General Synthetic Approach to Heterostructured Nanocrystals

Solid metal precursors (alloys or monometals) can serve both as a starting template and as a source material for chemical transformation to metal chal...
22 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

General Synthetic Approach to Heterostructured Nanocrystals Based on Noble Metals and I−VI, II−VI, and I−III−VI Metal Chalcogenides Minghui Liu† and Hua Chun Zeng*,†,‡ †

Department of Chemical and Biomolecular Engineering, Faculty of Engineering, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260 ‡ Institute of Materials Research and Engineering (IMRE), 3 Research Link, Singapore 117602 S Supporting Information *

ABSTRACT: Solid metal precursors (alloys or monometals) can serve both as a starting template and as a source material for chemical transformation to metal chalcogenides. Herein, we develop a simple solution-based strategy to obtain highly monodisperse noble-metal-based heterostructured nanocrystals from such precursor seeds. By utilizing chemical and structural inhomogeneity of these metal seeds, in this work, we have synthesized a total of five I−VI (Ag2S, Ag2Se, Ag3AuS2, Ag3AuSe2, and Cu9S5), three II−VI (CdS, CdSe, and CuSe), and four I−III−VI (AgInS2, AgInSe2, CuInS2, and CuInSe2) chalcogenides, together with their fifteen associated heterodimers (Au− Ag2S, Au−Ag2Se, Au−Ag3AuS2, Au−Ag3AuSe2, Au−AgInS2, Au−AgInSe2, Au−CdS, Au−CdSe, Ag−Ag2S, Ag−AgInS2, Au−Cu9S5, Au−CuInS2, Au− CuSe, Au−CuInSe2, and Pt−AgInS2) to affirm the process generality. Briefly, by adding elemental sulfur or selenium to AuAg alloy seeds and tuning the reaction conditions, we can readily obtain phase-pure Au−Ag2S, Au−Ag2Se, Au−Ag3AuS2, and Au−Ag3AuSe2 heterostructures. Similarly, we can also fabricate Au−AgInS2 and Au−AgInSe2 heterostructures from the AuAg seeds by adding sulfur/selenium and indium precursors. Furthermore, by partial or full conversion of Ag seeds, we can prepare both single-phase Ag chalcogenide nanocrystals and Ag-based heterostructures. To demonstrate wide applicability of this strategy, we have also synthesized Aubased binary and ternary Cu chalcogenide (Au−Cu9S5, Au−CuSe, Au−CuInS2, and Au−CuInSe2) heterostructures from alloy seeds of AuCu and Pt chalcogenides (e.g., Pt−AgInS2) from alloy seeds of PtAg. The structure and composition of the above products have been confirmed with X-ray diffraction, high-resolution transmission electron microscopy, X-ray photoelectron spectroscopy, and energy-dispersive X-ray spectroscopy methods. A kinetic investigation of the formation mechanism of these heterostructures is brought forward using Au−AgInS2 and Ag−CuInS2 as model examples.

■

INTRODUCTION Heterostructured nanocrystals (HNCs) composed of at least two different inorganic materials in the same particle have become a research focus in recent years due to not only their multifunctionalities from individual components, but also their synergetic effects arising from effective coupling of the separate phases.1−14 In particular, colloidal hybrid nanoparticles with multifunctional properties have found wide applications in biological imaging, cancer therapy, solar energy conversion, heterogeneous catalysis, electronics, and magnetism. 1−14 Among the various types of HNCs, those composed of semiconductors and noble metals are of special interest for nanotechnology. For example, the noble metals can provide an anchor point for electrical and chemical connections to the semiconductors.15,16 The rationale of this type of material combination lies in their complementary optical properties such as long-lived excitations in semiconductors and localized electromagnetic modes in metal counterparts.17 The resultant heterostructures may acquire enhanced and/or new physicochemical behaviors owing to the overlapping of the electronic bands, which is not possible for any of their © 2014 American Chemical Society

components alone or for related macroscopic physical mixtures.18,19 In making such HNCs, metal chalcogenides are an important class of semiconducting materials that have undergone extensive investigation. For example, it has been reported that in Au−CdSe heterodimers the photoluminescence of CdSe nanorods was quenched due to charge transfer from the CdSe nanorods to the Au tips.19 The gold surface plasmon resonance (SPR) and the CdS exciton peaks were suppressed because of plasmon−exciton coupling.20 The reduction of the second harmonic generation (SHG) signal can be observed when the interfacial region of Au−CdSe is dominant over the whole hybrid nanoparticle. 1,21 The construction of HNCs requires the combination of two or more different materials into one system and more commonly one particle to nucleate directly on the surface of another. The phase junction between the two different material phases can be attained through epitaxial or nonepitaxial growth. In the Received: May 16, 2014 Revised: July 20, 2014 Published: July 29, 2014 9838

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

construction of such artificial crystals, a presynthesized seed (which also acts as the first material phase in the final HNCs) is normally used as a substrate, while the second material is chemically deposited onto the substrate, which remains unchanged during the entire process. In this connection, two typical structures result from such seeded growth processes: (i) centrosymmetric core/shell structures via the formation of a conformal, complete shell of the second material on the surface of the seed when the lattice mismatch of the two materials is small (e.g., within 5%)7,22−24 and (ii) heterodimer structures formed by deposition of the second material on specific site(s) of the seed template when larger lattice mismatches occur.25 This step-by-step addition in the solution phase is an extended version of wet-chemical growth with traditional molecular beam epitaxy (MBE) or chemical vapor deposition (CVD) techniques and has been commonly adopted as a main synthetic strategy to obtain HNCs with epitaxial or nonepitaxial techniques.16,22,23,26−30 Although various methods for the preparation of metal−semiconductor HNCs have been reported in recent years, the products are usually limited to only a few (one or two) types of HNCs due to the limitation of specific step-by-step addition techniques in individual syntheses.25,28,31−40 There is no general synthetic scheme which is able to produce a wide variety of HNCs.41−44 To overcome this limitation, multiple steps and fine-tuning of experimental parameters, together with different surfactants and solvents, to nucleate different materials onto the seeds are usually required in the fabrication processes.27,28,45 Recently, a more sophisticated strategy was developed to expand the family of HNCs containing both Ag and Au.27,28 Briefly, an amorphous Ag shell was deposited onto a presynthesized Au core and later was converted to the Ag2S phase; the resultant Ag2S shell could be further transformed to the desired material phases through anionic exchange.27,46 Despite the new shell phases formed, the pristine Au core remained intact, acting simply as a seed template. However, if it is not chemically inert, a templating seed (i.e., a metal, alloy, or solid-compound particle) may also serve as a starting reactant for the formation of the final products. In such cases, chemically more reactive elements (or phases) would be more diffusive than others when they are involved in surface reactions. As a result, the original seed is no longer intact. Instead, it can be thought of as a “dynamic seed” in view of its chemical reactivity and structural changes. From a wider point of view, the concept in this dynamic seeding strategy may not be entirely new, since the well-known cation exchange also makes use of the diffusion of reactant atoms and ions to create novel structures and materials, and partial cation exchange is also an important way to achieve HNCs.27,28,46−49 Nevertheless, the latter approach could be quite challenging in practical synthesis, because it requires careful calculation of reactant stoichiometry and fine adjustment of the experimental parameters to control the extent of reaction and to prevent total conversion of primary nanocrystals.47 Recently, we used AuI,AgI-dodecanethiolates as reactants and achieved large-scale preparation of pure Ag nanoparticles or AuAg alloy nanoparticles with composition and size control.50 Herein, we use these solid materials as a starting platform to exploit the processes of such dynamic seeds in the general synthesis of metal chalcogenides and their related HNCs. The overall synthetic routes are depicted in Scheme 1. When a metal-containing solid precursor (e.g., metal dodecanethiolate, alloy, or monometal) is exposed to an elemental species possessing a high electronegativity (e.g., group VI), more active

Scheme 1. Synthetic Strategy of Au−X Heterodimers Starting with Metal(I) Dodecanethiolate (MI-DDT)a

a

X is the metal chalcogenide phase. MI = Au+ or Ag+. See the Experimental Section for the seeds of AuAg-t1 (t1 = 3−5 min) and AuAg-t2 (t2 = 30−60 min). The Au phase is depicted as small yellow spheres.

atoms will segregate out to react with this element. A new solid phase that contains the out-diffusing metal will be formed, while the less active atoms will stay behind and rearrange themselves into a new metallic phase, resulting in final heterodimer structures. It is noteworthy that the synthetic scheme proposed in this work is facile, since it requires only alkylamines (e.g., oleylamine) in its synthetic environment. The adopted alkylamine acts both as a surfactant and as a solvent in the reactions. Because no toxic chemicals or high-cost surfactants are required, our synthetic scheme greatly simplifies complex chemical processes involved in other established techniques for the fabrication of HNCs.

