Amazing photo of chemical bonds

June 1, 2013 • 1:21 pm

I don’t have a lot of time to post today, but alert reader Ginger brought this cool item to my attention. It’s a press release of a finding to appear in the “early edition” of Science, and shows the formation of benzene-ringlike structures from a complex chemical reaction.  The details are given here. An excerpt:

When Felix Fischer of the U.S. Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) set out to develop nanostructures made of graphene using a new, controlled approach to chemical reactions, the first result was a surprise: spectacular images of individual carbon atoms and the bonds between them.

“We weren’t thinking about making beautiful images; the reactions themselves were the goal,” says Fischer, a staff scientist in Berkeley Lab’s Materials Sciences Division (MSD) and a professor of chemistry at the University of California, Berkeley. “But to really see what was happening at the single-atom level we had to use a uniquely sensitive atomic force microscope in Michael Crommie’s laboratory.” Crommie is an MSD scientist and a professor of physics at UC Berkeley.

What the microscope showed the researchers, says Fischer, “was amazing.” The specific outcomes of the reaction were themselves unexpected, but the visual evidence was even more so. “Nobody has ever taken direct, single-bond-resolved images of individual molecules, right before and immediately after a complex organic reaction,” Fischer says.

Here you go; these are unbelievable photos:

The original reactant molecule, resting on a flat silver surface, is imaged both before and after the reaction, which occurs when the temperature exceeds 90 degrees Celsius. The two most common final products of the reaction are shown. The three-angstrom scale bars (an angstrom is a ten-billionth of a meter) indicate that both reactant and products are about a billionth of a meter across. (Caption from press release.)

The technique used to get these images, called noncontact atomic force microscopy, which in effect uses a single atom as a camera lens, is equally amazing. We should all be proud that our species can do something like this:

The collaborators then turned to a technique called noncontact atomic force microscopy (nc-AFM), which probes the surface with a sharp tip. The tip is mechanically deflected by electronic forces very close to the sample, moving like a phonograph needle in a groove.

“A carbon monoxide molecule adsorbed onto the tip of the AFM ‘needle’ leaves a single oxygen atom as the probe,” Fischer explains. “Moving this ‘atomic finger’ back and forth over the silver surface is like reading Braille, as if we were feeling the small atomic-scale bumps made by the atoms.” Fischer notes that high-resolution AFM imaging was first performed by Gerhard Meyer’s group at IBM Zurich, “but here we are using it to understand the results of a fundamental chemical reaction.”

The single-atom moving finger of the nc-AFM could feel not only the individual atoms but the forces representing the bonds formed by the electrons shared between them. The resulting images bore a startling resemblance to diagrams from a textbook or on the blackboard, used to teach chemistry, except here no imagination is required.


The atomic “theory” is verified once again.  When I was younger I never thought we’d be able to see individual atoms, but now that’s almost routine. It’s stunning that we can even see the bonds between them, looking like the Tinkertoys I played with as a child.


Reference: “Direct Imaging of Covalent Bond Structure in Single-Molecule Chemical Reactions,” by Dimas G. de Oteyza, Patrick Gorman, Yen-Chia Chen, Sebastian Wickenburg, Alexander Riss, Duncan J. Mowbray, Grisha Etkin, Zahra Pedramrazi, Hsin-Zon Tsai, Angel Rubio, Michael F. Crommie, and Felix R. Fischer, will appear in Science and is now available on Science Express,

58 thoughts on “Amazing photo of chemical bonds

  1. For an old chemist like me this is almost unbelievable. But I’ve seen so many wonders in my lifetime that it’s logical to believe it. I remember the phone of my grandparents in the 40’s where you really had to ‘ring’ somebody up, and now…

  2. I saw these images earlier, and I was immediately attracted to them.

    But seriously, this wets your curiosity in an “oh wow wow” fashion, analogous to one’s appetite in “om nom nom” mode.

    First of all, I didn’t expect electron clouds of molecular bonds to be so visible, seeing how electron clouds of atom’s are so delocalized.

    But maybe I should have expected it? Only the spherical ground state electron cloud permeates the nucleus, the rest tend to form densities away from it.

    Second, I wonder if what looks like dangling bonds in the free corners of the pentagons in the products is an artifact or real?

    Third, the asymmetric pentagons seems compressed. Hopefully those are the actual angles as measured, showing the difference between reality and the chemistry idealized shorthand schematics.

      1. Looks like perspective distortion to me. I’ll bet you a cup of coffee that the molecules aren’t flat, but rather are curled with the outer edges closer to the “camera” than the centers.


        1. That’s pretty amazing, too. There clearly seems to be a convex 3D shape to the molecules. The question is though, to what degree is this an illusion? Our eyes decode the shading in a way tuned to macroscopic objects. Can we really assume the subjects of these images follow the same rules? These images must be the result of some digital image processing algorithms, translating the spatial and electrostatic sensing of the needle tip into something our eye expects to see. So I wonder if we can reliably assume the apparent shape distortion and shading really indicates three dimensional perspective. I suppose it really could, and there doesn’t seem a good reason to assume the molecules would be perfectly flat.

