News & Views | Published:


No turning back for motorized molecules

Nature volume 534, pages 187188 (09 June 2016) | Download Citation

Two molecular motors have been developed that use chemical energy to drive rotational motion in a single direction. The findings bring the prospect of devices powered by such motors a tantalizing step closer. See Letter p.235

The conversion of chemical energy to mechanical motion drives movement in all living things, from bacteria to whales. An intricate array of molecular ratchets and motors allows cells to extract mechanical work from chemical reactions, for example to drive muscle contraction, or to twist the helical appendages that propel some bacteria. Two papers, one by Collins et al.1 in Nature Chemistry and another by Wilson et al.2 on page 235, report the design and construction of artificial molecular motors that achieve the same outcome using much simpler, purely synthetic structures. Both pieces of work show that chemical reagents can drive the unidirectional motion of one part of a molecule (the rotor) relative to another (the stator), and thus provide direct functional analogues of biological motors.

It is not easy to design a synthetic molecular motor3. As was pointed out nearly 20 years ago4,5, molecular motors are characterized by movement that must be more than random Brownian motion. Furthermore, angular momentum cannot be used to maintain a constant directionality on the molecular scale as it can in everyday electric motors. The thermodynamic landscape of a molecular system must be repeatedly altered to force concerted movement in a single direction, to prevent mere shuttling forwards and backwards between two states. The greatest successes in the field so far have used light energy to drive a molecular system away from equilibrium, followed by a directionally defined relaxation process; motors capable of megahertz rotational speeds have been designed and built using this approach6.

The two new motors both use chemical energy to drive rotation. Collins and colleagues' motor is remarkably simple in conception. The rotor and stator are each a benzene ring, connected by a single bond that forms a rotatable axle. Systems of this sort can rotate freely about the axle, but rotation in the authors' motor is partly restricted by groups or atoms attached next to the bond that connects the two benzene rings.

Collins et al. added alternating sets of reagents to a solution containing their motor, which allowed first one side of the rotor ring and then the other to slip past a sulfur-containing group (a sulfoxide; Fig. 1) bonded to the stator. The alternating reagents insert a palladium atom into a carbon–hydrogen (C–H) bond on one side of the rotor, and then into a carbon–bromine (C–Br) bond on the other. Palladium's affinity for the sulfur atom of the sulfoxide (SOR) group lets it form a bridge between the rings that lowers the energy barrier to rotation, allowing the rings to slip past one another. On its own, shuttling the metal between the C–H and C–Br bonds would simply cause random rotation clockwise or anticlockwise, but the chirality (handedness) of the sulfoxide group imparts directionality to the slippage mechanism, and so also to the rotation of the motor.

Figure 1: A unidirectional molecular motor incorporating a rotating axle.
Figure 1

Collins et al.1 report a system consisting of two benzene rings (green hexagons) connected by a single bond. One ring acts as a rotor, and has a hydrogen atom on one side and a bromine atom on the other. The other ring is a stator and has a sulfoxide group on one side and a fluorine atom on the other (fluorine atom not shown because it is not involved in the motor mechanism). The connecting bond acts as an axle. The rings are also viewed here from above, along the axis of the axle (top right in each panel). a, The system's rotation cycle begins with the rotor and stator perpendicular to each other. b, Addition of a palladium (II) reagent allows the side of the rotor carrying the hydrogen atom to pass the sulfoxide. A palladium atom bridges the two rings. c, The rings then relax to the alternative perpendicular arrangement. d, Conversion of palladium (II) to palladium(0) allows the side of the rotor carrying the bromine atom to pass the sulfoxide group, and a palladium atom again bridges the rings. The cycle continues if reagents are added to toggle the palladium between the two oxidation states. Br, bromine; SOR, sulfoxide (where R is a benzene-ring-containing group); Pd, palladium.

