Skip to Content

Running Your Fingers Over A Single Molecule

Readers will remember the extraordinary pictures of individual pentacene molecules last fall. Well, the same IBM team, working with a group at Aberdeen, has struck again.
This time they’ve imaged a much more complex organic molecule, cephalandole A. As that link details, the structure of this natural product has recently been revised – it’s one of those structural-isomer problems that NMR won’t easily solve for you. Here’s a single molecule of it, imaged by the same sort of carbon-monoxide-tipped atomic force microscope probe used in the earlier work>
AFM image 2
Now, it’s not like you can just look at that and draw the structure, although it is vaguely alarming to see the bonding framework begin to emerge. If you calculate the electon densities around the structure, though, it turns out that the recently revised one is an excellent fit to what the AFM tip picks up, while the other structural possibilities lead to different expected contours.
It’s quite possible that as this technique goes on that it could become a real structure-determination tool. These are early days, and it’s already being applied to a perfectly reasonable organic molecule. Of course, the people applying it are the world’s experts in the technique, using the best machine available (and probably spending a pretty considerable amount of time on the problem), but that’s how NMR was at the start, and mass spec too. Both of those are still evolving after decades, and I fully expect this technology to do the same.

13 comments on “Running Your Fingers Over A Single Molecule”

  1. El says:

    I spent many years hunched over one of these torturous instruments and I can guarantee they will have spent a ‘considerable’ amount of time on this (well some poor grad student will have anyway).
    Amazing stuff. I truly never thought this sort of thing would be possible with an AFM.

  2. Wavefunction says:

    Determining stereochemistry using STM/AFM by just “looking” at the molecule was in the offing, and I can see this paper pushing the goal a significant step further. Some papers were published in the 90s by a couple of guys like George Flynn on this. In addition, a couple of years ago there was a Nature paper which used AFM to distinguish cis and trans alkenes.
    Remember, Richard Feynman predicted something exactly like this in 1959 in “There’s plenty of room at the bottom”. He had the following to say:
    “If you have a strange substance and you want to know what it is, you go through a long and complicated process of chemical analysis. You can analyze almost anything today, so I am a little late with my idea. But if the physicists wanted to, they could also dig under the chemists in the problem of chemical analysis. It would be very easy to make an analysis of any complicated chemical substance; all one would have to do would be to look at it and see where the atoms are. The only trouble is that the electron microscope is one hundred times too poor. (Later, I would like to ask the question: Can the physicists do something about the third problem of chemistry—namely, synthesis? Is there a physical way to synthesize any chemical substance?”
    I am waiting to see when the last goal, namely physical synthesis of a molecule, would be achieved. That would be the day.

  3. Tok says:

    There is one thing to be said for traditional characterization techniques: They measure a boatload of molecules at once. Say you’ve isolated something that’s 99% pure, but you happen to pick one of those molecules that’s in the extra 1%.
    Similar problem with the physical synthesis of a molecule. I think it’ll happen someday, but it will probably be on the order of molecules, rather than the 10^20-26 molecules that we need for anything useful.
    Chemists don’t hang up your labcoats on this one quite yet. Well at least for this reason.

  4. ProteinChemist says:

    As the above posters have said, I wouldn’t hang my coat up quite yet. However, coming from someone who looks at x-ray crystallography of complex proteins on a regular basis, that is one of the prettiest little images I have ever seen. Just the bond effects alone are staggering. I cannot even imagine where this will be in 10 years.

  5. This work is stunning. As an NMR spectroscopist by training, and someone who has performed small molecule structure elucidations for many years (earlier in my career), this really sets a new bar. It indicates what is coming. Natural product structure elucidations are very problematic and I recently coauthored an article about many of the errors that find there way into the literature (DOI: 10.1039/c002332a) The elucidation of cephalandole A was mentioned in there and we correctly elucidated it using software. Also, I blogged on this Nature article and summarized our own efforts on this molecule here: http://tinyurl.com/2dc3pf4 . Software algorithms CAN help reduce the number of errors that can make their way into the literature but when this AFM approach goes mainstream (not if, but when…and no time line defined of course!) then who needs them!

  6. lynn says:

    Beautiful and … thrilling.

  7. Steve says:

    A few years ago, I did some tutoring of A-level chemistry in my spare time. I recall on more than one occasion telling my tutee that no-one had yet made a ‘microscope’ that would let us see the structures of individual molecules in much detail. There were the kind of fuzzy images that hinted at molecular structure, but nothing quite like this.

  8. Anonym says:

    Unrelated topic but has anyone noticed this paper
    “Organocatalysis in Cross-Coupling: DMEDA-Catalyzed Direct C-H Arylation of Unactivated Benzene”
    http://pubs.acs.org/doi/abs/10.1021/ja103050x
    Seems quite iffy, similar to the “NaH” oxidation debacle.

  9. Ed says:

    Don’t know about that – maybe a butoxide mediated benzyne formation (elimination of HI enhanced by the K+ chelator DMEDA) followed by a regular Friedel-Crafts-type electrophilic substitution of the solvent (benzene)? Realistic?

  10. anon says:

    The benzyne mechanism appears unlikely in light of the high regioselectivity for arylation of monosubstituted aryliodides.
    They argue for a radical mechanism due to the negative effects of TEMPO and other radical traps but whether its transition metal free radical mechanism or ppm Ni etc is maybe open for debate / sure to cause a massive blog fight.

  11. tj says:

    My understanding is that these images are created by dragging the afm tip over the molecule very slowly and recording the variation in the attractive force acting on the tip. If this is true, won’t the ultimate possible resolution be effectively limited by how “small” the incremental movements of the tip can be made?
    Is motion quantized on the smallest scale? Is it even possible to move an afm tip by arbitrarily smaller and smaller units “to the right”? Or is there a lower bound for the “shortest” possible movement? What about the uncertianty principle?

  12. eb says:

    11- My understanding is its dependent on how fine the tip is. (thus why they’ve had so much success with a single CO molecule at the end instad of say a single atom of Au). The recorded force is an average of all the possible interactions on the tip. Finer tips lead to smaller areas relating to the averages of interactions and thus more unique data points are captured within a Xsq angstrom ara. And the more data points you have the sharper the image (think pixels).

  13. mmc says:

    Really, an amazing image. I’m sure all of us who have spent our life trying to imagine the molecular scale get a little tingle in the spine looking at this one.

Comments are closed.