■

EXPERIMENTAL SECTION

The following descriptions are only focused on the synthetic procedures for making various HNCs. The general processes are illustrated in Scheme 1. Detailed information on the chemicals used, treatments, and characterization of the products can be found in the Supporting Information (section SI-1). AuAg Seeds. The AuAg alloy seeds were synthesized at room temperature using supramolecular solids MI-dodecanethiolate (where MI = Au+ or Ag+, named Au-DDT and Ag-DDT, respectively).50 For the synthesis of Au-DDT, 2.0 mL of DDT in ethanol (0.010 M) was added to 1.0 mL of ethanol, followed by addition of 0.5 mL of HAuCl4·3H2O in ethanol (0.100 M), and the mixture was stirred for 5 min. The resultant yellow-brown precipitate was washed with ethanol and then redispersed into 0.5 mL of ethanol. For the synthesis of AgDDT, 4.0 mL of DDT in ethanol (0.100 M) was added to 8.0 mL of AgNO3 solution in ethanol (0.010 M), and the mixture was stirred for 5 min. The resultant white precipitate was washed with ethanol and then redispersed into 1.5−2.0 mL of ethanol (0.080 mmol of Ag). The Ag-DDT suspension was divided into two or four portions (0.040 or 0.020 mmol of Ag) for future use. Similarly, the above Au-DDT in ethanol suspension (0.050 mmol of Au) was divided into two or four portions; one portion (0.025 or 0.0125 mmol of Au) was then added to 5.0−6.0 mL of oleylamine (OLA), and the resultant Au-DDT suspension was heated to 200 °C in a silicon oil bath. Immediately upon detection of a color change, usually about 3−5 min after heating, the Ag-DDT suspension (0.020, 0.040, or 0.080 mmol) was added to the heated Au-DDT, and the resultant system was further refluxed for a desired period of time (ranging from 4 min to 1 h) under constant stirring. The final reaction products (which served as seed precursors, 9839

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

Au−AgInS2. Reflux Method. In a typical synthesis, 0.4 mL of InOLA suspension (0.040 mmol of InCl3) and 0.6 mL of S-OLA suspension (0.120 mmol of S) were added to the heated AgAu-4 seed suspension (0.040 mmol of Ag). The mixed suspension was refluxed at 200 °C for 30 min. Solvothermal Method. After the AuAg-40 suspension (0.040 mmol of Ag) was transferred to a Teflon beaker and cooled to room temperature, 0.4 mL of In-OLA suspension (0.040 mmol of InCl3) and 0.50 mL of S-OLA suspension (0.100 mmol of S) were added, and the mixture was stirred for 30 min at room temperature. Finally, the mixture in the Teflon beaker was sealed in an autoclave and put into an electric oven at 200 °C for 1 h. Synthesis from Au−Ag2S (Two Steps). In the first step of synthesis, 0.1 mL of S-OLA suspension was added to a AuAg-4 suspension (0.080 mmol of Ag). The mixed suspension was refluxed at 200 °C for 30 min. In the second step of synthesis, 0.2 mL of In-OLA suspension (0.020 mmol of InCl3) was further added, and the mixed suspension was refluxed at 200 °C for 30 min. Au−AgInSe2. Reflux Method. For a typical synthesis, 0.4 mL of InOLA suspension (0.040 mmol of InCl3) and 1.2 mL of Se-OLA suspension (0.120 mmol of Se) were added to the above heated AuAg30 precursor suspension (0.040 mmol of Ag). The mixture was refluxed at 200 °C for another 30 min. Solvothermal Method. For a typical synthesis, after the AuAg-30 suspension was transferred to a Teflon beaker and cooled to room temperature, 0.4 mL of In-OLA suspension (0.040 mmol of InCl3) and 0.8 mL of Se-OLA suspension (0.080 mmol of Se) were added, and the mixture was stirred for overnight at room temperature. The mixture in the Teflon beaker was sealed in an autoclave and put into an electric oven at 200 °C for 1 h. Au−CdS. In a typical synthesis, the Au−Ag2S heterodimers obtained via the above methods were dispersed into 5−10 mL of toluene and mixed with 2.5 mL of 0.10 mol/L Cd(NO3)2 in methanol to form a uniform suspension. After the suspension was refluxed at 75 °C for 10 min, 0.5 mL of tri-tert-butylphosphine (TBP) was also added. The mixture was stirred for another 2 h at 75 °C for cation exchange between Ag+ and Cd2+. Au−CdSe. In a typical synthesis, the as-synthesized Au−Ag2Se heterodimers were dispersed into 5 mL of toluene and mixed with 2.5 mL of 0.1 mol/L Cd(NO3)2 in methanol. After the suspension was refluxed at 75 °C for 10 min, 0.8 mL of TBP was added. The mixture was stirred for another 2 h at 75 °C for cation exchange between Ag+ and Cd2+. Ag Seeds. The required amount of Ag-DDT (0.080 mmol for binary compounds and 0.040 mmol for ternary compounds unless otherwise specified) was injected into 5−6 mL of OLA at 200 °C and refluxed for 30 min to 1 h. The final products (which served as seeds) are named Ag-30 and Ag-60. Ag−Ag2S. After the above Ag-60 seed (0.080 mmol of Ag) was transferred to a Teflon beaker and cooled to room temperature, 0.1 mL of S-OLA suspension (0.020 mmol of S, molar ratio Ag:S = 4:1) was added, and the mixture was stirred for 40 min at room temperature. The mixture in the Teflon beaker was sealed in an autoclave and put into an electric oven at 200 °C for 1 h. Ag2S. In a typical experiment, the above Ag-60 seed (0.080 mmol of Ag) was transferred to a Teflon beaker and cooled to room temperature. Then 0.4 mL of S-OLA suspension (0.080 mmol of S, molar ratio Ag:S = 1:1) was added, and the mixture was stirred for 40 min at room temperature. The final mixture in the Teflon beaker was sealed inside an autoclave and put into an electric oven at 200 °C for 1 h. Ag2Se. For a typical synthesis, 0.4 mL of Se-OLA suspension (0.040 mmol of Se) was added to the above Ag-30 seed (0.080 mmol of Ag). The suspension was refluxed at 200 °C for another 15 min. AgInS2. After the Ag-60 seed suspension (0.040 mmol of Ag) was transferred to a Teflon beaker and cooled to room temperature, 0.4 mL of In-OLA suspension (0.040 mmol of InCl3) and 0.4 mL of SOLA suspension (0.080 mmol of S) were added, and the mixture was stirred for 40 min at room temperature. Finally, the mixture in the

AuAg-t1 and AuAg-t2, in Scheme 1) are named AuAg-4, AuAg-30, AuAg-40, and AuAg-60 (where the number represents the reflux time (min); t1 = 3−5 min, and t2 = 30−60 min). Note that the chemical conversion of Au-DDT and Ag-DDT to AuAg alloy nanoparticles depended strongly on the reaction (reflux) time. In a general synthesis, the molar ratio of input Au-DDT and Ag-DDT was set at 1:3.2 (i.e., 0.0125 mmol of Au and 0.040 mmol of Ag) and 1:6.4 (i.e., 0.0125 mmol of Au and 0.080 mmol of Ag) in the prepared seeds, unless otherwise specified. Other alloy seeds with different Au:Ag atomic ratios were also prepared in a similar way. Stock Suspensions. Suspensions of S-OLA, Se-OLA, and In-OLA were prepared by dissolving a certain amount of S, Se, and InCl3 in an appropriate volume of OLA to obtain concentrations of 0.20, 0.10, and 0.10 mol/L, respectively. A stock suspension of Cu-OLA was also prepared by dissolving a weighed amount of copper(II) acetate monohydrate (CuAc) in an appropriate volume of OLA to form a uniform suspension at a concentration of 0.10 mol/L. Au−Ag2S. Reflux Method (One Step). For a typical synthesis, 0.05 mL of S-OLA suspension (0.010 mmol of S, i.e., half the amount of sulfur required to completely convert Ag atoms in the system to Ag2S; otherwise Au−Ag3AuS2 would also be formed if a stoichiometric amount of sulfur (0.02 mmol) was added at once to a AuAg-4 suspension (0.040 mmol of Ag)). This mixed suspension was refluxed at 200 °C for 30 min. In this case, only half of the Ag atoms in the suspension were converted to Ag2S; i.e., the final products coexisted with unreacted AuAg alloy nanoparticles. Reflux Method (Two Steps). To avoid the simultaneous formation of Au−Ag3AuS2, the stoichiometric amount of sulfur was added in two steps: (1) A 0.05 mL volume of S-OLA suspension was added to a AuAg-4 suspension (0.040 mmol of Ag). The mixed suspension was refluxed at 200 °C for 30 min. (2) After that, another 0.05 mL of SOLA suspension was added, and the suspension was refluxed at 200 °C for another 30 min. In this way, all Ag was converted to Ag2S. Solvothermal Method. Only the AuAg-t2 seeds were used to obtain Au−Ag2S with this method. For example, after the AuAg-30 seed (0.080 mmol of Ag) was transferred to a Teflon beaker and cooled to room temperature, 0.2 mL of S-OLA suspension (S 0.040 mmol) was added, and the mixture was stirred for another 30 min at room temperature. The mixture was sealed in an autoclave and put into an electric oven at 200 °C for 0.5 h. Au−Ag2Se. Phase-pure Au−Ag2Se could only be obtained by the reflux method at the present stage of study. For a typical synthesis, 0.4 mL of Se-OLA suspension (0.040 mmol of Se) was added to the above heated AuAg-30 seed suspension (0.080 mmol of Ag), and the mixed suspension was refluxed at 200 °C for 30 min. Au−Ag3AuS2. Reflux Method. The amount of sulfur should be strictly controlled with this method; otherwise Au−Ag2S could be formed. In a typical synthesis, 0.075 mL of S-OLA suspension (0.015 mmol of S) was added to the AuAg-4 seed suspension (0.040 mmol of Ag). The mixed suspension was refluxed at 200 °C for 30 min. Solvothermal Method. In a typical synthesis, the AuAg-4 seed suspension (0.025 mmol of Au and 0.020 mmol of Ag) was transferred to a Teflon beaker and cooled to room temperature, and 0.2 mL of SOLA suspension (0.040 mmol of S) was added. The mixed suspension was stirred for another 30 min at room temperature. Finally, the mixture in the Teflon beaker was sealed in an autoclave and put into an electric oven at a temperature between 120 and 200 °C for 0.5 h. Au−Ag3AuSe2. Reflux Method. The amount of selenium should be strictly controlled for the reflux method; otherwise Au−Ag2Se could be formed. In a typical synthesis, 0.1 mL of Se-OLA suspension (0.010 mmol of Se) was added to the heated AuAg-4 seed precursor (0.025 mmol of Au and 0.020 mmol of Ag). The mixed suspension was refluxed at 200 °C for 30 min. Solvothermal Method. After the AuAg-4 seed suspension (0.025 mmol of Au and 0.020 mmol of Ag) was transferred to a Teflon beaker and cooled to room temperature, 0.25 mL of Se-OLA suspension (0.025 mmol of Se) was added, and the mixture was refluxed for 60 min at room temperature. Finally, the mixture in the Teflon beaker was sealed in an autoclave and put into an electric oven set at 140 or 200 °C for 30 min to 1 h. 9840

