14 Synthetic strategies to multi- material hybrid ...

Transworld Research Network 37/661 (2), Fort P.O., Trivandrum-695 023, Kerala, India

Advanced Wet-Chemical Synthetic Approaches to Inorganic Nanostructures, 2008: 407-453 ISBN: 978-81-7895-361-8 Editor: P. Davide Cozzoli

14

Synthetic strategies to multimaterial hybrid nanocrystals Gianvito Caputo, Raffaella Buonsanti, Marianna Casavola and Pantaleo Davide Cozzoli Scuola Superiore ISUFI, Distretto Tecnologico, University of Salento, National Nanotechnology Laboratory of CNR-INFM, Unità di Ricerca IIT, Via per Arnesano Km 5 73100 Lecce, Italy

Abstract Wet-chemical approaches to multi-component hybrid nanocrystals (HNCs), that incorporate nanoscale domains of different semiconductor, metallic and/or oxide materials interconnected through inorganic junctions, are illustrated and discussed. It is shown how control of interfacial lattice strain and surface energy in liquid media can be achieved within the frame of seeded-growth synthesis techniques, leading to structurally complex HNCs with purposely engineered compositional and geometric parameters. Various topological configurations are analyzed, including concentric core/shell architectures and hetero-oligomers grouping spherical and anisotropically shaped material sections. The most significant chemical-physical properties and technological advantages offered by such multifunctional HNCs are also briefly highlighted. Correspondence/Reprint request: Dr. Pantaleo Davide Cozzoli, Scuola Superiore ISUFI, Distretto Tecnologico University of Salento, National Nanotechnology Laboratory of CNR-INFM, Unità di Ricerca IIT, Via per Arnesano Km 5, 73100 Lecce, Italy E-mail: [email protected]

408

Gianvito Caputo et al.

1. Introduction Colloidal inorganic nanocrystals (NCs) are propelling substantial part of the current scientific revolution of nanoscience and nanotechnology. They are not only essential to the understanding of the dimensionality-dependent laws of nanoscale matter [1-25], but also to the implementation of new processes and the bottom-up fabrication of unprecedented functional materials and devices, in which the individual building units and their spatial arrangement are engineered down to the nanometer level [26-37]. This progress has been stimulated by significant advancement in the wet-chemical syntheses of robust and easily processable NCs in a wide range of sizes and shapes for a number of technologically valuable materials. This has allowed materials science to benefit from an incredibly wide range of appealing solid-state physical-chemical properties, boosting applications in technologic areas as diverse as optoelectronics, photovoltaics, sensing, environmental remediation, organic synthesis, advanced ceramics and functional coatings fabrication, and biomedicine [26-37]. Among the various wet-chemistry approaches to nanostructures, surfactant-assisted routes have especially distinguished for their capability to provide finely size- and shapetailored NCs of semiconductor, oxide and metal materials by careful regulation of thermodynamically and kinetically driven growth processes in liquid media [2, 3]. While refinement of this synthetic expertise is far from being exhausted, further efforts are currently pursued to fabricate novel breeds of NCs with a higher level of structural and compositional complexity and, consequently, of functionality. In recent years, nanochemistry research has made remarkable advances with the solution-phase synthesis of new generations of so-called hybrid nanocrystals (HNCs) with a topologically controlled distribution of their composition [4, 38-40]. HNCs are structurally elaborated multi-component NCs, consisting of two or more different material sections interconnected through permanent chemical bondings, eventually forming heteroepitaxial interfaces. The technological horizons that these HNCs can potentially spam are incredibly vast. First, the grouping of luminescent, magnetic and catalytic sections without bridging molecules or embedding matrices yields individually processable particles in which different properties are simultaneously available. This is particularly beneficial to boost those applicative fields, which naturally demand for “smart” platforms able to accomplish multiple tasks, such as biomedicine, environmental clean-up, catalysis, sensing, composite material fabrication. Second, due to the electronic contact that is established across adjacent material domains, HNCs can display modified or even unexpected physical-chemical responses, compared to those normally exhibited by the individual components. This suggests that forming an intimate junction among dissimilar nanocrystalline portions can be exploited as an additional tool to engineer the performances of complex nanomaterials. Third, exchange coupling mechanism among magnetic, optical and electrical properties may be operative in HNCs, thereby laying the foundations for exploiting novel spintronic concepts and devices. In this Chapter, we will describe and discuss state-of-art surfactant-assisted strategies that have been proposed to synthesize nanoheterostructures with a topologically controlled distribution of their chemical composition. We will show how the ability to prepare homomaterial NCs with finely tuned geometric parameters has been extended into elegant “seeded growth” approaches for accessing multimaterial HNCs by controlling

Hybrid nanocrystals

409

parameters, such as interfacial lattice strain and surface energy, in liquid media. Various topological configurations will be analyzed, including concentric core/shell architectures, hetero-oligomers grouping spherical material domains, and more spatially elaborated HNCs based on anisotropically shaped material sections. Some relevant properties and technological advantages offered by such newly conceived generations of complex nanocrystals will be also succinctly highlighted.

2. General concepts in surfactant-assisted synthesis 2.1. Single-material nanocrystals Colloidal NCs are synthesized in solution in the presence of some stabilizing organic ligands or of amphiphilic surfactants. As extracted from their growing medium, NCs typically comprise a crystalline core with the desired chemical composition, which is responsible for their intrinsic chemical-physical properties, and a monolayer of tightly bound capping molecules, conferring solubility and colloidal stability to the particles. This basic structure makes NCs technologically appealing for two main reasons. First, their production can be cheaply scaled up to milligram-to-gram quantities for many materials by using a relatively inexpensive equipment. Second, they are surface accessible, which allows them to be processed and implemented into current technologies. In particular, surfactant-assisted syntheses have proven to be the most powerful wetchemistry tools to tailor the size, the shape and the composition of NCs [2-3, 5]. Indeed, judicious adjustment of a few experimental parameters, such as the type and the relative concentration of molecular precursors, and organic stabilizers, in combination with temperature, transcribes into a control over a number of complex phenomena occurring during crystal formation in liquid media, including solution supersaturation, reactant diffusion, relative stability of crystal polymorphs, and crystallographic-directiondependent growth rates [2-3, 5]. The general scheme of a colloidal synthesis involves thermally activated reaction of organometallic precursors (that carry the atomic species necessary to build the desired material) in the presence of selected surfactants (e.g., alkyl amines, thiols, carboxylic and phosphonic acids, phosphines, phosphine oxides, etc). Once the synthesis is activated, the highly reactive species that are generated, commonly referred to as the “monomers”, induce the nucleation of NCs and sustain their subsequent enlargement. The surfactants play several key roles during the synthesis. Indeed, they form complexes with the monomers, dictating their actual chemical potential in solution. Simultaneously, they participate in an adsorption/desorption dynamics on the surface of the growing clusters, which prevents aggregation and uncontrolled growth. Mechanistic studies have shown that monodisperse NCs are produced when a burst of nucleation that enables separation of the nucleation and growth processes is combined with diffusioncontrolled growth. Satisfactory tailoring of the NC size and size distribution can be achieved by balancing the relative depletion of monomers between the nucleation and the growth stages either with the aid of suitable techniques (e.g., the “hot injection”), or profiting from the particular reactivity of the system (e.g., “delayed nucleation”), or exploiting digestive-ripening [2-3, 5]. It is also worth mentioning that surfactants can affect the specific surface energy of the growing NCs, which has important implications in the tuning of their shape [2-4]. Indeed, facet-preferential ligand adhesion can modify the relative growth rates along the various crystallographic directions and/or can favour

410


selective elimination of unstable surfaces by triggering oriented attachment of particles along defined crystallographic directions. In the absence of additional circumstances that can interrupt growth symmetry (e.g., formation of soft micelle templates, the presence of foreign particle catalysts, or the application of external electric or magnetic fields), surfactants remain mostly responsible for the formation of NCs in a variety of anisotropic shapes, such as cubes, polyhedrons, rods, wires, polypods, and rings. These topics are addressed in detail in other chapters of this book.

2.2. Multi-component hybrid nanocrystals The most recent developments of colloidal techniques involve elegant extensions of the above described surfactant-assisted routes to the fabrication of more elaborate hybrid

Scheme 1. Comparative sketches illustrating heterogeneous growth of nanoheterotructures: (a-c) MBE/CVD techniques from gas-phase precursors; (d-f) Seeded-growth approaches in the solution phase (reproduced from ref. 39 with permission, copyright 2008 Wiley-VCH Verlag GmbH & Co. KGaA).

Hybrid nanocrystals

411

nanocrystals (HNCs). The most widely exploited solution-based strategies for the synthesis of HNCs rely on a simple principle of the Classical Nucleation Theory (CNT), according to which the activation energy for the growth of pre-existing particles in a liquid (i.e., heterogeneous nucleation and growth) is considerably lower than the barrier for the generation of new material embryos (i.e., homogeneous nucleation) [41-43]. This concept has been applied in a simple and general “seeded growth” synthesis scheme, in which preformed NCs of one material (referred to as the “seeds”) are already present or introduced into a solution environment containing the molecular precursors necessary to build a second different material. Appropriate reaction conditions may be held, under which the new material preferentially deposits and grows onto such seeds, rather than nucleating homogenously. Despite of the apparent simplicity of this strategy, the preparation of HNCs in liquid media requires an exceptionally higher degree of synthetic ingenuity. In fact, the CNT neither takes into account growth constraints related to lattice incompatibility nor changes in the liquid-solid interface tension induced by adhesion of surfactants to the growing clusters. Therefore, in the preparation of nanoheterostructures comprising structurally dissimilar materials, the thermodynamic and kinetic processes, which determine the size and shape of the individual domains, are further complicated by material miscibility, interfacial strain, and facet-dependent reactivity. In order to rationalize such a complex growth dynamics, it is helpful to consider that the creation of NC-based hybrid architectures by seeded-growth in the solution phase has a close analogy with the growth processes involved in vacuum- and vapour-deposition techniques, such as Molecular Beam Epitaxy (MBE) and Chemical Vapour Deposition (CVD) techniques, which have been traditionally employed for the fabrication of heterostructured systems (e.g., sandwiched multilayer structures, quantum dot or quantum wells onto substrates) starting from gaseous precursors [43-45]. Indeed, the way in which two materials attain an inorganic heterointerface is ultimately related to fulfilment of the same fundamental energy requirements. This parallelism is addressed in Scheme 1. Consider a substrate with a well-defined crystallographic orientation (referred to as “1”in the scheme), onto which a second different material (labelled as “2”) has to be grown. The change in the total surface energy, ∆γ, that accompanies the overall deposition process is given by: ∆γ = σ1 – (σ 2 + γ1,2)

(1)

where σ 1 and σ 2 are the surface energies of the respective materials, and γ1,2 is the solidsolid interfacial energy, which is mostly related to the lattice-misfit-induced strain between the two lattices. Which growth mode will be adopted in given system will depend on the balance between these terms. When the material to be added is characterized by lower surface energy (σ 2 < σ 1) and good lattice fit (i.e., γ1,2 is low) with respect to the substrate, such that ∆γ > 0, then the deposition can proceed layer-by-layer toward a uniform coverage (panel a: Franck - van der Merwe mode). When the foreign material possesses high surface energy (σ 1 < σ 2) and/or it is significantly latticemismatched (i.e., γ1,2 is high), such that ∆γ < 0, then its deposition can take place only via formation of island-like features (panel b: Volmer-Weber mode). An additional possibility includes a mixed deposition regime, in which growth initially takes place

412


layer-by-layer, and then continues by producing islands in response to sudden increases in interfacial strain along the course of the deposition (panel c: Stranski-Krastanov mode). All of these growth regimes can be equally transcribed to the context of a seeded growth synthesis of HNCs, whereby NC seeds act as the substrates for deposition of a foreign material. For instance, if the conditions for complete wetting were either satisfied for any of the facets exposed by a starting seed, or selectively for only one or a few of them, the resulting HNCs could assume an onion-like or a dimer-/oligomer-like configuration, respectively (panel d). On the other hand, under establishment of an incomplete wetting regime, HNCs could be originated, for which one or more sufficiently extended facets of the original seeds are decorated with multiple domains of the new material (panel e). As intermediate case, transformation from metastable onion-like architectures to phase-segregated heterostructures could be expected to be a convenient path toward lowering of interfacial strain (panel f). At this point, it deserves emphasis that the creation of nanoscale heterointerfaces in colloidal solutions can benefit from the possibility of modulating the solution/solid interfacial tension (i.e., the σ1 and σ2 terms) by means of surfactants, an opportunity that is prohibited to MBE/CVD techniques. Indeed, the outstanding results that have been so far reported have highlighted that HNCs made of a given combination of structurally dissimilar materials can be accessed in more than one topological arrangement, when the interface energy term is properly balanced by the surface energy contributions. Moreover, in the liquid-phase synthesis of HNCs unconventional mechanisms (not fully understood yet) can come into play, through which misfit-induced interfacial strain can be accommodated (e.g. by local curving of nearinterface junction planes) or compensated for (e.g., by a decrease in the total surface energy due to solvent or surfactant binding, or via elimination of unstable facets following the interface formation). This may lead to heteroepitaxial interfaces with poor defect density, which is a critical condition to guarantee reproducible physical-chemical properties and intense exchange coupling effects [46].

3. Core/shell geometries The most common topology in which HNCs have been reported is the so-called core@shell geometry, in which a NC is uniformly enveloped by one or more layers of other materials. As a general advantage, associations of various semiconductors, metals and oxides in onion-like configurations often exhibit distinct behaviour as compared to those inherent to the individual components, such as tunable optical and/or magnetic properties, or enhanced catalytic activity, depending on the specific combination. Core@shell HNCs with epitaxially fused core and shell sections can form when the involved materials have similar crystal structure and lattice parameters (with differences of the order of less than 1-3 %), so that overall structure experiences negligible strain as long as the coating thickness is small enough. Nevertheless, when the synthesis conditions allow the interfacial energy to be somewhat counterbalanced by a proportional decrease in surface energy (see eq. 1), then the requirements of lattice compatibility can be fairly less restrictive. Colloidal strategies to access core@shell HNCs have proliferated in the last decade. They can be classified into six main categories, all of which are regardable as variants of the general seeded-growth scheme: (a) direct heterogeneous nucleation and growth of the

Hybrid nanocrystals

413

Scheme 2. Sketch of mechanisms leading to the formation of core@shell HNCs: (a) direct heterogeneous nucleation and growth of the shell material onto preformed nanocrystal seeds; (b) shell growth after chemical activation of the seed surface; (c) sacrificial red-ox replacement of the seed surface followed by vacancy coalescence; (d) self-controlled nucleation-growth; (e) thermally driven phase-segregation; and (f) solid-state diffusion (reproduced from ref. 39 with permission, copyright 2008 Wiley-VCH Verlag GmbH & Co. KGaA).

shell material onto preformed NC seeds; (b) shell growth after chemical activation of the seed surface; (c) sacrificial red-ox replacement of the outer seed layers; (d) one-pot approaches by self-controlled nucleation-growth; (e) thermally driven phase-segregation; and (f) solid-state diffusion. These mechanisms are sketched in panels a-f of Scheme 2, respectively.

3.1. Direct heterogeneous nucleation-growth of the shell Disparate classes of core@shell structures, consisting of combinations of metal, semiconductor and dielectric materials, have been obtained by selective heterogeneous nucleation and growth of a second material on preformed NC seeds (Scheme 2, path a).

