Metal-polymers have shining future

By mixing metal compounds with polymers in a suitable solvent, it is possible to produce low cost printed patterns of ceramic on the nanoscale. The potential applications for this include high temperature superconducting circuitry and intricate three dimensional nanostructures. No commercial products have yet been produced, but the technologies being developed are the subject of a number of collaborations aimed at doing just that. These include novel optical, electrical and electronic devices and even decorative finishes. The most immediate possibilities in sight include: lower cost, more efficient, painted or sprayed on photovoltaic cells, less expensive and more efficient catalytic converters, and the sorts of multi-layered actuators that were originally made by hand for the octopus arm inspired devices described in Eureka's September 2008 edition. "We started with metal nitrates plus polymer in solvent mixtures with the intention of making optical coatings," says Professor Ullrich Steiner in the University of Cambridge's Cavendish Laboratory. The precursor mixture is then deposited on the substrate of interest, either by spin coating or with the help of a rubber stamp, and then baked. The rubber stamp is made with capillary holes that draw solvent into the rubber, leaving the polymer behind, loaded with metal. The polymer keeps the metal precursor molecules in position, until it is burned away, leaving the metal behind in the form of a pattern of ceramic, usually oxides. In this way, starting with titanium nitrate, it is possible to produce arrays of columns, 250nm in diameter and 200nm high, to produce photonic crystals or even visible spectrum cloaking devices. It was then found that if the mixture contained lead and titanium nitrates and it was deposited on a substrate of strontium titanate, the final result is a patterned deposit of the piezoelectric material, lead titanate. The crystals are closely aligned along all three axes because they are grown on the strontium titanate matrix. This could be used to make arrays of micro actuators at very low cost. While the movement delivered by a single piezoelectric crystal is very small, there are many ways of multiplying it. Taking the idea one stage further, mixing the nitrates of bismuth, strontium calcium and copper with polymethacryclic acid in dimethylfuran, spin coating it on strontium titanate, drying and baking it results in an oriented growth of crystals of the high temperature superconductor, Bi2212. "But does it work?" says Professor Steiner. "Yes it does!" The material has zero resistance at 80°K, and a critical temperature of 81 to 83°K, versus the 92°K quoted in the literature for bulk material. Critical current is 105 to 106A/cm2 as opposed to 3x106A/cm2 in the literature. And the material has been printed as lines just 500nm across. High temperature superconducting printed circuits would be a breakthrough in many parts of power electrical and electronic engineering. But whereas these printed structures are what Professor Steiner described as, 'top down', even more dramatic are the three dimensional, self assembling structures he and his colleagues have developed, that he describes as, 'bottom up'. By linking two immiscible polymers at certain points with chemical groups so that they form block copolymers, it is possible to form three dimensional structures where one polymer is dispersed within the other as spheres, rods, lamella, or a regular interpenetrating network, according to the relative amounts. Block copolymers have many uses. But what Professor Steiner's team has been doing is to dissolve one of the polymers away then plate in a metal, or incorporate a metal alkoxide into one of the polymers and burn it away in an oven to leave a metal oxide structure. The soluble polymer can be made in the form of rods, which can be used to make wires. These will normally lie down in the plane of the polymer. However, applying an electric field makes them stand up. After plating in the metal rods, the other polymer can then be dissolved or baked away to leave an array of free standing wires. This makes it possible to make an array of free standing platinum wires just 10nm in diameter. But these cannot easily be used separately as they tend to stick to each other owing to Van der Waals forces and the effects of moisture, although they could be used as field emitters. The lamellae also have the potential to make actuators in the form of a stack of elastomer sheets. These are interleaved with conducting layers so that when a voltage is applied to alternate layers, they attract each other electrostatically. The thinner the interleaving layers of elastomer, the better they work. And Professor Steiner's method would potentially allow them to be made a lot thinner. But the form that appears to have the most immediate commercial application is the inter penetrating structure described as, gyroids. By making up a block copolymer of polyimide and polyethylene oxide with titanium isopropoxide, it is possible to produce a regular structure of titanium oxide that greatly improves the efficiency of dye sensitised, photovoltaic solar cells. In these devices, which can potentially be sprayed or painted onto large areas of substrate, light is absorbed by a dye layer only about 1nm thick. The electrons are then conducted away by crystalline titanium oxide, which is n-type, while the holes are conducted away through a p-type matrix material that can be either a liquid or polymeric. In order to be more efficient, Professor Steiner says: "We want the dye to coat the structure so it has a lot of interface." Solar cells are currently made by buying titanium oxide nanoparticles and sintering them together. But Professor Steiner says: "The trouble with this is that there are parts that are either not connected or are connected in the wrong way." This causes problems when introducing both the dye and the conducting material. But this can be overcome by the gyroid construction, which ensures a very regular mesh structure on a 10nm cell size, which also, is an ideal structure for filtration. Making solar cells using this approach, 3.5mm thick cells achieve a solar to electrical conversion efficiency of 6.4% up to 8.5%, and 400nm thick cells achieve 1.7% efficiency. So why not make them thicker? The answer is, the high temperature experienced during the manufacturing process tends to induce stresses in the film that begin to fracture. "But," Professor Steiner says, "we can overcome that." Low efficiencies do not greatly matter if the solar cells can be made cheap enough to cover the sides of buildings. The aim now is to come up with cells that are as near as possible to 10% efficiency on a roll to roll process. Research continues. Pointers * It is possible to produce patterned piezo electric materials, advanced optical photonic devices and even high temperature superconducting circuitry on a sub micron scale using low cost, wet chemical technology. * By extending the methodology into three dimensions, it is possible to produce arrays of spheres, rods, lamellae, or interpenetrating gyroid networks. * The rods can be formed as nanowires, the lamellae may form the basis of novel actuators and the gyroids improve titanium oxide networks for dye sensitised solar cells.