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

Teflon beaker was sealed in an autoclave and put into an electric oven at 200 °C for 1 h. AgInSe2. In a typical synthesis, 0.4 mL of In-OLA suspension (0.040 mmol of InCl3) and 0.8 mL of Se-OLA suspension (0.080 mmol of Se) were added to the above Ag-30 seed suspension (0.040 mmol of Ag). The mixture was refluxed at 200 °C for 30 min. AuCu Seeds. In this synthesis, 0.4 mL of Cu-OLA suspension (0.040 mmol of CuAc) was injected into 5−6 mL of OLA at 200 °C containing 0.020 mmol of Au-DDT, and the mixture was refluxed for 30 min at the same temperature. Au−Cu9S5. For a typical synthesis, 0.1 mL of S-OLA suspension (0.020 mmol of S) was added to the above AuCu seed suspension. The suspension was refluxed at 200 °C for another 30 min. Au−CuSe. In this synthesis, 0.4 mL of Se-OLA suspension (0.040 mmol of Se) was added to the above AuCu seed suspension. The suspension was refluxed at 200 °C for another 30 min. Au−CuInS2. In this synthesis, 0.4 mL of In-OLA suspension (0.040 mmol of InCl3) and 0.6 mL of S-OLA suspension (0.120 mmol of S) were mixed into the above AuCu seed suspension. The mixture was refluxed at 200 °C for another 30 min. Au−CuInSe2. Similarly, 0.4 mL of In-OLA suspension (0.040 mmol of InCl3) and 1.0 mL of Se-OLA suspension (0.100 mmol of Se) were added to the above AuCu seed suspension. The suspension was refluxed at 200 °C for another 30 min. Cu9S5. In this control experiment, 0.4 mL of Cu-OLA suspension (0.040 mmol of CuAc) was injected into 5−6 mL of OLA at 200 °C and refluxed for 30 min. Then 0.1 mL of S-OLA suspension (0.040 mmol of S) was added to this mixture, which was refluxed at 200 °C for another 30 min. CuSe. In this control experiment, 0.4 mL of Cu-OLA suspension (0.040 mmol of CuAc) was injected into 5−6 mL of OLA at 200 °C and refluxed for 30 min. Then 0.4 mL of Se-OLA suspension (0.040 mmol of Se) was added to this mixture, which was refluxed at 200 °C for another 30 min. CuInS2. In this control test, 4 mL of Cu-OLA suspension (0.040 mmol of CuAc) was injected into 5−6 mL of OLA at 200 °C and refluxed for 30 min, after which 0.4 mL of In-OLA suspension (0.040 mmol of InCl3) and 0.6 mL of S-OLA suspension (0.12 mmol of S) were added to the mixture. The mixture was refluxed at 200 °C for another 30 min. CuInSe2. In this control experiment, 0.4 mL of CuAc-OLA (0.040 mmol of CuAc) was injected into 5−6 mL of OLA at 200 °C and refluxed for 30 min. Then 0.4 mL of In-OLA suspension (0.040 mmol of InCl3) and 1.0 mL of Se-OLA suspension (0.100 mmol of Se) were added to the above CuAc-OLA suspension. The suspension was refluxed at 200 °C for another 30 min. Pt−AgInS2. Typically, a solution of 45 mg of AgNO3 (0.26 mmol of Ag) in 10 mL of OLA was heated at 160 °C under magnetic stirring for 1 h. Then 65 mg of Pt(acac)2 (0.16 mmol of Pt) was added to this Ag-OLA suspension, and the mixture was stirred at 240 °C for 1 h to form the PtAg seed. After that, equivalent amounts of InCl3 (0.26 mmol of InCl3) and sulfur powder (0.52 mmol of S) were added to the above heated PtAg seed suspension. The mixture was refluxed at 240 °C for another 2 h.

■

2). By tuning the elemental ratio of S to Ag or Se to Ag and process parameters (see the Experimental Section), Au−Ag2S or Au−Ag3AuS2 and Au−Ag2Se or Au−Ag3AuSe2 could be formed from different types of seeds with varying aging times (Scheme 1). Heterodimers composed of ternary I−III−VI compounds (i.e., Au−AgInS2 and Au−AgInSe2) can also be formed. Their synthesis is similar to that of the above four I−VI compounds except that, besides sulfur or selenium, an equivalent amount of indium precursor with respect to Ag is included. Furthermore, the resultant Au−Ag2S and Au−Ag2Se heterodimers can be transformed chemically to Au−CdS and Au−CdSe (i.e., their X phases now are II−VI chalcogenides) through simple cation exchange with the assistance of a common phase-transfer agent, TBP (section SI-2, Supporting Information).27,46 Due to the high reaction complexity involved in these syntheses, competing reactions which lead to different products must be recognized: (i) Au−Ag2S and Au−Ag3AuS2 with addition of sulfur, (ii) Au−Ag2Se and Au−Ag3AuSe2 upon addition of selenium, (iii) Au−Ag2S, Au−Ag3AuS2, and Au− AgInS2 with addition of both indium and sulfur, and (iv) Au− Ag2Se, Au−Ag3AuSe2, and Au−AgInSe2 with addition of both indium and selenium. Therefore, achieving phase purity for the products is the foremost issue of this synthetic scheme. In Figure 1, we first confirm the phase purity of our products by the XRD technique.

Figure 1. Powder XRD patterns of the Au−Ag2S, Au−Ag2Se, Au− Ag3AuS2, Au−Ag3AuSe2, Au−AgInS2, Au−AgInSe2, Au−CdS, and Au−CdSe samples (section SI-2, Supporting Information).

