The macromolecular matrix present in the composite shell of the blue mussel, Mytilus edulis, accounts for less than 1% of the shell by weight but is theorized to play a significant role in controlling the growth, morphology, and orientation of the CaCO3 that makes up the shell. The presence of several proteins in this matrix, only some of which have affinity for calcium, suggests a hierarchical structural model for the shell. Proteins were isolated under denaturing, reducing conditions and separated by centrifugation, gel electrophoresis, and high performance liquid chromatography. The major matrix proteins, both soluble and insoluble, were evaluated for amino acid composition, calcium binding, and glycosylation. Some N-terminal sequence data was collected. Non-proteinaceous components of the matrix were also analyzed. Comparison of the mussel shell matrix with the protein matrix of other molluscan systems suggests that this complexity is not unique to the mussel and may provide a key to the understanding of more generic biomineralization processes necessary for such applications as biomimetic ceramics.

The structure and crystallography of the nacre of red abalone, Haliotis rufescens, was studied by transmission electron microscopy imaging and diffraction. We found that the nacre structure is based upon hierarchical {110} twinning in aragonite with the following organization: (i) first generation twins between platelets having incoherent boundaries, (ii) second generation twins between domains having coherent boundaries within a given platelet, and (iii) nanometer-scale third generation twins within domains. Since the aragonite platelets nucleate and grow as separate crystals, this long-range crystallographic relationship between the inorganic units of a biological hard tissue indicates that the nucleation and growth process of crystals may be mediated by the organic matrix and that the organic template structure may also be long-range ordered. We propose a superlattice structure based on the possible twin variants and suggest that the organic matrix structure, or the arrangement of nucleation sites, is compatible with the superlattice. Multiple tiling based upon this superlattice allows all of the crystallographic and morphological platelet configurations observed in nacre.

In natural systems, structural macromolecules undergo prescribed recognition and assembly steps during synthesis and processing. These associations lead to more complex assemblies that exhibit useful multifunctional properties. Many of these processes are not well understood. Some aspects of these processes are presented using the fibrous protein polymer silk as an example. Issues such as polymer chain biosynthesis, chain interactions, processing into fibrils, and complex engineering into supra-assemblies are addressed and biochemical, spectroscopic and modeling studies are reviewed. Genetic level controls of chain composition, crystalline/amorphous domain distribution, chain aggregation, chain registry, silk I-silk II phase transitions, nematic liquid crystalline phase, loss of water, global molecular alignment, and solution spinning are some of the characteristics of this biological system that are addressed. Although some information is available at the molecular and macro-scale levels, a key issue is the paucity of information at the meso-scale level to fully understand the role of structural hierarchy in the silk fiber assembly process.

Nature has developed exceptional designs for resilient “soft” materials and for interfacing “soft-hard” materials which show capabilities well-beyond our present day technology. Most of the properties of biological materials – and a great deal of biological function – emanate from chemical and physical characteristics of interfacial structures (membranes). Although material interfaces in biology are complicated, it appears that these structures have been invariably optimized (with respect to harsh-environmental requirements) by the slow “evolutionary-driven” engineering. Some “hierarchical” features and characteristic length scales for biological membranes are directly related to architecture; but others are more subtle and arise because conformations of much of the molecular structure are randomized by thermal fluctuations.

Cylindrical-shaped cells of Bacillus subtilis (0.7 by 4 μm) are the building blocks of macrofibers, highly organized, helically twisted, multifilament structures millimeters to centimeters in length. The forces responsible for self-assembly and the cylinder-helix deformation trace to the assembly of cell wall polymers and restraint of the motions generated by cell growth. An hierarchical relationship exists involving: (i) molecular level events associated with cell surface assembly, which in turn govern, (ii) cellular level events concerned with motions that accompany cell growth, and these in turn drive, (iii) multicellular level events such as the folding and plying of cell filaments to form a mature macrofiber. Cell growth generates new material and engenders twisting of the cell cylinder along a screw axis as it elongates. The helix hand and degree of twist at the cellular level eventually dictate the hand and twist of the mature multifilament macrofiber. Although several different routes can lead to the initiation of macrofiber production, once initiated a repetitive cycle of folding and plying becomes established. The self-assembly proceeds until mechanical and geometrical factors preclude further folding cycles.

