Carbon nanotubes can be pictured as individual sheets of graphite rolled into tubes a few nanometers in diameter.

Electrons on the nanotube behave as if they live in a one-dimensional (1D) world, and the remarkable electrical as well as mechanical properties of carbon nanotubes have been much studied since they were first described in 1991. Now their optical properties are being investigated and show similarly remarkable properties. Shining light on the nanotubes creates excitons (pairs of positive and negative charges), which move randomly along the nanotube until they collide with each other or reach a defect whereupon they annihilate, and it is known that these drunken walks have a different nature in 1D compared to 3D.

The nanotubes we have studied are wrapped in DNA to make them soluble in water and prevent them aggregating. We studied the decay of the exciton population on timescales between 10 -13 and 10 -9 s, and have found that exciton annihilation on these nanotubes has archetypal characteristics of a 1D reaction-diffusion system, with different power-law decays according to different regimes of exciton density and the presence of defects.

By varying the light intensity we can generate an average of less than one exciton per nanotube, or generate excitons whose spacing approaches the exciton size. We also study the optical response of plasmons which are excited simultaneously with the excitons. The decay of the exciton population is an important factor in determining the potential usefulness of carbon nanotubes in various optical applications, and also gives us information on exciton transport and interactions in strongly excitonic 1D molecular semiconductors.

I shall touch on some of the measurement challenges associated with these measurements. A major barrier to the application of nanotubes grown by common techniques is the variability in their diameter. I shall end with a brief survey of techniques being developed to control the size of nanotubes.