Many cellular processes are driven by biological motors, and a range of synthetic mimics have been developed that function as nanomachines. However, there has tended to be a notable difference between naturally occurring and artificial machines. Biological motors — for example those responsible for flexing muscles — drive chemical systems away from equilibrium, whereas machines synthesized by chemists move them towards it.

David Leigh had been constructing molecular machines for a decade when research in his lab at the University of Edinburgh, UK, took a new course. At a group meeting in 2003, Leigh asked PhD student Euan Kay to present the mechanisms that theoretical physicists had proposed for biomolecular motors. “He did a great job of explaining the different mechanisms in a language that organic chemists could understand,” recalls Leigh. “By the end of the presentation, it was as if the scales had fallen away from our eyes.” It was then that the team first came up with a plan for a synthetic molecular machine that could transport particles away from their natural distribution (see page 523).

Its design was inspired by a thought experiment proposed 140 years ago by physicist James Clerk Maxwell. Maxwell imagined a tiny demon able to open a trap-door between two containers filled with gas. This demon would open the door only to allow particularly fast molecules to move from the container on the right to that on the left. Eventually, this would increase the temperature of the container on the left (which would contain all of the faster-moving molecules) and decrease the temperature of the other. Such spontaneous heating and cooling in the absence of energy input conflicts with the second law of thermodynamics, which (among other things) forbids an increase in the order of a system without any energy being added from outside.

Leigh's team designed a molecule that, like Maxwell's demon, moves and sorts particles according to their relative positions. This molecule, known as a rotaxane, consists of a ring threaded onto a linear unit and held in place by two bulky chemical groups (stoppers). A 'gate' located in the thread, closer to the left stopper than the right, blocks the movement of the ring along the thread.

When the ring was to the left of the gate and light was shone on it, the ring transferred the light's energy to the gate, changing its shape. The ring could then pass to the other side before the gate closed. But when the ring was on the right side, it was positioned too far away from the gate to transfer light energy to open it. As a result, the particles accumulated on the right. Because light energy was put into the system, it did not violate the second law of thermodynamics.

It took two lab members, PhD candidate Viviana Serreli and postdoc Chin-Fa Lee, more than three years to construct a rotaxane able to carry out this task. During this time the team synthesized and tested many smaller components, each time refining the design, before putting together the complete molecule. “The first time we shone the light and saw the distribution of the rings change, being driven away from equilibrium, was a very humbling moment,” says Leigh.

Now he plans to make a molecular motor that functions like a biological pump in a membrane. “We want to make things that are far beyond the current state of the art,” he explains.