Friday, April 26, 2013

The Enantiomers’ Swimming Competition

Image Source.
The spatial arrangement of some large molecules can exist in two different versions which are mirror images of each other, yet their chemical composition is entirely identical. These mirror versions of molecules are said to have a different “chirality” and are called “enantiomers.” The image to the right shows the two chiralities of alanine, known as L-alanine and D-alanine.

Many chemical reactions depend not only on the atomic composition of molecules but also on their spatial arrangement, and thus enantiomers can have very different chemical behaviors. Since organisms are not chirally neutral, medical properties of drugs made from enantiomers depend on which chirality of the active ingredient is present. One enantiomer might have a beneficial effect, while the other one is harmful. This is the case for example for Ethambutol (one enantiomer treats tuberculosis, the other causes blindness), or Naproxen (one enantiomer treats arthritis pain, the other causes liver poisoning).

The chemical synthesis of molecules however typically produces molecules of both chiralities in approximately equal amounts, which creates the need to separate them. One way to do this is to use chemical reactions that are sensitive to the molecules’ chirality. Such a procedure has the disadvantage though that it is specific to one particular molecule and cannot be used for any other.

Now three physicists have shown, by experimental and numerical analysis, that there may be a universal way to separate enantiomers
It’s strikingly simple: chiral particles swim differently in a stream of water that has a swirl to it. How fast they travel with the stream depends on whether their chirality is the same or the opposite of the water swirl’s orientation. Wait far enough downstream, and the particles that arrive first will almost exclusively be the ones whose chirality matches that of the water swirl.

They have shown this as follows.

Molecules are typically of the size of some nanometers or so, and the swimming performance for molecules of different chirality is difficult to observe. Instead, the authors used micrometer-sized three-dimensional particles made of a type of polymer (called SU-8) by a process called photolithography. The particles created this way are the simplest example of configurations of different chirality. They labeled the right-handed particles with a blue fluorescent dye, and the left-handed particles with a green fluorescent dye. This allows taking images of them by a fluorescent microscope. Below you see a microscope image of the particles

Next you need a narrow channel through which water flows under some pressure. The swirl is created by gratings in the wall of the channel. The length of this channel is about a meter, but its height and width is only of the order 150 μm. Then you let bunches of the mixed chiral particles flow through the channel and photograph them on a handful of locations. From the amount of blue and green that you see in the image, you can tell how many of each type were present at a given time. Here’s what they see (click to enlarge)

This figure is an overlay of measurements at 5 different locations as a function of time (in seconds). The green shade is for molecules with the chirality that matches the water swirl orientation, the blue shade is for those with the opposite chirality. They start out, at x=32.5mm, in almost identical concentration. Then they begin to run apart. Look at the left tail of the x=942.5 mm measurement. The green distribution is almost 200 seconds ahead of the blue one.

If you aren’t impressed by this experiment, let me show you the numerical results. They modeled the particles as rigidly coupled spheres in a flow field with friction and torque, added some Gaussian white noise, and integrated the equations. Below is the result of the numerical computation for 1000 realizations (click to enlarge)

I am seriously amazed how well the numerical results agree with the experiment! I’d have expected hydrodynamics to be much messier.

The merit of the numerical analysis is that it provides us with understanding of why this separation is happening. Due to the interaction of the fluid with the channel walls, the flow is slower towards the walls than in the middle. The particles are trying to minimize their frictional losses with the fluid, and how to best achieve this depends on their chirality relative to the swirl of the fluid. The particles whose chirality is aligned with the swirl preferably move towards the middle where the flow is faster, while the particles of the opposite chirality move towards the channel walls where the flow is slower. This is what causes them to travel at different average velocities.

This leaves the question whether this study of particles of micrometer size can be scaled down to molecules of nanometer size. To address this question, the authors demonstrate with another numerical simulation that the efficiency of the separation (the amount of delay) depends on the product of the length of the channel and the velocity of the fluid, divided by the particle’s diffusion coefficient in the fluid. This allows one to estimate what is required for smaller particles. If this scaling holds, particles of about 120 nm size could be separated in a channel of about 3cm length and 3.2 μm diameter, at a pressure of about 108 Pa, which is possible with presently existing technology.

Soft matter is not anywhere near by my area of research, so it is hard for me to tell whether there are effects at scales of some hundred nanometers that might become relevant and spoil this simple scaling, or whether more complicated molecule configurations alter the behavior in the fluid. But if not, this seems to me a tremendously useful result with important applications.


  1. Thank you for so wonderfully elucidating the contents of the paper, Bee. News such as this usually has very little chance of popping up in my feed reader. Despite the attraction of discussing very abstract entities whose physical reality may be called into question, there's something equally attractive with "simple(er)" experiments like this -- with results!, from a rather "boring/un-sexy" branch of the sciences.