414


Strategies to achieve this goal aims at circumventing parasitic homogeneous nucleation of the shell material by performing a slow addition of the relevant molecular precursors to the core seeds at mild temperature, and/or by using ligands to control the reactivity of the system. Representative transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) galleries of core@shell HNCs obtained by this mechanism are reported in Figures 1-3. II-VI and III-V semiconductors. The most developed examples of core@shell HNCs have been based on associations of fluorescent II-IV and III-V semiconductors to target applications ranging from the realization of optoelectronic devices (e.g., lightemitting diodes, lasers, solar cells) to biological labelling and sensing [14-17]. Typically, organometallic precursors (e.g, Cd(CH3)2, Zn(C2H5)2, Se and S powders, metal oxides and acetates, S[Si(CH3)3]2) and highly boiling coordinating solvents, such as tri-noctylphosphine (TOP), tri-n-octylphosphine oxide (TOPO), hexadecylamine (HDA), dodecylamine (DDA), or their mixtures, are the key ingredients to synthesize highquality HNCs. In these associations, the involved materials are often characterized by a close structural similarity, which favours good epitaxial growth of the outer shell, in turn enhancing the robustness and the photoluminescence quantum yield of the core. Depending on the relative staggering of the band edges of the involved material components, the charge carriers can be forced to recombine radiatively either in the core region (e.g., in CdS@ZnS and CdSe@CdS HNCs with type I alignment [47-53]) or at the core/shell interface (e.g., in InAs@CdSe and CdTe@CdSe HNCs with type II alignment [54-58]). In the latter case, emission can be obtained in a wavelength range inaccessible to their isolated components (e.g. infrared emission from a NC made of a high-band gap material), which is a clear demonstration of a novel property that can be de-novo introduced by proper engineering of core@shell arrangements. The readers are referred to Chap. 11 for an exhaustive treatment of these topics. Increasingly sophisticated core@shell architectures have been synthesized in order to circumvent the dramatic photoluminescence abatement that is normally observed as the shell exceeds a critical thickness. Such deterioration is correlated with the occurrence of structural defects, such as dislocations, which form when the strain fields at the latticemismatched heterojunction region cannot be tolerated any further. To avoid this drawback, heterostructured shells with a more elaborate composition have been proposed, which incorporate consecutive layers of semiconductors possessing gradually differing lattice parameters (e.g., in CdSe@CdS@ZnS, CdSe@ZnSe@ZnS, or InAs@CdSe@ZnSe associations) [59-61]. Such sandwiched configurations are synthesized by sequential additions of the respective material precursors and guarantee elevate and temporally stable photoluminescence quantum yields, since the lattice strain can be smoothly accommodated across the graded shell in a quite efficient way (Fig. 1a-f). Other examples include HNCs featuring quantum confinement prototypes, like the “quantum well”-in-“quantum dot” systems structures, such as CdS@HgS@CdS onion-like structures, in which a layer of small band-gap HgS is buried between a spherical core and an outer shell, both made of higher band-gap CdS [62-64]. In this case, carriers are confined in the intermediate HgS domain and recombine radiatively from this region. Though the shell habit normally conforms to the symmetry of the starting core, remarkable deviations from this growth mode can be achieved by sophisticated manipulation of the particle formation kinetics. For example, if the shell deposition is

Hybrid nanocrystals

415

Figure 1. Gallery of TEM and HRTEM images showing examples of core@shell HNCs, composed of combinations of semiconductor materials, which have been synthesized by direct heterogeneous nucleation of the shell material onto preformed NC seeds. seeds. (a) Starting CdSe seeds, (b) CdSe@CdS, and (c) CdSe@CdS@ZnS spherical HNCs prepared by consecutively growing CdS and ZnS shells around the CdSe cores. TEM and HRTEM images of: (d) Starting CdSe seeds, (e) CdSe@ZnSe, and (f) CdSe@ZnSe@ZnS HNCs prepared by consecutively growing ZnSe and ZnS shells around the CdSe cores (reproduced in part from ref. 60 with permission, copyright 2004 American Chemical Society). (g-h) CdSe@CdS HNCs obtained by growing a rod-like CdS shell onto spherical CdSe seeds. The inset reports a “mean dilatation” mapping, obtained from the analysis of the corresponding HRTEM image, which shows an area within the rod where lattice parameters are altered by 4.2% with respect to the reference area, due to the presence of the CdSe core (reproduced in part from ref. 66 with permission, copyright 2007 American Chemical Society).

accomplished under conditions promoting anisotropic crystal growth, initially spherical CdSe seeds can be coated with one elongated CdS section, forming noncentrosymmetric CdSe@CdS nanorods [65-66]. These hybrid nanorods (Fig. 1g-h) can be prepared with excellent size and shape monodispersity, which promotes their lateral or vertical

416


organization in close-packed assemblies over square micrometer areas onto solid substrates upon application of controlled-solvent-evaporation techniques or under assistance of electric fields [65-67]. Alternatively, it has been shown that phasecontrolled CdSe seeds can individually serve as branching points, onto which multiple straight arms of a second material (such as of CdS or CdTe) can be grown, providing access, for example, to tetrapod-like heteroarchitectures [67-68]. These HNC arrangements show optoelectronic properties, such as large Stokes shifts, linearly polarized emission, or enhanced charge carrier separation, which are indeed characteristic of one-dimensional nanostructures [65-68]. Transition Metals. The impact of shell deposition on the resulting physical-chemical properties is even more dramatic in core@shell HNCs based on noble metals (such as Ag, Au, Cu, Pd, Pt) and other transition elements (e.g. Co, Ni), in which a significant degree of interfacial alloying often leads to strong exchange coupling or electric-field enhancement effects. Depending on the specific material association, remarkably altered surface plasmon oscillations, modified magnetic parameters, improved SERS responses or extreme catalytic activity are observed, with respect to those offered by the individual components (a more extensive treatment of this topic can be found in Chap.6 and Chap. 7). Especially HNCs including coinage and magnetic metals represent bifunctional platforms whereby interesting magnetic and optical properties are simultaneously available, which therefore hold great promise in nanoparticle-based diagnostic and therapeutic applications [69-76]. Such bimetallic HNCs are generally synthesized in aqueous media, in which preformed core seeds of the desired material are mixed with the required foreign metal ion precursors in the presence of suitable surfactants and of a mild reducing agent (e.g., ascorbic acid, hydroxyl amine) [69-72, 74-77]. The seeds behave as powerful redox catalysts, selectively enhancing heterogeneous nucleation by fast ion reduction at their surface. Under the assistance of facet-preferential adhesion of surfactants (e.g. tetraalkylammonium bromide salts) or additives (e.g. NaOH, gaseous NO2), this scheme has been generally applied to achieve the growth of monometallic (e.g, Au, Ag) nanorods and nanowires [78-81]. More recently, a refined level of shape control has been reached in the synthesis of Me@Pd (where Me=Au, Pt) [76-77] and Au@Me (where Me=Pt, Ni) [82-83] heteronanostructures, in which relatively thick and uniform shells have been successfully deposited onto faceted cores (Fig. 2a-c). In these systems, the evolution of the overall HNCs toward a rod-, cubic-, or polyhedral-morphology has been explained in terms of competing overgrowth on the different crystal facets that the nascent shell exposes to the solution precursors [76-77]. In the particular case of Pt@Pd [76], the attainment of a fully conformal Pd shell around the starting cores (as assessed by advanced transmission electron microscopy imaging) has been correlated to the negligible lattice misfit experienced by the two interfaced materials. This explanation is indirectly supported by the observation that Au deposition onto cubic Pt NCs does not follow a core@shell growth mode, but instead leads to heterostructures made of a Au rod-like section partially embedding a single Pt cube at its sidewall. Actually, since the Au:Pt mismatch is as high as 4.08%, Au can wet the Pt seeds to a limited extent only, and subsequent Au growth proceeds preferentially on the high-energy Au domain initially nucleated [76]. Finally, a few examples of nanoheterostructures incorporating both transition-metals and fluorescent semiconductors are worth mentioning. These include Co@CdSe HNCs

Hybrid nanocrystals

417

Figure 2. Gallery of representative scanning electron microscopy (SEM), TEM and HRTEM images showing examples of core@shell HNCs, made of metal and semiconductor associations, which have been synthesized by direct heterogeneous nucleation and growth of the shell onto preformed NC cores. SEM and TEM studies of: (a) Au nanorod seeds and (b-c) Au@Pd HNCs prepared by growing a Pd shell onto the Au cores (reproduced from ref. 77 with permission, copyright 2006 American Chemical Society); (d) Co@CdSe HNCs (reproduced from ref. 84 with permission, copyright 2005 American Chemical Society); (e) FePt@CdS and (f) FePt@CdSe HNCs synthesized by sequential addition of Cd/S and Cd/Se precursors, respectively, onto FePt seeds (reproduced from ref. 854 with permission, copyright 2007 American Chemical Society).

(Fig. 2d) and FePt@CdX (X=S, Se) HNCs with a polycrystalline CdX shell (Fig. 2e-f), which have been produced by one-pot reaction schemes, whereby magnetic seeds are first nucleated in-situ and subsequently allowed to react with the cadmium chalcogenide precursors in conventional coordinating mixtures [84-85]. Interestingly, these nano-objects still retain the superparamagnetic properties of the respective cores (though little changes in the blocking temperature values are observed), while complete quenching of the shell luminescence due to metal-driven charge carrier separation is not observed [84, 85]. Material associations including metal oxides. In the past decade, a variety of core@shell HNCs incorporating at least one oxide component have been made available by direct heterogeneous nucleation approaches. Metal@oxide HNCs, in which the core is made of either Au or Ag, and the shell is composed of either TiO2, ZrO2, or SnO2, have been grown in mixed organic/aqueous media by two-step procedures, generally involving hydrolysis-condensation of transition metal alkoxides in the presence of surfactantcapped metal NCs, provided by an independent synthesis. By this way, shells with either a smooth [86] or flower-like morphology [87-88] are obtained (Fig. 3a-b). With a few exceptions, all the aforementioned routes typically yield amorphous coatings, which often

418


exhibit discontinuities, nanoporosity and/or even inner cavities. Profiting from the chemical accessibility of their inner cores, these metal@oxide HNCs have been exploited as catalytically active nanoreactors [29] as efficient electronic capacitors [89] as well as sacrificial templates for the creation of hollow oxide capsules upon selective metal oxidation [90, 91]. In addition, these systems reveal shell-tuneable surface plasmon absorption and interesting nonlinear optical properties with high laser damage threshold [28, 86]. Another class of core@shell structures that have been obtained by this scheme is represented by oxide@metal HNCs combining ZnO or TiO2 with either Au, Ag, Pt, or Cu. Typical preparations start from an alcoholic suspension of weakly organic-passivated oxide NCs onto which either chemical [92-95] or oxide-photocatalyzed reduction [96-97] of the desired metal precursors is accomplished. Depending on the specific experimental conditions used, the deposition process can achieve various degrees of surface coverage, leading to oxide NCs decorated with small island-like metal patches or entirely coated with a shell of irregular thickness. The pronounced light sensitivity of such noble metal nanostructures has been verified in laser irradiation experiments, during which initially well-separated oxide@metal HNCs undergo photoinduced fusion into larger composite structures, each comprising multiple oxide domains embedded in metallic matrix [94]. Although detailed information on the shell structure has not been provided for such nanostructures, the attainment of true inorganic nanojunctions has been proven to have a remarkable impact on the applications of these HNCs in solar energy harvesting. Indeed, under band-gap photoexcitation of the semiconductor section, the metal component acts as a sink for the photogenerated electrons, thereby circumventing their recombination with holes. The altered carrier separation dynamics in such HNCs can be advantageously manipulated in order to modulate the metal surface plasmon absorption as well as to benefit either from higher photocatalytic reaction yields or from enhanced charge storage capability under Fermi level equilibration. A few attempts to oxide-semiconductor core@shell associations have been also reported. These involve the coupling of ZnO with CdSe or CdS [98-99]. These HNCs are appealing candidates for photovoltaic and photocatalytic applications, since relative band edge staggering of these materials allows the cadmium chalcogenide to act as visible sensitizer for the oxide material. One example is represented by ZnO nanorods decorated with irregularly shaped crystalline CdS domains that have been grown by ultrasonication of aqueous suspensions containing oxide seeds, CdCl2 and thiourea [98]. This method permits only a rather limited control over the heterostructure topology, as CdS nucleation is not sitespecific. Although neither mechanistic aspects nor the involved epitaxial relationships have been clarified, the superior transport and gas sensing properties of these ZnO@CdS HNCs relative those of ZnO alone suggest that good electronic connection should exist between the two materials, which could have implication in the design and fabrication of highly performing optoelectronic devices [98]. An inverted CdSe@ZnO configuration has been reported by means of an approach that somewhat resembles a phase transfer procedure [99]. Highly hydrophobic surfactant-capped CdSe seed NCs, initially soluble in nonpolar media, are mixed with alcoholic solution of zinc acetate under ultrasonication, which favours detachment of the native ligands and subsequent adsorption of the zinc monomer species. This CdSe surface activation process has been considered critical to guarantee the growth of a ZnO layer upon hydrolysis of the precursor. Since acetate moieties ultimately protrude out

Hybrid nanocrystals

419

Figure 3. TEM gallery of core@shell HNCs based on oxide materials, which have been synthesized by direct heterogeneous nucleation and growth of the shell material onto preformed nanocrystal cores. Representative low-magnification TEM images of: (a) Au@TiO2 nanoreactors with inner vacant space (reproduced from ref. 88 with permission, copyright Wiley-VCH Verlag GmbH & Co. KGaA); (b) Ag@TiO2 HNCs (reproduced from ref. 86 with permission, copyright 2006 American Chemical Society); (c-e) Fe-oxide@Au HNCs in panel e obtained by thermally induced coalescence of Au NCs in panel d onto the Fe-oxide seeds in panel c (reproduced from ref. 101 with permission, copyright 2007 American Chemical Society); (f) Au@Fe3O4 HNCs (reproduced from ref. 104 with permission, copyright 2006 American Chemical Society); (g) Pt@Fe2O3 HNCs (reproduced from ref. 106 with permission, copyright 2005 IOP Publishing).

