novel properties for nanoplasmonics and nanophotonics.38

ASSOCIATED CONTENT Supporting Information. Experimental details (Chemicals and materials, synthetic procedures, characterization equipment) and Figure S1-S11. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author

ACS Paragon Plus Environment

16

Page 17 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

*E-mail: [email protected], [email protected] Author Contributions #These authors contributed equally. Funding Sources This research was supported by baseline research funds to Y.H. from King Abdullah University of Science and Technology. Notes The authors declare no competing financial interest. REFERENCES 1.

Huang, X.; Li, S.; Huang, Y.; Wu, S.; Zhou, X.; Li, S.; Gan, C. L.; Boey, F.; Mirkin, C. A.; Zhang, H. Nat. Commun. 2011, 2, 292.

2.

Huang, F.; Baumberg, J. J. Nano Lett. 2010, 10, 1787-1792.

3.

Lombardi, A.; Grzelczak, M. P.; Crut, A.; Maioli, P.; Pastoriza-Santos, I.; Liz-Marzan, L. M.; Del Fatti, N.; Vallee, F. ACS nano 2013, 7, 2522-2531.

4.

Cheng, Y.; Wang, M.; Borghs, G.; Chen, H. Langmuir 2011, 27, 7884-7891.

5.

Bidault, S.; Abajo, F. J.; Polman, A. J. Am. Chem. Soc. 2008, 130, 2750-2751.

6.

Li, W.; Camargo, P. H.; Au, L.; Zhang, Q.; Rycenga, M.; Xia, Y. Angew. Chem. Int. Ed. 2010, 49, 164-168.

7.

Kleinman, S. L.; Sharma, B.; Blaber, M. G.; Henry, A. I.; Valley, N.; Freeman, R. G.; Natan, M. J.; Schatz, G. C.; Van Duyne, R. P. J. Am. Chem. Soc. 2013, 135, 301-308.

ACS Paragon Plus Environment

17

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

8.

Page 18 of 20

Chen, G.; Wang, Y.; Yang, M.; Xu, J.; Goh, S. J.; Pan, M.; Chen, H. J. Am. Chem. Soc. 2010, 132, 3644-3645.

9.

Gao, C.; Goebl, J.; Yin, Y. J. Mater. Chem. C 2013, 1, 3898-3909.

10. Carbone, L.; Cozzoli, P. D. Nano Today 2010, 5, 449-493. 11. He, J.; Liu, Y.; Hood, T. C.; Zhang, P.; Gong, J.; Nie, Z. Nanoscale 2013, 5, 5151-5166. 12. Feng, Y.; He, J.; Wang, H.; Tay, Y. Y.; Sun, H.; Zhu, L.; Chen, H. J. Am. Chem. Soc. 2012, 134, 2004-2007. 13. Lee, S. U.; Hong, J. W.; Choi, S. I.; Han, S. W. J. Am. Chem. Soc. 2014, 136, 5221-5224. 14. Zeng, J.; Zhu, C.; Tao, J.; Jin, M.; Zhang, H.; Li, Z. Y.; Zhu, Y.; Xia, Y. Angew. Chem. Int. Ed. 2012, 51, 2354-2358. 15. Lee, J. H.; Kim, G. H.; Nam, J. M. J. Am. Chem. Soc. 2012, 134, 5456-5459. 16. Huang, J.; Liu, C.; Zhu, Y.; Masala, S.; Alarousu, E.; Han, Y.; Fratalocchi, A. Nat. Nanotechnol. 2015, DOI: 10.1038/nnano.2015.228. 17. Lim, D. K.; Jeon, K. S.; Hwang, J. H.; Kim, H.; Kwon, S.; Suh, Y. D.; Nam, J. M. Nat. Nanotechnol. 2011, 6, 452-460. 18. Peng, Z. M.; Yang, H. Nano Today 2009, 4, 143-164. 19. Huang, J.; Zhu, Y.; Lin, M.; Wang, Q.; Zhao, L.; Yang, Y.; Yao, K. X.; Han, Y. J. Am. Chem. Soc. 2013, 135, 8552-8561. 20. Huang, J.; Zhu, Y.; Liu, C.; Zhao, Y.; Liu, Z.; Hedhili, M. N.; Fratalocchi, A.; Han, Y. Small 2015, 11, 5214-5221. 21. Busbee, B. D.; Obare, S. O.; Murphy C. J. Adv. Mater. 2003, 15, 414-416. 22. Chen, H. J.; Shao, L.; Li, Q.; Wang, J. F. Chem. Soc. Rev. 2013, 42, 2679-2724.

ACS Paragon Plus Environment

18

Page 19 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

23. Sohn, K.; Kim, F.; Pradel, K. C.; Wu, J. S.; Peng, Y.; Zhou F. M.; Huang, J. X. ACS Nano 2009, 3, 2191-2198. 24. Sardar, R.; Heap, T. B.; Shumaker-Parry, J. S. J. Am. Chem. Soc. 2007, 129, 5356-5357. 25. Busson, M. P.; Rolly, B.; Stout, B.; Bonod, N.; Larquet, E.; Polman, A.; Bidault, S. Nano Lett. 2011, 11, 5060-5065. 26. Hu, Y.; Sun, Y. J. Am. Chem. Soc. 2013, 135, 2213-2221. 27. Yuan, X.; Zhang, B.; Luo, Z.; Yao, Q.; Leong, D. T.; Yan, N.; Xie, J. Angew. Chem. Int. Ed. 2014, 53, 4623-4627. 28. Gole, A.; Murphy, C. J. Chem. Mater. 2004, 16, 3633-3640. 29. Watari, M.; McKendry, R. A.; Vogtli, M.; Aeppli, G.; Soh, Y. A.; Shi, X.; Xiong, G.; Huang, X.; Harder, R.; Robinson, I. K. Nat. Mater. 2011, 10, 862-866. 30. Branicio, P. S.; Zhang, J. Y.; Srolovitz, D. J. Phys. Rev. B 2013, 88, 064104. 31. Jahnátek, M.; Hafner, J.; Krajčí, M. Phys. Rev. B 2009, 79, 224103. 32. Paine, D. C.; Howard, D. J.; Stoffel, N. G. J. Electron. Mater. 1991, 20, 735-746. 33. Hÿtch, M. J.; Snoeck, E.; Kilaas, R. Ultramicroscopy 1998, 74, 131-146. 34. Kobayashi, K.; Hogan, L. M. Philos. Mag. A 1979, 40, 399-407. 35. Gleiter, H.; Klein, H. P. Philos. Mag. 1973, 27, 1009-1026. 36. Liu, J.; Liu, W.; Sun, Q.; Wang, S.; Sun, K.; Schwank, J.; Wang, R. Chem. Commun. 2014, 50, 1804-1807. 37. Zheng, Y.; Luo, M.; Tao, J.; Peng, H. C.; Wan, D.; Zhu, Y.; Xia, Y. J. Phys. Chem. B 2014, 118, 14132-14139. 38. Sun, Y. Natl. Sci. Rev. 2015, 2, 329-348.

ACS Paragon Plus Environment

19

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 20

Table of Contents

ACS Paragon Plus Environment

20