Hierarchical structures are possible in polymer liquid crystals (PLCs) since each molecule contains at least two kinds of building blocks that are not homeomorphic to each other. We discuss some examples of molecular structures and phase structures of monomer liquid crystals (MLCs) and PLCs: smectic phases formed by interdigitated MLC molecules; PLC molecule classification based on increasing complexity – and its consequences on properties of the materials; and formation and phase structures of LC-rich islands in PLCs and in PLC blends. Some rules pertaining to hierarchical structures are formulated. The knowledge of hierarchies is neccessary – but not sufficient – for intelligent procesing of PLCs and their blends and for achieving properties defined in advance. Computer modelling represents another important element of building materials to order.

It is well known that the structure of highly oriented liquid crystalline polymers (LCPs) can be characterized by a hierarchical fibrillar structural model. Structure models were developed for the lyotropic aramid fibers and the thermotropic aromatic copolyester fibers during the last two decades showing the existence of fibrillar hierarchies. Hierarchies of structure have also been commonly observed for the biological materials. Concepts learned from the latter are useful in materials science studies today. The nature of the smallest nanostructure that aggregates, the combination of these small structures, typically microfibrils, into larger structures and the interaction of these hierarchical entities are important to understanding their behavior. The architecture of the whole of the polymer or the biological material is a further important variable as is the relation of the process with that architecture. This paper discusses details of the structure of LCPs and draws an analogy between the materials science and biological hierarchies.

Conductive composite fibers are prepared by dry-jet wet-spinning of a strong-acid solution of rod-like polymers and phthalocyanine(Pc) compounds. A model yielding a hierarchical structure for a pure rigid-rod fiber is described first. The inclusion of NiPc in the spinning dope leads upon coagulation to isolated NiPc domains that are predicted to correspond to body centered cells (exhibiting various degrees of disorder). The observed dependence of fiber conductivity of composition is explained in terms of tunneling and percolation processes.

The crystal structures and crystalline morphologies of natural and synthetic polymers are briefly reviewed. Analogies and differences between these two systems are presented. Several examples of cross-fertilization of research in the two fields are presented, with emphasis, among natural polymers, on fibrous polypeptides and proteins.

Diverse microstructures observed in rigid natural composites have been linked to mechanical properties that are equal to and, at times, superior to those which have been designed into synthetic composite materials comprised of organic, metallic, and ceramic constituents. The constitution, properties, and microstructures of the natural and synthetic materials are described, along with what is known and what is unknown about the constituent components, and interfaces. Hierarchical approaches to designing with and the mechanical analyses of both natural and synthetic composites will be described.

Rigid rod polymers are of potential use in fibers for composites because of their high specific tensile strength and modulus as well as their high thermal stability. Some other common properties of systems meeting these criteria are a nonlinear elasticity (increasing Young's modulus with increasing tensile stress) and a negative coefficient of thermal expansion along the fiber axis. In composites, these two properties can combine to cause an increase of Young's modulus with increasing temperature and can lead to yielding or interfacial failure. The source of the nonlinear elasticity resides at two length scales: 1) the scale of the crystallites, (the reversible realignment of the crystallites along the fiber axis with increasing stress), and 2) at the scale of the molecules, (an inherent nonlinear elasticity of the molecules themselves). The former effect is reviewed and computational results for the latter are presented along with computational results for the molecular origin of the negative coefficient of thermal expansion.

Semi-crystalline thermoplastic polymers are being used increasingly as matrices in high performance fiber reinforced composites. The crystallization kinetics and morphology of these polymers have been studied extensively, but relatively little attention has been given to the effects of the reinforcing fibers on the crystallization process.

We have studied the effects of glass, carbon and aramid fibers on the rates of crystallization, the degree of crystallinity, and the glass transition temperature of such typical thermoplastics as poly(phenylene sulfide) and poly(ethylene terephthalate). Based on the isothermal crystallization studies using DSC, we find that, in general, reinforcing fibers increase the rates of crystallization and decrease the degree of crystallinity, the extent of these effects being dependent on the weight fraction of fiber in the composite, the specific type of fiber, and the nature of surface finishes (sizes) that may have been applied.

