by CKR
My creativity is lagging. It could be campaign overload, or a series of difficult interactions, or the recent unsettled weather, or the beginnings of my head fogging up with reactions to juniper pollen.
There is something that has been bothering me lately that isn’t politics in some form, so let’s see if I can write about it. The February 8 Science magazine carried a research report about a chemical reaction selectively driven by lasers.
Back in the 1970s, as lasers became more available and as research proceeded into how chemical reactions might be harnessed to produce lasers, it seemed an intriguing prospect to excite a particular chemical bond with laser radiation, which would incline that bond, of all bonds in a molecule, to react chemically in a particular way.
Chemical bonds vibrate at characteristic frequencies, those frequencies being high enough that they are expressed as infrared light. Many lasers operated in the infrared range, so it seemed that one might match up the laser with a particular bond vibration.
And it worked, to a point. You could match the laser to the bond vibration, and the molecule absorbed the energy. But that energy was quickly distributed through all the bonds in a molecule, not just the bond that absorbed the energy. So the molecule as a whole was more ready to react, but not in particularly selective ways. You might just as well heat the thing on a Bunsen burner.
My contribution to the discussion was to suggest that having reactants available in the mixture could help to provide a differential effect; the energized bond would be able to react with other molecules just a bit more quickly, so you would get a bit more of the desired product than you might otherwise. I also decided to look at higher-frequency electronic transitions and had some success there, but that’s another story.
But differential effects were of little interest to most of the scientists looking at the issue. Zap the molecule and make it do our bidding! All or nothing. And, indeed, it might be hard to verify my approach. Differential effects, unless they are very large, require large numbers of experiments to show up. Nonetheless, many industrial processes are based on differential effects, and no industrial chemical synthesis provides a 100% yield.
Chemical reactions almost always take place with collections of molecules. That makes the problem worse than I’ve said so far: you’ve always got lots of molecules around, bouncing off each other and off the walls of the container, in both liquids and gases. Those collisions transfer the laser energy to other molecules. That’s even more like heating the thing up with a Bunsen burner.
Selective laser excitation and reaction could produce some very desirable outcomes: allowing difficult syntheses of rare molecules that might be used to treat disease; replacing catalysts in industrial processes with laser light, which could make the processes cheaper in multiple ways; and, yes, separating uranium isotopes. But all these require that selective reaction be generally applicable.
So the problem of bond-selective chemistry continued to intrigue. One way of getting at a problem is to simplify its conditions. If we take out most of the gas so that there are fewer collisions or line the molecules up in a molecular beam; if we play with various combinations of laser zaps to make the molecule react by itself before a collision can take place; if we create special molecules that behave in peculiar ways, then maybe we will see bond-selective chemistry.
And, eventually, a selective reaction was induced with lasers.
H + HOD —> OD + H2Exciting the OH stretch in HOD made the first reaction predominate; exciting the OD stretch gave more of the second reaction. This was done in a molecular beam, the molecules neatly spit, one by one, out of a very small aperture.
H + HOD —> OH + HD
The more recent achievement also depended on the differences between hydrogen and deuterium atoms, this time in methane rather than water. The experimenters excited CD3H molecules, methane with deuterium substituted for three of the normal hydrogens, and smashed them into a metal surface. The C—H bond broke more frequently than the C—D bonds.
So what, you may well ask. That is precisely what has been bothering me. We started with a big question, whether lasers could promote bond-selective chemistry. The big answer was that, in most cases that might be of interest at an industrial scale, not really. But the problem has been specialized and whittled down until systems were found that allowed bond-selective chemistry.
These experiments may provide some insights that will make it possible to extend bond-selective chemistry. They may provide some insights into something entirely different. But, for the life of me, I don’t see any just now. Some basic understandings of molecular energy transfer may be refined. In the second set of experiments, the role of the solid metal surface was a bit different than might have been expected. Perhaps this will eventually contribute to a better understanding of catalysis, which underlies all of petroleum chemistry.
But it’s not clear to me that the scientists doing these experiments and those admiring them recognize the limitations they’ve imposed on themselves in the service of getting to laser-enhanced bond-selective chemistry. Just reaching that goal seems to have been enough for them, that original big question forgotten.