Molecular machines have been in the spotlight recently with the award of the 2016 Nobel Prize in Chemistry for progress in producing the key structures and movements – rings, axles and rotation – that are required in a molecular-scale machine. The ability to design, synthesize and study mechanically functioning molecule-machines has been explored for almost 20 years at CEMES-CNRS in Toulouse, France, where Christian Joachim has created the PicoLab. However, rather than studying molecules in solution and following the ensemble averages of molecular motion – as is natural from a chemist’s point of view – the ultimate aims of the work at the PicoLab are to develop single-molecule machines that are powered to move across a surface individually.

From molecules in solution to race cars

Moving from a solution to a surface requires a significant rethink of how the molecules are designed. “On the whole you need fewer degrees of freedom and they need to be more rigid,” says Eric Masson, a researcher in the Department of Chemistry and Biochemistry at Ohio University, who is part of a team working on a design to compete in the world’s first nanocar race in 2017. “But there are advantages too – you’re not competing with water molecules so the affinity between the wheel and chassis is greater.”

Masson’s route into nanomotorsports was through Cucurbit[b]urils, which he describes as pumpkin-shaped hollow molecules that display outstanding recognition properties and often extreme affinities towards their guests in water. His colleague Saw Hla from the Physics Department, who was aware of plans at CEMES-CNRS to organize the world’s first nanocar race, saw the molecule and, thinking it looked like a wheel, entered it into the race.

“The decision to make a frame with four wheels rather than just race a wheel was aesthetic – it looks more like a car,” Masson tells However, there is a balance to be struck between large molecules that are cumbersome and stick to the track, and small molecules that bounce around with little control.

Gwénaël Rapenne and Claire Kammerer, representing the CEMES-CNRS nanocar race entry, are also working with a molecule-car that has wheels and axles for the race. In fact they had already developed the boron-subphthalocyanine double-wheel molecule in 2012 when they published a paper on it in Nanotechnology. When race organiser Joachim proposed the competition in 2013, they were naturally entered to compete.

The work on the CEMES-CNRS molecule actually began at the end of the 1990s with planar rotators, and as a joke Joachim had laid down the challenge to make the rotation vertical like a car wheel. This challenge was duly met with the design and synthesis of the first molecule-vehicle made of two front wheels, two rear legs and a central board, the famous molecular wheelbarrow published in Nanotechnology. The molecule has since been developed into their race entry molecule and into many other molecule-vehicles, including the nano-car of the 2016 Nobel Prize in Chemistry winner Bernard Feringa.

Tip structure and tip track interactions

In the initial 2007 report of the CEMES-CNRS team’s molecule, the researchers describe moving a two-wheel-axle molecule machinery with an STM on a gold surface. The manipulation of single atoms and molecules with an STM first reached the state of the art for its time in work by Don Eigler and Erhard K. Schweizer at IBM in their manipulation of xenon atoms resulting in some of the most iconic images in scanning probe microscopy.

The world’s first nanocar race will also use STM technology with unique new combined and independent four-STMs-in-one instruments designed by ScientaOMicron. Each team has their own Au(111) race track on a disk providing a track length of around 100 nm per team. Gold was chosen because its surface microstructure gives natural nanoscale tracks without the need for atomic-scale patterning. It also offers the convenience of adding single gold atoms to the end of the STM tip when necessary just by touching the tip to the surface – this functionalization of the tip makes it work better for the race.

Interactions between tip and surface are incredibly sensitive to tip features. In a recent Nanotechnology paper, Philip Moriarty and colleagues at the University of Nottingham and Quantulus Technology Ltd in the UK, and Institut National des Sciences Appliquéees de Toulouse, report detailed experiments on the extraction of hydrogen atoms from a surface with an STM tip and how this extraction is affected by the tip’s initial condition.

Curiously they found that tips that image with less detail – that is, rows as opposed to individual atoms – are more effective for desorbing single atoms, and avoiding the production of clusters. “The reason for this remains as yet unclear, but an obvious candidate explanation relates to the density of states of the tip apex,” suggest the researchers in their report. The results highlight how the influence on the contrast mechanism due to the atomistic structure and density of states of the tip – as opposed to the substrate – plays an essential role.