The XRD patterns reveal that all the metal chalcogenide phases are indeed phase pure, and they can be assigned respectively to monoclinic Ag2S (space group P21/c, lattice constants a0 = 4.229 Å, b0 = 6.931 Å, and c0 = 7.862 Å, JCPDS card no. 14-72), orthorhombic Ag2Se (space group P212121, lattice constants a0 = 4.333 Å, b0 = 7.062 Å, and c0 = 7.764 Å, JCPDS card no. 24-1041), tetragonal Ag3AuS2 (space group P4122, lattice constants a0 = 9.75 Å and c0 = 9.85 Å, JCPDS card no. 33-587), cubic Ag3AuSe2 (space group I4132, lattice constants a0 = 9.967 Å, JCPDS card no. 72-392), orthorhombic AgInS2 (space group Pna21, lattice constants a0 = 7.001 Å, b0 = 8.278 Å, and c0 = 6.698 Å, JCPDS card no. 25-1328), orthorhombic AgInSe2 (space group Pna21, lattice constants a0 = 7.223 Å, b0 = 8.489 Å, and c0 = 6.920 Å;51 section SI-2, Supporting Information), CdS (space group P63mc, lattice constants a0 = 4.141 Å and c0 = 6.720 Å, JCPDS card no. 411049), and CdSe (space group P63mc, lattice constants a0 =

RESULTS AND DISCUSSION

Synthesis of Au−X Heterodimers. With this synthetic scheme, six different types of complex Au−X heterodimers (X = I−VI and I−III−VI metal chalcogenides (i.e., Ag2S, Ag2Se, Ag3AuS2, Ag3AuSe2, AgInS2, and AgInSe2)) have been synthesized. The silver metal (and a small portion of gold in the cases of Ag3AuS2 and Ag3AuSe2) was readily converted to the X phase by tuning the experimental parameters. Revealed by X-ray diffraction (XRD) and transmission electron microscopy (TEM), the starting seed of AuAg-t1 is essentially a layered hybrid of MI-dodecanethiolate (MI = Au+ or Ag+), and with a longer reaction time, AuAg-t2 is highly monodisperse AuAg alloy nanoparticles (Supporting Information, section SI9841

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

4.299 Å and c0 = 7.01 Å, JCPDS card no. 8-459). The broadening of diffraction peaks also indicates that the product particles are on the nanoscale. Although in most cases, except for the (220) peak located at a 2θ angle of 64.5o, other characteristic peaks of Au (space group Fm3m ̅ , lattice constants a0 = 4.079 Å, JCPDS card no. 4-784), i.e., (111) and (200), could not be detected or are “shadowed” by the peak broadening of the metal chalcogenide phases. The existence of metallic Au in all these heterostructures can be confirmed unambiguously with high-resolution TEM (HRTEM), X-ray photoelectron spectroscopy (XPS), and energy-dispersive X-ray spectroscopy (EDX) methods as reported in section SI-2 (Supporting Information). The general structural views of these biphasic heterodimers can be found in the TEM images of Figure 2. They are

Figure 3. HRTEM images of HNCs (a) Au−Ag2S, (b) Au−Ag2Se, (c) Au−Ag3AuS2, (d) Au−Ag3AuSe2, (e) Au−AgInS2, and (f) Au− AgInSe2 corresponding to the TEM images shown in Figure 2.

SI-1, Supporting Information) are much larger than the threshold value (≤5%) expected for normal epitaxial growth. By taking advantage of the intrinsic high mobility of Ag+ ions in Ag2S nanocrystals, for example, Ag2S-based heterostructures have been fabricated by a seeded growth method in which Ag2S nanocrystals acted as a catalyst or source host.52 However, the largest lattice mismatch of the reported heterostructures is still within 10%. It has been reported that, in the preparation of HNCs comprising different materials with large lattice mismatch, the growth process is complicated by material miscibility, interfacial strain, and facet-dependent chemical reactivity.7,30 As mentioned earlier, in such cases, core/shell structures with large lattice mismatches can also be synthesized with nonepitaxial growth. The hypothesis is that the monocrystalline growth of the semiconductor shell is fully directed by the thermodynamic properties of the chemical reaction within the matrix. Therefore, the lattice structure of the shell can be independent of that of the core, circumventing the limitation of epitaxial growth.27,28 However, in our present cases, the nonepitaxial growth of the I−VI or I−III−VI phase occurred gradually from a reactive Au-containing seed, and the compound inherently attached to the Au phase was left behind; hence, heterogeneous nucleation and subsequent crystal growth were separated. Therefore, these transformable seeds served both as hard templates for the nucleation and as active reactants for the generation of the chalcogenide phase through solid-state diffusion. We further carried out an XPS analysis to verify the composition and oxidation states of elements in the heterostructures. The respective elements of Au, Ag, In, S, and Se are all present in the samples, except that Au could not be detected for the samples Au−CdS and Au−CdSe (section SI-2, Supporting Information), which were derived from Au− Ag2S and Au−Ag2Se by cation exchange, although Au can still be found in some particles with the HRTEM technique (Figure

Figure 2. TEM images of HNCs (a) Au−Ag2S, (b) Au−Ag2Se, (c) Au−Ag3AuS2, (d) Au−Ag3AuSe2, (e) Au−AgInS2, and (f) Au− AgInSe2: (a, c−e) by solvothermal synthesis and (b, f) by reflux reaction (section SI-2, Supporting Information).

monodisperse with high morphological uniformity. In particular, biphasic features of the heterodimers can be viewed discernibly in different image contrasts. In Figure 3, the darker gold component (as a small tip at one side of a particle) can be more clearly observed at higher magnifications. HRTEM images of individual particles reveal the highly crystalline nature of the X phases of all the samples, while the remaining Au phase can be either single crystalline (Figure 3e) or multiply twinned (Figure 3a). The lattice distances of the darker part match well with the d111 of Au. All the d spacings between the lattice planes of the X phases (the lighter part) are also in good agreement with the crystal planes of their respective metal chalcogenide compounds in the heterodimers (section SI-2, Supporting Information). The common structural feature of the above Au−X products is that their chalcogenide phases were grown nonepitaxially with respect to the attached Au phase. The lattice mismatches of all these heterodimers (ranging from 11.1% to 47.2%; section 9842

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

4; see section SI-2, Supporting Information, for details on the lattice parameters). This indicates that most leftover Au atoms

Figure 4. HRTEM images of HNCs (a) Au−CdS and (b) Au−CdSe obtained from cation exchange of Au−Ag2S and Au−Ag2Se, respectively.

were actually detached from CdS and CdSe with this synthetic route due to huge lattice mismatches in the latter two samples [Au−CdS (42.7%) vs Au−Ag2S (31.1%) and Au−CdSe (49.1%) vs Au−Ag2Se (13.6%); section SI-1, Supporting Information]. Additionally, since two Ag+ ions were replaced by one Cd2+ ion, a significant reconstruction within the initial chalcogenide phase was involved. Therefore, it would be difficult for the CdS and CdSe phases to remain in contact with the Au in their final heterostructures. Nevertheless, it is important to recognize that the reaction processes in Scheme 1 are designed to be continuous, and thus, lattice mismatches between the Au and X phase as large as 47.2% (in the case of Au−AgInSe2) could still be tolerated, which is an advantage of this dynamic seeding strategy. Formation of Superstructures. The as-prepared Au− Ag2S, Au−Ag3AuS2, and Au−Ag3AuSe2 heterodimers in Figure 2a,c,d exhibit spherical morphology and a monodisperse particle size distribution. Such particles have a strong tendency to assemble into two-dimensional (2D) superlattices or 3D supracrystals. In Figure 5, interestingly, all three types of heterodimers can be assembled into 2D or multilayered superlattices on the TEM sample grid in nonpolar solvents such as cyclohexane and toluene (Figure 5a,c,e). These heterodimers can even organize themselves into supracrystals in polar solvents such as ethanol with large dimensions on the order of micrometers (Figure 5b,d,f). This type of artificial crystals is unprecedented, because they are constructed from complicated biphasic building blocks for the first time. To further demonstrate the flexibility of this strategy, both solvothermal and reflux methods were carried out in our study (Experimental Section; section SI-2, Supporting Information). It is thus proved that both synthetic methods can lead to the formation of the above six types of Au−X heterodimers (except that phase-pure Au−Ag2Se and Au−AgInSe2 can only be obtained by the reflux method in the present stage of our study) at specific conditions. Generally speaking, the size distribution of the HNCs obtained via the solvothermal method is more uniform than that obtained via the reflux method. More details on the comparison of the two synthetic methods can be found in the Supporting Information (section SI-2). We note that, for the reported HNC fabrications in the literature,8,27,28,45 several process steps are required. For example, the presynthesized seeds normally need to be separated from their synthetic solutions and washed or subjected to additional purification processes, and then the seeds react with precursors to form the secondary compounds on the seeds. In our present cases, we can simply add sources/precursors directly to the

Figure 5. 2D superlattices and 3D supracrystals of (a, b) Au−Ag2S, (c, d) Au−Ag3AuS2, and (e, f) Au−Ag3AuSe2 heterodimers (section SI-2, Supporting Information).

