484

NANOCOMPOSITES, METAL-FILLED

Vol. 10

NANOCOMPOSITES, METAL-FILLED Introduction Atomic clusters of transition metals embedded into polymeric matrices represent a very attractive class of materials which combine properties belonging to the nanometer-sized metal phase (such as certain magnetic, electronic, optical, or catalytic properties) and the polymer phase (such as processability and film-forming properties) (1). By the careful selection and combination of both components, tailor-made functional materials exhibiting unique characteristics (2) can be obtained, with the advantage of using the well-established, low cost technologies available for polymer processing, such as printing, spraying, or spin-coating. Polymeric dispersions of nanometer-sized metal particles offer the possibility of functionalizing the polymer by properties coming from the large number of surface atoms (3) and the quantum-size effects (4). Nanometric metals show properties that differ significantly from that of bulk metals, which makes these nanocomposite systems intriguing for scientific study and potentially useful for a number of technological applications (5–12). The control of nanoparticle morphology becomes a very important aspect, since morphology profoundly influences the material performance. As a long-term goal the ability to control and vary particle size, distributions, shapes, and composition independently from one another is very desierable, in order to allow the tuning of nanocomposite properties. A broad area of nanoparticle features, ranging from nanosized single crystals to somewhat larger (in the range of about 100–200 nm) yet well-stabilized nonagglomerates, gain significant technological importance. Polymer-embedded nanostructures are potentially useful for a number of technological applications, especially as advanced functional materials (eg, high energy radiation shielding materials, microwave absorbers, optical limiters, polarizers, sensors, hydrogen storage systems) (5–12). In addition to the intrinsic nanoscopic material properties, the presence of a very large interface area in these polymer-based nanocomposites can affect significantly polymer characteristics (eg, glass-transition temperature, crystallinity, free volume content, ignition temperature), allowing the appearance of further technologically exploitable mechanical and physical properties (eg, fire-resistance, low gas diffusivity).

Historical Background The fundamental knowledge on the preparation and nature of metal/polymer nanocomposites looks back at a long history which is connected to the names of many illustrious scientists (2). The oldest technique for the preparation of metal/polymer nanocomposites that can be found in the literature was described in detail in an abstract in 1835 (13). The original article appeared in 1833 (J. Erdmann, p. 22). In an aqueous solution, a gold salt was reduced in the presence of gum-arabic, and subsequently a nanocomposite material was obtained in the form of a purple solid simply by coprecipitation with ethanol. Encyclopedia of Polymer Science and Technology. Copyright John Wiley & Sons, Inc. All rights reserved.

Vol. 10

NANOCOMPOSITES, METAL-FILLED

485

Around 1900, widely forgotten reports indicated the preparation of polymer nanocomposites with uniaxially oriented inorganic particles, and their remarkable optical properties (14,15). Dichroic plants and animal fibrils (eg, linen, cotton, spruce, or chitin, amongst others) were prepared by impregnation with solutions of silver nitrate, silver acetate, or gold chloride, followed by reduction of the corresponding metal ions under the action of light (16). Dichroic films were also obtained using gold chloride treated gelatin which was subsequently drawn, dried, and finally exposed to light (17). Similar results were obtained when gelatin was mixed with colloidal gold before drying and drawing (18). The gold or silver content in such systems amounted typically to ca 1% by weight (19). In 1904, Kirchner and Zsigmondy (Nobel Laureate in Chemistry, 1925) reported that nanocomposites of colloidal gold and gelatin reversibly changed color from blue to red upon swelling with water (20). In order to explain the mechanism of nanocomposite color change, they suggested that the material absorption must also be influenced by the distance between the embedded particles. In addition, around the same time, the color of gold particles embedded in dielectric matrices was subject of detailed theoretical analyses by Maxwell Garnett who explained the color shifts upon variation of particle size and volume fraction in a medium (21,22). During the following three decades, dichroic fibers were prepared with many different elements (ie, Os, Rh, Pd, Pt, Cu, Ag, Au, Hg, P, As, Sb, Bi, S, Se, Te, Br, I) (23–27). The dichroism was found to depend strongly on the employed element, and optical spectra of dichroic nanocomposites, made of stretched poly(vinyl alcohol) films containing gold, silver, or mercury, were presented in 1946 (however, the preparative scheme used is not really clear) (28). It was assumed already in the early reports that dichroism was originated by the linear arrangement of small particles (29) or by polycrystalline rod-like particles (30) located in the uniaxially oriented spaces present in the fibers. An electron micrograph depicted in 1951 showed that tellurium needles with typical dimensions of ca 5 × 50 nm were present inside a dichroic film made of stretched poly(vinyl alcohol), however also in this case, just a limited number of details were given about the technique used for sample preparation (31). In 1910, Kolbe was the first to prove that dichroic nanocomposite samples based on gold contained the metal indeed in its zero-valence state. Such affirmation was confirmed a few years later by X-ray scattering. In particular, it was shown that zero-valence silver and gold were present in the respective nanocomposites made with oriented ramie fibers, and the ring-like interference patterns of the metal crystallites showed that the individual primary crystallites were not oriented (32). Based on Scherrer’s equation, which was developed just in this period, the average particle diameter of silver and gold crystallites was determined in fibers of ramie, hemp, bamboo, silk, wool, viscose, and cellulose acetate to be between 5 and 14 nm (33).