          1. Annoyingly I can’t get access to the paper because my university doesn’t subscribe to science xpress. But I can clear up some of the things you are asking. ncAFM images are not images in the conventional “lens” way, normally there is a feedback loop to maintain a constant force (determined by a change in resonant frequency or amplitude) between the tip and the sample, the colour represents the height of the tip. These images, however, are probably constant height images so the colour directly represents the strength of the force. So why is the force higher at the edge, this could easily be that the molecules are buckled, thus are closer to the tip, and therefore there is more force. However, this phenomenon has been discussed in detail by Leo Gross (et al.), as a measure of bond order (number of bonds). It think this is most likely the cause of the bright bonds, as I would expect the lowest energy configuration would have the molecules flat on the surface.

          2. The image is entirely the result of what a computer program does with the “mechanical” response of the tip to the intraatomic Pauli repulsion forces between the single lowest atom of the tip and the atoms of the molecule. So while I wouldn’t call it an “illusion” the appearance of convexity can not be reliably interpreted as reflecting the molecule’s curvature. The entire image is the result of a bunch of “theories” and therefore no more real than Jesus.

            Not to be a spoil sport, but AFM (and STM – scanning tunneling microscopic) images like these are kinda ‘old-hat’ now. The cool thing for the investigators is that they were able to do this just after the molecules had formed in a reaction.

            1. Perhaps I’m missing your point, but how does computer processing negate the validity of these images? All digital images, including those produced by medical imaging devices, the Hubble telescope, or your phone’s camera, are the result of computer processing of some physical input according to some theory.

              Sure, processing artifacts are possible, and maybe the apparent curvature here is such an artifact, but that doesn’t make the images “no more real than Jesus”.

              Personally, if I’m sick and I’m given a choice between Jesus and a CT scan, I’ll take the CT scan.

              1. It doesn’t negate the validity of the images in the least – I certainly never meant to leave the impression that I thought it did. The Jesus bit was a joke. If creationists can deny the fact of evolution with “it is just a theory” then one could just as easily deny any evidence one can present for atomic theory as well.

                I’m a solid state chemist who reads a lot of materials science. STM and AFM are truly great techniques!

            2. I would disagree on the ‘old-hat’ for ncAFM and STM images ‘like these’. The first images like this were 2009, so that is only 4 years ago. The bond order explanation was not published until September 2012 less than a year ago (though it was presented at the ncAFM conference that July). Such resolution is exceptionally rare in ncAFM even in the best groups in the world and I have never seen STM images of the same resolution.

              Secondly, the Pauli repulsion is almost never what is being measured for non-contact AFM, most images are in the attractive not the repulsive regime. Also such clear images come from the forming of chemical bonds, not Van der Walls/Pauli repulsion.

              Finally “The entire image is the result of a bunch of “theories””? So is a camera image, based on theories of how light refracts in the camera. AFM images are built from the theory that atoms in close proximity exhibit forces, that forces on an oscillating cantilever affect resonant frequency, hardly controversial. It is as “real” as what your eyes see. It must be interpreted carefully, because it does not “see” everything and there are imaging anomalies. But then that is true of your eyes, they can’t see clean glass but it is definitely there when you run into the patio doors!

        2. If my layman’s thinking and perception is in the ballpark, the electrons composing the cloud around the single oxygen atom-tip, having a negative charge, would push against the target molecule’s collective electron cloud negative charge,

          [“The tip is mechanically deflected by electronic forces very close to the sample”]

          causing some mechanical movement and a bit of distortion of the target molecule’s shape. Also, in such a complicated molecule, perhaps the electron cloud distribution/density is not uniform, and that affects just what image we see.

          (I’m no chemist, but I remember such talk from general and organic chemistry days, and in my older age I much more appreciate the subject for its own sake, as contrasted with it at that time being a necessary gauntlet to successfully run as part of what I thought at the time was a serious interest in getting admitted to medical school.)

    1. I wondered about the asymmetric shapes of the 5 & 6 carbon rings as well. My working hypothesis, until I see it more comprehensively explained by someone who knows, is that the molecules aren’t flat due to ring strain.

    2. As for your dangling bonds on the 5 carbon atom rings, I suspect that that is not an artefact but rather the C-H bond that should be present in that position. A carbon atom can form 4 bonds and I think in the locations you refer to there are a single bond and a double bond, leaving 1 bond unaccounted for, which by convention attaches to hydrogen atom which is not depicted.

      You should however perhaps take my musings with a pinch of salt as I dropped out of my chemistry degree course 25 years ago.

    3. The term to search under is ring pucker. I suspect that the 4- and 5- membered rings suppress complete resonance of the structure, but that’s really only a slightly intelligent guess.

    4. Bear in mind that the molecules aren’t floating in solution, they’re on a silver surface. So they have to conform to some extent to the structure of the crystal beneath them.

    5. Regarding the visibility of the bonds, I think these images would be a great teaching tool for chemistry/quantum physics professors. What better way to convince your students that the electron cloud is a tangible thing and not just an abstract concept than to show them a picture?

