Bioresource Technology 35 ( 1991 ) 275-282 An Integrated Model for the Technical and Economic Evaluation of an Enzymatic Biomass Conversion Process Q. A. Nguyen* & J. N. Saddler: Forintek Canada Corporation, Biotechnology and Chemistry Department, 800 Montreal Road, Ottawa, Ontario, Canada K1G 3Z5 (Received 1 September 1989; revised version received 7 February 1990; accepted 7 April 1990) Abstract A process simulation model was constructed, using the Lotus 123 spreadsheet, to evaluate the technical and economic feasibility of a lignocellulosic-to- ethanol bioconversion process. The major com- ponents of the process were: steam pretreatment, fractionation, enzyme production, enzymatic hydrolysis, ethanol fermentation of glucose and xylose, product recovery, and waste treatment. The model provided an estimation of the price of ethanol and evaluated the interdependence of pro- cess parameters and their effect on the production cost of ethanol. A case study of a plant processing 500 tonnes of aspenwood per day showed that the parameters which had the most significant impact on the price of ethanol were: the cost of wood, cost of enzymes, efficiency of cellulose hydrolysis, ethanol yield from xylose, efficiency of the frac- tionation process, and the selling price of the lignin by-product. The benefits of developing a model included the identification of areas for technical improvement, reduced time and cost for scale up and pilot plant testing. Key words: Process simulation model, biomass, aspenwood, bioconversion processes, ethanol. INTRODUCTION of lignocellulosic wastes to ethanol. During this period there have been significant advances in some of the component parts of a bioconversion process, resulting in lower overall costs for the projected price of fuel ethanol. However, most of the current economic analyses still predict an ethanol price of S0.50-S0.95 per liter which is uncompetitive in terms of today's oil prices. In the past, our research group has examined various pretreatment methods i'Brownell & Sadler, 1984), sources of cellulase enzymes (Schwald et al., 1987), fermentation options (Mes- Hartree et al., 1984; Yu et al., 1984) and by- product credits to try and reduce the cost of producing ethanol. This has proven to be considerably more difficult than originally appre- ciated because of the complex interdependency of the various process steps. For example, we had originally planned to utilize a portion of the pre- treated lignocellulosic material as the substrate for enzyme production, however other groups (Vandecasteele & Pourquie, 1989) have achieved high enzyme productivities by using soluble sub- strates such as lactose. In our work we had improved the rate of cellulose hydrolysis by increasing the amount of enzyme added. How- ever, this raised the cost of enzyme production, diverted more of the substrate away from hydro- lysis and consequently lowered the overall glucose yield per unit mass of wood. Another variation involved the use of SO2 during steam pretreat- ment, which greatly reduced sugar loss due to degradation and increased cellulose hydrolysis, but also resulted in solubilization of a consider- able amount of the original cellulose (Schwald et aL, 1987). All of these variations have multiple ramifications on both the individual steps and the final cost of the product. During the last 5 years the relative stability of oil prices has reduced the emphasis on alternative sources of liquid fuels such as the bioconversion *Present address: TEMBEG Inc., Temiscaming,PQ, Canada J0Z 3R0. -~Towhom correspondence should be addressed at: Chair of Forest Products Biotechnology, Faculty of Forestry, Vancouver, British Columbia, Canada V6T 1W5. 275 Bioresource "lechnology 0960-8524/91/S03.50 © 1991 Elsevier Science Publishers Ltd, England. Printed in Great Britain