Common Aspects of Acid Prehydrolysis and Steam Explosion for Pretreating Wood Hans E. Grethlein
Michigan Biotechnology Institute, PO Box 27609, Lansing, Michigan 48909, USA
& Alvin O. Converse
Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire 03755, USA
The initial rate of hydrolysis using cellulase from Trichoderma reesei for various wood samples is directly proportional to the surface area available to the enzyme. Both dilute acid hydrolysis in a continuous flow reactor or autohydrolysis in a steam exploder are similar pretreatment methods in that they increase the pore volume of the wood by removing hemicellulose which increases the surface area available to the enzyme. Thus, pretreated wood samples by either method with the same available surface area give essentially the same initial rate of hydrolysis; and, the increase in rate of hydrolysis is controlled by the severity of the pretreatme~t. Key words': Pretreatment, enzymatic hydrolysis, acid hydrolysis, autohydrolysis, steam explosion, Trichoderma reesei, wood hydrolysis.
INTRODUCTION It is generally recognized that native cellulosic substrates cannot be hydrolyzed rapidly with cellulase without a pretreatment (Millet et al., 1975, 1979; Grethlein, 1980; Chang et al., 1981; Ladisch et al., 1983). This is true for any type of cellulase system (Hart et al., 1981; Ryu et aL, 1982; Avgerinos & Wang, 1983; Bachmann et al., 1983; Holtzapple & Humphrey, 1984). A recent review (Grethiein, 1984) indicates that there are many pretreatment methods and a better understanding of why pretreatments work or do not
work is desired. The objective of this paper is to help define what is required of a pretreatment and how to select the best one. Whether one uses cellulase or acid as the catalyst for the hydrolysis of a heterogeneous substrate such as cellulose, the catalyst must be able to contact the glucosidic bonds. Fortunately cellulose in biomass is very porous as indicated by the data in Table 1. There are about 600-800 m 2 of surface area per gram of substrate, but approximately 98% of this area is in pores that are available only to molecules smaller than 51 A. More details on the structure of wood can be found in Fengel and Wegener (1984) and Cowling (1975). The remarkable thing about acid hydrolysis is that, at a given temperature and acid concentration, the rates of hydrolysis of a wide variety of cellulosic substrates are within a factor of two of being equal (Grethlein & Converse, 1983). This suggests that the catalyst, the hydrated hydrogen ion, which is about 4 A, is able to get at all the glucosidic bonds with relatively the same ease in all substrates. With cellulase, however, the rate of hydrolysis of cellulosic substrates can vary by orders of magnitude. While not much information on the shape and overall size of cellulase is known, what is reported suggests that if the protein is essentially a sphere in aqueous solution, the diameter is about 50 A and if it is an ellipsoid, it is about 30 x 180 A (Cowling, 1975). Although the size and shape of the enzyme are important in understanding the details of accessible surface area, this issue is not addressed here but an empirical size of 51 A is used for the cellulase because some useful correlations develop below.
Bioresource Technology 0960-8524/91/S03.50 © 1991 Elsevier Science Publishers Ltd, England. Printed in Great Britain
Hans E. Grethlein, Alvin O. Converse
Table 1. Calculated surface area from solute exclusion measurements for various woods (m2/g)a Substrate Area available to various sized molecules
Mixed hardwood Poplar White pine Steam extracted pine 846 645 800 745
10-5 6"5 13'7 13"9
2.0 4.0 5"8 2.6
"Data from Grethlein (1985). b4 A is the size of the water molecule.
The problem is even more complex than this because, in fact, there are endo- and exo-cellulase...