Nanoscale swimmers

If you were able to make a nanoscale submarine to fulfill the classic “Fantastic Voyage” scenario of swimming through the bloodstream, how would you power and steer it? As readers of my book “Soft Machines” will know, our intuitions are very unreliable guides to the environment in the wet nanoscale world, and the design principles that would be appropriate on the human scale simply won’t work on the nanoscale. Swimming is good example; on small scales water behaves, not as the free flowing liquid we are used to on the human scale; viscosity is much more important on small scales. To get a feel for what it would be like to try and swim on the nanoscale, one has to imagine trying to swim in the most viscous molasses. In my group we’ve been doing some experiments to demonstrate the realisation of one scheme to make a nanoscale object swim, the results of which are summarised in this preprint (PDF), “Self-motile colloidal particles: from directed propulsion to random walk”.

The brilliantly simple idea underlying these experiments was thought up by my colleague and co-author, Ramin Golestanian, together with his fellow theoretical physicists Tannie Liverpool and Armand Adjari, and was analysed theoretically in a recent paper in Physical Review Letters, “Propulsion of a Molecular Machine by Asymmetric Distribution of Reaction Products” (abstract here, subscription required for full paper). If one has a particle that has a patch of catalyst on one side, and that catalyst drives a reaction that produces more product molecules than it consumes in fuel molecules, then the particle will end up in a solution that is more concentrated on one side than the other. This leads to an osmotic pressure gradient, which in turn results in a force that pushes the particle along.

Jon Howse, a postdoc working in my group, has made an experimental system that realises this theoretical scheme. He coated micron-sized polystyrene particles, on one side only, with platinum. This catalyses the reaction by which hydrogen peroxide is broken down into water and oxygen. For every two hydrogen peroxide molecules that take part in the reaction, two water molecules and one oxygen molecule results. Using optical microscopy, he tracked the motion of particles in four different situations. In three of these situations – with control particles, uncoated with platinum, in both water and hydrogen peroxide solution, and with coated particles in hydrogen peroxide solution, he found identical results – the expected Brownian motion of a micron-sized particle. But when the coated particles were put in hydrogen peroxide, the particles clearly moved further and faster.

Detailed analysis of the particle motion showed that, in addition to the Brownian motion that all micro-size particles must be subject to, the propelled particles moved with a velocity that depended on the concentration of the hydrogen peroxide fuel – the more fuel that was present, the faster they went. But Brownian motion is still present, and it has an important effect even on the fastest propelled particles. Brownian motion makes particles rotate randomly as well as jiggle around, so the propelled particles don’t go in straight lines. In fact, at longer times the effect of the random rotation is to make the particles revert to a random walk, albeit one in which the step length is essentially the propulsion velocity multiplied by the characteristic time for rotational diffusion. This kind of motion has an interesting analogy to the kind of motion bacteria do when they are swimming. Bacteria, if they are trying to swim towards food, don’t simply swing the rudder round and propel themselves directly towards it. Like our particles, they are actually doing a kind of random walk in which stretches of straight-line motion are interrupted by episodes in which they change direction – this kind of motion has been called run and tumble motion. Counterintuitively, it seems that this is a better strategy for getting around in the nanoscale world, in which the random jostling of Brownian motion is unavoidable. What the bacteria do is change the length of time for which they are moving in a straight line according to whether they are getting closer to or further away from their food source. If we could do the same trick in our synthetic system, of changing the length of the run time, then that would suggest a strategy for steering our nanoscale submarines, as well as propelling them.

12 thoughts on “Nanoscale swimmers”

  1. So why try and make them self propelling in the Human body anyway?
    The energy necessary and difficulties presented in swimming through the body suggest to me that in any situation where you need nanobots inside your body, they should actually be small enough to be swept along like cells, but intelligent enough to recognise when they are in teh correct place to throw out an anchor.

    Hang on, isn’t that how our cells work anyway….

  2. Congratulations on your interesting paper, Richard. You might also like to cite Vicario, J., Eelkema, R., Browne, W. R., Meetsma, A., La Crois, R. M., Feringa, B. L. Catalytic molecular motors: fuelling autonomous movement by a surface bound synthetic manganese catalase. Chem. Commun. 2005, 3936-3938 which appears, at least to me, to operate by a similar principle to yours.

    With previous catalyst-propelled systems, the motion of objects less than micron scale was indistinguishable from Brownian motion [“Synthetic Molecular Motors and Mechanical Machines”, E R Kay, D A Leigh and F Zerbetto, Angew Chem Int Ed, 46, 72-191 (2007)]. It will be fascinating to see if you can significantly break this scale barrier!

  3. Dave, thanks for the comment and thanks for the pointer to the Vicario paper.

    We do have data for the size dependence of the effect, which we have yet to write up. The story is interesting in that Ramin’s initial prediction was that the propulsion velocity should not depend on size at all, but this is not what the experiments show. I think the smallest particle Jon looked at was half a micron or so, and we certainly saw propulsion here. The problem, though, is that the particle tracking gets difficult for optical microscopy on this scale and smaller.

    Guthrie, I’m not at all sure at the moment under what circumstances propulsion is a better option than letting things drift or diffuse. The interesting thing that bacteria teach us is that they don’t necessarily propel themselves to go faster, but in order to have a mechanism to swim towards or away from concentration gradients (chemotaxis).

  4. Hydrogen per oxide and platinum, wow, nano-rockets!

    For further work, might I suggest a new topology? Expose the platinum on one half of the the inside of a ring / cylinder / donut. That should allow for continuous, directional flow of reactants and products.

