In previous blogs, I tried to outline some of the pros and cons of ethanol as a biofuel, with fermentable sugars largely being provided either from crops, or from lignocellulose by fermentation. The difficulty with using lignocellulose is that lignin has evolved to protect cellulose from such enzymatic attack, with the result that processing plant is very large, and there are massive volumes of water and wet spent biomass to process. Further, there is almost as much hemicellulose as cellulose, and because of the variety of linkages in hemicellulose, a set of enzymes is required, and there is a significant additional cost involved in supplying these enzymes. The alternative is acid hydrolysis, and for some reason, this appears to have been discarded as an option. Is that premature?
The basic problem with acid hydrolysis is that employing dilute acid results in an unacceptably slow reaction unless heated, but glucose (the desired product) reacts with hot acid to produce hydroxymethyl furfural (HMF), which further reacts to produce levulinic acid plus dark polymeric material. The net result is that in the Madison process, which involves heating wood chips with dilute sulphuric acid, the recovered yield was about 35% of theoretical, which is clearly not good enough. Interestingly, however, there are two options that are seemingly not currently considered.
The first is flash hydrolysis. As shown by Chen and Grethlein (Biomass 23: 319-326, 1990) if you heat the biomass for seconds at over 200 degrees C, conditions they argue are reasonably reached in a specially designed cyclonic reactor, then yields of up to 87% are claimed. The concept is, of course, you make the glucose and quench it before the conversion to HMF can get underway. Actually, that in turn may not matter that much because if the HMF can be recovered, it can also be converted into useful chemicals/fuel. These conditions will also hydrolyse hemicellulose, and the pentoses, which do not so easily ferment to ethanol, should be recoverable as furfural, which is valuable in its own right. One problem might involve size reduction; the acid has to get at the cellulose to hydrolyse it, seconds do not permit much diffusion, so the interior of chips may not be reached. Size reduction is possible, but the work done doing it may be too costly.
The second option is hydrolysis with chilled 40% hydrochloric acid. Provided the concentration of acid is high enough, the cellulose simply dissolves. The cellulose converts smoothly to glucose, and further reaction is apparently trivial. The hydrochloric acid is removed by vacuum distillation and recycled. You may be skeptical, however this is one of the very few processes that were ever deployed at scale: the Germans made ethanol this way during World War II. As far as can be determined, this worked well even on reasonably sized chips, and there was only one problem: corrosion. However, while biomass-processing may not have advanced much since then, materials have become far more advanced.
Would either of these processes solve any problem? I do not know, but what concerns me is that there are no data around that are readily available that would permit these processes to be either eliminated from consideration, or provisionally considered so that data can be obtained to address questions that currently have no answers. Sending public funds after random guesses seems wrong to me. Why not analyse the problem and publish the findings so that funding can be deployed on a more rational basis? A very large amount of money must be invested to compensate for declining oil supplies. Given the current financial constraints, surely it is better not to waste it, and any progress that can be made from theory will save a lot of wastage. Theory is cheap – why not use it?