  2. Hi Navneeth,

    You have to thank Ralf (one of the authors) for this, because without him I'd never have heard of this paper either. Certainly not a journal that I read... Best,


  3. All molecules have mirror images. A chiral mass distribution is not superposable upon its mirror image only by translation and rotation. It has no S_n improper rotation axes. DOS software SYMMETRY by Serguei Patchkovskii swallows an *.xyz coordinate file then outputs a complete symmetry element analysis plus point group.

    "chiral particles swim differently in a stream of water that has a swirl to it" Opposite shoes embedded within a surrounding medium left foot have different minimum action trajectories. You describe the geometric Eötvös experiment. It detects whether the vacuum is trace chiral anisiotropic toward fermionic matter. Massless boson photons see isotropic vacuum.
    The "swimmers" are La Coupe du Roi (or rotationally locked hydrogen peroxide molecules).

  4. Interesting but to be practical it would have to be applicable to typical chemical compounds with sizes well below 1 nanometer.

    Chiral column chromatography is a somewhat similar technique.

  5. The homochirality of life is explained in similar way with specific absorption of sugars and aminoacids to curved hydrophobic surfaces of lipidic bilayers: the chiral less polar molecules (D-sugars with OH groups) tend to adsorb into outer surface (it's sorta superhydrophobicity effect), the polar L-aminoacids tend to collect inside of it. Aminoacids are building material of proteins, whereas the sugars are source of energy for living cell.

    Note that chiral pairs of men and women are separated by phase gradients of family settlement: the women tend to stay inside of it and they collect building material (wood), whereas the men are collecting energy related material (food) from outside.

    The homochirality is therefore scale invariant geometry in AWT and it manifest itself at many scales.

  6. BTW The spin-oriented photons of polarized light swim in different speed through solids as well: the well know birefringence effect of island calcite is based on their swimming competition...

  7. Surely the process must work well enough or authors wouldn't be discussing it in such positive terms. However, I see ambiguity in the concept of "Wait far enough downstream, and the particles that arrive first will almost exclusively be the ones whose chirality matches that of the water swirl." What "rating" or priority etc for various organic groups can properly assign them to "right-hand" or "left-hand" chirality? It's not like real spin, but somewhat arbitrary decisions have to be made about a convention to consider -COOH, -Cl, -CHO, -CF3, -OCH3 etc as proper markers to assign to the categories.

    Maybe a molecule with three big groups coming out of the stereocenter, and a puny -H on the side would be "obvious" but I wonder how Nature in the form of the swirling water decides what should be "really RH or LH" etc. In any case, still a clever idea and reported here as best could be considering that AFAICT unanswered question.

  8. While I take pleasure in hydrodynamic flows.....this is a question about what begins?

    See: Models to explain homochirality


  9. This process is an example of "In Time."

    Darwinian evolutionary biology is the prototype for thinking in time because at its heart is the realization that natural processes developing in time can lead to the creation of genuinely novel structures. Even novel laws can emerge when the structures to which they apply come to exist. Evolutionary dynamics has no need of abstract and vast spaces like all the possible viable animals, DNA sequences, sets of proteins, or biological laws. Exaptations are too unpredictable and too dependent on the whole suite of living creatures to be analyzed and coded into properties of DNA sequences. Better, as Stuart Kauffman proposes, to think of evolutionary dynamics as the exploration, in time, by the biosphere, of the adjacent possible. See: Thinking In Time Versus Thinking Outside Of Time (bold added in quote by me for emphasis)

    Which of course sends me back to the review you have Bee of Lee Smolin's new book.


  10. I find it interesting that the human body recognizes 1 of 256 chiral forms if cholesterol in its carbons.
    Some deeper principle may explain separation by specific chemical pathways as well as this fluid flow.
    But as if an abstract sheet or background could spin of particles be separated in chiralities as fine structures over wider mirror symmetry into at least four forms?
    The handedness would be six forms of tetrahedral edges of two open hands that fit together of the same structural chirality. In a sense can a lot of effects not be controlled by polarized light.
    What does this say for the nano scale as if a Casmir effect between sheets or doubled shells of fat?
    Let us also consider the role of fibers like feathers or silica and fat itself that structurally or mechanically are vaguely implicated as gateways to cancerous cell differentiation.
    A 4D simplex would view these projections as achairal loops.
    (I respond to a Nordita feed to the backreaction blogspot one year from the current date. )


COMMENTS ON THIS BLOG ARE PERMANENTLY CLOSED. You can join the discussion on Patreon.

Note: Only a member of this blog may post a comment.