Skip to Content

Tie Your Crystals Into Knots

Chemists love crystals. We don’t do as much recrystallization as we used to, since there are higher-throughput (and less labor-intensive) ways of purifying things these days, but I don’t think I’ve ever met an organic chemist who isn’t happy when a product crystallized out nicely. And we all know what crystals are like – straight-sided, hard, brittle, prone to fracture under stress into smaller shards, etc.

Or not. People who do grow a lot of crystals (typically for X-ray structures) can tell you that while most things fit that description, there are some oddball outliers. You see some compounds that grow long, curved crystals, for one, which makes you think that one face is behaving differently in solution than another. Some of the longer ones are surprisingly twangy, and can take a good deal of bending before shattering, and even the chunky, faceted ones vary quite a bit in how hard and friable they seem to be if you expose them to stress. There’s a lot going on in the crystalline solid state, and not all of it is well understood.

Credit: Nature Publishing Group

This paper illustrates the point, for sure. It’s looking at what you’d think would be an unremarkable coordination compound, copper (II) acetylacetonate, known to many chemists (as are its kin) as “copper ack-ack”. I would not want to guess how many (acac) metal complexes have been crystallized, alone or with other ligands, but let’s stipulate that it is a large heap, and that it’s a very well known species. But you can grow long needle-like crystals of the plain Cu(II) complex that act very weirdly indeed. They’re long and flexible, and as that photo shows, can actually be threaded into knots and then untied. What’s going on at the atomic and molecular level to allow them to bend this much without breaking?

The paper presents a careful X-ray study that figures it out. The authors (from Queensland) mounted a bent crystal in a synchrotron beam and carefully focused in on different parts of bent shape. They found that a particular crystalline axis was significantly elongated on the outside of the curve, and compressed on the inside, as you’d well imagine. It turns out that while each individual copper-acac molecule is identical as the crystals bends, the arrangement they make with each other certainly changes. The distances between the crystal planes don’t change, but the molecules themselves rotate with respect to each other (here’s a movie from the paper’s SI files illustrating that). The changes are not large at all, but when you add them up across several zillion molecules in a long crystal, it gives you some real wiggle room. A philosophical question comes up: if such a sample is not a regularly arranged array of molecules across its width (or not any more), is it still a crystal, or not? If not, do we have a word for what it is?

This is a nice piece of crystallography, of course, but people into materials science will appreciate that there are a lot of interesting features that might emerge from some changes. The optical and magnetic properties of the different sides of such a crystal could well be different (in fact, might almost have to be), and these changes could well be valuable in real-world applications. These results apparently also go against at least one theory of what sorts of crystals can undergo such deformations and how they do it, which would make you think that there could be several different mechanisms available. There’s a whole world in between crystals and amorphous matter, and there’s a lot being discovered in it.

 

 

Note: All opinions, choices of topic, etc. are strictly my own – I don’t in any way speak for my employer

16 comments on “Tie Your Crystals Into Knots”

  1. colintd says:

    I still find glide twinning a magical process, given that it effectively involves rearrangement of atoms with atomic precision from one orientation to another.

  2. Andy J says:

    > A philosophical question comes up: if such a sample is not a regularly arranged array of molecules across its width (or not any more), is it still a crystal, or not? If not, do we have a word for what it is?

    Presumably a quasicrystal of some sort: there’s a pattern, but it can’t be translated

    https://en.wikipedia.org/wiki/Quasicrystal

  3. One Man CRO says:

    I’m certainly happy when a product crystallizes out of solution……provided that does not happen (unexpectedly) during chromatograhpic purification!

    1. Nick K says:

      In my hands, the first time a compound crystallizes is usually while I am pipetting it onto the top of a column.

  4. Uncle Al says:

    Is Jahn-Teller distortion operative here? Does ß-Cu(II) phthalocyanine cooperate in kind?

    Cu(acac)2
    DOI:10.1038/nchem.2848
    space group P2(1)/n

    ß-CuPc
    DOI:10.1021/nn100782w, arXiv:1012.2141
    DOI: 10.1063/1.1766295
    space group P2(1) /a
    http://www.pcimag.com/articles/83452-copper-phthalocyanines

    1. Rhenium says:

      Oh my god, Uncle Al is still metabolizing… (!)

      1. Uncle Al says:

        Viam sapientiae mundi, per quam pervenitur. If it isn’t empirically predictive, it is only mathematics. I deal in Pyrex.

        1. industrychemist says:

          Uncle Al… he’s everywhere you want to be. (And some places you don’t either)

  5. HGMoot says:

    “There’s a whole world in between crystals and amorphous matter, and there’s a lot being discovered in it”.
    Exactly! And remembering about this can be quite important in the art of process chemistry product purification and isolation, especially when it comes to defining the state of the matter isolated. It happened to us a few times that complex polyamine salts could be nicely purified by a “crystallization”, yet crystallographers repeatedly told us that the structure was a long shot from anything resembling a crystal… yet it looked to me and behaved like a crystal…

  6. David says:

    That is…absolutely fascinating. I wonder if such an effect could be used to design multifocal contact lenses, that would change optical properties in specific ways in response to deformation as the eye attempts to focus on objects at different distances.

  7. Barry says:

    “The optical and magnetic properties of the different sides of such a crystal could well be different (in fact, might almost have to be), and these changes could well be valuable in real-world applications.”
    I for one certainly expect birefringeance in a bent crystal, for exactly the same reasons we use polarized light to assess stress/annealing in glass.

  8. gippgig says:

    This reminds me of an item in Science that stated that a reaction product “crystallized as an amorphous solid”.
    I remember reading once that bent crystals can be used to steer ion beams (don’t remember details).
    I suspect that frozen liquid crystals might be very flexible.

    1. tangent says:

      Steering ion beams might be similar to the “ridged mirror” used to reflect neutral particles at grazing angles? In this case a curved mirror which the particles follow.

  9. Chris Phoenix says:

    Piezoelectric crystals change shape due to electric fields, or vice versa. Apparently, this happens when the atoms move around. I’m not sure whether this is directly comparable to what’s reported here (molecules moving vs. atoms) but here’s a link in case it’s interesting.
    http://www.sas.upenn.edu/rappegroup/publications/Papers/Cooper02p26.pdf

  10. Anon says:

    That’s no crystal. Somebody shed a pube.

  11. George says:

    It actually looks to me like the while the distance between (some of) the crystal planes is changing, the distance between the planes defined by the molecules (which I think is the 11-1 plane) is what’s staying the same. Alas, I have no access, but in the video at 0:30s, the unit cell box in the b direction is getting smaller as you rotate perpendicular to it.

    I’m not quite sure what I would call it either. It doesn’t quite fit the formal definition of a quasicrystal because it lacks any exact symmetry about any given point, but it doesn’t fit any of the experimentally defined weird guys that show up with modulation.

    Looking at this article,
    http://journals.iucr.org/b/issues/2015/03/00/dq5011/index.html

    I would want to call it…a continuous crystal?

Leave a Reply to Uncle Al Cancel reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.