dynamic seeds in the solution phase and trigger the secondary phase formation in a consecutive manner. In many reported cases, the synthesis of colloidal nanoparticles usually involves the cointeraction of several types of surfactants and solvents.8,27,28,45 In contrast, in our present approach, alkylamine (oleylamine) has multiple roles as a mild reducing agent, a solvent for reaction, and a capping surfactant for the product, since no other toxic chemicals or surfactants were involved in the preparation of our HNCs. It is also noteworthy that our syntheses were carried out at atmospheric conditions without any inert gas protection, which greatly reduced the process complexity. Seed Reactivity and Product Formation. According to Scheme 1, Au−X heterodimers are prepared through phase segregation of silver from the AuAg seeds. Two types of seeds were used in this work (Experimental Section; section SI-2, Supporting Information): (i) short-aged seeds (AuAg-t1) and (ii) long-aged seeds (AuAg-t2).50 In particular, AuAg-t1 is still in the form of MI-DDT (MI = Au+ or Ag+), while AuAg-t2 is essentially a AuAg alloy due to a longer reaction time (XRD/ HRTEM results; section SI-2, Supporting Information). Different reactivities of these seeds lead to two categories of products (Scheme 1). Although both types of seeds can produce Au−Ag2S and Au−AgInS2, Au−Ag3AuS2 and Au− Ag3AuSe2 can only be obtained from the short-aged seeds (i.e., AuAg-t1), whereas the phase-pure Au−Ag2Se and Au−AgInSe2 can only be achieved with the long-aged ones (i.e., AuAg-t2, t2 = 30 min or longer). When starting with AuAg-t1, for example, the synthetic conditions in making Au−Ag2S become more stringent due to possible formation of Au−Ag3AuS2. The shortaged seeds can also lead to formation of Au−Ag2Se, but coexistence of Au−Ag3AuSe2 cannot be avoided, because of easier formation of the Ag3AuSe2 compound (ΔfG°(Ag3AuSe2) = −86450 ± 320 J/mol) compared to Ag2Se (ΔfG°(Ag2Se) = −35018 − 41.6T (405 < T (K) < 457) J/mol).53,54 On the 9843

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

other hand, we could obtain phase-pure products of Au−Ag2S and Au−Ag3AuS2 from the short-aged seeds by controlling the experimental conditions (Experimental Section), because of a smaller difference in the formation free energies of Ag2S (Δ f G°(Ag 2 S) = −43880 + 20.8T J/mol and Ag 3 AuS 2 (ΔfG°(Ag3AuS2) = −40410 + 20.8T J/mol).55 It has been reported that the formation of bulk Ag3AuS2 and Ag3AuSe2 usually requires a high reaction temperature (1050 °C) and long reaction time (e.g., 3 days).56 In comparison, the same ternary chalcogenides could be formed under our lowtemperature conditions (Experimental Section) in a much shorter time, because of the readiness of Au+ in the short-aged seeds (AuAg-t1; section SI-2, Supporting Information). In other words, some of the existing Au+ ions could be transformed into Au0 and some of the Au+ ions readily combine with chalcogenide elements, resulting in Au−Ag3AuS2 and Au− Ag3 AuSe 2 (Experimental Section). However, the same chalcogenides could not be formed with the AuAg-t2 seed under such moderate conditions due to the metallic state of gold in this alloy seed. Extension to Other Types of Seeds. As demonstrated above, apart from the intrinsic compositional difference in the starting seeds (i.e., MI-DDT versus AuAg alloy nanoparticles; section SI-2, Supporting Information), structural defects within the same seed particle can also lead to a variation of reactivity. To establish this point further, we also used as-prepared Ag nanoparticles as starting dynamic seeds to make silver chalcogenides and heterostructures. As shown in Figure 6, phase-pure chalcogenides Ag2S, Ag2Se, AgInS2, and AgInSe2 could be obtained by full conversion of the Ag seeds (Ag-t2; Supporting Information, section SI-3). Through controlling the reaction kinetics, heterodimers of Ag−Ag2S and Ag−AgInS2

could also be prepared by partial conversion of the silver metal, in which more reactive Ag atoms were converted to the chalcogenide phases and less reactive ones retained in the metallic phase. Apparently, structural inhomogeneity is the cause of the differentiation of the atomic reactivity of Ag. Detailed information on the characterization of these Ag chalcogenides and their metal composites can be found in section SI-3, Supporting Infromation). Using AuCu alloy seeds (Supporting Information, section SI4), along the lines of Scheme 1, we have also synthesized four Au−X with Cu-based chalcogenides (where X = Cu9S5, CuInS2, CuSe, and CuInSe2). Our preliminary XRD, TEM, XPS, and EDX results (section SI-4, Supporting Information) confirm the formation of these Au-attached I−VI and I−III−VI chalcogenides, although there is a need to attain additional morphological control for these samples. Interestingly, without the Au phase, it was only possible to obtain phase-pure CuInS2 (section SI-4, Supporting Information). It is well-known that the metallic seeds often participate in the reaction as catalysts and lower the activation energy for heterogeneous nucleation.9 Thus, this observation implies that the residual Au atoms are essential and may also act as a catalyst for the observed chemical transformation. Herein, we point out that, besides the above AuAg and AuCu alloy seeds, many other alloy nanoparticles reported in the literature could also be utilized as dynamic seeds. Using the reported PtAg nanoparticles,57 for example, we have also synthesized heterostructures of Pt− AgInS2 following a synthetic route similar to that of Scheme 1 (Experimental Section). The preparative details and material characterization (with XRD, TEM, XPS, and EDX techniques) of this heterodimer product can be found in the Supporting Information (section SI-5). Manipulation of the Phase Ratio in HNCs. Typically, the shell thickness in core/shell HNCs is controlled by varying the molar ratio of the seed and shell materials together with the reaction time. However, only a few examples have been successful with this conventional appraoch.32,40 In comparison, an advantage for the present dynamic seeding strategy is that the ratio of the two phases in the heterostructures can be tuned facilely by the initial input ratio of Au and Ag. Herein we will use the syntheses of Au−Ag2S and Au−AgInS2 (i.e., gold with binary and tertiary metal chalcogenides; Supporting Information, section SI-6) to demonstrate this phasic control. In Figure 7a, the XRD patterns and TEM images of Au−Ag2S heterodimers are displayed. The variation in the relative intensities in the XRD patterns can be related qualitatively to different phase ratios of Au and Ag2S in the heterodimers. In agreement, different ratios of the dark (Au) and the light (Ag2S) contrasts can be observed in their corresponding TEM images. On the basis of these results, it can be concluded that the Au phase increases while the Ag2S phase decreases in the HNCs in accordance with their Au:Ag ratio in the alloy seeds. A similar trend can also be found for Au−AgInS2 HNCs in Figure 7b. For both Au−Ag2S and Au−AgInS2, heterodimers are formed when the Au amount is lower. However, core/shell structures (although they are not strictly centrosymmetrical) tend to form at a higher Au content, i.e., both Au:Ag = 8:4 and Au:Ag = 4:4 for Au−AgInS2. Apparently, the formation of such pseudocore/shell structures in Au−AgInS2 with a high content of Au can be attributed to a larger lattice mismatch (Au−Ag2S at 31.1% and Au−AgInS2 at 42.5%, section SI-1, Supporting Information). According to literature reports for the conventional seeded growth, large lattice mismatches usually lead to

Figure 6. TEM images and powder XRD patterns of (a) Ag2S, (b) Ag2Se, and (c) AgInSe2 nanocrystals and (d) Ag−Ag2S and (e) Ag− AgInS2 (note that the real formula should be Ag−AgxInySz, which matches well with the AgInS2 XRD pattern) heterodimers prepared from pure Ag seeds (section SI-3, Supporting Information). 9844

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

tunable composition control offers us a convenient platform to investigate the optical properties of HNCs. Figure 7c shows a series of UV−vis absorption spectra of Au−AgInS2 HNCs synthesized from AuAg seeds with different Au:Ag ratios. Interestingly, the spectra exhibit different SPR absorption bands in the visible region with respect to variable ratios of the two phases. It can be observed that, at higher Au ratios (Au:Ag = 4:4 and 8:4), the absorption spectra of the Au−AgInS2 HNCs are dominated by the SPR effect of the Au nanoparticles, as the absorption features of AgInS2 are not pronounced due to superimposed SPR absorption of the tail of Au.59 There is no obvious change in the location of the SPR band in these two cases (Au:Ag = 4:4 and 8:4), and they both show an absorption maximum at ∼538 nm, which is a characteristic peak of Au nanoparticles in both their single-phase form50 and their HNCs.60 Obviously, the SPR band is more prominent for Au− AgInS2 with Au:Ag = 8:4, because of a higher Au content in the HNCs. The SPR absorption spectrum of the Au−AgInS2 HNCs with Au:Ag = 2:4 is widened, and a shoulder absorption can be noted. However, the absorption spectrum of Au−AgInS2 HNCs with Au:Ag = 1:4 shows typical features of AgInS2 NCs, but the SPR feature of Au is not very visible. Therefore, through adjustment of the starting seed composition, the phase ratio and optical properties of the product HNCs can be finely tuned. Investigation of the Formation Mechanism. Timedependent UV−vis absorption spectroscopy can be used as a probe to monitor the formation process of Au nanostructures. This method was also employed to explore the dynamic growth of our HNCs. In particular, the formation of Au−AgInS2 heterostructures from Au:Ag = 8:4 seeds was chosen as a model system for this study, in view of its prominent SPR peak among the others (Figure 7c). As shown in Figure 8a, before addition of indium and sulfur precursors, the SPR band is located at ∼485 nm, which is the characteristic peak of AuAg alloy nanoparticles.50 After addition of the two precursors (the sample for measurement was withdrawn within 1 min after addition of the two precursors; see the Experimental Section), the original SPR band disappears immediately. However, as the reaction proceeds, the metal SPR band gradually gains intensity and is finally located at ∼538 nm,50 indicating that phase-pure Au is formed in the Au−AgInS2 product. Consistent with SPR band development, the evolution of Au and AgInS2 phases was clearly recorded in Figure 8b. A similar trend can also be found during the formation of Au−CuInS2 except that the SPR band is much broader and red-shifted due to the existence of Au nanorods in the Au−CuInS2 product (derived from nonuniform AuCu seeds; Supporting Information, sections SI-4 and SI-7). To understand the reaction sequences, Figure 8 displays some representative XPS spectra of Au 4f, Ag 3d, S 2p, and In 3d of the reaction intermediates (i.e., 1 min after addition of the indium and sulfur precursors) and final Au−AgInS2 product (after 2 h of reaction). It can be found that, at the beginning stage, the intensity of the In 3d peaks is much higher than those of the other three elements and the signal-to-noise ratio of the In 3d peaks is also the highest, which reveals that the In3+ ions are more abundant in the surface region upon reaction. As the reaction proceeds, the intensity of the In 3d photoelectrons decreases while those of Au 4f, Ag 3d, and S 2p increase. This XPS analysis, together with the UV−vis and XRD results, provides us valuable information for assessing the formation kinetics of HNCs. On the bsis of these observations, a

