Does anyone check a paper's claims?
One interesting paper from the not too distant past involved the reduction of carbon dioxide to either methanol or methane (J. Am. Chem. Soc., 2015, 137, 5332) using lithium o-phenylbisborate as a catalyst. What the catalyst is claimed to have done is to bend the CO2 molecule (highly plausible) and thus form an aromatic ring. It is this last part that I find hard to stomach, because it brings us back to the question, what causes "aromaticity"? Now, I should issue a warning here: I have published what I think causes aromaticity, so I am not exactly unbiased.
So, where is my problem? The authors seem to have argued that a six-membered ring is formed (correct) and there will be 6 π electrons in it, therefore the system will show aromaticity. I suppose if you construct molecular orbitals and then place the electrons in them, there is a case for this. However, my argument about aromaticity is there has to be 2n 1 double bonds that alternate with single bonds, which is not quite the same thing. The reason for aromaticity in this case lies in the phase of the waves. Similarly to thinking about the Woodward Hoffmann rules, run the phases around the ring, then keep going. What you find is that with aromaticity, the second round cancels the first, which cancels the double bond amplitude, and since the charge has to go somewhere, it goes to the single bonds (the other major canonical structure). But that has the same problem, and as such, molecules such as cyclohexadiene cannot exist. Cyclobutadiene, however, finds that the second cycle reinforces the displacement of the first cycle, and so it is locked into the classical structure. Now, the reason I find this reduction paper of interest is that in principle it offers an alternative. My model predicts no aromaticity because the double bonds in carbon dioxide are orthogonal. The double bond orbitals cannot overlap with each other and therefore cannot form an extended wave with one polarization.
Does it matter? I think so. I think it is important that chemists try to understand what is going on. Oddly enough, when I started my career with physical organic chemistry, by at large chemists thought they understood tolerably well most of the reactions of which they were aware. Now there are so many additional reactions, but I am far from convinced the understanding has increased.
One final point. The paper ends with a statement that "further studies" are required to adopt the transformation for "practical applications". Methanol and methane will not be there. This catalyst merely bends the CO2, the actual reductant is either triethylsilane or pinacolborane. These would be more than somewhat more expensive and harder to get than methane and methanol. That hardly seems likely to be "useful", at least from what was demonstrated in this paper.
Posted by
Ian Miller on Sep 28, 2015 5:06 AM Europe/London