Basic Concepts A nanocomposite is a material made of two or more phases one of which has at least one dimension in a nanometric size range. Metal/polymer nanocomposites

486

NANOCOMPOSITES, METAL-FILLED

Vol. 10

Fig. 1. Structure of atomic clusters of metals: (a) elemental cluster, (b) small cluster, (c) large cluster, (d) thiol-derivatized cluster, and (e) alloyed cluster.

are made of a continuous polymeric matrix embedding nanosized metal domains. Thermoplastic polymers, elastomers, and thermosetting resins can be potentially used as matrix, but amorphous linear polymers, like optical plastics (eg, polystyrene, poly(methyl methacrylate)) and conductive polymers are the most frequently used. About the metallic filler, very small clusters and nanoparticles with a pseudospherical shape are frequently employed (see Fig. 1). The use of nanorods and nanowires has also been described (2). In these materials, the polymer has the function of protecting the nanostructures and allowing their manipulation, whereas the nanosized filler provides the polymeric matrix with unique properties coming from “small-size” effects (ie, mesoscopic properties). To allow the appearance of mesoscopic properties, the nanoparticle size is required to be very small; usually a dimension less than 30 nm is necessary for most metallic materials. In particular, the metal domain size, where electrons move, needs to be comparable to the critical lengths of physical phenomena (eg, wavelength of the electrons at the Fermi edge, mean free paths of electrons or phonons, coherency length, screening length). Such phenomenon is named electron confinement (4). In addition, on this size scale, also phenomena related to the high percentage of surface atoms appear in the material (3). Different topologies can result in the preparation of polymer-embedded metal clusters. To prepare materials characterized by the properties of surface atoms or with characteristics coming from the confinement effect, a contact-free dispersion of clusters must result in the polymer matrix, since only in this case the large amount of surface atoms present allows the surface properties of matter to prevail on that of bulk. Both regular (eg, uniaxially oriented pearl-necklace type of arrays of nanoparticles) and irregular metal cluster distributions are used in technological applications. The properties of nanometric particles strictly depend on their microscopical structure (ie, chemical composition, shape, size, percentage of defects, microstrain concentration, etc). For example, the characteristic surface plasmon absorption of a system of metal nanoparticles dispersed into a dielectric matrix is related to the particle shape and size (34). To prepare a color filter, identical particles should be used, otherwise the material will appear black. The presence of a single type of microscopic structure allows each particle to provide the same contribution to the composite properties. From a theoretical point of view, an ideal nanostructured composite should be made of identical metal domains uniformly dispersed into the polymeric matrix. However, since it is very difficult to prepare a sample of

Vol. 10

NANOCOMPOSITES, METAL-FILLED

487

identical metal grains, the only practically possible situation is a system of single crystals with a narrow size distribution. Because of the high surface chemical reactivity, metal clusters can be easily oxidized or contaminated by nucleophilic molecules. The surface reactivity of small clusters is so high that even noble metals (eg, Pt, Pd, Ag) can be oxidized by air. Since small molecules present in the air (eg, oxygen, water, sulfur dioxide) are dissolved in polymers and can diffuse through it, the polymeric matrix cannot prevent metal from surface reactions. Finally, the polymer matrix has only the function to freeze particles, avoiding their diffusion and aggregation by sintering. For such a reason the nanoparticles need to be passivated before their embedding in polymers. The surface treatment (eg, thiol-derivatization, see Fig. 1d) prevents particle aggregation by short-range steric repulsion and stabilizes the metal core principally by electronic effects. Surface passivation offers also the possibility to disperse filler into most of liquid monomers and organic solvents where polymers are soluble.

Classification Nanocomposites are biphasic materials that can be classified on the basis of their microstructure by the self-connectivity concept, that is the number of space directions (X, Y, Z) in which each phase inside the composite physically contacts itself. Composite classification based on self-connectivity has been proposed for the first time by Newnham in 1978 (35). Because of the discontinuous nature of most nanocomposite materials, the self-connectivity concept needs to be applied to a limited but representative composite portion (ie, a local self-connectivity definition should be used). In particular, like in macro- and microcomposite systems, each phase in a nano-composite material can be locally self-connected in zero, one, two, or three dimensions. It is natural to confine the attention to three perpendicular axes since all property tensors are generally referred to such a system. In general, for a n-phase system the number of possible self-connectivity patterns is given by (n + 3)!/3!n!. Therefore, inside a biphasic system (n = 2) there are 10 different self-connectivity patterns: (0–0)∗ , (1–0)∗ , (2–0)∗ , (3–0)∗ , (1–1)∗ , (2–1)∗ , (3–1)∗ , (2–2)∗ , (3–2)∗ , and (3–3)∗ . The first number in the notation represents the physical connectivity of one of the two phases and the second number refers to connectivity of the other one. A schematic representation of these 10 types of selfconnectivities is given in Figure 2, using a cube as the basic building block. Arrows are used in Figure 2 to indicate the connected directions, and an asterisk has been introduced in the symbol to indicate the local connectivity.

Properties of Nanosized Metals Nanosized metals are characterized by novel thermodynamic, chemical, catalytic, optical, magnetic, and transport properties which are much different from those of corresponding massive metals. For this reason a 3-D periodic table of elements has been frequently proposed by chemists working in this material science field (36).

488

NANOCOMPOSITES, METAL-FILLED

Vol. 10

Fig. 2. The connectivity patterns of a biphasic solid (35).