Figure 7. (a) XRD patterns and their corresponding TEM images of Au−Ag2S samples prepared from seeds (AuAg-t2) with different Au:Ag atomic ratios (section SI-6, Supporting Information): (i) Au:Ag = 4:2, (ii) Au:Ag = 2:2, (iii) Au:Ag = 1:4, (iv) Au:Ag = 1:6. (b) XRD patterns and their corresponding TEM images of Au−AgInS2 samples prepared from seeds (AuAg-t2) with different Au:Ag atomic ratios (section SI-6, Supporting Information): (i) Au:Ag = 1:4, (ii) Au:Ag = 2:4, (iii) Au:Ag = 4:4, (iv) Au:Ag = 8:4. To make a comparison, the four XRD peaks were normalized with the (121)/(201) peak of AgInS2. (c) UV− vis absorption spectra of Au−AgInS2 samples prepared from seeds with different Au:Ag atomic ratios (section SI-6, Supporting Information). The scale bar for all the TEM images is 5 nm.

phase segregation of the core and shell, and it is difficult to form core/shell structures between large lattice mismatch materials.7,27,28 In this regard, our present approach seems to be able to overcome this limitation because it involves a continuous intimate phase evolution process, which can be considered a methodic advantage for the general preparation of core/shell structures with large lattice mismatches among the different phases. It is well-known that noble metals and their alloy nanoparticles often exhibit SPR properties at specific wavelengths in the visible range.58 The SPR feature of the Au nanoparticles is sensitive to the composition, size, shape, and dielectric environment.17 In fact, our synthetic strategy with 9845

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

Figure 8. Time-dependent (a) UV−vis absorption spectra and (b) XRD patterns of Au−AgInS2. The arrow indicates the Au(111) peak. XPS spectra of (c) Au 4f, (d) Ag 3d, (e) S 2p, and (f) In 3d right after the In and S precursors were added to the AuAg seed (AuAg-t2) solution (1 min) and fullfledged Au−AgInS2 after 2 h of reaction. (g) Proposed formation mechanism of Au−AgInS2 HNCs.

peaks; S 2p3/2 has a higher intensity, whereas S 2p1/2 is seen as a shoulder, Figure 8e) can be found after a very short reaction time (1 min), indicating the formation of Ag−S and In−S bonds in Ag2S and AgInS2. The sulfur doublet peaks become sharper and better resolved due to higher crystallinity in the final AgInS2 phase upon a longer reaction time (2 h, Figure 8e). Regarding the optical properties, we notice that SPR is a localized phenomenon which persists and penetrates into its surrounding dielectrics with a limited detection range.32 It may be quenched when the metal/alloy nanoparticles are surrounded by atomic or ionic clouds. Nevertheless, the SPR could also disappear or shift when metal/alloy nanoparticles undergo chemical transformation. In our present case, no SPR feature can be observed at the beginning stage of the reaction (1 min), which suggests two possibilities: (i) quenching due to the presence of ionic and/or atomic “clouds” (S2− and In3+ ions) and (ii) reconstruction of metal/alloy nanoparticles due to chemical transformation of Ag2S to AgInS2, during which the noble-metal nanoparticles become numerous tiny metallic

schematic reaction mechanism for the formation of Au−AgInS2 HNCs is proposed in Figure 8g, which will be explained below. Right after addition of indium and sulfur precursors into the AuAg seed suspension, elemental sulfur is attracted to the surface Ag atoms, while the Ag atoms in the core region also migrate toward the surface to form Ag−S bonds (surface reaction 2Ag + S → Ag2S). The original AuAg alloy now becomes Au(Ag); that is, phase separation takes place, and Ag is now surface-rich. Due to this rearrangement of metal atoms, the Ag content in the bulk alloy continues to decrease until all the Ag is exhausted. Accompanied by the formation of S2− and thus Ag2S, In3+ ions are also adsorbed onto the surfaces of Au(Ag)−Ag2S, which leads to instant formation of the AgInS2 phase by cation exchange (2Ag2S + In3+ → AgInS2 + 3Ag+). Note that there is no direct reaction between In3+ and elemental S0. This explains why the highest XPS peak intensity of In 3d was observed at the beginning. The changes in the intensity of sulfur XPS peaks (Figure 8e) also support this reaction kinetics. The S2− anions (S 2p1/2 and S 2p3/2 doublet 9846

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

■

clusters embedded in the semiconductor matrix (e.g., AgInS2). Therefore, the quenching effect of the matrix and/or the strong metal−semiconductor interaction must be taken into account. In fact, the XRD phase analysis can resolve this problem. As shown in Figure 8b, no peaks are observed in the first minute of reaction, while the Au(111) peak starts to appear at 3 min. This observation indicates that the starting AuAg seed has indeed undergone a severe chemical transformation (Figure 8g); that is, Ag atoms have reacted with S, leaving disordered Au atoms or clusters in the Ag2S or AgInS2 matrix, according to possibility ii, that is, amorphization of the AuAg seed due to chemical transformation. With longer reaction times, both Au and AgInS2 are better crystallized, as these two material phases can be clearly resolved (2 h, Figure 8b). Finally, consistent with this observation, we have also observed an SPR peak at ∼538 nm (Figure 8a) for the gold phase in the Au−AgInS2 HNC product. On the basis of the mechanism proposed in Figure 8g, it can be deduced that Au−Ag2S(···In3+) is a reaction intermediate during the formation of Au−AgInS2, although we could not observe Au−Ag2S as an intermediate from time-dependent XRD patterns (Figure 8b), which might be due to the fact that the transformation of Ag2S to AgInS2 is a fast process and many complicated steps are involved, resulting in noncrystalline solid intermediates (3 min, Figure 8b). To confirm this, we also carried out some control experiments by adding In3+ ions to the as-prepared Au−Ag2S heterodimers (Experimental Section). Indeed, the product is pure Au−AgInS2 through this cation exchange. The above proposed mechanism is thus elucidated.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge financial support from the National University of Singapore and GSK Singapore.

