Chain reaction
You wouldn’t construct a car or a washing machine by carving it out of a huge chunk of metal. But this is exactly how we produce things at the scale of nanometres (a billionth of a metre), including the billions of tiny switches that make up modern computer processors.
This process, which uses beams of light to carve nanoscale structures into a material such as silicon, is known as photolithography. It is fiddly, expensive and probably approaching its fundamental limits and so is ripe for replacement by something else. Ideally, that something else will involve tiny, nanoscale structures building themselves.
Unfortunately, such self-assembly, as it is known, is much easier said than done. You can’t just hand nanoparticles a set of instructions and let them get on with it; you have to set up the conditions in just such a way that the nanoparticles naturally come together to build the desired structure. And this has proved hard to do. Like any belligerent builder, nanoparticles have a tendency to do their own thing.
But now, by taking their cue from the production of plastics, a team of US and Canadian chemists led by Eugenia Kumacheva from the University of Toronto has identified a potential way for scientists to get more control over the self-assembly process. They unveil the details in an article in this week’s Science.
Plastics are made from polymers, which comprise lots of copies of the same or similar molecular units, known as monomers, linked together into long chains. Proteins, which are made up of long chains of amino acids, are a natural type of polymer.
Because plastics are so widely used, scientists understand quite a bit about how their component monomers link together to form polymers, allowing them a high degree of control over the process. One of the simplest ways in which a polymer can form is via step-growth polymerisation, in which single monomers naturally form chemical bonds with each other, leading to longer and longer chains; just like clicking magnetic beads together.
Kumacheva and her team have now developed nanoparticles that link together via a process very similar to step-growth polymerisation. Their nanoparticles consist of gold nanorods just 30–50 nanometres in length, with ends shaped like a crosshead screwdriver. To each of these ends, the chemists attached a special kind of polystyrene molecule, which is hydrophobic, meaning it has a strong aversion to water.
The chemists then dispersed a load of these gold nanorods in an organic liquid. Adding water to the organic liquid caused the gold nanorods to start linking together into chains, as the polystyrene molecules bound to each other to limit their exposure to the water.
The vast majority (over 90%) of the nanorods linked up with just two other nanorods, one at each end, either lining up with each other or coming together at a 90° angle. As a result, the nanorods formed a whole variety of different shape chains, from straight lines to circles. The process was also very slow, taking more than 10 minutes for a single nanoparticle to add itself to a chain.
Obviously, producing a load of slowly forming, randomly-shaped chains is not at the moment very useful. But the fact that scientists can now apply their extensive knowledge of polymerisation to self-assembly represents an important advance. Using the same approach with other, more complex nanoscale building blocks potentially offers a way for scientists to construct a whole range of interesting and useful structures.
Indeed, it might not just be computer processors that self-assemble in the future, but perhaps cars and washing machines as well.