The thermodynamic properties of matter are classically described as natural constants; however, they change significantly when the dimension approaches a few nanometers (37). The first small-size effect which has been observed for metals is the change in the melting point (38–43). When a solid is heated, it melts, and the melting temperature is normally considered one of the most important characteristics of a material. Nanometer-size crystals melt at much lower temperatures than extended ones. This can be understood in two ways. First, the cohesive energy of a crystal arises by the sum of all pairwise interactions between the atoms. In a very small crystal the number of surface atoms is large, and therefore the cohesive energy per atom has not yet converged on the bulk value. A second picture is more thermodynamic: as a solid is heated, melting takes place at the temperature where the chemical potentials of solid and liquid are equal. In addition to the usual term in the chemical potential, a very small crystal also has a term in it for the surface energy. In general, the surface energy of liquids is less than that of solids, because a liquid can readily assume the lowest surface area shape, a sphere. As a consequence, the smaller the crystal the lower is the melting temperature, and the reduction in the melting temperature is proportional to the surface to volume ratio, or inversely proportional to the nanocrystal radius. This scaling law for melting temperature reduction has been verified in many metals (37). Several interesting chemical properties arise as the grain size of a metal is decreased to a nanometer-size range (eg, enhanced reactivity, stoichiometric behavior of heterogeneous reactions, new reaction routes). Theoretical computations

Vol. 10

NANOCOMPOSITES, METAL-FILLED

489

(37) showed that the fraction of metal atoms residing on the surface of a 100-nm grain increases from 1–2% up to 85–95% for grain sizes in the range of 1–2 nm. The importance of this observation is that the chemistry of such surfaces is much different from that of the same atoms contained within larger grains. If the size of a cubic crystal is decreased from 100 nm down to the 3–5 nm range, the fraction of atoms contained in edge sites as compared to those contained in basal planes increases to 70%. The consequence of this observation is that as the crystal size decreases, the fraction of atoms located in low coordination sites increases sharply, which inherently imparts a higher chemical reactivity to such materials (activated metals). High surface areas and intrinsically high surface reactivities of hyperfine reactants allow surface reactions to approach stoichiometric conversion. With the development of hyperfine powders, the heterogeneous phase reaction can be carried out much faster and at lower temperatures (44,45). Finally, gas–solid reactions and liquid–solid reactions take on a new dimension, and hyperfine solids enter now in the chemist’s arsenal as novel chemical reagents. The properties of nanosized metals in heterogeneous catalysis are well established, since heterogeneous catalysts represent one of the pioneering fields of nanotechnology (3). The increasing portion of surface atoms with decreasing particle size, compared with bulk metals, makes small metal particles as highly reactive catalysts, since surface atoms are the active centers for catalytic elementary processes. Among the surface atoms, those sitting on the crystal edges and corners (see Fig. 3) are more reactive than those on basal planes. The percentage of edge

Fig. 3. SEM micrograph of 3-D self-organized poly(methyl methacrylate) particles. If beads are considered as atoms, the picture can be used to visualize the different catalytic sites on the surface of a metal crystal.

490

NANOCOMPOSITES, METAL-FILLED

Vol. 10

and corner atoms also increases with decreasing size and this is the reason for the enhanced catalytic activity (ie, supercatalytic effect) and different selectivity of very small metal catalysts. If metal particles become very small, reaching the nanometer-size scale, a color may occur. This is a typical phenomenon of nanometric metals. Actually, optical absorption may result in the ultraviolet or visible part of the spectrum, and this arises from a surface plasmon resonance. This is due to a collective electron plasma oscillation (plasmon) that is coupled to an external transverse electromagnetic field through the particle surface. It is possible to quantitatively relate the absorption coefficient to the wavelength of the exciting radiation by the Mie theory for spherical inclusions in a dielectric matrix (34). Far-IR luminescence is another optical phenomenon frequently observed with nanosized metals (46). New magnetic properties appear in metals when the size approaches the nanometric regime (47). The so-called odd–even effect is a phenomenon observed when diamagnetic metals are reduced to a nanometric-size regime. In particular, diamagnetic materials have only spin-paired electrons. However, in practice it cannot be assumed that a macroscopic piece of a diamagnetic metal does not have one or more unpaired electrons. This cannot be measured because of the effectively infinite number of atoms and electrons. However, if the particle size is small enough to make one unpaired electron measurable, the odd–even effect should become visible. Among small diamagnetic metal particles there should be an equal percentage of odd and even numbers of electrons. For magnetic materials such as Fe, Co, and Ni, the magnetic properties are size-dependent (47). In particular, the coercivity force H c needed to reverse an internal magnetic field within the particle changes with particle size and it is maximum for single-domain particles. Further, the strength of the internal magnetic field of a single particle can be size-dependent. Giant magnetostriction, magnetoresistivity, and magnetocaloric effects represent further examples of new properties arising from the small size of magnetic domains (48). Also transport properties, which are strictly related to the electronic structure of a metal particle, critically depend on size (37). For small particles, the electronic states are not continuous, but discrete, because of the confinement of electron wave function. Consequently, also properties like electrical and thermal conductivity may exhibit quantum-size effects. Most of these unique chemical and physical characteristics of nanosized metals can be used for the functionalization of plastic materials.

Preparation Methods A limited number of methods have been developed for the preparation of metal/polymer nanocomposites. Usually, such techniques consist of highly specific approaches, which can be classified as in situ and ex situ methods. In the in situ methods two steps are needed: firstly, the monomer is polymerized in solution, with metal ions introduced before or after polymerization. Secondly, metal ions in the polymer matrix are reduced chemically, thermally, or by UV/γ - irradiation. In the ex situ processes, the metal nanoparticles are chemically synthesized,

