Turning Molecules into Motors

While molecular machines driven by chemical, light or thermal energies can be found throughout nature, little progress has been made toward creating synthetic counterparts. The gap between nature and nanotechnology remains due to the limited fundamental understanding of the transfer of energy to mechanical motion at the nanoscale.

Understanding and actuating the rotation of individual molecules on surfaces is a crucial step towards the development of nanoscale devices such as fluid pumps, sensors, delay lines, and microwave signaling applications. Recently a new, stable and robust system of molecular rotors consisting of thioether molecules (RSR) bound to metal surfaces has offered a method with which to study the rotation of individual molecules as a function of temperature, molecular chemistry, proximity of neighboring molecules and electrical current.

Our initial studies used the simple, symmetric thioether dibutyl sulfide. These molecules adsorb to metal surfaces via the central sulfur atom and rotation of the alkyl tails occurs around the central S-metal bond. These molecules appear hexagonal as they rotate due to the superposition of three equivalent orientations with respect to the hexagonally-packed surface below.

Rotation of these molecules at 80 K occurs faster than the time-scale of scanning tunneling microscopy (STM) imaging (~2 min/image), so it was not possible to decouple the superimposed orientations at this temperature. Upon cooling the system further (to 5 K) it was possible to observe the single dibutyl sulfide molecules in static positions, as their rotation was halted.

STM image of three molecular rotors, just 1 nanometer wide, spinning at over 1,000,000 times per second when heated to a temperature of 78 Kelvin (-320 F).

Low-temperature STM allowed the measurement of the rotation of the dibutyl sulfide molecules as a function of temperature and the quantification of both the energetic barrier and pre-exponential factor for their motion. In these measurements the feedback loop (which is normally used to modulate the STM tip height in order to maintain a constant tunneling current) is turned off and the tunneling current is monitored with respect to time (I vs. t). By measuring the rotational rate as the temperature is increased, it is possible to create Arrhenius plots to further understand the rotational energetics of the individual molecules.

To understand the interaction between the alkyl tails of rotor and the surface we studied the rotation of thioether molecules as a function of chain length. Dimethyl-, diethyl-, dibutyl- and dihexyl sulfides were studied, and it was found that all of the rotors except dimethyl sulfide were static at 7 K. Each of the molecular species was then heated until they were visibly rotating within STM images.

Interestingly, the thermal onset to rotation was found to be nearly identical for studied thioether molecules with alkyl tails of two carbons or more. It is proposed that this plateau in thermal onset was due to an interplay between degrees of freedom in the alkyl tail vs. S-metal bond length, which was supported by subsequent molecular dynamics calculations^1,2.

Unlike longer alkyl-chained thioethers, dimethyl sulfide molecules were seen to rotate too quickly to measure at 7 K. Experimental studies showed a very low barrier to rotation, as dimethyl sulfide molecules rotated >10³ Hz at 5 K. In order to further understand this molecule's rotation, DFT studies were used to calculate the rotational barrier. Also, using theoretical methods the minimum energy adsorption site was determined and the mechanism of rotation was elucidated. These theoretical results indicate that the rotation of a small, simple molecule is actually rather complex; as the CH₃ groups of dimethyl sulfide rotate around the Au-S bond, the central S atom precesses around a surface Au atom².

STM images showing how a spinning molecular rotor can be "braked" by physically moving it towards a chain of static molecules.

Through a series of single-molecule manipulation experiments, we have mechanically switched the rotation on and off reversibly by moving the molecules toward or away from one another. If two rotors are pushed close together, they stop rotating due to van der Waals attraction between the alkyl chains. One of the major goals for the field of molecular rotors is creating ordered arrays with which to study rotational energy propagation.

Our mechanical deactivation of dibutyl sulfide molecules demonstrates that there will be a complex interplay between sterics and electrostatics that will mediate the rotational coupling of neighboring molecules. Towards this end, we have created ordered 2D arrays of dibutyl sulfide rotors on a Ag/Cu(111) surface alloy. This alloy forms a very regular hexagonal array of hcp and fcc stacked atoms, and allows the thioether rotors to be precisely self-assembled, with a spacing of 2.6 nm^3,4.

While our previous studies revealed that small amounts of thermal energy are capable of inducing rotation, thermodynamics dictates that thermal energy alone cannot be used to perform useful work in the absence of a temperature gradient. Therefore, for molecules to meet their full potential as components in molecular machines, methods for coupling them to external sources of energy that selectively excite the desired motions must be devised.

To this end, we have studied using an electrical current to rotate individual dibutyl sulfide molecules on command. For these studies the source of energy is supplied via high energy electrons from the STM tip. It was found that at temperatures below 8 K the molecules were static and could be stably imaged for many hours at tunneling voltages less than ±0.35 eV. However, either imaging or positioning the STM tip over the molecules at biases above ±0.35 eV caused them to switch between their three distant orientations. With isotopic labeling experiments we were able to show that the mechanism for this electrically-induced rotation is a C-H stretch which decays to the rotation of the molecule^4,5.

While these studies are mostly fundamental in nature, they have made great strides forward for the field of molecular rotors. Fifty years ago Richard Feynman expressed his dreams to see the miniaturization of useful machines. Today, while we have the tools to study many of these systems, we still need the imagination which Feynman displayed in his famous address to realize all of the potential for the use of nanomachines across every field of science and engineering.

While this study and others like it have not put Feynman's "swallowable surgeons" on the shelves in any stores, we (the scientific research community) are just beginning to understand the necessary fundamental science toward achieving these goals. Our studies are continuing in the field of molecular rotors with chiral molecular rotors^5,6 and the influences of chiral scanning probe tips on their rotation in order to further diversify the utility of this class of single molecule rotors^6,7.

References

1. Baber A. E. . et al. ACS Nano 2, 2385-2391 (2008).
2. Tierney, H. L. et al. J. Phys. Chem. C 113, 10913-10920 (2009).
3. Tierney, H. L. et al. J. Phys. Chem. C 114, 3152-3155 (2010).
4. Bellisario, D. O.; Baber, A. E.; Tierney, H. L.; Sykes, E. C. H. J. Phys. Chem C 113, 5895-5898 (2009).
5. Tierney, H. L. et al. Chem. Eur. J. 15, 9678-9681 (2009).
6. Tierney, H. L.; Han, J. W.; Jewell, A. D.; Iski, E. V.; Baber, A. E.; Sholl, D. S.; Sykes, E. C. H. J. Phys. Chem. C (2010) In Press.
7. Tierney, H. L.; Jewell, A. D.; Baber, A. E.; Iski, E. V.; Sykes, E. C. H. Submitted to Phys. Rev. Lett. (2010).