3/24/2007

Defying Woodward-Hoffmann with a Yank


As Paul from TotallySynthetic has mentioned, this issue of Nature is packed with organic chemistry- rather unusual for the journal. Reviewed in TotSyn was Baran's protecting group-free total synthesis of Ambiguine H and other molecules, which is quite a read. Also in the issue are papers by Bergman on C-H activation and one on relativistic effects in gold catalysis by demi-god F. Dean Toste. Like I said, jam-packed with organic-y goodness.

Then there's the cover story. This one's particularly near and dear to my heart, as it comes from my own group. The Moore group, in collaboration with profs from materials science and aerospace engineering, has developed a system which displays new reactivity when mechanical energy is utilized. The article has a much more thorough description than I could butcher out, so instead, I'll just attempt to give a general description. (You should read the article anyway.) If you're reading this blog, chances are good that you're at least partially familiar with the Woodward-Hoffmann rules, which predict the geometry of products of pericyclic reactions. The pericyclic reaction in question is the 4-electron concerted ring-opening of 1,2-disubstituted benzocyclobutenes. The Woodward-Hoffmann rules dictate that heating of this system should undergo conrotary ring-opening, and illumination of this system should result in disrotary ring-opening, regardless of the substituents. However, as computational studies have predicted and experiment has confirmed, the ring-opening geometry when mechanical energy is used is dependent on the geometry of the subtituents- in other words, whether conrotary or disrotary ring-opening occurs is dependent on whether the disubstituted benzocyclobutene is cis or trans.

This new reactivity isn't terribly hard to comprehend. You can do this experiment at home, without even having to resort to messy density functional theory like they did. Simply make a model of benzocyclobutene, and attach strings to each sp3 carbon. Now yank on the strings, hard. Whether you attach the strings on the same side (cis) or on different sides (trans), and assuming the model doesn't break (don't yank too hard), you end up with the same product, in stark contrast to the Woodward-Hoffmann rules. It's completely intuitive, again in stark contrast to the Woodward-Hoffmann rules. :)

So, great. It works in theory. The clever bit was proving this happens on a molecular level. The Moore group, being the deft polymer chemists they are (we are? though I don't do polymers), decided to utilize polymer tethers (namely poly(ethylene glycol)) as the "strings." A fluorescent, 13C-enriched dienophile was used to "capture" the ring-opening products via Diels-Alder cycloadditions, which were purified and analyzed by GPC and NMR.

The source for the mechanical energy was ultrasound. Pioneered in part by Ken Suslick in the same department, ultrasound is a viable source of mechanical energy on the nanoscale. Ultrasound induces small cavitation bubble formation in liquids which subsequently (and quickly) collapse. The collapse causes liquid molecules to "rush" into the space, resulting in frictional forces against the reactant molecules. (neat, huh?) Using ultrasound as the mechanical energy source, both cis and trans-polymer substituted benzocyclobutenes react to give the same stereoisomer, agreeing with theory. A very clever experiment!

So yes, I am tooting my own horn here, as it's my own group's work, even though I had nothing to do with it. But it is very novel science, indeed (plus, it probably won't happen again that I get to brag about my group making the cover of Nature). Although this system probably doesn't serve any practical purpose, it might open the doors to new systems which utilize mechanical stress as their main energy source. For more info, there are some fun videos on the Nature website with the Bossman explaining the lure of mechanochemistry and the development of the system.

18 comments:

Ψ*Ψ said...

Very cool! Carmen also covered the Baran paper over at She Blinded Me with Science.
I got such a horrible and incoherent explanation for the Woodward-Hoffmann rules that it hurts a little just to hear that phrase. Needless to say, I'm sitting in on that class again under someone who can actually teach.

Chemgeek said...

HA!!! Take that Corey... I mean Woodward!!!

Anonymous said...

Wouldn't using ultrasound be like using pressure to break bonds? So where is the novelty there?

Mitch

Excimer said...

Mitch, that is the novelty, or at least part of it. Not only are you using mechanical energy to cause a reaction to occur, you're also using it to direct the course of the reaction, in a manner dissimilar to heat and light. So the novelty is threefold, I guess: using mechanical energy to break bonds (rarely used, esp. in organic chemistry) to direct a new course in a reaction (hardly ever seen except in polymers) in a predictable, reproducible manner (why this paper is in Nature).

ElwoodCity, Ph.D. said...

Aaaah, you are AT the U of I. I guess all the love for the U of I I see here turns out to be a bit of Homerism. I just thought that perhaps you had just stumbled onto Ken Suslick's witty homepage by accident.

Excimer said...

Yes yes elwoodcity, I am at the U of I, but not everything I've written about it on this blog has been particularly positive. It is hard, though, to not like Ken Suslick. (Also fun: I know who YOU are... but you don't know who I am- not yet, anyway. Not many people around here who are obsessed with Unwiches. ;)

Wavefunction said...

4 pi thermal con: Remember that and the rest is apple pie.

Anonymous said...

Do the authors know that the ring opening isn't radical mediated (that physical force isn't generating intermediate radicals which yield the observed products)?

Excimer said...

I'm not sure, Hap. I think the stereospecificity of the product suggests a concerted mechanism and not a radical one.

Anonymous said...

I think that stereospecificity refers to reactions in which two distinct stereoisomeric reactants give two distinct stereoisomeric products (Sn2 is the canonical example). Stereoselective reactions are those in which is one stereoisomeric product is preferred over another - it doesn't necessarily matter what one you start with. (epimerizing an a-substituted ketone in base).

I didn't know if physical manipulation might lead to radical intermediates - you might see mixtures of stereoisomers rather than just one in that case, based on which one is most stable (I figure that if formed the radicals could equilibrate, but maybe not). On the other hand, with a single molecule or small numbers, you might not see statistical behavior. It could be as said, too - I don't know how many people figured that light and heat would give opposing results in reactions.

ElwoodCity, Ph.D. said...

I guess I was just thinking of one post, in particular.

http://coronene.blogspot.com/2007/01/rankings-we-dont-need-your-stinkin.html

Excimer said...

hap, (yeah ok I get the two terms mixed up a lot. i'm not alone though.) As for the radical intermediates, i'll have to look into it. iirc it was a possibility that was eliminated, i just don't remember how.

Anonymous said...

Sorry - wasn't trying to be annoying.

Anonymous said...

I saw a grad student get RIPPED apart for mixing those two up at a seminar. It made everyone pretty uncomfortable, and then there was a huge blowup between the ripper and the rippee's boss.

Chemgeek said...

Sure, Excimer, you didn't get that job with C&EN, but at least you scooped them on this and the alkyne/nitrile metathesis post.

Their lose.

Carmen said...

Cool post. It's Nature, so I think it's OK to toot your own horn here. I don't know much about AFM; has it been used to effect similar reactivity on a surface? Not sure what that'd be useful for, but just curious.
I had never heard of Woodward (or the rules) until I got to grad school. My adviser almost cried when I told him.

Excimer said...

I talked to the Bossman, Hap, and the answer is "We don't know." The DFT calculations don't distinguish a radical vs. a concerted intermediate in this case, so a non-concerted pathway is possible. Additional tests would have to be done. Lord knows we're not done with this beast quite yet.

Anonymous said...

Radical intermediates would be tough to find there anyway. Thank you for asking.