Vol. 10

NANOCOMPOSITES, METAL-FILLED

491

and their surface is organically passivated. The derivatized nanoparticles are dispersed into a liquid monomer which is then polymerized. In situ Methods. A simple, direct, and versatile in situ method was proposed by Watkins and McCarthy (49). In this method, an organometallic precursor is dissolved in supercritical fluid (SCF) carbon dioxide and infused into a solid polymer as SCF solution. Chemical or thermal reduction of the precursor to the zero-valence metal either in presence of SCF or subsequent to its removal produces metal domains within the solid polymer matrix. Such technique has been used in the synthesis of nanoscale platinum clusters embedded into poly(4-methyl1-pentene) and poly(tetrafluoroethylene), using dimethyl(cyclooctadiene)Pt(II) as metal precursor. There are multiple advantages associated with this approach: first, the high permeation of CO2 in virtually all polymers and the wide range of organic and organometallic reagents which are soluble in CO2 render this technique a generally useful scheme for the synthesis of polymer composites. Neither the polymer substrate nor the reaction product needs to be soluble in CO2 . Second, the sorption of CO2 is a very fast process which significantly enhances the kinetics of the penetrant absorption. The degree of polymer swelling, diffusion rates within the substrate, and the partitioning of penetrates between the SCF and the swollen polymer can be controlled by density mediated adjustments of solvent strength via changes in temperature and pressure. Coupled with manipulation of reaction rates, SCFs offer unprecedented control over composite composition and morphology. In addition, SCFs such as CO2 are gases at ambient conditions and the solvent dissipates rapidly upon the release of pressure. In the case of metalation using dimethyl(cyclooctadiene)Pt(II), the process effluent consists of CO2 and light hydrocarbons (methane and cyclooctane) derived from the precursor ligands. Bronstein and co-workers (50) reported an in situ method for the preparation of polymer-cobalt nanocomposites by mixing CO2 (CO)8 with a polyacrylonitrile copolymer or an aromatic polyamide in dimethylformamide (DMF). The cobalt carbonyl interacts with DMF giving the complex [Co(DMF)6 ]2+ [Co(CO)4 ]2 − , which is then converted to nanodispersed Co particles by thermolysis. Analogously, silver/polyimide nanocomposites have been prepared by Fragala` and co-workers (51) by thermolysis of a mixture of a metallorganic silver complex and polyamidic acid (PPA). The PAA precursor transforms in polyimide (PI) at a curing temperature compatible with the metallorganic precursor reduction temperature, so that at the same time the precursor decomposes, giving metallic silver particles, and the PAA transforms itself in PI, allowing the silver particles to remain trapped in the polymeric cage. Copper/polymer nanocomposites have been prepared in the solid state by Lyons and co-workers (52). A soluble precursor was synthesized by complexing poly(2-vinylpyridine) with copper formate (ie, Cu(HCO2 )2 ) in methanol. The thermal decomposition of the complex results in a redox reaction whereby Cu(II) is reduced to copper metal and the formate anion is oxidized to CO2 and H2 . By incorporating a reducing agent into the complex, the thermal decomposition reaction is not diffusion-limited, and nanocrystalline copper particles can be prepared in the solid state. Chen and co-workers (53) describe the synthesis of iron nanoparticles dispersed in poly(4-vinylpyridine) homopolymer and vinylstyrene-4-vinylpyridine copolymers by chemical reduction of thin films of iron chloride/polymer

492

NANOCOMPOSITES, METAL-FILLED

Vol. 10

nanocomposite precursors swollen by DMF. Idrazine was used as reducing agent. Such approach can be used because the metal ions have a high tendency to form ionic clusters in a polymeric matrix. The produced microscopic phases, which are separated by the hydrophobic portions of the polymer, have a diameter ranging from 2 to 10 nm. The chemical reduction of the dispersed ionic metal domains produces ultrafine metal particles. Huang and co-workers (54) prepared copper/poly(itaconic acid-co-acrylic acid) nanocomposites via in situ chemical reduction of the Cu2+ –polymer complex by hydrazine hydrate aqueous solution. Metal/polymer nanocomposites were prepared by Chen and co-workers (55) using dispersion of metal chlorides in polyurethane. Both polyurethane and metal salts were dissolved in N,N  -dimethylacetamide, followed by film casting and reduction of the metal salts by sodium borohydrate. The metal particle size depended on the type of metal salt used and on its concentration. Zhu and co-workers (56) developed a method for the preparation of metal/polymer nanocomposites based on γ -irradiation. In this method, the metal salt was dissolved in the organic monomer, and the formation of nanocrystalline metal particles and monomer polymerization was obtained simultaneously in solution by irradiation with a 60 Co γ -ray source, leading to a homogeneous dispersion of nanocrystalline metal particles in the polymer matrix. Also UV-irradiation has been used for the reduction of metal ions dispersed in polymer matrices (57). Wizel and co-workers (58) have described the use of ultrasounds in the in situ synthesis of composite materials made of polystyrene and iron. The propagation of ultrasound waves through a fluid causes the formation of cavitation bubbles. The collapse of these bubbles, described as an implosion in the hot-spot theory, is the origin of extreme local conditions: high temperatures (5000–25,000 K) and high pressures (1000 atm). The cooling rates obtained during the bubble collapse are greater than 107 K · s − 1 . These high cooling rates have been utilized in the sonication of Fe(CO)5 as a neat liquid or in solution, to produce amorphous iron nanoparticles. In addition, ultrasound radiation has been widely used for the preparation of polymers without the use of initiators (59). These two chemical processes have been combined for the fabrication of metal/polymer nanocomposites by sonication of styrene solution of iron carbonyl. Ex Situ Methods. Because of the high optical purity that can be achieved in the final product, the ex situ synthesis of metal/polymer nanocomposites is a very attractive technique, especially in the preparation of materials for optical applications. The particulate material of the required size, as obtained by a solution chemistry route, is stabilized by legand chemisorption (eg, thiols) in order to reduce their surface reactivity and tendency to agglomeration, and then it is incorporated into a castable polymeric matrix. Usually, the passivated nanoparticles are dispersed into a liquid monomer–initiator mixture (eg, styrene or methyl methacrylate activated by benzoyl peroxide), which is then thermally polymerized. Because a little amount of filler is required, the polymerization behavior of the monomer is not significantly influenced by the presence of passivated nanoparticles (the reaction rate is only slightly decreased (60)). Gonsalves and co-workers (60,61) have prepared surface-derivatized gold particles by phase-transfer reaction of gold ions with dodecanthiol and used these particles for the synthesis of poly(methyl methacrylate) based nanocomposites. Carotenuto (62) has developed