The spherulitic morphology that develops in these polymers during the crystallization process, as characterized by polarized light microscopy, is also affected by the reinforcing fibers. In many cases, transcrystalline regions develop near the fiber surface due to nucleation effects.

Continuous microlayer composites of polycarbonate (PC), a ductile glassy polymer, and styrene-acrylonitrile copolymer (SAN), a glassy relatively brittle polymer, were discussed. Microlayered systems composed of 49 to 776 continuous layers, in which the primary variable was the variation in thickness of the continuous layers between 30 and 2μm, were emphasized. The bulk properties of these microlayered composites showed a dramatic improvement in the toughness or ductility and in the fatigue properties as the layer thickness decreased. Investigation of the irreversible deformation processes revealed that when the layers were thicker (30μm, 49 layers) the SAN layers crazed and the PC shear banded in the usual manner. Subsequently the composite system fractured in a relatively brittle manner due to the development of large voids in the SAN layers. When the layer thickness was reduced to 2μm (776 layers) the entire system behaved in a ductile manner and both the PC and SAN shear banded due to a new cooperative process. This was analyzed by considering the micromechanics of these irreversible processes at the PC-SAN interface. As a result of this work, it is expected that ultra-thin layered structures of other alternating composite systems will reveal synergistic properties if the interfacial properties are designed so that the ductile component will dominate the yield and failure characteristics of the entire system.

This paper presents a detailed investigation of the hierarchical and structural organization of the collagen-based aggregates in ivory. Ivory from African elephants is selected as the prototype in this study. A sophisticated architecture composed of collagen fibers and hydroxyapatite-like particles is revealed by optical microscopy, scanning electron microscopy and transmission electron microscopy. X-ray diffraction and TEN with selected area diffraction are employed to analyze the structure. Electron spectroscopy for chemical analysis and infrared absorption spectroscopy give information about the composition and chemical environment of the atoms in ivory. It is found that the structure of ivory has a three-level hierarchical organization, which includes both organic and inorganic materials. In the structure the inorganic material exists inside an organic framework, located outside of the collagen fibrils in the extrafibrillar volume. This inorganic structure has a polycrystalline form. Both the chemical compositions and the chemical environment of the atoms in the hydroxyapatite-like particles in ivory are different from those in natural hydroxyapatite.

The insect cuticle is an excellent example of a natural, fiber-reinforced, polymeric composite consisting of chitin fibers embedded in a protein matrix. Optical and electron microscopy have been used to examine the structure and interaction of the constituents of the bessbeetle (Odontotaenius disjunctus) cuticle from the molecular to the macroscopic levels.

Molecular chains of the polysaccharide chitin (N-acetylglucosamine) are grouped together to form “fibrils” which are either dispersed throughout the matrix or combined to form larger “fibers”. The fibers are unidirectionally oriented within individual sheets or laminae which are stacked on top of one another at various angles forming a laminated structure.

The protein matrix is ductile upon initial deposition but then undergoes a crosslinking process which increases its shear stiffness, thereby improving load transfer between fibers. The matrix is bound to the chitin via beta linkages holding it together at both the fibril and fiber levels. The matrix has a fibrous morphology which provides adequate toughness in spite of the high degree of crosslinking.

Reference is made to designs observed in the bessbeetle cuticle which could be applied to man-made composites for improved performance primarily in the areas of damage tolerance and strength and stiffness coupled with low weight. For these designs to be implemented using synthetic materials, new or modified processing and fabrication methods are needed.

The structure-mechanical property relationships were studied in nacre, a laminated ceramicpolymer biocomposite found in seashell. Four-point bending strength and three-point bend fracture toughness tests were performed, and the results averaged 180 ± 30 MPa and 9 ± 3 MPa·m1/2, respectively, indicating that the composite is many orders of magnitude stronger and tougher than monolithic CaCO3,, which is the primary component of nacre. Fractographic studies conducted with a scanning electron microscope identified two significant toughening mechanisms in the well-known “brick and mortar” microstructure of nacre: (i) sliding of the aragonite platelets and (ii) ligament formation in the organic matrix. These toughening mechanisms allow for high energy absorption and damage tolerance and thereby prevent catastrophic failure of the composite. The structure of the organic matrix and the interfacial structure between the organic and inorganic components were studied with transmission electron microscopy by using both ion milled and ultramicrotomed sections with and without the intact aragonite platelets. We found that the organic matrix is indeed a multilayered composite at the nanometer scale but is thinner (about 100 Å) than reported in the literature. The morphology of the interfacial region between the organic and the inorganic layers suggests the presence of a structural “transitory” region that interlocks the two dissimilar phases.

