The present credit crunch is forcing everyone to save money, and chemists are no exception. A good cost-cutting measure is to perform reactions on a small scale, thereby reducing the outlay on raw materials and minimizing the energy required to drive the reactions. Reporting in Nature Chemistry, Anzenbacher and Palacios describe the ultimate in miniature reaction vessels — junctions formed when two polymer nanofibres are fused together (P. Anzenbacher Jr & M. A. Palacios Nature Chem. doi:10.1038/nchem.125; 2009).

The authors prepared nanofibres — each hundreds of times narrower than a human hair — from readily available polymers, and loaded them with various chemical reactants. They then laid fibres containing different reactants across each other and exposed them to either heat or solvent vapour. This caused the fibres to fuse together, forming junctions that defined discrete chemical reactors with attolitre-scale volumes (1 attolitre is 10−18 litres). The junctions contained as few as 1,500 molecules of each reactant.

To illustrate the principle of their ultra-small reactors, Anzenbacher and Palacios doped fibres with two non-fluorescent compounds that form a fluorescent product when they react. When the two types of fibres were overlapped to form a random mat and then heated, fluorescence at the fused junctions clearly indicated the formation of the reaction product (pictured).

The authors showed that several types of reactions can be performed in their attoreactors, including those in which one of the reactants is polymeric (although reactions with two polymeric reactants are expected to be problematic, because polymers can't easily diffuse through the nanofibre matrix). Furthermore, the products can be analysed directly within the junctions using fluorescence measurements or mass spectrometry.

Several applications for these attoreactors suggest themselves. For example, libraries of nanofibres could be prepared in which each nanofibre is loaded with a different compound from the same chemical class. Selected libraries could then be reacted with each other, as in combinatorial chemistry, to prepare many different products quickly and easily. If the products could be screened directly for biological activity, this would be useful for the high-throughput preparation and testing of compounds in drug-discovery programmes.