Vol. 10

NANOCOMPOSITES, METAL-FILLED

493

a new technique for the preparation of bulk quantities of atomic clusters of gold, derivatized by thiols by a modification of the well-known Polyol Process technique (63). The ex situ approach has been used also in the synthesis of silver/polyethylene nanocomposites (64,65). A variety of high nuclearity metal cluster compounds which can be used for the polymeric nanocomposite preparation have been recently synthesized and studied (66–76). Clusters of noble and easily reducible metals (eg, Ag, Au, Pd) can be obtained by alcoholic reduction of metal ions in presence of polymeric stabilizers. Poly(vinylpyrrolidinone), poly(vinyl alcohol), poly(methylvinyl ether), etc are usually utilized. High quality polymer-protected metal clusters can be obtained at the end of reaction simply by addition of a nonsolvent liquid which causes the nanocomposite separation from the reactive mixture (coprecipitation) (77–79). However, the use of such nanocomposites is limited because of their excessive moisture sensitivity.

Characterization Techniques For the comprehension of mechanisms involved in the appearance of novel properties in polymer-embedded metal nanostructures, their characterization represents the fundamental starting point. The microstructural characterization of nanofillers and nanocomposite materials is performed mainly by transmission electron microscopy (TEM), large-angle X-ray diffraction (XRD), and optical spectroscopy (UV–vis). These three techniques are very effective to determine particle morphology, crystal structure, composition, and grain size (48). Of the many techniques which have been used to study the structure of metal/polymer nanocomposites, TEM has undoubtedly been the most useful. This technique is currently used to probe the internal morphology of nanocomposites. As visible in Figure 4, high quality images can be obtained because of the presence, in the sample, of regions that do not allow the high voltage electron beam passage (ie, the metallic domains) and region perfectly transparent to the electron beam (ie, the polymeric matrix). High resolution TEM (HRTEM) allows morphological investigations with a resolution of 0.1nm, and thus this technique makes possible to accurately image nanoparticle sizes, shapes, and in some cases even inner atoms (80) (see Fig. 5). Large-angle X-ray powder diffraction (XRD) has been one of the most versatile techniques utilized for the structural characterization of nanocrystalline metal powders. The modern improvements in electronics, computers, and X-ray sources have allowed it to become an indispensable tool for identifying nanocrystalline phases as well as crystal size and crystal strain (see Fig. 6). The comparison of the crystallite size obtained by the XRD diffractogram using the Scherrer formula with the grain size obtained from the TEM image allows to establish if the nanoparticles have a mono- or polycrystalline nature (11). Metal clusters are characterized by the surface plasmon resonance, which is an oscillation of the surface plasma electrons induced by the electromagnetic field, and consequently their microstructure can be indirectly investigated by optical spectroscopy (UV–vis spectroscopy). The characteristics of this absorption (shape, intensity, position, etc) are strictly related to the nature, structure, topology, etc

494

NANOCOMPOSITES, METAL-FILLED

Vol. 10

Fig. 4. TEM micrograph of metal gold clusters embedded in poly(vinylpyrrolidinone).

of the cluster system. In fact, the absorption frequency is a fingerprint of the particular metal, the eventual peak splitting reflects aggregation phenomena, the intensity of the peak is related to the particle size, the absorption wavelength is related to the particle shape, the shift of the absorption with increasing of temperature is indicative of a cluster melting, and so on. For bimetallic particles information about inner structure (intermetallic or core/shell) and composition can be obtained from the maximum absorption frequency (81). Differently from out-line techniques (eg, TEM, XRD) this method allows on-line and in situ cluster sizing and monitoring of morphological evolution of the system (see Fig. 7). This method has been used also in the study of cluster nucleation and growth mechanisms (82,83). In addition to multielectron transition (ie, plasmon absorption) also single-electron transition (interband transition) can be detected and studied by optical spectroscopy in order to obtain important microstructural information. This outline of the principal characterization techniques for nanocomposite materials and nanosized metal fillers is far from being complete. Advances in Raman spectroscopy, energy dispersive spectroscopy, infrared spectroscopy, and many other techniques are of considerable importance as well. In fact, the success that nanostructured materials are having in the last few years is strictly related to the advanced characterization techniques which are available today.

Vol. 10

NANOCOMPOSITES, METAL-FILLED

Fig. 5. HRTEM image of a very small metal cluster.

Fig. 6. XRD pattern of nanosized gold clusters.

495

496

NANOCOMPOSITES, METAL-FILLED

Vol. 10

Fig. 7. UV–vis spectra of a colloidal gold suspension during the cluster growth. Temperature = 60◦ C; —— 5 min, – – – – 10 min, - - - - 15 min.

Applications Applications of metal/polymer nanocomposites have already been made in different technological fields. However the use of a much larger number of devices based on these materials is predicted for the near future. Because of the plasmon surface absorption band, atomic clusters of metals can be used as pigments for optical plastics. The color of the resulting nanocomposites is light-fast and intensive, and these materials are perfectly transparent, since the cluster size is much lower than light wavelength. Gold, silver, and copper can be used for color filter application. Also UV absorbers can be made for example by using Pd clusters. The plasmon surface absorption frequency is modulated by making intermetallic particles (eg, Pd/Ag, Au/Ag) of adequate composition. As shown in Figure 8, polymeric films containing uniaxially oriented pearlnecklace type of arrays of nanoparticles exhibit a polarization-dependent and tunable color (64,65,84–86). The color of these systems is very bright and can change strongly, modifying the light polarization direction. These materials are obtained by dispersing metal nanoparticles in polymeric thin films and subsequently reorganizing the dispersed phase into pearl-necklace arrays by solid-state drawing at temperature below the polymer melting point. The formation of these arrays in the films is the cause of a strong polarization-direction-dependent color which can

