Steam-explosion pretreatment of wood: effect of chip size, acid, moisture content and pressure drop

Material balances for pentosan, lignin, and hexosan, during steam‐explosion pretreatment of aspenwood, showed almost quantitative recovery of cellulose in the water‐insoluble fraction. Dilute acid impregnation resulted in more selective hydrolysis of pentosan relative to undesirable pyrolysis, and gave a more accessible substrate for enzymatic hydrolysis. Thermocouple probes, located inside simulated aspenwood chips heated in 240°C‐saturated steam, showed rapid heating of air‐dry wood, whereas green or impregnated wood heated slowly. Small chips, 3.2 mm in the fiber direction, whether green or airdry gave approximately equal rates of pentosan destruction and solubilization, and similar yields of glucose and of total reducing sugars on enzymatic hydrolysis with Trichoderma harzianum. Partial pyrolysis, destroying one third of the pentosan of aspenwood at atmospheric pressure by dry steam at 276°C, gave little increase in yield of reducing sugars on enzymatic hydrolysis. Treatment with saturated steam at 240°C gave essentially the same yields of glucose and of total reducing sugars, and the same yields of butanediol and ethanol on fermentation with Klebsiella pneumoniae, whether or not 80% of the steam was bled off before explosion and even if the chips remained intact, showing that explosion was unnecessary.