The alternating C–H and C–Br insertions needed to drive this process require palladium to be in the +2 and 0 oxidation states, respectively. This means that, in its current form, the motor cannot work autonomously, because different reaction conditions are needed to shuttle the palladium between these states. However, metal redox processes can be driven electrochemically, raising the intriguing possibility that future versions of the motor could be electrically powered.

Wilson and colleagues' chemically powered motor overcomes the autonomy problem by using a different, more complex design than that of Collins and co-workers. In Wilson and colleagues' motor, a small ring is threaded onto a larger one (the track), and travels like a train around the track by constantly advancing in the same direction from one of two 'stations' to the other. Crucially, and in contrast to Collins' and co-workers' motor, only one set of reaction conditions is needed to drive the motor forward: the authors use a reactive 'fuel' known as Fmoc-Cl, which continuously breaks down to carbon dioxide and other by-products as the motor runs.

The small ring's progress is powered by a mechanism that channels random kinetic motion into movement in a single direction (Fig. 2). Immediately after each station, an unstable carbonate group (which becomes attached to the track by reaction with Fmoc-Cl) provides a 'stop' signal. If the carbonate group is removed, the signal switches to 'go'. The authors designed the chemistry of the system such that the signal switches from 'stop' to 'go' at a more or less constant rate at both stations, but changes from 'go' to 'stop' more rapidly after the small ring has passed through to the other station. The stop signal therefore tends to follow the train around the track, ensuring that forward motion is always faster than reverse.

Figure 2: A unidirectional molecular motor incorporating a ring travelling round a track.
Figure 2

a, Wilson et al.2 report a system in which a small molecular ring is threaded onto a larger one (the track). The track has two 'stations' at which the ring can dock, and two molecular groups that act as signals and can be set to 'stop' or 'go'. When the ring is docked at a station, its proximity to the nearby signal forces that signal to stay in the 'go' position, allowing the ring to travel to the second station. b, When the ring docks at the second station, the first signal changes to 'stop', preventing reversal of the direction of travel, while the second signal switches to 'go', allowing the ring to carry on around the track to the first station. The signals are switched from 'go' to 'stop' using the reagent Fmoc-Cl.

The choice of Fmoc-Cl as the fuel is ingenious, because the chemical mechanisms that involve Fmoc-Cl in switching from 'stop' to 'go' and vice versa are different, which means that the rates of the switching steps can be independently controlled. The energy that powers the constant forward movement of the train is provided by the consumption of Fmoc-Cl, so the train keeps moving until all of the Fmoc-Cl has been consumed.

Wilson and colleagues' work constitutes an important step in the construction of a chemically propelled, autonomous molecular device, but there is still a long way to go. The small ring typically takes 12 hours to travel around the track, and the Fmoc-Cl fuel is used rather inefficiently — the chemical conditions required for the fuel to power the motor also cause the fuel to decompose wastefully. Both motors currently work in solution, with at least 1018 molecules working in tandem. But the translation of chemical energy into macroscopic motion is likely to require motor components to be constructed in the solid phase, and to be individually controllable.

The story of artificial molecular motors is still in its opening pages, but chemists' attempts to mimic cellular motors reveal how many challenges biological systems have overcome to evolve the machinery that powers movement. The design principles that work are becoming clearer, however, and although the possibility of molecular motors routinely powering artificial devices in the future is still distant, it is now distinct.



  1. 1.

    , , & Nature Chem. (2016).

  2. 2.

    et al. Nature 534, 235–240 (2016).

  3. 3.

    & Nature Nanotechnol. 1, 25–35 (2006).

  4. 4.

    , & J. Org. Chem. 63, 3655–3665 (1998).

  5. 5.

    Angew. Chem. Int. Edn 37, 909–910 (1998).

  6. 6.

    et al. J. Am. Chem. Soc. 130, 10484–10485 (2008).

Download references

Author information


  1. Jonathan Clayden is at the School of Chemistry, University of Bristol, Bristol BS8 1TS, UK.

    • Jonathan Clayden


  1. Search for Jonathan Clayden in:

Corresponding author

Correspondence to Jonathan Clayden.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing