INTRODUCTION
Polymer nanocomposites are composites with a polymer matrix and a filler with at least one dimension less than 100 nm. The fillers can be plate-like (clays), high aspect ratio nanotubes, and lower aspect ratio or equiaxed nanoparticles. While some nanofilled composites (carbon black[1] and fumed silica[2,3]-filled polymers) have been used for over a century, in recent years the dedicated research and development of nanofilled polymers has greatly increased. This is due to our increased ability to synthesize and manipulate a broad range of nanofillers and significant investment by government and industry in this field.
Current interest in nanocomposites has been generated and maintained because nanoparticle and carbon nano-tube-filled polymers exhibit unique combinations of properties not achievable with traditional composites. For example, the inclusion of equiaxed nanoparticles in thermoplastics, and particularly semicrystalline thermoplastics, increases the yield stress, the tensile strength, and Young’s modulus1-41 compared to pure polymer. Other examples include scratch-resistant transparent amorphous thermoplastic coatings.[5] These combinations of properties can be achieved because of the small size of the fillers, the large surface area the fillers provide, and in many cases the unique properties of the fillers themselves. As will be shown, in many cases these large changes in the material properties require small to modest nanofiller loadings. Unlike traditional micron-filled composites, these novel fillers often alter the properties of the entire polymer matrix while, at the same time, imparting new functionality because of their chemical composition and nanoscale size.
This article will give a general introduction to polymer nanocomposites and address what is unique to nanofillers compared to traditional micron-scale fillers. The second section will briefly address nanofiller surface modification and the third will provide specific examples of mechanical, electrical, and optical properties in nanoparticle-filled polymers. The last section provides a detailed description of the mechanical properties of nanotube-filled polymers and a brief description of some electrical and optical properties that have been reported.
WHAT MAKES NANOCOMPOSITES UNIQUE
The small size of nanofillers leads to several factors that distinguish nanocomposites from traditional composites. First of all, nanofillers are small mechanical, optical, and electrical defects compared to micron-scale fillers. This means that the addition of nanofillers to a polymer does not necessarily lead to a decrease in the ductility of the polymer and in some cases can increase it.[6,7] It also means that below about 50 nm,[8] many fillers do not scatter light significantly. Thus it is possible to make composites with altered electrical or mechanical properties that maintain their optical clarity. Finally, as small electrical defects, nanofillers do not concentrate electromagnetic fields as sharply as micron-scale fillers and indeed may act to trap charge and increase the electrical breakdown strength of polymers.
Secondly, although many properties of a material are said to be intrinsic, they often depend upon matter being assembled above a critical length scale. When the nano-particles decrease below this size, the properties of the particles can differ significantly from the bulk material; thus variations in melting temperature, color, magnetization, and charge capacity are often observed.1-9-1
Third, the small size of the fillers leads to an exceptionally large interfacial area in the composites. Fig. 1a shows the surface area per unit volume as a function of particle size for spherical particles that are ideally dispersed. If one compares the surface area of a 10-pm carbon fiber to that of a 1-nm single-walled nanotube (SWNT) for the same total volume of the two, the surface area increases by a factor of 10,000. In addition (Fig. 1b), the interparticle spacing decreases such that at small volume fractions of filler, the interparticle spacing is similar to the radius of gyration of the polymer 100 A). The high surface area becomes even more significant when one considers that there is an interaction zone (IZ) surrounding the filler. This is a region in which the structure and properties have been altered because of the presence of the filler. It could be a region of altered chemistry, polymer conformation, chain mobility, degree of cure, or crystallinity. This zone of affected polymer has been approximated to be between 2 and 9 nm thick,[10] but may be much larger. If we assume that this IZ is about 10 nm in thickness, then at 2.5 vol.% of a 20-nm equiaxed nanoparticle well dispersed, 37% of the polymer has different properties from the bulk polymer. Therefore the IZ can be a significant portion, if not the entire bulk, of the matrix. Thus the nanofillers can alter the expected properties of the composite considerably.
Fig. 1 (a) The surface area per unit volume as a function of particle size for spherical particles showing the large surface area in nanoparticles. (b) The interparticle spacing of nano-particles arranged on a simple cubic lattice showing the variation with particle size (15, 50, and 100 nm diameter) and volume fraction.
Fig. 2 Polymer-coated nanotubes observed in the fracture surface of MWNT reinforced polycarbonate composite.
An example of the influence of the IZ on behavior can be seen by monitoring the glass transition temperature, Tg. The Tg of a bulk part can be raised and lowered with the addition of nanoparticles due to the immobilization of polymer chains by the particles or, conversely, an increase in polymer mobility due to noninteracting particles. Both increasing[11-13] and decreasing[14,15] Tg cases have been shown. The physical nature and extent of this IZ has recently been probed through some recent work on multi-walled carbon nanotubes (MWNTs).[16] In this study, a solvent processing method was used to make MWNT polymer nanocomposites. Upon observation of the composite fracture surface, a polymer layer was observed on the nanotubes (Fig. 2) that had pulled out of the opposing side of the fracture. This ”sheath” was confirmed to be polymer from the matrix, but with obviously altered thermal and mechanical properties from the bulk. The thickness of this interfacial layer increased with chemical modification of the nanotubes.
Fig. 3 Atomic force micrographs showing the change in crystalline morphology for unfilled and nanofilled low-density polyethylene, (a) neat-low density polyethylene, (b) low-density polyethylene with 5 wt.% titania nanoparticles.
Nanoparticles can also influence the polymerization, curing, or crystallization aspects of polymer synthesis. For example, nanoparticles can serve as nucleation sites in semicrystalline polymers and result in changes in crystalline content and spherulite structure.[17- Fig. 3 shows an AFM micrograph of unfilled polyethylene and nano-particle-filled polyethylene. Note that in the case of the nanofilled polyethylene, the crystalline structure is much less organized.[18- Similar results have been seen in polyethylene terephthalate (PET).[19]
SURFACE MODIFICATION OF NANOFILLERS
The nanoscale sizes and subsequently higher surface energies of nanofillers lead to some unique challenges in the processing of these materials. The most critical of these challenges is dispersion of the nanofiller. Any agglomeration of the filler reduces the interfacial area in the composites and thus reduces the opportunity to take advantage of the unique nanofiller properties mentioned in the previous section. Thus aggregated nanoparticles are simply micron fillers.
Controlling the size and degree of agglomeration of nanoparticles is difficult due to their large radius of curvature and subsequent increase in surface energy. As these high surface area fillers tend to aggregate, silanes and organotitanates are used extensively both to tailor the particle surface properties to mimic the surrounding matrix and to lower their surface energy and reduce their tendency to agglomerate.[20-22] Other methods to alter the surface properties of the nanoparticles include radiation grafting,[23] chemical vapor deposition, and a host of complicated synthesis procedures that attempt to polymerize polymer chains off of initiating agents coupled to the surface.[24-26] An excellent review by Caruso[27] provides an extensive background on the modification of nano-particle surfaces. In addition to achieving better dispersion, these techniques control the nature of the interaction between the nanofillers and the polymer and thus the properties and size of the IZ.
Fig. 4 An SEM micrograph showing the bundled nature of multiwalled carbon nanotubes grown using a chemical vapor deposition process.
Carbon nanotubes not only tend to agglomerate but are often prepared in a bundle-like structure, as illustrated in Fig. 4. In order to take advantage of their high surface area for interacting with the polymer, the bundles have to be separated into individual nanotubes. For MWNT this can usually be accomplished with sonication.[28,29] For SWNT, exfoliation is a more difficult process but progress is being made.[30-34] The challenge is in exfoliating the bundles without shortening the SWNT and introducing significant numbers of defects. Once the nanotubes have been separated, it is important to disperse them uniformly in the polymer matrix, preventing agglomeration of nanotubes. This is accomplished with functional groups which also mediates the interaction of the nanotubes with the polymer.
Noncovalent surface modification of carbon nanotubes includes all treatments that cause a change in the functional groups that face the solvent (or the polymer), without modifying the chemical nature of the nanotube. The advantage of the noncovalent surface modification is that the basic structure and hence the mechanical and electrical properties of the tubes are not affected due to the modification.1-35-1 Noncovalent attachment is possible if there is a secondary bonding between these groups and the surface of the nanotube.[36,37] For example, wrapping of the nanotubes by polymer chains, in particular conjugated polymers, has been observed.[36,38-40]
Covalent attachment of chemical groups to the outer wall of the nanotubes can occur particularly at defect sites. One example is the attack of the defect sites by concentrated nitric acid, in order to form carboxylic acid groups.[41] The reaction with nitric acid also eliminates the catalysts that are left from the nanotubes preparation process. This reaction has been applied on SWNT and MWNT. The resultant carboxylated nanotubes can then be further covalently modified by means of reactions based on the carboxylic acid groups.[42,43] There are also other variations of the chemical oxidation of carbon nanotubes (for example, applying a mixture of sulfuric acid and H2O2[44]). The presence of carboxylic acid groups on the nanotube walls enables various reactions for the further attachment of functional groups.[43,45-50]
Other types of covalent surface modification of carbon nanotubes are based on chemical reactions between the carbon-carbon bond structures and specific reagents such as fluorination[51] or on radical attachments.[52,53] These reactions and many others[54-58] enable tailoring of the nanofillers to the specific application and environment. The potential for using surface modification to improve the properties of nanotube-reinforced polymer composites has just begun to be explored.