Vol. 10

NANOCOMPOSITES, METAL-FILLED

497

Fig. 8. (a) UV–vis spectra of drawn polyethylene/silver nanocomposites containing 4 wt% dodecanethiol coated silver nanoparticles in linearly polarized light. (b) Absorption spectra of drawn polyethylene/silver nanocomposite film annealed for 15 h at 180◦ C. The angle between the polarization direction of the light and the drawing direction of the film is given (64).

be used in the fabrication of liquid-crystal color display (see Fig. 9) and special electrooptical devices (see Fig. 10). Surface plasmon resonance has been used to produce a wide variety of optical sensors, eg, systems which are able to change their color in presence of specific analytes. These devices can be used as sensors for immunoassay, gas, and liquid (see Fig. 11) (48). Metals are characterized by ultrahigh/low refractive indices and therefore they can be used to modify the refractive index of optical plastics (87–89). Ultrahigh/low refractive index optical plastics can be used in the waveguide technology (eg, planar waveguides and optical fibers). Plastics doped by atomic clusters of ferromagnetic metals show magnetooptical properties (ie, when subject to a strong magnetic field, they can rotate the vibration plane of a plane-polarized light) and therefore they can be used

Fig. 9. Schematic representation of (a) a conventional color twisted nematic liquid-crystal display; and (b) display set-up containing a color polarizing filter (64).

498

NANOCOMPOSITES, METAL-FILLED

Vol. 10

Fig. 10. Twisted nematic liquid-crystal cell equipped with a polyethylene/silver nanocomposite filter (65).

as Faraday rotators (90,91). These devices have a number of important optical applications (eg, magneto-optic modulators, optical isolators, optical shutters). Nanosized metals (eg, gold, silver) have attracted much interest because of the nonlinear optical polarizability, which is caused by the quantum confinement of the metal’s electron cloud (92) (see NONLINEAR OPTICAL PROPERTIES). When irradiated with light above a certain threshold power, the optical polarizability deviates from the usual linear dependence on that power. By incorporating these particles into a clear polymeric matrix, nonlinear optical devices can be made in a readily processable form (61). These materials are used to prepare a number of devices for photonics and electrooptics (see ELECTROOPTICAL APPLICATIONS). Contact-free dispersion of noble metal clusters in polymer can be also used as nonporous catalytic membranes (93–95). Traditionally, nonporous catalytic separation layers are composed of palladium or palladium alloy foils. Reduction in

Fig. 11. Nanosized gold-based optical sensor for biological assay.

Vol. 10

NANOCOMPOSITES, METAL-FILLED

499

thickness by metal film deposition on special supports increases the hydrogen permeability and reduces the costs for the precious metal. Polymeric matrix filled by a catalytic nanosized metal offers in addition to the above advantages, the possibility of using the catalytic properties of nanoscale metal catalysts. One of the most important characteristics of polymers filled by metal microclusters is their ability to absorb microwave radiation, producing heat (96). These plastic materials can be processed by using new technologies based on the microwave heating and can be used for food packaging, microwave shielding, RADAR camouflage, plastic welding by microwaves, etc. Metal/polymer nanocomposites can have many other important applications. For example, nanoparticles embedded into poly(vinylpyrrolidinone) can be used for the electroless plating of polymeric, ceramic, and semiconductor substrates (93–98). These materials have also been used for the preparation of “smart” systems that experience a reversible alteration of their properties upon exposure to light. They are used as infrared barriers against exposures to intense solar light or fires (99).

Summary Metal cluster–polymer systems are receiving increased attention because of the interesting properties and large potentialities for technological applications. In particular, these nanocomposite systems can provide simple, low cost options for obtaining tailored materials with high promise for various catalytic, optical, magnetic, and electronic applications. Usually, the control of the nanoparticle morphology, ie, particle size, size distribution, shape, and composition, is of main interest. Nanoparticle features can be varied by selection of preparation method and variation of experimental conditions. In addition to the versatility of the nanoparticle features and the polymer morphologies, options can be provided to tailor the topology of the metal–polymer systems. This includes for example the uniaxially oriented pearl-necklace type distribution of metal nanoparticles within the polymer matrix. For a variety of applications special topologies are important and determining factors for tuning the composite material performance. In conclusion, further research activity in this field is extremely important for the development of advanced devices for functional applications based on these materials.

BIBLIOGRAPHY 1. A. B. R. Mayer, Mater. Sci. Eng., C 6, 155 (1998). 2. W. Caseri, Macromol. Rapid Commun. 21, 705 (2000). 3. W. R. Moser, Advanced Catalysts and Nanostructured Materials: Modern Synthetic Methods, Chapman & Hall, New York, 1996. 4. R. C. Ashoori, Nature 379, 413 (1996). 5. G. Carotenuto, Polym. News 25, 191 (2000). 6. G. Carotenuto, Polym. News 25, 260 (2000). 7. G. Carotenuto, Polym. News 25, 365 (2000).

500 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49.