420


of the shell into the solution, then CdSe cores are rendered dispersible in the polar alcoholic medium. As an indirect proof for the formation of a core@shell structure, an increase in the CdSe photoluminescence quantum yield has been reported, although more precise structural confirmations have not been supplemented [99]. Another interesting category of bifunctional core@shell HNCs includes nanostructures in which a layer of a noble metal, such as Au or Ag, surrounds an interior made of either iron or iron oxide [100-103]. A key advantage of these Fe/Fe-oxide@metal HNCs resides in the association of a biocompatible core, potentially useful as magnetic resonance imaging (MRI) contrast agent or exploitable in magnetically driven separations, with an optically active surface that is easy to functionalize. In one approach, organic-capped Fe3O4 NCs act as substrates for the alkyldiol reduction of gold acetate in the presence of oleic acid (OLAC) and oleyl amine (OLAM) at ~180°C [100]. Such temperature is needed to regulate the adsorption/desorption dynamics of the surfactant on the Fe3O4 surface of the seeds, such as to render them accessible to the reactive gold species. An appealing variant of this procedure relies on the thermally induced fusion of pre-synthesized OLAC/OLAM-capped Fe3O4 and thiol-capped tiny Au NCs codissolved in toluene in the presence of an ammonium bromide salt [101]. During the heating of this mixture at 150°C in a close vessel, the less stable Au NCs coalesce onto the Fe3O4 NCs, forming a uniform metal coverage on such seeds as a means of reducing the overall surface energy of the system (Fig. 3c-e). Room temperature aqueous syntheses have been also reported, although the quality of the resulting nanostructures in terms of size distribution and control over the aggregation degree is lower. Water-soluble γ-Fe2O3@Au HNCs have been obtained by hydroxylamine reduction of a gold salt in aqueous tetrabutylammonium-stabilized suspensions of oxide NCs [102]. In another approach to Fe3O4@Ag HNCs, water-in-oil micro emulsion droplets have been used to confine the reaction of silver nitrate with borohydride in the presence of suspended Fe3O4 NCs [103]. In both cases, since the seeds are only weakly surface passivated, the deposition of a rather thick shell can readily occur under mild conditions. Occasionally, structures made of multiple cores embedded within a single shell matrix have been found [100-103]. Fe-oxide@metal HNCs exhibit the characteristic surface plasmon absorption of nanosized Au/Ag, which correlates with the shell thickness [100-102]. As for what concerns their magnetic properties, lower blocking temperatures and higher coercive fields have been measured relative to those found for the corresponding Fe3O4 cores, which has been explained by a decreased coupling of the particle magnetic moments due to screening effect of the metal shell. Notably, the versatile surface chemistry of gold and the magnetism of iron oxide have paved the way to prototypical bio-applications of these hybrid nanostructures after water transfer and shell functionalization with proteins, such as in bioassays and magnetic-field-assisted separation [101]. The use of Fe3O4@Ag HNCs as easily recyclable antibacterial agents has been also explored [103]. Sets of inverted metal@Fe-oxide HNCs have been also devised [104-106]. Monodisperse Au@Fe3O4 HNCs (Fig. 3f) with either spherical or cube-like shell and good epitaxial connection between the two materials have been synthesized by decomposing Fe(CO)5 at 200-300°C in the presence of small Au seeds and OLAC/OLAM/alkyl ether media [104]. The interactions between the two components in these heterostructures are manifested by a pronounced red-shift of the Au plasmon

Hybrid nanocrystals

421

resonance due to the dielectric coating, and by an increased magnetic coercitivity due to extra surface anisotropy. By using a growing environment analogous to that described above, highly regular Pt@Fe2O3 HNCs have been obtained in OLAC/OLAM/octyl ether mixtures at 290°C by performing one-pot sequential reaction scheme, in which Pt NCs are first nucleated in-situ upon diol reduction of platinum acetylacetonate, and subsequently combined with the iron precursor for shell growth (Fig. 3g) [105]. The relative geometric parameters of the HNCs are easily engineered by adjusting the relative reactant concentration ratio. The Pt@Fe2O3 HNCs have been converted by a high temperature solid-state reaction under reducing atmosphere into bi-magnetic FexPt1-x@Fe heterostructures possessing exchange-coupled soft and hard phases as well as high coercitivity, which are potentially appealing to fabricate hard magnets and data storage media [106]. Bi-magnetic FePt@MFe2O4 HNCs (where M= Fe, Co) with a coherent core-shell interface have been also accessed directly by surfactant-assisted two-step solution synthesis [107-108]. In the reported procedure, FexPt1-x seeds, obtained by simultaneous platinum acetylacetonate reduction and iron carbonyl decomposition, are covered with a MFe2O4 shell upon co-decomposition of cobalt and iron acetylacetonate in suitable proportion in the presence of OLAC and OLAM stabilizers. The relevant magnetic data indicates that effective exchange interactions are established between the hard FePt and the soft MFe2O4 phases, whose relative proportions dictate the overall coercitivity value. Therefore, the chemical engineering of geometric parameters of these HNCs allows their magnetic properties to be finely tuned for tailored applications. By using a slight modification of the above reported procedures to Fe-oxide shell growth, other material combinations have been accessed by seeded techniques [109-111]. For example, biocompatible Me@Fe2O3 HNCs (where Me = Co or SmCo5) have been conjugated via a robust dopamine bridging to nitrilotriacetic acid possessing high specificity for binding histidine-tagged proteins [109]. In another report, CoFe2O4 NCs have been used as seed substrates to be coated with Fe3O4, unexpectedly leading to CoFe@Fe3O4 HNCs [110]. Although the involved formation mechanism has not been clarified, this work proposes an accurate methodology to gather in-depth structural and compositional information on core@shell nanostructures, which is a challenging task when the shell is either polycrystalline or shares an incoherent or lattice-mismatched interface with the core [110]. Finally, a technologically appealing approach to hard nanomagnets has been devised, in which solution-phase synthesis is coupled with a solidstate reaction. The proposed methodology relies on subjecting colloidal Co@Sm2O3 HNCs to high-temperature reductive annealing, which yields hexagonal close-packed nanocrystalline SmCo5 exhibiting large magnetic coercitivity [111].

3.2. Shell growth on functionalized seeds A broad category of procedures to encapsulate NCs of a variety of materials with a SiO2 or ZrO2 shell relies on a “priming” step to activate the seed surface [15, 63-73, 110, 112-123]. To this aim, appropriate silicon or zirconium organometallic compounds are selected which can bind to the surface of NC seeds by means of their functional moiety (e.g., an amine, carboxyl or thiol group), while exposing alkoxide groups. The latter are hydrolyzed, forming a primary polymerization layer from which a metal oxide network is progressively built upon sequential hydrolysis-condensation reactions with metal

422


alkoxide monomers (see Scheme 2, path b). The procedure for growing a silica shell, based on the so-called Stöber method [124], has been the most widely investigated and various modifications (e.g. the use of alternative coupling agents, such as polymers or gelatine, or the exploitation of micelle reactors) have been introduced [15, 114-118, 122, 124-134]. The outer shell surface can be itself provided with additional chemical moieties, thus serving as platform for implanting functionalities which would be otherwise hard to anchor onto as-synthesized NCs. Due to the amorphous nature and porous structure that characterize this type of oxide coatings, they do not cause accumulation of significant strain within the resulting hybrid nanostructure, so that remarkably thick shells are straightforwardly obtained. Examples of core@shell HNCs prepared by means of this techniques are reported in Figure 4. So far, NC encapsulation with oxide has enabled the realization of disparate objectives. Upon SiO2 coating, high-quality semiconductor NCs (e.g., CdSe, CdSe@ZnS, CdTe, PbSe), which are usually synthesized in organic media and are therefore hydrophobic, have been made soluble in water and buffered solutions with retention of their fluorescent properties and reduced susceptibility toward photodegradation [15, 114118]. A tailored functionalization of the shell surface has further allowed interactions between such HNCs and biological environments to be controlled, while paving the way to the development of refined optical detection techniques [131]. Noble metal@SiO2 HNCs constitute another interesting family of functional nanosized building blocks, for which the shell not only enables dispersion in polar media, but also modifies the surface plasmon oscillations of the metal [124-134]. These robust nanostructures (Fig. 4a-d) provide a framework for creating a variety of space-filling nanostructures (such as inverse lattices, opals, and photonic crystals) and can also act as nanoreactors, whereby reactants can diffuse through the shell, causing modification of the electron density of the cores or its dissolution. These achievements hold clear promise for addressing new catalytic and sensing applications [118, 122, 124, 127, 134]. Also magnetic Fe-oxide, FePt, and Co NCs have been provided with a SiO2 coating (Fig. 4e-f), which does not depress their magnetic properties and preserves them from coalescing under post-synthesis thermal processing [117, 130-132, 134-136]. Moreover, the SiO2 shell effectively prevents toxic heavy metals from being released into the external environment, which is an especially important requirement for these nanomaterials to be practically exploited in biological environments. Consequently, these magnetic@SiO2 structures, conjugated with suitable receptor molecules, have been proposed as a smart platform able to accomplish multiple-tasks, such as specific cell targeting and sorting, and imaging. Complex three-component systems have been recently reported by multiplexed approaches [135, 136]. For example, a reverse microemulsion method has been exploited to synthesize fluorescent-magnetic SiO2 spheres embodying Fe3O4 particles and CdTe quantum dots for immunofluorescence assay purposes [136]. Furthermore, it has been possible to fabricate multilayered nanostructures that consist of a magnetic iron oxide core surrounded by a thick SiO2 shell and further covered with an outer Au layer. They display strong visible and near-infrared resonance absorption and can be manipulated by using an external magnetic field, two appealing characteristics that hold promise in biomedical applications [135].

Hybrid nanocrystals

423

Figure 4. Gallery of core@shell HNCs synthesized by chemically activating the surface of seed cores for silica shell growth. The panels are low-magnification TEM images of: (a-b) Au nanorod@SiO2 HNCs (reproduced from ref. 129 with permission, copyright 2006 American Chemical Society); (c-d) Au@SiO2 HNCs (reproduced from ref. 125 with permission, copyright 1996 American Chemical Society) (e-f) FePt@SiO2 HNCs (reproduced from ref. 132 with permission, copyright 2006 American Chemical Society).

Finally, a case of ZrO2 coating accomplished by a procedure analogous to that used for silica shell growth process, is worth mentioning [137]. In the reported work, a priming of Ag nanoparticles with mercaptobenzoic has been utilized to help subsequent reaction with zirconyl chloride, thereby initiating the formation of a ZrO2 shell. The resulting Ag@ZrO2 HNCs can incorporate metal ions through the permeable oxide, which can be indirectly probed by monitoring corresponding changes in the surface plasmon resonance of Ag. This finding anticipates the possible use of these materials in ion sensing.

3.3. Shell formation by red-ox reactions Another commonly pursued strategy towards core@shell HNCs relies on the sacrificial conversion of the outermost exposed layers of a starting NC core into a

424


different material by a galvanic replacement reaction (Scheme 2, path c) [138-148]. In this respect, many transition metal NCs are potentially useful substrates, since they are easily oxidized when exposed to either air, solvated oxygen species, or other oxidizing reagents (e.g. other metal ions). In addition, such red-ox conversions even permit the creation of oxide core@shell HNCs where the mean oxidation state of metal atoms located in the core is different from that in the shell section. Examples of as-synthesized HNCs are shown in Figures 5-6. A set of magnetic-metal-based core@shell bimetallic HNCs has been prepared through performing transmetalation reactions in organic media [138-140]. Metal ion precursors (such as Au3+, Pt2+, Pd2+, Cu2+) are reduced at 140-180°C at the surface of surfactant-capped Co (or Fe) NCs to their corresponding zero-valence state at the expense of the topmost exposed Co0 atomic layers, which are, in turn, released as CoII-ligand (or Fe(III)-ligand) complexes to the solution. Providing a magnetic NC with a noble metal

Figure 5. Examples of core@shell HNCs synthesized by performing red-ox reactions on the surface of metal seed cores. The panels show low-magnification TEM images of: (a) Co@Pt HNCs (reproduced from ref. 138 with permission, copyright 2001 American Chemical Society); (b-c) Co@Au HNCs and Co@Cu HNCs (reproduced from ref. 139 with permission, copyright 2005 American Chemical Society); (d) Ag@Au HNCs (reproduced from ref. 141 with permission, copyright 2005 American Chemical Society).

Hybrid nanocrystals

425

shell changes its response from superparamagnetic to ferromagnetic, while protecting the core against oxidative corrosion and offering the shell material as a suitable surface for the attachment of molecules (e.g. to enable HNC transfer into water and biofunctionalization). By a conceptually similar approach, Ag@Au HNCs with tuneable plasmonic properties have been prepared [141], as well as a number of mono- and bimetallic nanostructures with hollow interior under extreme replacement reaction conditions [149-157]. Metal@metal-oxide systems achieved by calibrated oxidation of their respective metal cores deserve to be discussed more extensively. For example, monodisperse Ni@NiO HNCs [144] and amorphous Fe@Fe-oxide hybrid nanoparticles [146] have been derived from TOPO-capped Ni seeds and OLAM-capped Fe seeds, respectively, following the extraction/cleaning procedures that are carried out under ambient atmosphere. These superparamagnetic objects are potentially appealing in the biomedical field, for in vivo and in vitro applications, such as biomedical separation, MRI contrasting, magnetic-field-assisted drug delivery, hyperthermal treatment of cancer cells. Promising results in this direction have been authenticated for the Ni@NiO HNCs, whereby the NiO surface has been selectively functionalized with histidine-tagged proteins for magnetically assisted recovery [144]. By an analogous route, Co@CoO HNCs have been synthesized by subjecting pre-synthesized ferromagnetic (FM) Co NCs to either controlled bubbling of oxygen or simple exposure to air, which initiates the progressive deposition of an antiferromagnetic (AFM) CoO shell, until full particle oxidation is eventually reached [142-143]. Inverted AFM/ferrimagnetic(FiM) MnO@Mn3O4 HNCs have been derived from surface oxidation of MnO NCs, prepared by surfactant-assisted pyrolysis of manganese acetlyacetonate [145]. These types of HNCs are technologically appealing because the intimate communications between the FM (or FiM) section and the AFM portion produces enhanced unidirectional magnetic anisotropy, being responsible for the so-called “exchange bias”, that improves the thermal stability of the magnetic moments of the FM (or FiM) phase [142-143, 145, 158159]. These nanoheterostructures, for which exchange bias effects can be modulated by proper topological design [142-145] are candidates for high density magnetic and writing recording media [46, 160]. More recently, unprecedented advances toward simultaneous morphological and compositional control have been achieved with the synthesis of sophisticated “yolk-shell HNCs”, via a mechanism known in metallurgy as the “Kirkendall effect”, an atomic diffusion process that takes place through vacancy exchange rather than by direct interchange of atoms [161-162]. It is indeed established that, in a spherical object, made of a fast-diffusing core material and an outer layer or reservoir of slower diffusing material (e.g. a metal particle that has been surface passivated with an oxide layer), net transport of matter from the core outwards may occur, with consequent formation of voids. Under sufficient supply of thermal energy, a huge fraction of the vacancies can ultimately coalesce into a single large void (Scheme 2, path c). The Kirkendall effect has been first verified in nanoparticles in the case of reactions between Co NCs and O2, sulfur or selenium at moderate temperatures, upon which hollow CoO, CoSe or Co3S4 NCs are formed [163-165]. However, if surface oxidation and vacancy diffusion are allowed to take place selectively on the shell portion of pre-formed core@shell HNCs, then even more topologically complex yolk-shell architectures can be produced, in which a void

426


Figure 6. Structural evolution of core@shell Fe@Fe3O4 HNCs (a-b), to yolk-shell Fe-Fe3O4 HNCs (c-d), and finally to hollow Fe3O4 NCs (e-h) during controlled oxidation of Fe nanoparticles with molecular oxygen (reproduced from ref. 148 with permission, copyright 2007 American Chemical Society).

space separates the core from shell. Relevant demonstrations include Pt@CoO and FePt@CoS2 HNCs, obtained upon reacting Pt@Co or FePt@Co core@shell HNCs with O2 or sulfur, respectively [163, 166], and Fe@Fe3O4 yolk-shell HNCs, synthesized by temperature-controlled oxidation of monodisperse amorphous Fe nanoparticles with either O2/Ar flow or trimethylamine N-oxide in octadecene/OLAM solution [147-148]. For the latter system, the entire evolution from amorphous Fe@Fe-oxide core@shell to Fe@Fe3O4 yolk-shell nanostructures, and finally to hollow Fe3O4 NCs has been clearly

Hybrid nanocrystals

427

observed (Fig. 6). These findings have greatly deepened the understanding of inward/outward diffusion properties that can be induced in colloidal nanoparticles, suggesting useful routes to manipulate them to achieve improved synthesis design [162]. In addition to being ascertained by direct electron microscopy inspection, the internal structure of the yolk-shell HNCs has been assessed by further interesting studies. For example, like bare Pt, Pt@CoO yolk-shell HNCs have been found to be catalytically active in ethylene hydrogenation, thus confirming that small molecular reactants and products can permeate through grain boundaries of the CoO shell and reach the surface of the inner Pt core [163, 167]. In a set of biological experiments, FePt@CoS2 yolk-shell HNCs, dispersed in aqueous medium, have been observed to behave as powerful killing agents for HeLa cells. Indeed, after cellular uptake and exposure to the intracellular acidic environment, the highly reactive FePt cores are oxidized, releasing toxic Pt2+ ions, that diffuse out of CoS2 porous barrier and then enter the nucleus and mitochondria, ultimately leading to cell apoptosis [166]. These achievements are expected to propel future development of new nanomedicine tools for the treatment of cancers.

