8-44: The use of predictive models to optimize sugar recovery in the steam pretreatment step of a softwood-to-ethanol process

Acid catalyzed steam pretreatment is recognized as a very effective method for producing soluble hexose sugars from softwood hemicellulose. The recovery of this fraction contributes to high total sugar yields, after pretreatment and enzymatic hydrolysis, as well as a high ethanol yield following subsequent fermentation. Prediction of hemicellulose yields from steam pretreatment substrates is desirable for the purpose of process control and it is anticipated that predictive models may prove to be invaluable for the design of effective steam pretreatment reactors. In the past, several different strategies have been employed in the development of predictive pretreatment models. Many empirical models are based on thermal severity factors while some others were developed using response surface methodology (RSM). Theoretical models are based on reaction kinetics, mass and heat transfer, or both. To date, very few theoretical models have been developed for the steam pretreatment of softwoods. Of the theoretical models available, we anticipate that those containing both reaction kinetics and transport phenomena are most able to predict the direct outcomes of pretreatment. However, we also found that thermal severity factors closely approximated these same outcomes when steam pretreatment was operated under controlled conditions. As a result, we have established rough guidelines for the appropriate use of thermal severity factors. Using data from previously published work on the SO₂ catalyzed steam pretreatment of radiata pine (Pinus radiata), an RSM model was developed. In a first attempt to broaden the model’s applicability, it was then validated using whitewood chips of lodgepole pine (Pinus contorta).