Fiber reinforced polymer matrix composites have become very useful in chemical processing systems, in transportation applications such as automobiles, aircraft and boats, in electrical hardware and in sports equipment. Failures in these rarely involve gross cohesive fracture; usually leaks develop, the stiffness decreases, local delamination occurs, the dielectric properties degrade or the strength declines. Almost all of these failures can be traced to local cracking, either at the fiber matrix interface or within the matrix itself. The cracks are generated by mechanical loads, by thermal excursions or by fluid absorption, because the relevant properties of the fiber and the matrix often differ by orders of magnitude. Current technology attempts to avoid the cracks by maximizing the fiber-matrix adhesion and great progress has been made in this area. An alternative interposes a thin compliant layer, rubber, between the two constituents thereby reducing the stress concentrations which exist because of their greatly different properties. Cracking is inhibited, composite strength is increased and its energy absorption also rises; if the rubber layer is thin (a few thousand Angstroms) no loss of stiffness or heat resistance is evident.

The polycation conducting polymers, oxidized polypyrrole and polyalkylthiophene, possess the ability to form complexes with polyanionic DNA molecules through largely electrostatic interactions. This study demonstrated the solution uptake and binding of 32p radiolabeled DNA by conducting polymer thick films (50–100μm). Polypyrrole (PPy) was synthesized by electrochemical methods and poly(3-hexylthiophene) (PHT) and poly(3-undecylthiophene) (PUT) were synthesized by chemical methods. The DNA binding rates on PPy films were affected by DNA concentration and the oxidation state (measured as conductivity). The DNA kinetics support a diffusion limited model for binding. We measured DNA binding levels onto all three polymer films; PUT, PHT, and PPy. The binding levels increased in the same order as the conductivities of the polymer films. DNA binding onto oxidized PPy film was diminished upon electrochemical reduction. These observations showed, therefore, the binding may be linked with the positive charge sites responsible for conduction in the polymer films.

A selection of nanoscale membrane pores is being constructed by genetic manipulation of α-hemolysin (αHL), a 33.2 kDa polypeptide secreted by the bacterium Staphylococcus aureus, which can self-assemble into hexameric cylindrical channels -1 to 2 nm In Internal diameter. Ultimately, the new pores will be used to confer novel permeability properties upon materials such as thin films utilizing, for example, monolayer sheets of the hexamer. Recombinant αHL (r-αHL) has now been obtained in multimilligram amounts and purified to homogeneity after overexpression of the αHL gene in Escherichia coli. The properties of r-αHL are closely similar to those of αHL purified from S. aureus. Recent deletion mutagenesis experiments have given us new insight into the assembly mechanism of the pore. Three intermediates have been identified: a membrane-bound monomer; an oligomeric pore precursor; and the hexameric pore itself. Currently, point mutogenesis combined with chemical modification is being used to produce new pores of different internal diameter, with selectivity for the passage of molecules and Ions, and with gating properties (the ability to open and close in response to a physical stimulus, e.g. an electric field or light).

In the past few years, atom clusters with diameters in the range of 2–20 nm of a variety of materials, including both metals and ceramics, have been synthesized by evaporation and condensation in high-purity gases and subsequently consolidated in situ under ultrahigh vacuum conditions to create nanophase materials. These new ultrafine-grained materials have properties that are often significantly different and considerably improved relative to those of their coarser-grained counterparts owing to both their small grain-size scale and the large percentage of their atoms in grain boundary environments. Since their properties can be engineered during the synthesis and processing steps, cluster-assembled materials appear to have significant potential for the introduction of a hierarchy of both structure and properties. Some of the recent research on nanophase materials related to properties and scale are reviewed and some of the possibilities for synthesizing hierarchical nanostructures via cluster assembly are considered.