3.4. Unusual pathways: Self-controlled nucleation-growth, thermally induced phase segregation, and solid-state diffusion Core@shell HNCs have been also accessed following less conventional mechanistic pathways, which will be here briefly revisited on the basis of the representative examples reported in Figure 7. There have been successful efforts to devise one-pot approaches, in which all necessary material precursors are simultaneously present in the same growing solution since the beginning of the synthesis, thus simplifying the overall preparation procedure. Appropriate synthetic conditions may contribute to balance the relative reactivities of the respective material precursors in a way that: (i) two different materials are generated at distinct times; and (ii) the shell material is produced exclusively by heterogeneous nucleation (Scheme 2, path d). The search for such “smart” colloidal systems, in which the nucleation-growth of core and the shell coating stage are selfregulated, has been extremely challenging up to date. Actually, only a few cases are available in the literature. For example, metal@oxide HNCs (where Me=Ag or Au, and oxide=TiO2 or ZrO2) have been derived by refluxing a mixture of a titanium or zirconium alkoxide with the desired noble metal salt in dimetylformamide/water mixture [168-171]. Thermal activation of dimetylformamide-driven reduction of the metal ions leads to a fast in-situ nucleation of metallic seeds, which are subsequently passivated with a thin amorphous oxide shell upon slower hydrolysis-condensation reactions of the alkoxide in the presence of chelating/stabilizing agents. An important property of such Ag@TiO2 HNCs stands within their ability to accumulate electrons within the Ag core upon TiO2 photoexcitation, and in the possibility to utilize such charges to affect either the metal surface plasmon resonance or to accomplish mild reductive reactions [170-171]. Additional examples that nicely illustrate the concept of self-controlled nucleation/ growth mechanism are provided by the circumstances under which Cr@γ-Fe2O3 and Ag@Co HNCs are formed [172-173]. The Cr@γ-Fe2O3 HNCs have been synthesized by thermolysis of Cr(CO)6 and Fe(CO)5 in mesitylene in the presence of a polymer surfactant, which cause the two organometallic precursors to decompose at different rates. The compositional distribution of the resulting HNCs has been found to be

428


consistent with an initial Fe-cluster-catalyzed formation of the Cr cores, followed by the deposition of a Fe shell and its subsequent oxidation [172]. In an analogous way, Ag@Co HNCs have been accessed by combining Co2(CO)8 thermolysis with a trasmetalation reaction [173], namely cobalt-driven AgClO4 reduction, in OLAC/tridodecylamine mixtures, during which a selective nucleation of Ag NCs is followed by the deposition of Co on top of the formed nuclei (Fig. 7a-c). In yet another report, Au@CdSe HNCs have been produced by using Cd-Au alloy nanoparticles as reactants [174]. Heating the latter with Se in TOPO/TOP mixtures at 250°C yields CdSe, driving the coalescence of gold atoms in the precursor alloy to a single segregated Au domain (Fig. 7d). As a result, heterostructures are obtained, in which Au portion is almost fully surrounded by a CdSe shell (Scheme 2, path e). This process is likely driven by the poor miscibility of the nascent CdSe in the starting alloy as well as by the difference in lattice parameters between CdSe and Au [174]. Another interesting report deals with the reaction of organic-capped InAs NCs with a AuIII-alkylammonium bromide complex and DDA as the reducing agent in toluene at

Figure 7. Examples of core@shell HNCs achieved by less conventional mechanisms. (a) TEM images of Co@Ag HNCs obtained by self-controlled nucleation and growth and their electron energy loss spectroscopy (EELS) mapping of Ag (b) and Co (c) content (reproduced from ref. 173 with permission, copyright 2002 American Chemical Society); (d) Au@CdSe HNCs obtained by thermal phase segregation of an alloy (reproduced from ref. 174 with permission, copyright 2005 American Chemical Society; (e-f) formation of Au@InAs HNCs upon gold diffusion in Audecorated InAs HNCs (reproduced from ref. 175 with permission, copyright 2006 Wiley-VCH Verlag GmbH & Co. KgaA.

Hybrid nanocrystals

429

room temperature [175]. At a sufficiently high gold ion concentration, InAs NCs decorated with multiple Au islands are initially obtained, which then evolve into Au@InAs core@shell HNCs under inert atmosphere or into Au@In2O3 HNCs under air (Fig.7e-f). Control experiments have allowed the authors to conclude on the occurrence of a fast solid-state diffusion, according to which Au atoms migrate from the surface toward the interior of the semiconductor via an interstitial-substitution mechanism [176-177]. When the reaction is accomplished under inert atmosphere, the InAs NCs loose their crystalline order along the course of the Au diffusion process, so that the resulting amorphous-like InAs shell becomes readily susceptible to be converted to In2O3 once exposed to ambient oxygen (Scheme 2, path f). The driving force for the process should presumably be the elevate Au-InAs interface energy associated to the multiplyAu-decorated InAs NCs, which can be decreased upon coalescence of the Au patches into a single larger domain.

4. Oligomer-like heteroarchitectures Recently, tremendous advances in the chemical fabrication of increasingly complex architectures have been made with the development of spatially asymmetric HNCs that embody well-segregated size- and shape-controlled sections of chemically and structurally dissimilar materials into one continuous, all-inorganic hybrid particle. Such nanoheterostructures are characterized by an oligomer-like topology that clearly differs from core@shell geometries in that the component domains are interconnected by small junction regions. The elaboration of such types HNCs has a close analogy with the synthesis of progressively larger organic molecules bearing a number of functional groups. With respect to the core@shell arrangement, in which only the outer shell side is chemically accessible, the oligomer-like configuration provides a diversified surface platform onto which a spatially segregated distribution of functional moieties can be eventually implanted. In seeded-growth syntheses of HNCs, various circumstances can contribute to bypass the core@shell-type growth regime, leading to HNCs with a noncentrosymmetric arrangement of their component domains. This will depend on the ultimate surface energy balance accounted for by eq. 1, as discussed earlier. For example, materials that are even poorly miscible and/or quite lattice uncorrelated, could be assembled in the form of heterostructures by reducing the extension of the shared interfaces at the proportionally smaller cost of additional surface energy [178-184]. On the other hand, a small inorganic junction could form as a means to compensate for the high surface energy values which would otherwise characterize separate homoparticles of the respective materials, for example, in the case of ineffective ligand passivation [179, 185-189]. Other growth conditions leading to noncentrosymmetric HNCs could be dictated by the presence of seeds with a site-specific accessibility, arising from differences either in the strength of ligand adhesion on the respective facets, or in their inherent chemical reactivity, or in the degree of lattice matching achievable [16, 104, 187-188, 190-204].

4.1. Hetero-dimers and hetero-oligomers Hetero-dimers and hetero-oligomers grouping two or more nearly spherical NCs of different materials have been synthesized by seeded-growth based mechanisms, such as:

430


(a) direct or (b) surface-activated heterogeneous nucleation; (c) nonepitaxial deposition followed by thermally driven coalescence-crystallization; (d) reactions at liquid/liquid interfaces; (e) self-regulated homogeneous-heterogeneous nucleation; and (f) oriented NC attachment. These pathways are sketched in panels a-f of Scheme 3, respectively. Representative TEM images of therefrom derived HNCs can be found in Figures 8-12.

Scheme 3. Sketch of mechanisms leading to the formation of oligomer-type HNCs: (a) direct or (b) surface-activated heterogeneous nucleation onto preformed seeds; (c) initial nonepitaxial deposition followed by thermally induced coalescence-crystallization; (d) reaction at liquid/liquid interfaces; (e) self-regulated nucleation and growth; (f) reactions among preformed heterodimers (reproduced in part from ref. 39 with permission, copyright 2008 Wiley-VCH Verlag GmbH & Co. KGaA).

Hybrid nanocrystals

431

a) Direct heterogeneous nucleation. This route (Scheme 3, path a) has been widely applied to prepare HNCs made of various associations of magnetic, metal, and oxide materials in the form of binary and ternary oligomers, as illustrated in Figure 8. Such HNCs promise to offer new technological solutions, especially in the field of sensing, imaging, and therapy in the biomedicine. For example, it has been proposed that distinct

Figure 8. Examples of hetero-oligomer HNCs synthesized by direct epitaxial heterogeneous nucleation onto preformed seeds. The panels report TEM images of: (a-b) peanut-shaped and dumbbell-like Au-Fe3O4 HNCs, respectively (reproduced from ref. 190 with permission, copyright 2005 American Chemical Society); (c) Au-Fe3O4-PbS HNCs obtained by nucleating a rod-like PbS section onto Au-Fe3O4 heterodimers seeds (reproduced from ref. 104 with permission, copyright 2006 American Chemical Society); (d) FePt-Au heterodimer HNCs (reproduced from ref. 193 with permission, copyright 2006 American Chemical Society); (e-f) TEM and HRTEM images of CoPt3-Au heterodimer HNCs (reproduced from ref. 194 with permission, copyright 2006 American Chemical Society).

432


material domains in a HNC can be used as platforms for the site-specific anchoring of DNA strands, proteins, and peptides, by exploiting the affinity of specific chemical moieties of such biomolecules toward inorganic surfaces. Furthermore, while a metal or semiconductor portion can enable optical detection, a magnetic domain can be used for complementary purposes, such as MRI and magnetic separation [193, 205, 206]. HNCs with epitaxially joint Au-Fe3O4, PbS-Fe3O4 or FePt-CoFe2O4 sections have been synthesized by thermal decomposition of appropriate organometallic precursors in the presence of preformed Au, Fe3O4 or FePt seeds, respectively, in OLAC/OLAM/ octadecene (or phenyl ether) mixtures at 200-300°C [104, 190, 207]. Depending on the relative sizes of the two domains, these heterostructures show a peanut-, dumbbell- or brick-like morphological profile (Fig. 8a-b). The formation mechanism has been discussed for the case of Au-Fe3O4 HNCs [190]. Here the differences in lattice parameters have been presumed to drive the formation of the dimer-like topology, in which the two domains are oriented such that they share a coherent interface. In addition, the fundamental role of the solvent in regulating the nucleation sites on the Au seeds has been recognized. In a nonpolar medium, a polarization charge is induced at the Au seed location where Fe3O4 is initially grown, which depletes the electron density elsewhere and thus inhibits further nucleation events. This leads to a dumbbell-like configuration. In contrast, the use of a polar electron-donor solvent allows electron deficiency to be replenished over the Au surface, thus rendering it a suitable ground either for the nucleation of multiple Fe3O4 domains or even for the achievement of a uniform oxide coverage. Thus, either flower-like heterostructures, made of several Fe3O4 “petals” arranged on a central Au particle or Au@Fe3O4 core@shell HNCs have been obtained, depending on the specific solvent and on the temperature used [190]. These findings clearly highlight how decisive the impact of changes in the solution environment is on the setting of the overall energy balance of heterostructure growth and, in turn, on the final topology of the resulting HNCs. Different types of heterodimers, such as of Ag-Se [192], Au-FePt [193], and of AuCoPt3 [194] have been prepared via mild metal-catalyzed reduction of either SeO32- ions or AuI/AuIII complexes onto preformed Ag, FePt or CoPt3 seeds, respectively (Fig. 8d-f). In particular, the ability to tune independently the size of the two components in a heterodimer has been demonstrated for the Au-CoPt3 system, whereby CoPt3 seeds with different sizes provide substrates with varying degree of reactivity toward AuIII ions, while the CoPt3 to AuIII concentration ratio regulates the final Au domain sizes [194]. A general approach to the fabrication of dumbbell-like HNCs, comprising one noble metal domain (e.g. Au, Ag, Pt, Pd) and one section of a magnetic (e.g. Co, Ni, Fe Fe3O4, CoFe2O3, MnFe2O3, NiFe2O3) or semiconductor (e.g., PbSe, PbS, CdSe, CdS) materials, has been patented [208]. This strategy relies on performing a preliminary surface “activation” of the seeds by depositing a layer of either elemental S or Se, metal sulphide or metal selenide onto their surface. Such treatment facilitates subsequent heterogeneous growth of an additional material domain from the respective molecular precursors (Scheme 3, path b), while the structural difference between the materials that are joint promotes the hetero-dimer topology. A higher architectural complexity has been pursued by applying reiterated seeding steps (Scheme 3, path a), as a means to increase the number of component domains in a heterostructure. As a remarkable demonstration of this concept, ternary Fe3O4-Au-PbSe

Hybrid nanocrystals

433

hetero-oligomer HNCs have been achieved by inducing heterogeneous nucleation of PbSe on Fe3O4-Au dumbbell-shaped seeds from Pb/Se surfactant complexes [104]. Interestingly, it has been shown that anisotropically shaped PbSe nanostructures can be generated on top from Fe3O4-Au seeds (Fig. 8c) by a sort of solution-liquid-solid (SLS) growth mechanism, and eventually detach from them [209]. The properties of the attained magnetic-metal-semiconductor HNCs clearly deviate from those of the individual components alone. For example, the semiconductor luminescence is quenched due to the metal contact favouring electron migration, while the plasmon resonance is largely redshifted due to dielectric coating effects. Also, the relevant magnetic parameters are altered, for which a full explanation is still elusive [104, 190]. Finally, luminescent ZnO-Au heterodimers obtained by a room temperature aqueous route are worthy to be mentioned [186]. Here, the presence of weakly acetate-passivated facets on ZnO NC seeds is essential to aid epitaxial Au deposition upon citrate-induced AuIII reduction under mild reaction conditions. These ZnO-Au HNCs exhibit red-shifted Au surface plasmon absorption and ZnO band-edge UV emission that is about one order of magnitude more intense than that of pure ZnO NCs. Also, multiphonon Raman scattering is enhanced by the strongly localized electromagnetic field of the metal plasmon, as found for other types of composite systems [210-211]. These modified properties again reflect the mutual influence of interfacial electron communication that is achieved upon formation of an inorganic heterojunction. (b) Nonepitaxial deposition followed by thermally driven coalescencecrystallization. A systematic investigation of how minimization of interfacial strain energy governs the topological evolution of HNCs has been reported for heterostructures based on either FePt or γ-Fe2O3 and metal chalcogenides (i.e., CdS, ZnS, HgS, CdSe) [117, 181-182, 212]. Examples of HNCs synthesized by this mechanism are reported in Figure 9. The synthesis of these HNCs is seeded with either γ-Fe2O3 or FePt seeds, onto which a highly defective and amorphous MeS layer is initially deposited upon sequential addition of standard organometallic precursors at low temperature (lower than 100°C). Upon prolonged heating at 280°C, the amorphous MeS shell starts to crystallize, which determines a gradual strain intensification at the seed/MeS junction region owing to the poor lattice matching between the two materials. Over time, the shell coalesces and segregates into a discrete MeS grain, which is ultimately found attached at one side of the original seed (Scheme 3, panel c). Such evolution can be explained by considering that the large junction tension in the annealed core@shell nanostructure can be greatly relieved when the interfacial area between the two materials is reduced. This process can compensate for the increase in the overall surface energy that eventually accompanies the heterodimer formation. The topological evolution of the γ-Fe2O3-MeS system has been studied in detail as a function of the MeS type [181-182] and rationalized on the basis of the Coincidence Site Lattice Theory [213]. It has been proposed that, since the lattice fit at a γ-Fe2O3/MeS interface is better for Me=Zn than for Me=Cd, and much worse for Me=Hg, then the heterogeneous growth of MeS on γ-Fe2O3 NPs could result in multiple ZnS domains or in only one CdS domain on each oxide seed, respectively, whereas the formation of γ-Fe2O3-HgS junctions should be disfavoured. As for what concerns their chemical-physical properties, these HNCs retain both the fluorescence of the MeS part and the typical superparamagnetic behaviour of magnetic portion, which suggests their use as bifunctional probes, for instance, in biomedical applications. First efforts in this direction have been recently made with their

434


Figure 9. Examples of hetero-oligomer HNCs synthesized by nonepitaxial deposition followed by thermal coalescence-crystallization. The panels show TEM and HRTEM images of: (a-b) Fe3O4ZnS (c-d) Fe3O4-CdS heterodimers HNCs (reproduced from ref. 181 with permission, copyright 2005 American Chemical Society); (e-f) FePt-CdS heterodimer HNCs (reproduced from ref. 212 with permission, copyright 2004 American Chemical Society).