  3. Oh wow, they really are little hexagons! Amazing. And Im made of this stuff. The thought gives me a tingling sensation everywhere

    1. I’m with you. I’ve spent my life assuming that organic molecules don’t really look like that.

      1. Ditto. I tried hard to play devils advocate and debunk this hoax when my son sent it to me a couple of days ago. But, no, it seems the genuine article.

      2. The clincher for me, prior to “molecular photography” was that chemists can make catenanes – think one ring looped through another like links of a chain.

  4. This truly is revealed knowledge, in the sense of guessing (or deducing) an answer on a flash card, and then having it revealed that the answer was correct.

    How fantastic that our conception of the shapes of these molecules, and all of their properties, were deduced without being able to “see” them in this sense. To have such a visual confirmation is truly satisfying.

    Who has never heard a religious person say “Have you ever seen an atom? Your belief in atoms is no different than religious faith”. Ha! Eat a wafer and shut up zombie cannibal.

  5. I was kinda hoping—dreaming, really–they’d look more like snakes curling up and swallowing their tails.

  6. How long is it going to be before someone looks at a photograph of a molecule and declares that it looks like Jesus? Or the Verging Merry?

    1. I am hoping for a molecule that looks like a grilled cheese sandwich with Jesus’s face.

      1. Well, I have before me a bunch of molecules that look awfully like a grilled cheese sandwich when looked at under natural light. The less interesting aspect of this collection of molecules is that they form an actual grilled cheese sandwich, and no Jesus as yet.

  7. I am astonished to see how much they resemble the diagrams, and as AK said above, how hexagonal they actually are. Amazing!

    1. Bear in mind that we’re seeing actual quanta of matter here. The parts are completely interchangeable, and fit together only in highly specific ways. There isn’t a lot of scope for sloppiness at this scale. So if there’s such a thing as a perfect hexagon anywhere in the universe, this is where we should expect to find it.

        1. The very fact that you see a smooth outline of a figure where there really are only a handfull of separate electron particles moving around. Theres you quantum uncertainty!

          1. “Really” is a slippery word here. From some points of view, the probability cloud is the reality, and the appearance of discrete electrons is an illusion.

            1. Explain! I would say from a quantum mechanics point of view the number of electrons involved is a clearly defined quantity – you could remove them one by one until the thing falls apart completely. The objects you then have removed are from a macroscopic perspective a finite number of pointlike things.

              1. Better: the electrons have no classical postion. And the reason they look somewhat localized is because of the system they are in (network of benzene rings or whatever) and because of the time scale. I’ve been told, for example, that even different conformers of cyclohexane can be isolated given low enough temperature.

        2. Probability doesn’t have to be smooth or continuous. Electron energy levels are discrete and discontinuous (that’s why it’s called quantum theory), and therefore so are the types of bonds they can form.

      1. Indeed, I’m more used to biology where the diagrams are significantly cleaned up and reality very messy.

  8. How are the benzene rings in the reactant molecule bonded? That doesn’t look like graphene and I don’t recognize the symbols drawn between the rings, although they obviously correlate with something in the AFM images.

      1. Went to original article. They state there are 26 C atoms in the reactant image. Therefore, there are two C atoms creating the links between rings. Remembering that those images are showing electron density – i.e., where the bonds are – it now makes sense and there is indeed a triple bond (acetylene-like functional group) there. I just did’t remember those bonds’ being drawn in that “truncated” way.

  9. Last Year, Gerhard Meyer’s group at IBM Research Zurich published a paper on “Imaging the charge distribution within a single molecule” using AFM, and I found that amazing. Now this!

    Speaking of AFM, which was invented by Gerd Binnig in 1986, also at IBM Zurich: Binnig’s colleague, mentor and co-laureate for the invention of STM (Scanning Tunneling Microsopy), Heinrich Rohrer, died just a fortnight ago. I’m sure he would have enjoyed these results, which he anticipated.

    1. Ah yes, but if god is omnipresent then we ought to be able to see ‘bits of god’ in these pictures. Which we don’t. Unless, of course, he doesn’t interact with the oxygen atom at the tip of the microscope… which makes you wonder how he ‘does things’ in the real world.

      The god-struck will still coo slogans like ‘God is love’. Perhaps they should add ‘God is noncontact atomic force microscopy’ too?

  10. I dunno. To me, it seems pretty disappointing that imaging of the actual molecules looks so much like the stick diagrams. May as well just stick with the diagrams; reality has nothing to add, except blur. Ho hum.

    Although, I would’ve thought reality would at least draw in the damn hydrogens.

    1. “stick diagrams. May as well just stick with the diagrams; ”

      Hence the name!

      “reality has nothing to add, except blur. Ho hum.”

  11. WOW!

    It gives me chills up the spine to see this imagery. I never thought I would live to see such resolution & such an affirmation of what was always taught as a ‘conceptual model’ rather than ‘what it really looks like’.

    I’m blown away… thanks for posting this.


  12. Between this and aberration corrected high resolution tem atomic scale imaging is an awesome thing. When I did my graduate work, we started with image simulations at various microscope settings to get an idea of what to look for. Actually finding it is exhilarating!

Leave a Reply