Lignin is a macromolecule that most people don't like. It gets in the way of animals when they are trying to digest plant polysaccharides, and it gets in our way when we are trying to make wood cells into paper. Its structure is unpleasantly complex and varies in an awkward statistical manner from one location to another. Because lignin is polymerised not by enzymes but by a statistically controlled chemical process (after enzymic generation of the monomer free radicals), it cannot be dismantled enzymatically in the organised way that we expect of other biopolymers, that is by reversal of the synthetic route. It follows that a lignin molecule cannot be made soluble without destroying at least some of its characteristic structural features. Since most classical methods of structure determination work in solution, that's a problem.

We have used solid-state ¹³C NMR to look at the structure of lignins in situ, without the need to make them soluble. In particular we have done this on lignin from plants with genes for lignin synthesis downregulated. It turns out that solid-state NMR is not very good for qualitative identification of lignins with unusual structures, because the spectral resolution isn't good enough. However for structural features that we are able to identify, like G:S ratios and cinnamoyl esters, it gives useful quantitative information that is almost impossible to get in any other way. This is simply because we don't need to solubilise the lignin first and so we are looking at all the lignin present, not just the less cross-linked fraction that can be brought into solution.