γ-Fe2O3-CdSe heterodimers, which have been coated with a SiO2 shell and bioconjugated to perform the imaging of live cell membranes [117]. (c) Biphasic reactions. A technique based on performing seeded growth at a liquidliquid interface under mild conditions has been devised to synthesize hetero-dimers

Hybrid nanocrystals

435

coupling a magnetic section and a noble metal domain [185]. Examples of HNCs synthesized by this biphasic strategy are shown in Figure 10. In the reported procedure (Scheme 3, path d), an aqueous metal salt solution is brought in contact with an immiscible organic solvent (such as dichlorobenzene, dichloromethane, hexane, or octyl ether) in which surfactant-capped Fe3O4 or FePt NCs are dissolved. Upon ultrasonic irradiation under oxygen-free atmosphere, an emulsion is formed and stabilized by the hydrophobic seeds self-assembling at the organic/water interfaces of the microdroplets [214]. These NCs provide catalytic sites onto which the Ag+ or AuCl4─ ions can be reduced to Ag or Au by sonochemically produced radicals, respectively. As the seeds are only partially exposed to the aqueous phase, metal deposition is spatially restricted to a small surface region and proceeds self-catalytically, which explains why a single metal domain is ultimately found on each seed. Notably, the final heterodimer HNCs are fully soluble in the nonaqueous phase. An appealing feature of these HNCs is that they offer two material platforms with distinct surface properties. As a proof of concept, it has been demonstrated that each of two domains can be selectively functionalized with a specific biomolecule [185]. (d) Self-regulated homogeneous and heterogeneous nucleation. There have been also recent reports on the preparation of heterodimer HNCs by one-pot methods, in which all the reagents required to grow the heterostructures are present since the beginning of

Figure 10. TEM images of heterodimers synthesized at liquid-liquid interfaces. The panels are TEM images of: (a) as-prepared Fe3O4 seeds; (b-d) Fe3O4-Ag heterodimers with different Ag domain size; (e) FePt-Ag heterodimers; (f) Au-Ag heterodimers (reproduced from ref. 185 with permission, copyright 2005 American Chemical Society).

436


the synthesis in the same surfactant mixture. Both homogeneous and heterogeneous nucleation processes of different materials are self-governed at once, and permit simplification of overall synthetic procedure, since no separate seed preparation step is necessary. Nevertheless, the identification of experimental conditions under which selfregulated nucleation/growth is established, is extremely challenging, since it requires a rather delicate choice and ingenious manipulation of suitable reaction pathways. Examples of HNCs prepared by such approaches are illustrated in Figure 11. When attempting to create an alloy of compounds characterized by partial miscibility and large interfacial energy, then phase-segregation of two materials into separate domains may take place, leading to a dimer-like heterostructure (Scheme 3, top path e).

Figure 11. Examples of heterodimers synthesized by self-controlled nucleation-growth processes. The panels are TEM and HRTEM images of: (a-b) Co9S8-PdSx (reproduced from ref. 178 with permission, copyright 2004 American Chemical Society) and (c-d) Cu2S-In2S3 phase-segregated heterodimers (reproduced from ref. 180 with permission, copyright 2006 Wiley-VCH Verlag GmbH & Co. KGaA; (e) Ag-AgBr heterodimers (synthesized according to ref. 183); (f) FePt-Fe3O4 heterodimers (reproduced from ref. 184 with permission, copyright 2008 American Chemical Society).

Hybrid nanocrystals

437

This has been verified in the growth of Co9S8-PdSx (Fig. 11a-b) [178-180] and Cu2SIn2S3 (Fig. 11c-d) [183] heterodimers with acorn, bottle and larva shapes by codecomposition of the corresponding metal carboxylate precursors in the presence of various alkyl thiols. During the reaction, selective homogeneous nucleation of one material is followed by growth continuation of the second one, that emerges by developing an interface of graded composition. Ag-AgBr heterodimers have been derived upon methanol addition to an optically transparent toluene solution containing a AgNO3/alkylammoium bromide complex and dodecanthiol capping ligands (Fig. 11e) [183]. The alcohol here acts both as weak reducing agent for Ag+ ions and as AgBr precipitating agent, thus triggering simultaneous nucleation and growth of the Ag and AgBr portions of each heterodimer (Scheme 3, bottom path e). This particular evolution is corroborated by the statistically significant observation that the ratio between the dimensions of the two domains is approximately constant despite of the relatively large size distribution of the colloids. Another interesting case is represented by FePt-Fe3O4 heterodimers with tunable geometric parameters (Fig. 11f), that have been prepared by reacting Pt(acac)2 and Fe(CO)5 precursors in a OLAM/OLAC/octadecene environment [201]. The HNC portions are formed in tow sequential steps, involving the initial homogeneous nucleation of FePt NCs at T ≤ 200°C, and subsequent heterogeneous growth of iron oxide onto such seeds at T ≈ 295°C. Since each reaction stage is selectively activated under well-defined thermal conditions, then regulation of both the temperature and the heating time guarantees that the two material sections of the HNCs form at distinct times of the synthesis course. As a consequence of the magnetic exchange coupling between the two soft and hard materials, the HNCs exhibit tunable single-phase-like magnetic behavior, distinct from that of their individual components, which can be conveniently exploited in MRI techniques [104]. (e) Reactions between HNCs. An original strategy has been devised with the potential to access increasingly complex HNCs, in which hetero-dimers are regarded as small functional molecules that can be combined with each other to form larger molecules. Example of as-synthesized HNCs are shown in Figure 12. It has been reported [189] that peanut-shaped ternary Fe3O4-Au-Fe3O4 HNCs, in which an Au section bridges two Fe3O4 domains, can be obtained by inducing the coalescence of the Au domains that belong to separate Au-Fe3O4 hetero-dimers in the presence of sulphur (Scheme 3, panel f). Due to its high affinity to gold surfaces, it may be presumed that sulphur helps to displace the capping surfactants on the Au section of the heterodimers, thereby driving them to fuse with each other as a means of lowering their total surface energy (Fig. 12a-b). In another approach, PdSx-Co9S8-PdSx peanut-shaped HNCs, in which a central Co9S8 portion connects two PdSx domains (Fig. 12c-d), have been synthesized by seeding Co9S8 phase with PdSx NCs using cobalt acetate and noctadecanthiol as precursor in octyl ether at 230°C [178]. As demonstrated by timedependent growth investigation, the ternary PdSx-Co9S8-PdSx HNCs result from crystaloriented coalescence of the acorn-shaped PdSx-Co9S8 heterodimers [179, 189] that form at earlier reaction stages. The driving force of this process appears to be the weak organic passivation of the Co9S8 domains, which renders them particularly reactive and, hence, prone to fuse with each other in the absence of extra surfactants. Finally, an oriented crystal attachment route has been purposely for synthetic purposes in the preparation of binary

438


Figure 12. Examples of hetero-oligomer HNCs synthesized by thermally induced fusion of heterodimer seeds. The panels are TEM and HRTEM images of: (a-b) ternary Fe3O4-Au-Fe3O4 HNCs with a Au bridging section (reproduced from ref. 104 with permission, copyright 2006 American Chemical Society); (c-d) ternary PdSx-Co9S8-PdSx HNCs with an intermediate Co9S8 bridging portion (synthesised according to ref. 189).

TiO2-SnO2 HNCs. In this case, a mixture of rutile TiO2 nanorods and cassiterite SnO2 NCs has been thermally activated under hydrothermal conditions, yielding heterostructures characterized by a “pearl necklace” distribution of SnO2 domains onto TiO2 [215].

4.2. Asymmetric heterostructures based on spherical and rod-shaped sections HNCs with a spatially asymmetric arrangement of their component domains have been obtained by programming heterogeneous growth reactions onto anisotropically shaped seeds, such as nanorods and tetrapods. Such NCs commonly occur in highly asymmetric crystalline phases, which ultimately offer a pronounced facet-dependent reactivity. Variations in the strength of ligand binding, differences in the inherent chemical accessibility, and/or in the degree of lattice matching experienced at the relevant crystal surfaces can lead to site-selective nucleation of a foreign material on these seeds. (a) Site-selective heterogeneous nucleation. The first prototypes of asymmetric HNCs have been based on cadmium chalcogenide NCs, for which a high level of size and shape control is achievable [196-197]. Nanorods or tetrapods of these materials commonly comprise one-dimensional sections made of c-axis elongated wurtzite (hexagonal) lattice, whose apex facets exhibit chemical reactivity dissimilar from that of their longitudinal sidewalls, which is consistent with the mechanism of their anisotropic

Hybrid nanocrystals

439

Scheme 4. Sketch of nanocrystal heterostructures grown form anisotropically shaped seeds and of paths leading to their formation: (a) functionalization of a nanorod with spherical domains of a different material at selected locations; (b) tip-selective growth continuation of a nanorod into either a linear or a branched manner; (c) growth of either spherical, linear or branched section onto the tips of tetrapods (reproduced from ref. 40 with permission, copyright 2007 Bentham Science Publisher).

440


Figure 13. Examples of HNCs synthesized by tip-selective heterogeneous nucleation of a different material onto preformed nanorod seeds. The panels are TEM mages of: (a) CdS-PbSe nanomatchsticks (reproduced from ref. 198 with permission, copyright 2005 American Chemical Society); (b) Au-CdSe-Au nanodumbbells (reproduced from ref. 216 with permission, copyright 2004 AAAS); (c) Pt-CdS-Pt and (d) PtNi-CdS-PtNi nanodumbbells (reproduced from ref. 202 with permission, copyright 2008 American Chemical Society); (e) Co-TiO2-Co nanodumbbels (reproduced from ref. 187 with permission, copyright 2007 American Chemical Society); (f) Co-Au nanomatchsticks (reproduced from ref. 220, copyright 2007 Wiley-VCH Verlag GmbH & Co. KGaA).

growth [198, 199]. One additional peculiarity of the wurtzite structure is the absence of a plane of symmetry perpendicular to the c-axis, which makes the two basal sides of a rod not chemically equivalent. One can therefore expect significant differences in reactivity between the two apexes (Scheme 4a, path 1). Demonstration of such reactivity is provided in Figure 13. An interesting proof-of-concept is provided by the growth of spherical PbSe or CdTe sections either on one or on both tips of CdS or CdSe nanorods, which leads to matchstick-like or dummbell-like architectures, respectively [198-199]. The possibility of changing the overall topology of the resulting HNCs from a PbSe-CdS matchstick-like profile to a PbSe-CdS-PbSe dumbbell-like morphology is an indirect

Hybrid nanocrystals

441

striking evidence for the non-equivalence of two basal planes of the wurtzite CdS nanorod seeds (Fig. 13a). In these associations of semiconductors the energy band structure can be engineered such that either localization or separation of the charge carriers can be promoted, which can have important implications in the realization of photovoltaic devices capable to perform specific tasks [16]. An exceptionally high versatility has been further achieved in the manipulation of the formation kinetics of HNCs based on cadmium chalcogenides. By adjusting the reaction conditions, it has been possible to decide on which locations and in which manner the growth of an additional material can occur. The synthesis of complex HNCs based on mixtures of elongated and branched portions has been reported [200]. For various combinations of semiconductors (such as CdS-CdS, CdS-CdTe, CdSe-CdTe), it has been demonstrated that the tips of nanorod seeds can be continued either in a linear manner (Scheme 4b, paths 1-2) or by the introduction of branching (Scheme 4b, paths 3-4). Tetrapod seeds can be manipulated in a similar way, such that either spherical, rodshaped or branching domains can be added to their tips (Scheme 4c, paths 1-3, respectively). One aspect that discloses the technological potential of these elaborate heterostructures stands within their ability of enhancing charge carrier delocalization. For example, upon visible light absorption, photogenerated electrons and holes in CdTedecorated CdSe tetrapods can spatially separate from each other over distances as large as ~30 nm. The latter value can be eventually tuned by changing the geometric parameters of the heteroarchitectures. Exploitation of these properties can be envisaged in the field of solar energy conversion. Other relevant HNC prototypes obtained by the aforementioned mechanism are represented by semiconductor-metal heterostructures, whereby the apexes of anisotropic CdSe nanocrystals (Scheme 4a, path 1 and Scheme 4c, path 1) are joint to metal domains [216]. For example, CdSe nanorods or tetrapods have been provided with Au patches by reducing an Au(III)-alkylammonium bromide complex with DDA at room temperature (Fig. 13b) [217]. Instead, when a similar protocol is applied to CdS nanorods, multiple Au patches form on their longitudinal sidewalls, as DDA likely etches the surface and creates crystal defects that can catalyze numerous nucleation events. However, an apexselective secondary growth of Pt, PtNi or PtCo domains has been recently realized also onto CdS nanorods by means of high-temperature pyrolysis of organometallic precursors [202]. In all of the aforementioned cases, the strong electronic band coupling between the semiconductor and the metal sections is authenticated by a significant quenching of the luminescence, reflecting increased charge carrier separating properties [219]. Interestingly, the metal-tipped semiconductor HNCs can be utilized as a set of functional building blocks, whereby, for example, Au terminations are functionalized with biological molecules capable of molecular recognition. By this procedure, nanodumbbells and nano-matchsticks can be chained together through their “anchoring points”, opening new perspective in the self-assembly of NCs [218]. For other material combinations, the extent and the strength of ligand passivation on the respective seed facets and/or the intrinsic structure of the seeds have been recognized to play a decisive role in governing the topology of the final hybrid architectures. Relevant studies have provided precious mechanistic insights on this matter. A room-temperature UV-driven photocatalytic silver ion reduction approach has been exploited for preparing Ag-functionalized ZnO nanorods [196]. Due to the weak

442


organic passivation of the acetate-capped ZnO seeds employed, metallic Ag embryos are allowed to deposit and subsequently self-catalyze their enlargement, which avoids undesired homogenous nucleation in the bulk solution. The preferential deposition of Ag domains in proximity of either nanorod tips has been tentatively explained as arising from internal-dipole-driven localization of conduction electrons and/or from more favourable lattice match conditions on those regions. In another report, the tippreferential deposition of TiO2 has been observed onto similar acetate-coated ZnO nanorods [195]. In this case, a hydrothermal treatment of the ZnO nanorods with amorphous TiO2 powders has been used to synthesize binary oxide heterostructures, in which one amorphous TiO2 head cap is ultimately attached to either nanorod apexes. The latter have been identified to coincide to the polar basal planes of c-axis elongated ZnO wurtzite lattice, which have been regarded as more chemically reactive owing to their inherent atomic arrangement. Co-decorated TiO2 nanorods have been synthesized by thermal decomposition of Co2(CO)8 under assistance of octanoic acid (OCAC) and OLAM at 250-280°C [187]. A calibrated temporal variation of the OCAC/OLAM composition and/or concentration along the synthesis course has been conveniently used to switch heterogeneous Co growth from a tip-preferential to a nonselective deposition regime, in which the metal nucleates also onto the longitudinal sidewalls of the NRs (Scheme 5). Here, Co overgrowth proceeds as a means of compensating for the increase in the overall surface tension caused by ineffective organic passivation on the respective seed facets, which is realized at controllably low surfactant concentration. Since both types of TiO2-Co

Scheme 5. Sketch of surfactant-controlled mechanisms leading to site-specific heterogeneous nucleation of a second material on nanorod seeds (Reproduced in part from ref. 39 with permission, copyright Wiley-VCH Verlag GmbH & Co. KGaA).