NANOCOMPOSITES, METAL-FILLED

Vol. 10

G. Carotenuto, Polym. News 26, 48 (2001). G. Carotenuto, Polym. News 26, 119 (2001). G. Carotenuto, Polym. News 26, 234 (2001). G. Carotenuto, Polym. News 26, 278 (2001). G. Carotenuto, Polym. News 26, 342 (2001). C. U. Shepard, Sill. J. (Am. J. Sci. Arts) 28(1), 145 (1835). H. Ambronn, Kgl. Sachs. ¨ Ges. Wiss. 8, 613 (1896). H. Ambronn and R. Zsigmondy, Ber. Sachs. ¨ Ges. Wiss. 51, 13 (1899). H. Ambronn, Kgl. Sachs. ¨ Ges. Wiss. 8, 613 (1896). H. Ambronn, Kgl. Sachs. ¨ Ges. Wiss. 8, 613 (1896). H. Ambronn and R. Zsigmondy, Ber. Sachs. ¨ Ges. Wiss. 51, 13 (1899). E. Kolbe, Ph.D. dissertation, Jena, Germany, 1913 (cited by A. Frey-Wyssling, Protoplasma 27, 372 (1937)). F. Kirchner and R. Zsigmondy, Ann. Phys., vierte Folge (Drude’s Ann.) 15, 573 (1904). J. C. Maxwell Garnett, Philos. Trans. R. Soc. London, A 203, 385 (1904). J. C. Maxwell Garnett, Philos. Trans. R. Soc. London, A 205, 237 (1906). F. Braun, Ann. Phys., vierte Folge (Drude’s Ann.) 16, 238 (1905). A. Frey, Jahrbuch Wiss. Bot. 67, 597 (1927). A. Frey, Z. Wiss. Mikrosk 42, 421 (1925). A. Frey, Naturwiss 13, 403 (1925). F. Bion, Helv. Phys. Acta 1, 165 (1928). E. H. Land and C. D. West, in J. Alexander, ed., Colloid Chemistry, Reinhold, New York, 1946, p. 160. H. Ambronn, Z. Wiss. Mikrosk. 22, 349 (1906). A. Frey-Wyssling, Protoplasma 27, 563 (1937). E. H. Land, J. Opt. Soc. Am. 41, 957 (1952). S. Berkman, J. Bohm, and H. Zocher, Z. Phys. Chem. 124, 83 (1926). A. Frey-Wyssling, Protoplasma 27, 372 (1937). U. Kreibig and M. Vollmer, in J. P. Toennies, ed., Optical Properties of Metal Cluster, Springer-Verlag, Berlin, 1993. R. E. Newnham, D. P. Skinner, and L. E. Cross, Mater. Res. Bull. 13, 525 (1978). A. Ros´en, Adv. Quantum Chem. 30, 235 (1998). K. J. Klabunde, ed., Nanoscale Materials in Chemistry, Wiley-Interscience, New York, 2001. M. Wautelet, Solid State Commun. 74, 1237 (1990). Ph. Buffat and J. P. Borel, Phys. Rev. A 13, 2287 (1997). Q. Jiang, Z. Zhang, D. T. Hsu, H. Y. Tong, and M. Iskandar, J. Mater. Sci. 34, 5919 (1999). G. L. Allen, R. A. Bayles, W. W. Gile, and W. A. Jesser, Thin Solid Films 144, 297 (1986). C. R. M. Wronski, Br. J. Appl. Phys. 18, 1731 (1967). Z. L. Wang, J. M. Petroski, T. C. Green, and M. A. El-Sayed, J. Phys. Chem. B 102, 6145 (1998). Y. Zhang, S. Liao, Y. Fan, J. Xu, and F. Wang, J. Nanoparticle Res. 3, 23 (2001). R. D. Rieke and S. E. Bales, J. Am. Chem. Soc. 96, 1775 (1974). J. Zhang, Z. Wang, J. Liu, S. Chen, and G. Liu, Self-Assembled Nanostructures, Kluwer Academic, New York, 2003. J. S. Miller and M. Drillon, eds., Magnetism: Molecules to Materials III, Nanosized Magnetic Materials, Wiley-VCH, Weinheim, Germany, 2002. H. S. Nalwa, ed., Nanostructured Materials and Nanotechnology, Academic Press, San Diego, Calif., 2001. J. J. Watkins and T. J. McCarthy, Chem. Mater. 7, 1991 (1995).