■

REFERENCES

(1) Donega, C. d. M. Synthesis and Properties of Colloidal Heteronanocrystals. Chem. Soc. Rev. 2011, 40 (3), 1512−1546. (2) Cortie, M. B.; McDonagh, A. M. Synthesis and Optical Properties of Hybrid and Alloy Plasmonic Nanoparticles. Chem. Rev. 2011, 111 (6), 3713−3735. (3) Behrens, S. Preparation of Functional Magnetic Nanocomposites and Hybrid Materials: Recent Progress and Future Directions. Nanoscale 2011, 3 (3), 877−892. (4) Costi, R.; Saunders, A. E.; Banin, U. Colloidal Hybrid Nanostructures: A New Type of Functional Materials. Angew. Chem., Int. Ed. 2010, 49 (29), 4878−4897. (5) Cozzoli, P. D.; Pellegrino, T.; Manna, L. Synthesis, Properties and Perspectives of Hybrid Nanocrystal Structures. Chem. Soc. Rev. 2006, 35 (11), 1195−1208. (6) Choi, S.-H.; Kim, E.-G.; Hyeon, T. One-Pot Synthesis of Copper−Indium Sulfide Nanocrystal Heterostructures with Acorn, Bottle, and Larva Shapes. J. Am. Chem. Soc. 2006, 128 (8), 2520−2521. (7) Carbone, L.; Cozzoli, P. D. Colloidal Heterostructured Nanocrystals: Synthesis and Growth Mechanisms. Nano Today 2010, 5 (5), 449−493. (8) Buck, M. R.; Bondi, J. F.; Schaak, R. E. A Total-Synthesis Framework for the Construction of High-Order Colloidal Hybrid Nanoparticles. Nat. Chem. 2012, 4, 37−44. (9) Zeng, H.; Sun, S. Syntheses, Properties, and Potential Applications of Multicomponent Magnetic Nanoparticles. Adv. Funct. Mater. 2008, 18, 391−400. (10) Pineider, F.; de Julián Fernández, C.; Videtta, V.; Carlino, E.; al Hourani, A.; Wilhelm, F.; Rogalev, A.; Cozzoli, P. D.; Ghigna, P.; Sangregorio, C. Spin-Polarization Transfer in Colloidal MagneticPlasmonic Au/Iron Oxide Hetero-nanocrystals. ACS Nano 2013, 7 (1), 857−866. (11) Pisanello, F.; Leménager, G.; Martiradonna, L.; Carbone, L.; Vezzoli, S.; Desfonds, P.; Cozzoli, P. D.; Hermier, J.-P.; Giacobino, E.; Cingolani, R.; De Vittorio, M.; Bramati, A. Non-Blinking SinglePhoton Generation with Anisotropic Colloidal Nanocrystals: Towards Room-Temperature, Efficient, Colloidal Quantum Sources. Adv. Mater. 2013, 25 (14), 1974−1980. (12) Casavola, M.; Falqui, A.; García, M. A.; García-Hernández, M.; Giannini, C.; Cingolani, R.; Cozzoli, P. D. Exchange-Coupled Bimagnetic Cobalt/Iron Oxide Branched Nanocrystal Heterostructures. Nano Lett. 2009, 9 (1), 366−376. (13) Kostopoulou, A.; Thétiot, F.; Tsiaoussis, I.; Androulidaki, M.; Cozzoli, P. D.; Lappas, A. Colloidal Anisotropic ZnO-Fe@FexOy Nanoarchitectures with Interface-Mediated Exchange-Bias and BandEdge Ultraviolet Fluorescence. Chem. Mater. 2012, 24 (14), 2722− 2732. (14) Deka, S.; Falqui, A.; Bertoni, G.; Sangregorio, C.; Poneti, G.; Morello, G.; Giorgi, M. D.; Giannini, C.; Cingolani, R.; Manna, L.; Cozzoli, P. D. Fluorescent Asymmetrically Cobalt-Tipped CdSe@CdS Core@Shell Nanorod Heterostructures Exhibiting Room-Temperature Ferromagnetic Behavior. J. Am. Chem. Soc. 2009, 131 (35), 12817−12828. (15) Mokari, T.; Aharoni, A.; Popov, I.; Banin, U. Diffusion of Gold into InAs Nanocrystals. Angew. Chem., Int. Ed. 2006, 45 (47), 8001− 8005.

■

CONCLUSIONS In summary, we have developed a modular dynamic seeding method for the synthesis of a total of 12 different I−VI, II−VI, and I−III−VI chalcogenides. By utilizing the intrinsic composition or structural defects of metal nanoparticles, 15 noble-metal-based heterodimers have been further obtained. The ratio of metal and metal chalcogenide phases in the heterostructured nanocrystals can be systematically tuned by adjusting the ratio of the seed elements. Among them, complex heterodimers and core/shell heterostructures composed of ternary chalcogenide compounds such as Au−Ag3AuS2, Au− Ag3 AuSe 2 , Au−AgInS 2, Au−AgInSe 2 , Au−CuInS 2 , Au− CuInSe2, and Pt−AgInS2 are the first products, and some of them can even build into supracrystals owing to the high monodispersity of the heterostructured nanocrystals. As demonstrated in this work, abundant alloy or monometal precursors can serve as dynamic seeds in these chemical transformations. A general reaction mechanism has been put forward for the formation of these heterostructured nanocrystals using both Au−AgInS2 and Au−CuInS2 as model cases. As shown for their tunable optical properties, the availability of such a wide variety of metal chalcogenide compounds, together with their nanocomposites with metals, could offer higher flexibility in the design and synthesis of this class of semiconducting materials for future applications in the fields of optoelectronics, photovoltaics, heterogeneous catalysis, water splitting, and gas sensing.

■

Article

ASSOCIATED CONTENT

S Supporting Information *

Chemicals, treatments, and materials characterization. This material is available free of charge via the Internet at http:// pubs.acs.org. 9847

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

(16) Mokari, T.; Rothenberg, E.; Popov, I.; Costi, R.; Banin, U. Selective Growth of Metal Tips onto Semiconductor Quantum Rods and Tetrapods. Science 2004, 304 (5678), 1787−1790. (17) Achermann, M. Exciton-Plasmon Interactions in Metal-Semiconductor Nanostructures. J. Chem. Phys. Lett. 2010, 1, 2837−2843. (18) Schaadt, D. M.; Feng, B.; Yu, E. T. Enhanced Semiconductor Optical Absorption via Surface Plasmon Excitation in Metal Nanoparticles. Appl. Phys. Lett. 2005, 86, 063106-1−3. (19) Costi, R.; Saunders, A. E.; Elmalem, E.; Salant, A.; Banin, U. Visible Light-Induced Charge Retention and Photocatalysis with Hybrid CdSe−Au Nanodumbbells. Nano Lett. 2008, 8 (2), 637−641. (20) Khon, E.; Mereshchenko, A.; Tarnovsky, A. N.; Acharya, K.; Klinkova, A.; Hewa−Kasakarage, N. N.; Nemitz, I.; Zamkov, M. Suppression of the Plasmon Resonance in Au/CdS Colloidal Nanocomposites. Nano Lett. 2011, 11 (4), 1792−1799. (21) Shaviv, E.; Banin, U. Synergistic Effects on Second Harmonic Generation of Hybrid CdSe−Au Nanoparticles. ACS Nano 2010, 4 (3), 1529−1538. (22) Danek, M.; Jensen, K. F.; Murray, C. B.; Bawendi, M. G. Synthesis of Luminescent Thin-Film CdSe/ZnSe Quantum Dot Composites Using CdSe Quantum Dots Passivated with an Overlayer of ZnSe. Chem. Mater. 1996, 8 (1), 173−180. (23) Li, J. J.; Wang, Y. A.; Guo, W.; Keay, J. C.; Mishima, T. D.; Johnson, M. B.; Peng, X. Large-Scale Synthesis of Nearly Monodisperse CdSe/CdS Core/Shell Nanocrystals Using Air-Stable Reagents via Successive Ion Layer Adsorption and Reaction. J. Am. Chem. Soc. 2003, 125 (41), 12567−12575. (24) Peng, X.; Schlamp, M. C.; Kadavanich, A. V.; Alivisatos, A. P. Epitaxial Growth of Highly Luminescent CdSe/CdS Core/Shell Nanocrystals with Photostability and Electronic Accessibility. J. Am. Chem. Soc. 1997, 119 (30), 7019−7029. (25) Zeng, J.; Huang, J.; Liu, C.; Wu, C. H.; Lin, Y.; Wang, X.; Zhang, S.; Hou, J.; Xia, Y. Gold-Based Hybrid Nanocrystals through Heterogeneous Nucleation and Growth. Adv. Mater. 2010, 22 (17), 1936−1940. (26) Wang, C.; Daimon, H.; Sun, S. Dumbbell-like Pt−Fe3O4 Nanoparticles and Their Enhanced Catalysis for Oxygen Reduction Reaction. Nano Lett. 2009, 9 (4), 1493−1496. (27) Zhang, J.; Tang, Y.; Lee, K.; Ouyang, M. Nonepitaxial Growth of Hybrid Core−Shell Nanostructures with Large Lattice Mismatches. Science 2010, 327 (5973), 1634−1638. (28) Li, M.; Yu, X.-F.; Liang, S.; Peng, X.-N.; Yang, Z.-J.; Wang, Y.-L.; Wang, Q.-Q. Synthesis of Au−CdS Core−Shell Hetero-Nanorods with Efficient Exciton−Plasmon Interactions. Adv. Funct. Mater. 2011, 21 (10), 1788−1794. (29) Woggon, U.; Herz, E.; Schöps, O.; Artemyev, M. V.; Arens, C.; Rousseau, N.; Schikora, D.; Lischka, K.; Litvinov, D.; Gerthsen, D. Hybrid Epitaxial-Colloidal Semiconductor Nanostructures. Nano Lett. 2005, 5 (3), 483−490. (30) Casavola, M.; Buonsanti, R.; Caputo, G.; Cozzoli, P. D. Colloidal Strategies for Preparing Oxide-Based Hybrid Nanocrystals. Eur. J. Inorg. Chem. 2008, 6, 837−854. (31) Pang, M.; Hu, J.; Zeng, H. C. Synthesis, Morphological Control and Antibacterial Properties of Hollow/Solid Ag2S/Ag Heterodimers. J. Am. Chem. Soc. 2010, 132 (31), 10771−10785. (32) Lee, J.-S.; Shevchenko, E. V.; Talapin, D. V. Au−PbS Core− Shell Nanocrystals: Plasmonic Absorption Enhancement and Electrical Doping via Intra-Particle Charge Transfer. J. Am. Chem. Soc. 2008, 130 (30), 9673−9675. (33) Vaneski, A.; Susha, A. S.; Rodriguez−Fernandez, J.; Berr, M.; Jackel, F.; Feldmann, J.; Rogach, A. L. Hybrid Colloidal Heterostructures of Anisotropic Semiconductor Nanocrystals Decorated with Noble Metals: Synthesis and Function. Adv. Funct. Mater. 2011, 21 (9), 1547−1556. (34) Franchini, I. R.; Bertoni, G.; Falqui, A.; Giannini, C.; Wang, L. W.; Manna, L. Colloidal PbTe−Au Nanocrystal Heterostructures. J. Mater. Chem. 2010, 20, 1357−1366.