Hybrid nanocrystals

443

heterojunction configurations are quite favourable in terms of interfacial lattice strain, then the facet-preferential adhesion of the surfactants is concluded to be mainly responsible for regulating the TiO2 seed accessibility and reactivity. Another aspect deserving emphasis is that the magnetic properties of these HNCs do not follow a simple dependence on the Co domain size. One possible explanation could be that, as the two materials communicate through a rather extended interface, proximity effects and/or exchange bias at the interface can unusually influence the magnetic anisotropy term. It is also expected that these Co-TiO2 HNCs could work as nanosized bifunctional catalysts for a variety of gas and liquid phase reactions benefiting from light-induced enhanced charge separation and/or cooperative metal-oxide interactions. Appealing sets of heterometallic HNCs have been reported recently. A clear influence of the growing environment on the ultimate location of secondary material domains has been also found in the synthesis of Au-decorated Co nanorods from AuCl(tht) (where tht=tetrahydrothiophene) precursors in the presence of lauric acid (LA) and hexadecylamine (HDA) [220]. In this system, the Co seeds trigger the otherwise kinetically prohibited AuI reduction to Au0, while surface adsorbed LA and HDA confer a robust protective coating on the longitudinal sidewalls of the nanorods, which explains why the deposition of gold remains mainly confined to their tips instead. In yet another case, Pt-tipped Au nanorods have been prepared by careful manipulation of PtCl42─ reduction with ascorbic acid under assistance of facet-selective adhesion of a cationic cetylammonium bromide surfactant and of Ag+ ions, the latter dictating the Pt deposition mode on the seeds [82]. Another report has demonstrated the possibility to prepare segmented Ag-Au-Ag rod-like heterostructures. The key strategy applied in this case relies on performing AgNO3 reduction in diethylen glycol at 260°C in the presence of multiply twinned dodecahedral Au seeds and of PVP as shape-directing agent [204]. Under such conditions, Ag can nucleate selectively on Au developing two rod-shaped sections per seed which depart out at opposite directions. Interestingly, each material segment of these HNCs exhibit distinct chemical reactivity that can be exploited for further synthetic purposes. For example, the Ag segments can be stoichiometrically converted to Ag2S upon reaction with Na2S, leading to Ag2S-Au-Ag2S that retain the morphology of the starting Ag-Au-Ag heterostructures [204]. (b) Heterogeneous nucleation followed by intra-particle Ostwald ripening. Deeper mechanistic investigations on the growth of Au-tipped CdSe nanorods/tetrapods and of Au-decorated CdS nanorods have revealed that these heterostructures are unstable and can transform into matchstick-like architectures with a single Au tip via an intraparticle Ostwald-ripening promoted by high gold precursor concentrations [216-217, 221]. This is shown in Figure 14. The driving forces for this evolution arise from the surface energy and the size dependence of the red-ox potential (Scheme 6, path a). Indeed, Au-CdSe-Au nanodumbbells inevitably possess Au domains with slightly dissimilar sizes, with the smaller Au tip being less stable due to its higher surface energy and stronger susceptibility to oxidation compared to the larger one. For growth or ripening to occur, the smaller Au tip has to be oxidized to metal ions that would be consequently released to the solution. When this occurs, electrons shuttle all the way across the dumbbell by hopping through surface states and reaching the opposite bigger Au tip, where they are used to reduce gold ions and make that tip grow further. The process stops when the smaller Au gold tip disappears completely. Interestingly, this

444


Figure 14. Evolution of Au-CdSe-Au nanodumbbells into singly Au-tipped CdSe nanorods (synthesized according to ref. 221 and reproduced with permission, copyright 2005 Nature Publishing).

Scheme 6. Sketch of red-ox driven mechanisms allowing manipulation of heteroepitaxial growth on nanorod seeds: (a) intra-particle Ostwald ripening; (b) selective replacement reaction of the tip material.

evolution is not observed when electrons cannot travel through the CdSe section, as in the case of a nano-tetrapod. The central section of the tetrapod contains several defects that act as a barrier for the free movement of electrons. Therefore, only simultaneous growth of four gold tips takes place in tetrapods without any ripening being induced. This mechanism has been also observed in the transformation of multiply-Au-decorated CdS nanorods into Au-CdS matchstick-like structures [217]. (c) Galvanic replacement reactions. Galvanic replacement reactions have been employed to prepare heterostructured HNCs that can hardly be synthesized by direct heterogeneous nucleation mechanisms. As an example, the growth of Au-tipped semiconductor nanorods has been performed through the selective oxidation of either CdTe or PbSe sacrificial domains, initially grown on the apexes of CdSe and CdS NRs, with a AuIII-alkylammonium bromide complex, which results in the partial substitution of the CdTe/PbSe material for Au (Scheme 6, path b) [222]. By applying a similar strategy, localized oxidation of Pd nanorods has been accomplished with aqueous AuCl4─ ions, leading to either Au-Pd-Au nanodumbbells or to Pd-Au tadpole-like heterostructures (in

Hybrid nanocrystals

445

which one Au head is attached to a Pd tail) via an Ostwald ripening process similar to that previously reported for the CdSe(S)-Au system [203]. (d) Strain-driven heterostructure formation Interesting NC heterostructures, as those shown in Figure 15, have been engineered by taking advantage of the generation of strain fields during heteroepitaxial growth. One case that illustrates the role of strain is represented by the synthesis of asymmetric binary HNCs, each made of one rod-like anatase TiO2 section and one γ-Fe2O3 spherical domain attached together [191]. These heterostructures have been obtained by decomposing Fe(CO)5 at 240-300°C in an

Figure 15. Examples of HNCs synthesized by strain-driven heteroepitaxial growth. The panels report TEM and HRTEM images of: (a-b) TiO2 nanorod-iron oxide heterostructures (reproduced from ref. 191 with permission, copyright 2006 American Chemical Society); (c) CdS nanorods before and (d-e) after growth of a linear array of Ag2S domains (reproduced from ref. 225 with permission, copyright 2007 AAAS).

446


octadecene solution containing hydrolytically prepared TiO2 nanorods [197], and three surfactants, namely OLAC, OLAM and 1,2-hexadecandiol, in well calibrated proportions. The proposed synthetic approach demonstrates to be highly versatile in that it allows the mean dimension of the γ-Fe2O3 that are heterogeneously nucleated on the TiO2 seeds (Fig. 15a-b) to be finely tailored by adjusting the reaction parameters. Such γ-Fe2O3 size tuning has been rationalized in terms of competition between homogeneous and heterogeneous nucleation processes, and on the basis of balance attained between the nucleation and growth stages. Temperature influences the exchange dynamics between the surfactants bound to the seed surface and the free ligands in the solution, and hence dictates the number of TiO2 nanorods that become activated for γ-Fe2O3 deposition. Simultaneously, the TiO2 to Fe(CO)5 precursor ratio regulates the availability of monomers for sustaining the growth of the Fe-oxide that nucleates heterogeneously. From the structural point of view, the most striking feature of these HNCs is that the γ-Fe2O3 domain grows exclusively on the longitudinal sides of the TiO2 nanorod seeds. Owing to the high lattice misfit (8-11%) that characterizes the relevant γ-Fe2O3/TiO2 junction planes, the TiO2 section is deformed and bowed toward the γ-Fe2O3 sphere. Detailed structural analyses have indeed disclosed the involvement of a peculiar heteroepitaxial growth mechanism, that has been encountered during MBE growth of quantum dot islands on highly mismatched substrates (Scheme 7, path a). The TiO2 and γ-Fe2O3 domains are coherently connected via a rather limited junction area at which the near-surface planes of the respective materials are locally curved. This allows the strain accumulated at the interface to be accommodated to a great extent, paying only a proportionally smaller cost of additional surface energy [223-224]. The possibility of tuning the dimension of iron oxide section translates into a fine control over the magnetic properties of the HNCs, which resemble those of free γ-Fe2O3 particles owing to the limited contact area shared with TiO2. Another step forward in the engineering of complex colloidal heterostructures by strain manipulation has been made by a procedure that spontaneously creates a linear array of quantum-confined semiconductor domains within a nanorod of a different material. This goal has been proven by performing partial Cd2+ exchange for Ag+ ions upon incubating CdS nanorods with AgNO3 at ~60°C [225]. In its early advancement stages, the metal replacement reaction initially produces randomly distributed tiny Ag2S islands on the CdS nanorods, which then evolve into regularly spaced Ag2S domains (Fig. 15c-e) via an intra-rod Oswald ripening process sustained by fast diffusion in CdS (Scheme 7, path b). Two factors have been assumed to drive this structural changes. First, owing to the high energy cost for CdS-Ag2S interface formation, the coalescence of small Ag2S patches into larger domains is energetically convenient, and indeed occurs spontaneously. Second, because of the CdS-Ag2S lattice misfit, there exist elastic repulsions between two Ag2S segments, which arise from strain fields in the intervening CdS regions. These forces can prevent further ripening after the Ag2S domains have reached a size spanning the nanorod diameter. The obtained Ag2S-CdS striped nanorods display characteristic optical properties, such as the quenching of visible CdS luminescence and retention of near-infrared Ag2S fluorescence, which are actually expected for linear arrays of Ag2S quantum dots electronically coupled to confining regions of CdS in a type-I band arrangement [225].

Hybrid nanocrystals

447

Scheme 7. Sketch of mechanisms exploiting lattice strain fields to manipulate heteroepitaxial growth on nanorod seeds: (a) bending of near-surface planes at the interface junction (reproduced from ref. 39 with permission, copyright Wiley-VCH Verlag GmbH & Co. KGaA); (b) partial cation exchange followed by ripening driven by intra-rod cation diffusion.

5. Conclusions and perspectives The chemical synthesis of nanocrystal heterostructures combining sections of oxide, metal and semiconductor materials in a single complex particle, represents a fascinating research direction that has recently emerged in nanochemistry research. The preparation of these new generations of multi-material inorganic nanoparticles in the liquid phase is a rather demanding task, as the ability to tailor specific nanosized materials must be integrated with the understanding of the parameters which determine heterogeneous and/or heteroepitaxial growth at the nanoscale. Although HNCs can not yet be engineered in such a way as to prevent possible detrimental effects on the properties of individual material components, nevertheless they embody a new paradigm in nanoscience whereby the purpose-tailored design and fabrication of elaborate hybrid nanosized objects with topologically defined composition holds a new key for manipulating the chemicalphysical behavior of nanosized matter in a predictable and efficient way. It can be expected that future progress in synthetic ability will open up access to a new breeds of colloidal nanoheterostructures which could enable optoelectronic, magnetic, biomedical, photovoltaic and catalytic applications with a high level of performance.

Acknowledgments The authors gratefully thank Prof. Uri Banin and Dr. Aaron Edwar Saunders (Institute of Chemistry and Center for Nanoscience and Nanotechnology, Faculty of Science, Edmund J. Safra Campus, The Hebrew University, Jerusalem, Israel) for providing a TEM image of Ag-AgBr heterodimers (Fig. 11e) and Prof. Toshiharu Teranishi (Department of Chemistry, Graduate School of Pure and Applied Sciences, University of Tsukuba, Tsukuba, Japan) for providing TEM and HRTEM images of PdSx-Co9S8-PdSx heterostructures (Fig. 12c-d).

448


References 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.