Race rules

Other rules of the competition include no touching with the tip – the nanocars must be propelled purely by the tunnelling current. Joachim adds, “Formula 1 drivers don’t push their cars.” This means that each team must determine what voltage pulses they need to avoid damage while scanning and moving their vehicles, as well as making any necessary repairs. The Rice-Graz team had a very mobile molecule and aimed for currents less than a 100 fA, making full use of the high specifications available with the race STM systems. These include a few picometer noise in the z direction and current control down to 800 fA, which the Picolab researchers are still pushing to reduce for even more refined current control.

Determining the voltage needs of each vehicle touches on a number of open questions around the energy use and locomotive mechanisms. “It is not understood how the delivered power is used by the nanocar itself,” says Joachim. “The actual provided 1 nW per molecule is huge for one molecule and it seems that only about 10–21 W are effectively used for example by the Dresden-registered molecule-vehicle to move across the surface.” In addition, most teams still don’t know if their wheels roll or slide, a question that has been under investigation for well over a decade.

The “Rotation of a single molecule within a supramolecular bearing” was reported by Joachim and James K Gimzewski alongside colleagues at IBM in Zurich, Switzerland, CEMES-CNRS and Risø National Laboratory, in a paper by that name in the late 1990s. However, determining the mechanism of motion for different molecules is no mean feat. Joachim and colleagues had hoped to settle the question of whether their molecular wheel-barrow vehicle rolled or slid by designing a chassis that would move up and down if the wheels rolled. Unfortunately, with that design the molecule stuck to the substrate and didn’t move at all. Later work reported in 2007 by Joachim and Rapenne and colleagues at CEMES-CNRS and the Freie Universität Berlin was able to show the rolling of a single molecule equipped with two wheels for the first time.

Different nanocar designs

A number of other nuances in nanocar performance have also been detected during test-run training sessions at Toulouse. Rémy Pawlak’s Basel team found that their car moved more easily – about 20 nm per hour – when it picked up an atom, raising questions with the organizers as to whether such “doping” should be allowed. The car design of Francesca Moresco’s Dresden team resembles a windmill and moves at about 5 nm per hour, although like many of the competitors this can sometimes get stuck turning corners, taking several hours, as experienced by this team during a second training session in Toulouse. The race itself is expected to take around 38 hours.

Perhaps the simplest design is that by We-Hyo Soe and and the Japanese team directed by the chemist Waka Nakanishi. Soe is a MANA-sponsored researcher who has been working in STM and single-molecule mechanics for 10 years, winning a place in the Guinness Book of Records for the world's smallest working gear in 2011. He had been studying how to manipulate atoms and molecules and had initially approached this by pushing, although no direct contact is allowed for the nanocar race. The final design of the Japanese team’s entry looks less like a car and more “like an inch worm with paddles”, as Soe describes it.

The significant range in size in the competing vehicles introduces some interesting distinctions in their behaviour, as Jean-Pierre Launay – CEMES-CNRS researcher and race referee – points out. Larger cars like the Ohio team’s “nanobobcat” behave essentially classically, whereas small ones like Soe’s MANA paddle molecule-vehicle behave essentially like a quantum system. Intermediary sizes have some of both types of behaviour.

2017 – the nanocar race and beyond

Despite the extraordinary expertise among the race hosts and teams, the nanocar race has already seen a number of delays due to the shear difficulty of motorsports at this scale. Designing and testing the original molecules has taken longer than expected. To make matters worse, the four master computers for the race in Toulouse were hacked shortly before the initially programmed 2016 race date, causing it to be postponed.

It is now hoped the world will see its first nanocar race in spring 2017. However, Joachim suggests that a lot of what he hoped to accomplish with the race is already being achieved. Not only have tackling the challenges of the race progressed the understanding of these systems by a long way already, but the prospective race has brought together groups from around the world at the very forefront of their field in an invaluable exchange of ideas. While the jury is out as to which model of nanocar will be first past the finishing line, it seems the real winner in this venture is the world of science itself.