  5. Martin, thanks to your crew for tracking that down.

    Jim, that’s an interesting thought on topology, but I don’t think Jon’s going to welcome the thought of trying to work out how to make them. I also should say that it isn’t at all obvious what the best topology for maximising the propulsion is here – although it’s very tempting to think in terms of analogies with jet engines, the physics is actually quite a bit different.

  6. hello…why dont you publish a feed to your blog…it would be much easier to read your blog then

  7. Dear Richard:

    Your work is indeed quite interesting and significant. I should point out, however, that autonomous movement of nano- and micron-sized objects powered by catalytic decomposition of hydrogen peroxide was first described by us back in 2004 (“Catalytic Nanomotors: Autonomous Movement of Striped Nanorods,” Walter F. Paxton, Kevin C. Kistler, Christine C. Olmeda, Ayusman Sen, Sarah K. St. Angelo, Yanyan Cao, Thomas E. Mallouk, Paul E. Lammert, and Vincent H. Crespi, J. Am. Chem. Soc., 2004, 126, 13424) and highlighted in Chemical & Engineering News. Prof. Golestanian has also referenced this work in an earlier publication – I have not seen the Phys. Rev. Lett. paper.

    Since our first publication on the subject, we have continued to publish on this theme. The Galilean reverse of this phenomenon is of course pumping of fluids by nano- and micron-sized catalytic objects anchored on a surface, and we have also reported on this phenomenon. A list of our relevant publications are given below.



    1. “Catalytic Nanomotors: Autonomous Movement of Striped Nanorods,” Walter F. Paxton, Kevin C. Kistler, Christine C. Olmeda, Ayusman Sen, Sarah K. St. Angelo, Yanyan Cao, Thomas E. Mallouk, Paul E. Lammert, and Vincent H. Crespi, J. Am. Chem. Soc., 2004, 126, 13424.

    2. “Catalytic Nanomotors: Remote-Controlled Autonomous Movement of Striped Metallic Nanorods,” Timothy R. Kline, Walter F. Paxton, Thomas E. Mallouk, and Ayusman. Sen, Angew. Chem., Int. Ed., 2005, 44, 744.

    3. “Directed Rotational Motion of Microscale Objects Using Interfacial Tension Gradients Continually Generated via Catalytic Reactions,” Jeffrey Catchmark, Shyamala Subramanian, and Ayusman Sen, Small, 2005, 1, 202.

    4. “Motility of Catalytic Nanoparticles Through Self-Generated Forces,” Walter F. Paxton, Ayusman Sen, Thomas E. Mallouk, Chem. Eur. J., 2005, 11, 6462.

    5. “Catalytic Micropumps: Microscopic Convective Fluid Flow and Pattern Formation,”
    Timothy R. Kline, Walter F. Paxton, Yang Wang, Darrell Velegol, Thomas E. Mallouk, and Ayusman Sen, J. Am. Chem. Soc., 2005, 127, 17150.

    6. “Autonomously Moving Nanorods at a Viscous Interface,” Prajnaparamita Dhar, Thomas M. Fischer, Yang Wang, Thomas E. Mallouk, Walter F. Paxton, and Ayusman Sen, Nano Lett., 2006, 6, 66.

    7. “Chemical Locomotion,”Walter F. Paxton, Shakuntala Sundararajan, Thomas E. Mallouk, Ayusman Sen, Angew. Chem., Int. Ed., 2006, 45, 5420.

    8. “Catalytically Induced Electrokinetics for Motors and Micropumps,”Walter F. Paxton, Paul T. Baker, Timothy R. Kline, Yang Wang, Thomas E. Mallouk, Ayusman Sen, J. Am. Chem. Soc., 2006, 128, 14881.

    9. “Bipolar Electrochemical Mechanism for the Propulsion of Catalytic Nanomotors in Hydrogen Peroxide Solutions,” Yang Wang, Rose M. Hernandez, David J. Bartlett, Julia M. Bingham, Timothy R. Kline, Ayusman Sen, Thomas E. Mallouk, Langmuir, 2006, 22, 10451.

    10. “Catalytically-driven Colloidal Patterning and Transport,” Timothy R. Kline, Jodi Iwata, Paul E. Lammert, Thomas E. Mallouk, Ayusman Sen, and Darrell Velegol, J. Phys. Chem. B. 2006, 110, 24513.

    11. “Autonomously Moving Local Nano Probes in Heterogeneous Magnetic Fields,” Prajnaparamita Dhar, Yanyan Cao, Timothy Kline, Priya Pal, Cheryl Swayne, Thomas M. Fischer, Brian Miller, Thomas E. Mallouk, Ayusman Sen, and Tom H. Johanson, J. Phys. Chem. 2007, 111, 3607.

  8. Ayusman, thanks very much for your comment and for drawing attention to your important and pioneering work in this area. Of course, we are very much aware of your work; our paper was out a couple of weeks ago in PRL (, and this both references your work and discusses (so far as the very limited space permitted) ther relationship of our work to yours. Essentially, we suspect at the moment that the two sets of experiments use somewhat different mechanisms, with yours relying on electro-osmotic effects and ours exploiting self-diffusiophoresis. It’s a fascinating area and I’m looking forward to following how your work develops.

  9. When reading your book I became very tempted to do a reconstruction of your version of the ‘Fantastic Voyage’ – either as a stop-motion animation or by building a miniature ‘obstacle course’ for people to experience the unfamiliar conditions at that scale so that they can maybe ask different sorts of questions about new technologies. Am still tempted…

  10. Angela, I think this would be a great idea. The Nanomission game does this to some extent in a gaming environment, getting the general ideas across but without (at the moment) accurate physics. I think it would be quite fun to film a model submarine – getting the Reynolds number scaling right would be easy with a high viscosity fluid, the Brownian motion would be a bit more difficult.

Comments are closed.