Nature combines hard and soft materials, often in hierarchical architectures, to get synergistic, optimized properties with proven, complex functionalities. Emulating natural designs in robust engineering materials using efficient processing approaches represents a fundamental challenge to materials chemists and engineers. Our recent work combined sol-gel processing with molecular self-assembly during evaporative processes like spin-coating or ink-jet printing (we named this approach "evaporation-induced self-assembly" [EISA], which is now generally accepted) as a simple, general means to prepare porous and composite nanostructures in thin film and particulate forms. Highly ordered porous materials are of interest for membranes, low-dielectric-constant (low-k) insulators, sensors, and catalysts. Hybrid (organic/inorganic) nanocomposites are of interest for tough materials mimicking seashells or for endowing materials with environmental responsiveness or adaptability.

Click images for larger view    “…the simplicity of this approach allows these well-defined nano-materials to be readily integrated into devices, which I feel should broadly advance the nanotechnology field.”

I suspect our papers on EISA are cited for the simplicity of the approach—starting with a homogeneous solution of precursors plus two-sided molecules called surfactants, solvent evaporation drives their self-assembly into a variety of useful periodic nanostructured architectures. Using these solutions as an ink allows us to integrate nanostructures into almost any kind of device or platform with simple operations like ink-jet printing.

The original idea of combining so-called sol-gel processing (the formation of mainly inorganic oxides from molecular precursors; see my book Sol-Gel Science, Academic Press, 1990) with molecular self-assembly was developed by zeolite chemists who wanted to develop catalytic materials with larger pore sizes to perform reactions on larger substrates. However, these materials were formed as rather ill-defined powders by precipitation. I felt that these materials prepared as continuous thin films could be of much interest for size-selective membranes, low dielectric constant insulators, sensors, and optical coatings. As self-assembly depends on the concentration of the two-sided surfactant molecules, I thought I could use solvent evaporation to concentrate the evolving systems in surfactant and therefore drive the continuous formation of thin film versions of these powders. Having demonstrated that (Nature, 1997), I realized that I could prepare composites (Nature, 1998) and could use any type of evaporative process like aerosolization (Nature, 1999) or ink-jet printing (Nature, 2000) to form them. By introducing photosensitive or photopolymerizable components, I was able to use light to pattern them (Science, 2000) or polymerize polymers confined within them (Nature, 2001).

Again the simplicity of this approach allows these well-defined nano-materials to be readily integrated into devices, which I feel should broadly advance the nanotechnology field. They also represent well-defined materials of interest for understanding fundamental properties of nanostructured solids. These two aspects are evident in a recent paper (Science, 2004) where we self-assembled nanocrystals into a 3D array integrated into a metal-insulator-metal architecture. Using this integrated device, we were able to determine for the first time the current-voltage scaling relationships for a highly ordered 3D nanocrystalline array.

Self-assembly as a means to organize material at the nanoscale has seen explosive growth over the past 10 years as it is perceived that it could be the replacement for lithography, allowing the IC industry to continue to follow Moore’s law. (However, this will require very fundamental studies on self-assembly so that we can scale it up in an industrial setting—see below). Whether or not ICs will be fabricated using self-assembly, this area of research will continue to grow, as it remains one of the few reliable means to create well-defined nanostructured materials. As most modern devices are composed of films or layers, the EISA process we developed will continue to be an important sub-discipline within the self-assembly field.

Many challenges remain in the area of molecular self-assembly. As indicated above, self-assembly must be fundamentally understood in order for it to be more than a laboratory curiosity. Recently, sophisticated small-angle scattering and molecular simulations have been employed to understand and predict it.

I believe there will be progress in incorporating environmentally active components within self-assembled structures to develop new classes of sensors, drug delivery platforms, and transducers.

Our self-assembled nanostructures, although beautiful, are boringly repetitive. To better emulate biology, we need to break symmetry and develop structure and function on a broader range of length scales. I expect there to be much future progress in so-called directed assembly, where active intervention, such as the application of an external field, can be used to perturb or modulate the assembly process. Currently, there is much interest in structure formation by energy-dissipating systems, where large-scale patterns often emerge. Combined with smaller-scale molecular self-assembly, this may be a means of developing hierarchical structures. We should consider compartmentalized synthesis, for example, within membrane-bound molds, where selective ion channels are used to spatially control pH, ionic strength, and the introduction of precursor ions. Such an approach could allow the formation of synthetic analogues to natural structures like diatoms.