Vol. 10

NANOCOMPOSITES, METAL-FILLED

501

50. M. Bronstein, E. Sh. Mirzoeva, M. V. Seregina, P. M. Valetsky, S. P. Solodovnikov and R. A. Register, ACS Symp. Ser. 622, 102 (1996). ` G. Compagnini, G. Malandrino, C. Spinella, and O. Puglisi, Eur. Phys. 51. M. E. Fragala, J. D 9, 631 (1999). 52. A. M. Lyons, S. Nakahara, M. A. Marcus, E. M. Pearce, and J. V. Waszczak, J. Phys. Chem. 95, 1098 (1991). 53. L. Chen, W. J. Yang, and C. Z. Yang, J. Mater. Sci. 32, 3571 (1997). 54. C. Huang and C. Z. Yang, Appl. Phys. Lett. 74, 1692 (1999). 55. G. M. Nasr and M. M. Badawy, Polym. Int. 38, 249 (1995). 56. Y. Zhu, Y. Qian, X. Li and M. Zhang, Chem. Commun. 1081 (1997). 57. M. Zheng, M. Gu, Y. Jin and G. Jin, Mater. Res. Bull. 36, 853 (2001). 58. S. Wizel, S. Margel and A. Gedanken, Polym. Int. 49, 445 (2000). 59. P. Kruus, M. O’Neill and D. Robertson, Ultrasonics 28, 304 (1990). 60. K. E. Gonsalves, G. Carlson, X. Chen, S. K. Gayen, R. Perez and M. Jose-Yacaman, Nanostruct. Mater. 7, 293 (1996). 61. K. E. Gonsalves, G. Carlson, X. Chen, J. Kumar, F. Aranda, R. Perez and M. Jose-Yacaman, J. Mater. Sci. Lett. 15, 948 (1996). 62. G. Carotenuto and L. Nicolais, J. Mater. Chem. 13, 1038 (2003). 63. T. Sugimoto, ed., Fine Particles Synthesis, Characterization, and Mechanism of Growth, Marcel Dekker, Inc., New York, 2000, p. 460–496. 64. Y. Dirix, C. Bastiaansen, W. Caseri and P. Smith, J. Mater. Sci. 34, 3859 (1999). 65. Y. Dirix, C. Bastiaansen, W. Caseri and P. Smith, Adv. Mater. 11, 223 (1999). 66. R. L. Whetten, J. T. Khoury, M. M. Alvarez, S. Murty, I. Vezmar, Z. L. Wang, P. W. Stephens, C. L. Cleveland, W. D. Luedtke and U. Landman, Adv. Mater. 8, 428 (1996). 67. R. L. Whetten, M. N. Shafigullin, J. T. Koury, T. G. Scaaff, I. Vezmar, M. M. Alvarez and A. Wilkinson, Acc. Chem. Res. 32, 397 (1999). 68. W. D. Luedtke and U. Landman, J. Phys. Chem. 100, 13324 (1996). 69. P. C. Ohara, D. V. Leff, J. R. Heath and W. M. Gelbart, Phys. Rev. Lett. 75, 3466 (1995). 70. R. L. Whetten, M. N. Shafigullin, J. T. Khoury, T. G. Schaaff, I. Vezmar, M. M. Alvarez and A. Wilkinson, Acc. Chem. Res. 32, 397 (1999). 71. Z. L. Wang, S. A. Harfenist, R. L. Whetten, J. Bentley and N. D. Evans, J. Phys. Chem. B 102, 3068 (1998). 72. M. M. Alvarez, J. T. Khoury, T. G. Scheeff, M. Shafigullin, I. Vezmar and R. L. Whetten, Chem. Phys. Lett. 266, 91 (1997). 73. A. Badia, W. Gao, S. Singh, L. Demers, L. Cuccia and L. Reven, Langmuir 12, 1262 (1996). 74. C. B. Murray, S. Sun, W. Gaschler, H. Doyle, T. A. Betley and C. R. Kagan, IBM J. Res. Dev. 45, 47 (2001). 75. G. Katabi, Y. Koltypin, X. Cao and A. Gedanken, J. Cryst. Growth 166, 760 (1996). 76. Z. L. Wang, S. A. Harfenist, I. Vezmar, R. L. Whetten, J. Bentley, N. D. Evans and K. B. Alexander, Adv. Mater. 10, 808 (1998). 77. G. Carotenuto, G. P. Pepe, and L. Nicolais, Eur. Phys. J. B 16, 11 (2000). 78. G. Carotenuto, G. P. Pepe, L. Parlato, and L. Nicolais, Mater. Eng. 11, 261 (2000). 79. G. Carotenuto, Appl. Organomet. Chem. 15, 1 (2001). 80. M. J. Y. Yacaman, R. Perez, P. Santiago, M. Banaissa, K. Gonsalves, and G. Carlson, Appl. Phys. Lett. 69, 913 (1996). 81. H. Remita, J. Khatouri, M. Treguer, J. Amblard, and J. Belloni, Z. Phys. D 40, 127 (1997). 82. G. Carotenuto, S. DeNicola, G. P. Pepe, and L. Nicolais, Eur. Phys. J. B 24, 437 (2001). 83. G. Carotenuto, S. DeNicola, and L. Nicolais, J. of Nanoparticle Res. 3, 469 (2001). 84. N. A. F. Al-Rawashdeh, M. L. Sanfrock, C. J. Seugling, and C. A. Foss, J. Phys. Chem. 102, 361 (1998).

502 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99.

NANOCOMPOSITES, METAL-FILLED

Vol. 10

J. C. Wittmann and P. Smith, Nature 352, 414 (1991). A. H. Lu, G. H. Lu, A. M. Kessinger, and C. A. Foss, J. Phys. Chem. B 101, 9139 (1997). L. Zimmerman, M. Weibel, W. Caseri, and U. W. Suter, J. Mater. Res. 8, 1742 (1993). M. Weibel, W. Caseri, U. W. Suter, H. Kiess, and E. Wehrli, Polym. Adv. Technol. 2, 75 (1991). L. Zimmerman, M. Weibel, W. Caseri, U. W. Suter, and P. Walther, Polym. Adv. Technol. 4, 1 (1992). A. H. Lu, G. H. Lu, A. M. Kessinger, and C. A. Foss, J. Phys. Chem. B 101, 9139 (1997). K. Baba, F. Takase, and M. Miyagi, Opt. Commun. 139, 35 (1997). K. Puech and W. J. Blau, J. Nanoparticle Res. 3, 13 (2001). L. Troger, H. Hunnefeld, S. Nunes, M. Oehring, and D. Fritsch, Z. Phys. D 40, 81 (1997). D. Fritsch and K. V. Peinemann, Catal. Today 25, 277 (1995). M. Ohtaki, M. Komiyama, H. Hirai and N. Toshima, Macromolecules 24, 5567 (1991). D. Acierno, C. Leonelli, G. C. Pellacani, eds., Application of the Microwave Technology, Mucchi Editore, Modena, Italy, 2000. E. P. Boonekamp, J. J. Kelly, and L. G. J. Fokkink, Langmuir 10, 4089 (1994). A. M. T. Van der Putten, J. W. G. De Bakker, and L. G. J. Fokkink, J. Electrochem. Soc. 139, 3475 (1992). G. A. Gaddy, J. Mclain, S.-W. Huang, B. L. Slaten, and G. Mills, in Proceedings 67th Annual Meeting of the APS Southeastern Section, Mississippi State, 2000.

G. CAROTENUTO L. NICOLAIS Institute for the Composite Material Technology, National Research Council

NANOCOMPOSITES, POLYMER–CLAY.

See Volume 3.