Yin, Y., Alivisatos, A. P. 2005, Nature, 437, 664. Burda, C., Chen, X. B.; Narayanan, R.; El-Sayed, M. A. 2005, Chem. Rev., 105, 1025. Jun, Y.-w., Choi, J.-s., Cheon, J. 2006, Angew. Chem. Int. Ed., 45, 3414. Cozzoli, P. D., Pellegrino, T., Manna, L. 2006, Chem. Soc. Rev., 35,1195. Park, J., Joo, J. , Kwon, S.G., Jang, Y., Hyeon, T. 2007, Angew. Chem. Int. Ed, 46, 4630 Reed, J., Ceder, G. 2004, Chem. Rev., 104, 4513. Chiesa, M., Paganini, M. C., Giamello, E., Murphy, D. M., Di Valentin, C., Pacchioni, G. 2006, Acc. Chem. Res., 39, 861. Zhao, J., Li, B., Onda, K., Feng, M., Petek, H. 2006, Chem. Rev., 106, 4402. Osgood, R. 2006, Chem. Rev., 106, 4379. Chen, C. C., Herhold, A. B., Johnson, C. S., Alivisatos, A. P. 1997, Science, 276, 398. Goldstein, A. N., Echer, C. M., Alivisatos, A. P. 1992, Science, 256, 1425. Jacobs, K., Zaziski, D., Scher, E. C., Herhold, A. B., Paul Alivisatos, A.P. 2001, Science, 293, 1803. Alivisatos, A. P. 1996, J. Phys. Chem., 100, 13226. El-Sayed, M. A. 2004, Acc. Chem. Res., 37, 326. El-Sayed, M. A. 2001, Acc. Chem. Res., 34, 257. Eustis, S., El-Sayed, M. A. 2006, Chem. Soc. Rev., 35, 209. Ghosh, S. K., Pal, T. 2007, Chem. Rev., 107, 4797. Narayanan, R., El-Sayed, M. A. 2003, J. Am. Chem. Soc., 125, 8340. Narayanan, R., El-Sayed, M. A. 2004, J. Am. Chem. Soc., 126, 7194. Narayanan, R., El-Sayed, M. A. 2005, J. Phys. Chem. B, 109, 12663. Bratlie, K. M., Lee, H., Komvopoulos, K., Yang, P.; Somorjai, G. A. 2007, Nano Lett., 7, 3097. Pankhurst, Q. A., Connolly, J., Jones, S. K., Dobson, J. 2003, J. Phys. D: Appl. Phys., 36, R167. Lu, A.-H., Salabas, E.L., Schüth, F. 2007, Angew. Chem. Int. Ed., 46, 1222. Jun, Y.-w., Choi, J.-s., Cheon, J. 2007,Chem. Commun. (12), 1203. Fernández-García, M., Martinez-Arías, A., Hanson, J. C., Rodríguez, J. A. 2004, Chem. Rev., 9, 4063. J. A. Rodríguez, Fernández-García, M. 2007, Synthesis, Properties, and Applications of Oxide Nanomaterials, John Wiley & Sons, Inc., Hoboken, New Jersey, 2007, 717 pp. Niederberger, M. 2007, Acc. Chem. Res., 40, 793. Adler, S. B. 2004, Chem. Rev., 104, 4791. McIntosh, S., Goerte, R. J. 2004, Chem. Rev., 104, 4845. Chen, X., Mao, S. S. 2007, Chem. Rev., 107, 2891. Jun, Y.-w., Huh, Y.-M., Choi, J.-s., Lee, J.-H., Song, H.-T., Kim, S.-j., Yoon, S., Kim, K.-S., Shin, J.-S., Suh, J.-S., Cheon, J. 2005, J. Am. Chem. Soc, 127, 5732. Jun, Y.-w., Jang, J.-t., Cheon, J. 2006, Bull. Korean Chem. Soc., 27, 961. Hu, J. T., Li, L. S., Yang, W. D., Manna, L., Wang, L. W., Alivisatos, A. P. 2001, Science, 292, 2060. Bruchez, M., Moronne, M., Gin, P., Weiss, S., Alivisatos, A. P. 1998, Science, 281, 2013. Huynh, W. U., Dittmer, J. J., Alivisatos, A. P. 2002, Science, 295, 2425. Gur, I., Fromer, N.A., Geier, M.L., Alivisatos, A.P. 2005, Science, 310, 462. Palmisano, G., Augugliaro, V., Magliaro, M., Palmisano, L. 2007, Chem. Commun., (33), 3425. Kudera, S., Carbone, L., Carlino, E., Cingolani, R., Cozzoli, P.D., Manna, L. 2007, Physica E, 37, 128. Casavola, M., Buonsanti, R., Caputo, G., Cozzoli, P.D. 2008, Eur. J. Inorg. Chem., (6), 837. Buonsanti, R., Casavola, M., Caputo, G., Cozzoli, P. D. 2007, Recent Patents on Nanotechnology, 1, 224. Mullin, J.W. 1997, Crystallization, 3rd ed, Butterworth-Heinemann, Oxford.

Hybrid nanocrystals

449

42. Sugimoto, T. 2001, Monodisperse Particles, Elsevier, Amsterdam. 43. Markov, I. V. 2003, Crystal Growth for Beginners: Fundamentals of Nucleation, Crystal Growth, and Epitaxy, World Scientific, Singapore. 44. Markov, I. 1997, Mat. Chem. Phys., 49, 93. 45. Cho, A. Y. 1999, J. Crys. Growth, 202, 1. 46. Li, K., Wang, Y., Guo, Z., Zheng, Y., Han, G., Qiu, J., Lu, P., An, L., Zhou, T. 2007, J. Nanosci. Nanotechnol., 7, 1. 47. Dabbousi, B. O., Rodriguez-Viejo, J., Mikulec, F. V., Heine, J. R., Mattoussi, H., Ober, R., Jensen, K. F., Bawendi, M. G. 1997, J. Phys. Chem. B, 101, 9463. 48. Peng, X., Schlamp, M. C., Kadavanich, A. V., Alivisatos, A. P. 1997, J. Am. Chem. Soc., 119, 7019. 49. Li, J. J., Wang, Y. A., Guo, W. Z., Keay, J. C., Mishima, T. D., Johnson, M. B., Peng, X. G. 2003, J. Am. Chem. Soc., 125, 12567. 50. Chen, X. B., Lou, Y. B., Samia, A. C., Burda, C. 2003, Nano Lett., 3, 799. 51. Steckel, J. S., Zimmer, J. P., Coe-Sullivan, S., Stott, N. E., Bulovic, V., Bawendi, M. G. 2004, Angew. Chem. Int .Ed., 43, 2154. 52. Pan, D. C., Wang, Q., Jiang, S. C., Ji, X. L., An, L. J. 2005, Adv. Mater., 17, 176. 53. Mokari, T., Banin, U. 2003, Chem. Mater., 15, 3955. 54. Cao, Y. W., Banin, U. 1999, Angew. Chem. Int. Ed., 38, 3692. 55. Kim, S., Fisher, B., Eisler, H. J., Bawendi, M. 2003, J. Am. Chem. Soc., 125, 11466. 56. Yu, K., Zaman, B., Romanova, S., Wang, D. S., Ripmeester, J. A. 2005, Small, 1, 332. 57. Ivanov, S. A., Piryatinski, A., Nanda, J., Tretiak, S., Zavadil, K. R., Wallace, W. O., Werder, D., Klimov, V. I. 2007, J. Am. Chem. Soc., 129, 11708. 58. Blackman, B., Battaglia, D. M., Mishima, T. D., Johnson, M. B., Peng, X. 2007, Chem. Mater., 19, 3815. 59. Manna, L., Scher, E. C., Li, L.-S., Alivisatos, A. P. 2002, J. Am. Chem. Soc., 124, 7136. 60. Talapin, D. V., Mekis, I., Gotzinger, S., Kornowski, A., Benson, O., Weller, H. 2004, J. Phys. Chem. B, 108, 18826. 61. Aharoni, A., Mokari, T., Popov, I., Banin, U. 2006, J. Am. Chem. Soc., 128, 257. 62. Eychmuller, A., Mews, A., Weller, H. 1993, Chem. Phys. Lett., 208, 59. 63. Schooss, D., Mews, A., Eychmuller, A., Weller, H. 1994, Phys. Rev. B, 49, 17072. 64. Mews, A., Eychmuller, A. 1998, Phys. Chem. Chem. Phys., 102, 1343. 65. Talapin, D. V., Koeppe, R., Gotzinger, S., Kornowski, A., Lupton, J. M., Rogach, A. L., Benson, O., Feldmann, J., Weller, H. 2003, Nano Lett., 3, 1677. 66. Carbone, L., Nobile, C., De Giorgi, M., Sala, F. D., Morello, G., Pompa, P., Hytch, M., Snoeck, E., Fiore, A., Franchini, I. R., Nadasan, M., Silvestre, A. F., Chiodo, L., Kudera, S., Cingolani, R., Krahne, R., Manna, L. 2007, Nano Lett., 7, 2942. 67. Talapin, D. V.; Nelson, J. H.; Shevchenko, E. V.; Aloni, S.; Sadtler, B.; Alivisatos, A. P. 2007, Nano Lett., 7, 2951. 68. Zhong, H.; Zhou, Y.; Yang, Y.; Yang, C.; Li, Y. 2007, J. Phys. Chem. C, 111, 6538. 69. Cao, Y. W., Jin, R., Mirkin, C. A. 2001, J. Am. Chem. Soc., 123, 7961. 70. Srnova-Sloufova, I., Vlckova, B., Bastl, Z., Hasslett, T. L. 2004, Langmuir, 20, 3407. 71. Liu, M. Z., Guyot-Sionnest, P. 2005, J. Phys. Chem. B., 108, 5882-5888. 72. Rodriguez-Gonzalez, B., Burrows, A., Watanabe, M., Kiely, C. J., Liz-Marzan, L. M. 2005, J. Mater. Chem., 15, 1755. 73. Bao, Y., Calderon, H., Krishnan, K. M. 2007, J. Phys. Chem. C, 111, 1941. 74. Bao, F., Li, J.-F., Ren, B., Yao, J.-L., Gu, R.-A., Tian, Z.-Q. 2008, J. Phys. Chem. C, 112, 345. 75. Ferrer, D., Torres-Castro, A., Gao, X., Sepulveda-Guzman, S., Ortiz-Mendez, U., JoseYacaman, M. 2007, Nano Lett., 7, 1701. 76. Habas, S.E., Lee, H., Radmilovic, V., Somorjai, G.A., Yang, P. 2007, Nat. Mater., 6, 692. 77. Xiang, Y., Wu, X., Liu, D., Jiang, X., Chu, W., Li, Z., Ma, Y., Zhou, W., Xie, S. 2006, Nano Lett., 6, 2290.

450

78. 79. 80. 81. 82.


Jana, N. R., Gearheart, L., Murphy, C. J. 2001, Adv. Mater., 13, 1389. 2002, J. Mater. Chem. 12, 1765. Gou, L.F., Murphy, C.J. 2005, Chem. Mater., 17, 3668. Sun, Y., Yin,Y., Mayers, B.T., Herricks, T., Xia, Y. 2002, Chem. Mater., 14, 4736. Grzelczak, M., Pérez-Juste, J., Rodríguez-González, B., Liz-Marzán, L.M. 2006, J. Mater. Chem., 16, 3946. 83. Grzelczak, M., Rodriguez-Gonzales, B., Pérez-Juste, J., Liz-Marzán, L. M. 2007, Adv. Mater., 19, 2262. 84. Kim, H., Achermann, M., Balet, L.P., Hollingsworth, J.A., Klimov, V.I. 2005, J. Am. Chem. Soc., 127, 544. 85. Gao, J., Zhang, B., Gao, Y., Pan, Y., Zhang, X., Xu, B. 2007, J. Am. Chem. Soc. 129, 11928. 86. Sakai, H., Kanda, T., Shibata, H., Ohkubo, T., Abe, M. 2006, J. Am. Chem. Soc. 128, 4944. 87. Oldfield, G., Ung, T., Mulvaney, P. 2000, Adv. Mater. 12, 1519. 88. Li, J., Zeng, H.C., Size 2005, Angew. Chem. Int. Ed. 44, 4342. 89. Giersig, M., Ung, T., Liz-Marzán, L. M., Mulvaney, P. 1997, Adv. Mater. 9, 570. 90. Nair, A.S., Tom, R.T., Suryanarayanan, V., Pradeep, T. 2003, J. Mater. Chem., 13, 297. 91. Koktysh, D.S., Liang, X., Yun, B.-G., Pastoriza-Santos, I., Matts, R.L., Giersig, M., SerraRodríguez, C., Liz-Marzán, L.M., Kotov, N.A., 2002, Adv. Func. Mater., 12, 255. 92. Subramanian, V., Wolf, E.E., Kamat, P.V. 2003, J. Phys. Chem. B, 107, 7479. 93. Subramanian, V., Wolf, E.E., Kamat, P.V. 2003, Langmuir, 19, 469. 94. Dawson, A., Kamat, P.V. 2001, J. Phys. Chem. B, 105, 960. 95. Kamat, P.V., Flumiani, M., Dawson, A. 2002, Colloid Surf. A, 202, 269. 96. Wood, A.; Giersig, M.; Mulvaney, P. 2001, J. Phys. Chem. B, 105, 8810. 97. Hikov, T., Schroeter, M.K., Khodeir, L, Chemseddine, A., Muhler, M., Fischer, R.A. 2006, Phys. Chem. Chem. Phys. 8, 1550. 98. Gao, T., Li, Q., Wang, T. 2005, Chem. Mater., 17, 887. 99. Shan, G., Kong, X., Wang, X., Liu, Y. 2005, Surface Sci., 582, 61. 100. Wang, L.Y., Luo, J., Fan, Q., Suzuki, M., Suzuki, I.S., Engelhard, M.H., Lin, Y., Kim, N., Wang, J.Q., Zhong, C.J. 2005, J. Phys. Chem. B, 109, 21593. 101. Park, H.-Y., Schadt, M.J., Wang, L., Lim, I.-I., Njoki, P.N., Kim, S.H., Jang, M.-Y., Luo, J., Zhong, C.-J. 2007, Langmuir, 23, 9050. 102. Lyon, J.L., Fleming, D.A., Stone, M.B., Schiffer, P., Williams, M.E., 2004, Nano Lett., 4, 719. 103. Gong, P., Li, H., He, X.,Wang, K., Hu, J., Tan, W., Zhang, S., Yang, X. 2007, Nanotechnology, 18, 285604. 104. Shi, W., Zeng, H., Sahoo, Y., Ohulchanskyy, T.Y., Ding, Y., Wang, Z.L., Swihart, M., Prasad, P.N. 2006, Nano Lett., 6, 875. 105. Teng, X., Black, D., Watkins, N.J., Gao, Y., Yang, H. 2003, Nano Lett., 3, 261. 106. Teng, X.,Yang, H. 2005, Nanotechnology, 16, S554. 107. Zeng, H., Li, J., Wang, Z.L., Liu, J.P., Sun, S. 2004, Nano Lett., 4, 187. 108. Zeng, H., Sun, S., Li, J., Wang, Z.L., Liu, J.P. 2004, Appl. Phys. Lett., 85, 792. 109. Xu, C., Xu, K.; Gu, H., Zheng, R., Liu, H., Zhang, X., Guo, Z., Xu, B. 2004, J. Am. Chem. Soc., 126, 9938. 110. Li, J., Zeng, H., Sun, S., Liu, J.P., Wang, Z.L. 2004, J. Phys. Chem. B, 108, 14005. 111. Hou, Y., Xu, Z., Peng, S., Rong, C., Liu J. P., Sun, S. 2007, Adv. Mater. 19, 3349. 112. Mulvaney, P., Liz-Marzan, L. M., Giersig, M., Ung, T. 2000, J. Mater. Chem. 10, 1259. 113. Gerion, D., Pinaud, F., Williams, S.C., Parak, W.J., Zanchet, D., Weiss, S., Alivisatos, A.P. J. Phys. Chem. B, 2001, 105, 8861. 114. Correa-Duarte, M.A., Giersig, M., Liz-Marzan, L.M. 1998, Chem. Phys. Lett., 286, 497. 115. Wang, Y., Tang, Z., Liang, X., Liz-Marzan, L.M., Kotov, N.A. 2004, Nano Lett., 4, 225. 116. Zhelev, Z., Ohba, H., Bakalova, R. 2006, J. Am. Chem. Soc., 128, 6324. 117. Selvan, S.T., Petra, P.K., Ang, C.Y., Ying, J.Y. 2007, Angew. Chem. Int. Ed., 46, 2448.