(35) McDaniel, H.; Shim, M. Size and Growth Rate Dependent Structural Diversification of Fe3O4/CdS Anisotropic Nanocrystal Heterostructures. ACS Nano 2009, 3 (2), 434−440. (36) Habas, S. E.; Yang, P.; Mokari, T. Selective Growth of Metal and Binary Metal Tips on CdS Nanorods. J. Am. Chem. Soc. 2008, 130 (11), 3294−3295. (37) Zhang, Q.; Wang, J.-J.; Jiang, Z.; Guo, Y.-G.; Wan, L.-J.; Xie, Z.; Zheng, L. Au−Cu Alloy Bridged Synthesis and Optoelectronic Properties of Au@CuInSe2 Core−Shell Hybrid Nanostructures. J. Mater. Chem. 2012, 22, 1765−1769. (38) Li, M.; Yu, X.-F.; Liang, S.; Peng, X.-N.; Yang, Z.-J.; Wang, Y.-L.; Wang, Q.-Q. Synthesis of Au−CdS Core−Shell Hetero-Nanorods with Efficient Exciton−Plasmon Interactions. Adv. Funct. Mater. 2011, 21 (10), 1788−1794. (39) Liang, S.; Liu, X.-L.; Yang, Y.-Z.; Wang, Y.-L.; Wang, J.-H.; Yang, Z.-J.; Wang, L.-B.; Jia, S.-F.; Yu, X.-F.; Zhou, L.; Wang, J.-B.; Zeng, J.; Wang, Q.-Q.; Zhang, Z. Symmetric and Asymmetric Au−AgCdSe Hybrid Nanorods. Nano Lett. 2012, 12 (10), 5281−5286. (40) Yang, J.; Ying, J. Y. Room-Temperature Synthesis of Nanocrystalline Ag2S and Its Nanocomposites with Gold. Chem. Commun. 2009, 3187−3189. (41) Lu, W.; Wang, B.; Zeng, J.; Wang, X.; Zhang, S.; Hou, J. G. Synthesis of Core/Shell Nanoparticles of Au/CdSe via Au−Cd Bialloy Precursor. Langmuir 2005, 21 (8), 3684−3687. (42) Motl, N. E.; Bondi, J. F.; Schaak, R. E. Synthesis of Colloidal Au−Cu2S Heterodimers via Chemically Triggered Phase Segregation of AuCu Nanoparticles. Chem. Mater. 2012, 24 (9), 1552−1554. (43) Ding, X. G.; Zou, Y.; Jiang, J. Au−Cu2S Heterodimer Formation via Oxidization of AuCu Alloy Nanoparticles and in Situ Formed Copper Thiolate. J. Mater. Chem. 2012, 22 (43), 23169−23174. (44) Ding, X. G.; Zou, Y.; Ye, F.; Yang, J.; Jiang, J. Pt−CuS Heterodimers by Sulfidation of CuPt Alloy Nanoparticles and Their Selective Catalytic Activity toward Methanol Oxidation. J. Mater. Chem. A 2013, 1 (38), 11880−11886. (45) Shi, W.; Zeng, H.; Sahoo, Y.; Ohulchanskyy, T. Y.; Ding, Y.; Wang, Z. L.; Swihart, M.; Prasad, P. N. A General Approach to Binary and Ternary Hybrid Nanocrystals. Nano Lett. 2006, 6 (4), 875−881. (46) Son, D. H.; Hughes, S. M.; Yin, Y.; Alivisatos, A. P. Cation Exchange Reactions in Ionic Nanocrystals. Science 2004, 306 (5698), 1009−1012. (47) Sadtler, B.; Demchenko, D. O.; Zheng, H.; Hughes, S. M.; Merkle, M. G.; Dahmen, U.; Wang, L.-W.; Alivisatos, A. P. Selective Facet Reactivity during Cation Exchange in Cadmium Sulfide Nanorods. J. Am. Chem. Soc. 2009, 131 (14), 5285−5293. (48) Mews, A.; Eychmueller, A.; Giersig, M.; Schooss, D.; Weller, H. Preparation, Characterization, and Photophysics of the Quantum Dot Quantum Well System Cadmium Sulfide/Mercury Sulfide/Cadmium Sulfide. J. Phys. Chem. 1994, 98 (3), 934−941. (49) Robinson, R. D.; Sadtler, B.; Demchenko, D. O.; Erdonmez, C. K.; Wang, L. W.; Alivisatos, A. P. Spontaneous Superlattice Formation in Nanorods through Partial Cation Exchange. Science 2007, 317, 355−358. (50) Chng, T. T.; Polavarapu, L.; Xu, Q.-H.; Ji, W.; Zeng, H. C. Rapid Synthesis of Highly Monodisperse AuxAg1−x Alloy Nanoparticles via a Half-Seeding Approach. Langmuir 2011, 27 (9), 5633− 5643. (51) Ng, M. T.; Boothroyd, C. B.; Vittal, J. J. One-Pot Synthesis of New-Phase AgInSe2 Nanorods. J. Am. Chem. Soc. 2006, 128 (22), 7118−7119. (52) Zhu, G.; Xu, Z. Controllable Growth of Semiconductor Heterostructures Mediated by Bifunctional Ag2S Nanocrystals as Catalyst or Source-Host. J. Am. Chem. Soc. 2010, 133 (1), 148−157. (53) Echmaeva, E. A.; Osadchii, E. G. Determination of the Thermodynamic Properties of Compounds in the Ag-Au-Se and AgAu-Te Systems by the EMF Method. Geol. Ore Deposits 2009, 51 (3), 247−258. (54) Osadchii, E. G.; Echmaeva, E. A. The System Ag−Au−Se: Phase Relations below 405 K and Determination of Standard Thermody9848

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849

Langmuir

Article

namic Properties of Selenides by Solid-State Galvanic Cell Technique. Am. Mineral. 2007, 92, 640−647. (55) Pal’yanova, G.; Seryotkin, Y.; Kokh, K. Formation of Gold and Silver Sulfides from Melts in the Ag-Au-S System: Experimental Data. Dokl. Earth Sci. 2011, 436 (1), 42−46. (56) Seryotkin, Y. V.; Bakakin, V. V.; Pal’yanova, G. A.; Kokh, K. A. Synthesis and Crystal Structure of the Trigonal Silver(I) Dithioaurate(I), Ag3AuS2. Cryst. Growth Des. 2011, 11 (4), 1062−1066. (57) Yang, J.; Yang, J.; Ying, J. Y. Morphology and Lateral Strain Control of Pt Nanoparticles via Core−Shell Construction Using Alloy AgPd Core toward Oxygen Reduction Reaction. ACS Nano 2012, 6 (11), 9373−9382. (58) Alvarez, M. M.; Khoury, J. T.; Schaaff, T. G.; Shafigullin, M. N.; Vezmar, I.; Whetten, R. L. Optical Absorption Spectra of Nanocrystal Gold Molecules. J. Phys. Chem. B 1997, 101, 3706−3712. (59) Mao, B.; Chuang, C.-H.; Wang, J.; Burda, C. Synthesis and Photophysical Properties of Ternary I−III−VI AgInS2 Nanocrystals: Intrinsic versus Surface States. J. Phys. Chem. C 2011, 115 (18), 8945− 8954. (60) Shaviv, E.; Schubert, O.; Alves-Santos, M.; Goldoni, G.; Di Felice, R.; Vallée, F.; Del Fatti, N.; Banin, U.; Sönnichsen, C. Absorption Properties of Metal−Semiconductor Hybrid Nanoparticles. ACS Nano 2011, 5 (6), 4712−4719.

9849

dx.doi.org/10.1021/la501637m | Langmuir 2014, 30, 9838−9849