Pyrolysis of biomass is one of the oldest technologies, originally used to make charcoal for iron smelting, however recently it has been advocated as a means of making liquid fuels. I find this difficult to come to grips with and I question, from a theoretical point of view, is this process worth persisting with or should we move onto something else? Are we wasting valuable money pursuing a lost cause?The objective of gasification is excluded from this discussion.
 
Saleh et al. (Energy and Fuels, 23, 3767 (2009) has given an outline of how to optimize pyrolysis of biomass. The required conditions to optimize liquid products are a very high heating rate, finely ground feed and temperature of 500 degrees C. The yield of liquids can be as high as 44%, although these may have up to 40% water in them. The rest is char and gas. The problem then is, what to do with the liquids? The major constituents are often formic and acetic acids, 1-hydroxy-2-propanone, furfural and a range of phenols. There are also at least a hundred minor components, some of which reach a few per cent. Saleh et al. suggest that this oil can be converted to synthesis gas for Fischer Tropsch synthesis, but this seems somewhat wasteful – if synthesis gas was required, why not gasify the biomass as a whole? Similarly, the oils can be upgraded through hydrogenation technology, (at the cost of losing the small carbon fragments) but this becomes yet a further step, and if you are going to do that, why not hydrogenate the biomass in the first place? Some suggest using it as low-grade heating oil, but if you merely want to heat something, it might be more efficient to burn the original biomass. One thing that should not be done is to store it for a significant time because most pyrolysis oils appear to be somewhat reactive, and over time they darken and become much more viscous.
 
The composition is highly dependent on how you do the pyrolysis, and on the raw material. The reason for this is that while the pyrolysis initial steps are quite simple, the products further react. Lignins are based on the free radical condensation of phenols substituted at the 4-position with CH=CH-CH2-OH groups; the phenol may have 0, 1 or 2 methoxyl groups adjacent to the phenolic group, depending on the nature of the plant. During pyrolysis, lignins tend to break either at the 4-position or one carbon further, so the products tend to be methoxylated phenols, 4-methyl methoxylated phenols, aldehydes such as vanillin, together with a variety of other 'bits and pieces' from the alkyl group. As far as I am aware there is no preferred procedure that makes some more useful products. Pieces not attached to phenols are effectively lost.
 
Cellulose first decomposes almost entirely to levoglucosan (1,6-anhydroglucose), which is volatile at the decomposition temperature. However it is also highly reactive, and if it cannot be removed from the solid in a very short time, it reacts with something else and forms a cascade of materials, most of which would generally be described as 'tars'. The same can be said for a number of other products of biomass pyrolysis, which essentially then requires the material to be very much size-reduced (which is expensive) and the heat transfer rate to be very high.
 
To summarize, this technology converts about a third to a half of the biomass to form liquid with a wide a range of reactive components that have little direct use. Superficially, pursuing this technology would seem to be something of a last resort yet there is evidence that a lot of money is being spent attempting to make it work. There appear to be more papers produced in the energy-related journals  that I read on this topic than any other. Why? Someone is just plain wrong; the question is, who?