Hybrid nanocrystals

451

118. Burns, A., Ow, H., Wiesner, U. 2006, Chem. Soc. Rev., 35, 1028. 119. Wolcott, A., Gerion, D., Visconte, M., Sun, J., Schwartzberg, A., Chen, S., Zhang, J.Z., 2006, J. Phys. Chem. B, 110, 5779. 120. Darbandi, M., Lu, W., Fang, J., Nann, T. 2006, Langmuir 22, 4371. 121. Liu, S. H., Han, M. Y. 2005, Adv. Funct. Mater., 15, 961. 122. Ung, T., Liz-Marzan, L. M., Mulvaney, P. 1999, J. Phys. Chem. B, 103, 6770. 123. Obare, S.O., Jana, N.R., Murphy, C.J. 2001, Nano Lett., 1, 601. 124. Liz-Marzan, L.M 2006, Langmuir, 22, 32. 125. Liz-Marzan, L.M., Giersig, M., Mulvaney, P. 1996, Langmuir 12, 4329. 126. Ung, T., Liz-Marzan, L.M.; Mulvaney, P. 1998, Langmuir, 14, 3740. 127. Salgueirino-Maceira, V., Caruso, F., Liz-Marzan, L.M. 2003, J. Phys. Chem. B, 107, 10990. 128. Kobayashi, Y., Katakami, H., Mine, E., Nagao, D., Konno, M., Liz-Marzan, L.M. 2005, J. Colloids Interface Sci., 283, 392. 129. Pastoriza-Santos, I., Pérez-Juste, J., Liz-Marzán, L.M. 2006, Chem. Mater., 18, 2465. 130. Kobayashi, Y., Horie, M., Konno, M., Rodriguez-Gonzalez, B., Liz-Marzan, L.M., 2003, J. Phys. Chem. B, 107, 7420. 131. Yoon, T.J., Yu, K.N., Kim, E., Kim, J.S., Kim, B.G., Yun, S.-H., Sohn, B.-H., Cho, M.-H., Lee, J.K., Park, S.B. 2006, Small, 2, 209. 132. Lee, D.C., Mikulec, F.C.; Pelaez, J.M.; Koo, B., Korgel, B.A. 2006, J. Phys. Chem. B, 110, 11160. 133. Ung, T., Liz-Marzan, L.M., Mulvaney, P. 2001, J. Phys. Chem. B, 105, 3441. 134. Liz-Marzan, L.M., Mulvaney, P. 2003, J. Phys. Chem. B, 107, 7312. 135. Salgueiriño-Maceira, V., Correa-Duarte, M.A.; Farle, M. Lopez-Quintela, A., Sieradzki, K., Diaz, R. 2006, Chem. Mater., 18, 2701. 136. He, R., You, X., Shao, J., Gao, F., Pan, F., Cui, D. 2007, Nanotechnology, 18, 315601. 137. Eswaranand, V., Pradeep, T. 2002, J. Mater. Chem. 12, 2421. 138. Park, J.I., Cheon, J. 2001, J. Am. Chem. Soc., 123, 5743. 139. Lee, W.-r., Kim, M.G., Choi, J.-r., Park, J.-I., Ko, S.J., Oh, S.J., Cheon, J. 2005, J. Am. Chem. Soc., 127, 16090. 140. Cho, S.-J., Idrobo, J.-C., Olamit, J., Liu, K., Browning, N. D., Kauzlarich, S.M. 2005, Chem. Mater. 17, 3181. 141. Yang, J., Lee, J.Y., Too, H.P. 2005, J. Phys. Chem. B, 109, 19208. 142. Tracy, J.B., Weiss, D.N., Dinega, D.P., Bawendi, M.G. 2005, Phys. Rev. B, 72, 064404. 143. Verelst, M., Ely, T. O., Amiens, C., Snoeck, E., Lecante, P., Mosset, A., Respaud, M., Broto, J.M., Chaudret, B. 1999, Chem. Mater. 11, 2702. 144. Lee, I. S., Lee, N., Park, J., Kim, B.H., Yi, Y.-W., Kim, T., Kim, T.K., Lee, I.H., Paik, S.R., Hyeon, T. 2006, J. Am. Chem. Soc., 128, 10658. 145. Salazar-Alvarez, G., Sort, J., Surinach, S., Baro, M.D., Nogues, J. 2007, J. Am. Chem. Soc., 129, 9102. 146. Peng, S., Wang, C., Xie, J., Sun, S. 2006, J. Am. Chem. Soc., 128, 10676. 147. Peng, S., Sun, S. 2007, Angew. Chem. Int. Ed, 46, 4155. 148. Cabot, A., Puntes, V.F., Shevchenko, E., Yin, Y., Balcells, L., Marcus, M.A., Hughes, S.M., Alivisatos, A.P. 2007, J. Am. Chem. Soc., 129, 10358. 149. Sun, Y., Xia, Y. 2002, Science 298, 2176. 150. Sun, Y., Mayers, B.T., Xia, Y. 2002, Nano Lett. 2, 481. 151. Kim, S.-W., Kim, M., Lee, W.Y., Hyeon, T. 2002, J. Am. Chem. Soc., 124, 7642. 152. Mayers, B. J., X., Sunderland, D., Cattle, B., Xia, Y. 2003, J. Am. Chem. Soc., 125, 13364. 153. Liang, H.-P., Wan, L.-J., Bai, C.-L., Jiang, L. 2005, J. Phys. Chem. B, 109, 7795. 154. Schwartzberg, A. M., Olson, T. Y., Talley, C. E., Zhang, J. Z. 2006, J. Phys. Chem. B, 110, 19935. 155. Schwartzberg, A. M., Olson, T. Y., Zhang, J. Z., Huser, T., Talley, C. E. 2006, Anal. Chem. 78, 4732. 156. Yin, Y., Erdonmez, C., Aloni, S., Alivisatos, A. P. 2006, J. Am. Chem. Soc., 128, 12671.

452


157. Guo, S., Fang, Y., Dong, S., Wang, E. 2007, J. Phys. Chem. C, 111, 17104. 158. Nogués, J., Schuller, I. K. 1999, J. Magn. Magn. Mater., 192, 203. 159. Nogués, J., Sort, J., Langlais, V., Skumryev, V., Suriñach, S., Muñoz, J.S., Baró, M.D. 2005, Physics Reports, 422, 65. 160. Skumryev, V., Stoyanov, S., Zhang, Y., Hadjipanayis, G., Givord, D., Nogués, J. 2003, Nature, 423, 850. 161. Smigelskas, E. O., Kirkendall, A. D. 1947, Trans. AIME, 171, 130. 162. Fan, H. J., Gösele, U., Zacharias M. 2007, Small, 3, 1660. 163. Yin, Y., Rioux, R. M., Erdonmez, C. K., Hughes, S., Somorjai, G. A., Alivisatos, A. P. 2004, Science, 304, 711. 164. Yin, Y., Erdonmez, C. K., Cabot, A., Hughes, S., Alivisatos, A. P. 2006, Adv. Funct. Mater., 16, 1389. 165. Tu, K. N., Gösele, U. 2005, Appl. Phys. Lett., 86, 093111. 166. Gao, J. H., Liang, G. L., Zhang, B., Kuang, Y., Zhang, X., Xu, B. 2007, J. Am. Chem. Soc., 129, 1428. 167. Kim, S. Y., Ying, Y. D., Alivisatos, A. P., Somorjai, G. A., Yates, J. T., Jr. 2007, J. Am. Chem. Soc., 129, 9510. 168. Pastoriza-Santos, I., Koktysh, D. S., Mamedov, A. A., Giersig, M., Kotov, N. A., Liz-Marzan, L. M. 2000, Langmuir, 16, 2731. 169. Tom, R. T., Nair, A. S., Singh, N., Aslam, M., Nagendra, C. L., Philip, R., Vijayamohanan, K., Pradeep, T. 2003, Langmuir, 19, 3439. 170. Hirakawa, T., Kamat, P. V. 2004, Langmuir, 20, 5645. 171. Hirakawa, T., Kamat, P. V. 2005, J. Am. Chem. Soc., 127, 3928. 172. Lai, J., Shafi, K. V. P. M., Ulman, A., Loos, K., Popovitz-Biro, R., Lee, Y., Vogt, T., Estournes, C. 2005, J. Am. Chem. Soc., 127, 5730. 173. Sobal, N. S., Hilgendorff, M., Mohwald, H., Giersig, M., Spasova, M., Radetic, T., Farle, M. 2002, Nano Lett., 2, 621. 174. Lu, W., Wang, B., Zeng, J., Wang, X. P., Zhang, S. Y., Hou, J. G. 2005, Langmuir, 21, 3684. 175. Mokari, T., Aharoni, A., Popov, I., Banin, U. 2006, Angew. Chem. Int. Ed., 45, 8001. 176. Frank, F. C., Turnbull, D. 1956, Phys. Rev., 104, 617. 177. Gösele, U., Morehead, F., Frank, W., Seeger, A. 1981, Appl. Phys. Lett., 38, 157. 178. Teranishi, T., Inoue, Y., Nakaya, M., Oumi, Y., Sano, T. 2004, J. Am. Chem. Soc. 126, 9914. 179. Teranishi, T., Inoue, Y., Saruyama, M., Nakaya, M., Kanehara, M. 2007, Chem. Lett., 36, 490. 180. Choi, S. H., Kim, E. G., Hyeon, T. 2006, J. Am. Chem. Soc., 128, 2520. 181. Kwon, K. W., Shim, M. 2005, J. Am. Chem. Soc., 127, 10269. 182. Kwon, K.W., Lee, B. H., Shim, M. 2006, Chem. Mater., 18, 6357. 183. Saunders, A. E., Popov, I., Uri Banin, 2007, Z. Anorg. Allg. Chem., 633, 2414. 184. Figuerola, A., Fiore, A., Di Corato, R., Falqui, A., Giannini, C., Micotti, E., Lascialfari, A., Corti, M., Cingolani, R., Pellegrino, T. Cozzoli, P. D., Manna, L. 2008, J. Am. Chem. Soc., 130, 1477. 185. Gu, H. W., Yang, Z. M., Gao, J. H., Chang, C. K., Xu, B. 2005, J. Am. Chem. Soc., 127, 34. 186. Wang, X., Kong, X. G., Yu, Y., Zhang, H. 2007, J. Phys. Chem. C, 111, 3836. 187. Casavola, M., Grillo, V., Carlino, E., Giannini, C., Gozzo, F., Fernandez Pinel, E., Garcia, M. A., Manna, L., Cingolani, R. Cozzoli, P. D. 2007, Nano Lett., 7, 1386. 188. Pacholski, C., Kornowski, A., Weller, H. 2004, Angew. Chem. Int. Ed., 43, 4774. 189. Teranishi, T., Saruyama, M., Nakaya, M., Kanehara, M. 2007, Angew. Chem. Int. Ed., 46, 1713. 190. Yu, H., Chen, M., Rice, P. M., Wang, S. X., White, R. L., Sun, S. H. 2005, Nano Lett., 5, 379. 191. Buonsanti, R., Grillo, V., Carlino, E., Giannini, C., Curri, M. L., Innocenti, C., Sangregorio, C., Achterhold, K., Parak, F. G., Agostiano, A., Cozzoli, P. D. 2006, J. Am. Chem. Soc., 128, 16953.

Hybrid nanocrystals

453

192. Gao, X., Yu, L. T., MacCuspie, R., Matsui, H. 2005, Adv. Mater., 17, 426. 193. Choi, J. S., Jun, Y. W., Yeon, S. I., Kim, H. C., Shin, J. S., Cheon, J. 2006, J. Am. Chem. Soc., 128, 15982. 194. Pellegrino, T., Fiore, A., Carlino, E., Giannini, C., Cozzoli, P. D., Ciccarella, G., Respaud, M., Palmirotta, L., Cingolani, R., Manna, L. 2006, J. Am. Chem. Soc., 128, 6690. 195. Cheng, C., Yu, K. F., Cai, Y., Fung, K. K., Wang, N. 2007, J. Phys. Chem. C, 111, 16712. 196. Pacholski, C., Kornowski, A., Weller, H. 2002, Angew. Chem. Int. Ed., 41, 1188. 197. Cozzoli, P. D., Kornowski, A., Weller, H. 2003, J. Am. Chem. Soc., 125, 14539. 198. Kudera, S., Carbone, L., Casula, M. F., Cingolani, R., Falqui, A., Snoeck, E., Parak, W. J., Manna, L. 2005, Nano Lett., 5, 445. 199. Halpert, J. E., Porter, V. J., Zimmer, J. P., Bawendi, M. G. 2006, J. Am. Chem. Soc., 128, 12590. 200. Milliron, D. J., Hughes, S. M., Cui, Y., Manna, L., Li, J. B., Wang, L. W., Alivisatos, A. P. 2004, Nature, 430, 190. 201. Figuerola, A., Fiore, A., Di Corato, R., Falqui, A., Giannini, C., Micotti, E., Lascialfari, A., Corti, M., Cingolani, R., Pellegrino, T. Cozzoli, P. D., Manna, L. 2008, J. Am. Chem. Soc., 130, 1477. 202. Habas, S. E., Yang, P., Mokari, T. 2008, J. Am. Chem. Soc., 130, 3294. 203. Camargo, P. H. C., Xiong, Y., Ji, L., Zuo, J. M., Xia, Y. 2007, J. Am. Chem. Soc., 129, 15452. 204. Seo, D., Yoo, C. I., Jung, J., Song, H. 2008, J. Am. Chem. Soc., 130, 2940. 205. Xu, C., Xie, J., Ho, D., Wang, C., Kohler, N., Walsh, E. G., Morgan, J. R., Chin, Y. E., Sun, S. 2008, Angew. Chem. Int. Ed. 47, 173. 206. Quarta, A., Di Corato, R., Manna, L., Ragusa, A., Pellegrino, T. 2007, IEEE Trans. Nanobiosci., 6, 298. 207. Chaubey, G. S., Nandwana, V., Poudyal, N., Rong, C.-B., Liu, J. P. 2008, Chem. Mater., 20, 475. 208. Sun, S., Yu, H., Wang, S. X. 2006, patent US20060053971. 209. Shi, W., Sahoo, Y., Zeng, H., Ding, Y., Swihart, M. T., Prasad, P. N. 2006, Adv. Mater., 18, 1889. 210. Shan, G. Y., Xu L. H., Wang, G., Liu, Y. 2007, J. Phys. Chem. C, 111, 3290. 211. Song, W., Wang, Y., Hu, H., Zhao, B. 2007, J. Raman Spectr., 38, 1320. 212. Gu, H. W., Zheng, R. K., Zhang, X. X., Xu, B. 2004, J. Am. Chem. Soc., 126, 5664. 213. Randle, V., 1997, The role of the coincidence site lattice in grain boundary engineering, Woodhead Publishing Limited, Cambridge (England). 214. Dinsmore, A. D., Hsu, M. F., Nikolaides, M. G., Marquez, M., Bausch, A. R., Weitz, D. A. 2002, Science, 298, 1006. 215. Ribeiro, C., Longo, E., Leite, E. R. 2007, Appl. Phys. Lett., 91, 103105. 216. Mokari, T., Rothenberg, E., Popov, I., Costi, R., Banin, U. 2004, Science, 304, 1787. 217. Saunders, A. E., Popov, I., Banin, U. 2006, J. Phys. Chem. B, 110, 25421. 218. Salant, A., Saunders, A. E., Banin, U. 2006, J. Am. Chem. Soc., 28, 10006. 219. Steiner, D., Mokari, T., Banin, U., Millo, O. 2005, Phys. Rev. Lett., 95, 056805. 220. Wetz, F., Soulantica, K., Falqui, A., Respaud, M., Snoeck, E., Chaudret, B. 2007, Angew. Chem. Int. Ed., 46, 7079. 221. Mokari, T., Sztrum, C. G., Salant, A., Rabani, E., Banin, U. 2005, Nat. Mater., 4, 855. 222. Carbone, L., Kudera, S., Giannini, C., Ciccarella, G., Cingolani, R., Cozzoli, P. D., Manna, L. 2006, J. Mater. Chem., 16, 3952. 223. Eaglesham, D. J., Cerullo, M. 1990, Phys. Rev. Lett., 64, 1943. 224. Carlino, E., Giannini, C., Gerardi, C., Tapfer, L., Mäder, K. A., von Känel, H. 1996, J. Appl. Phys., 79, 1441. 225. Robinson, R. D., Sadtler, B., Demchenko, D. O., Erdonmez, C. K., Wang, L.-W., Alivisatos, A. P. 2007, Science, 317, 355.