Elsevier

Fuel

Volume 89, Supplement 1, 1 November 2010, Pages S20-S28
Fuel

Techno-economic comparison of process technologies for biochemical ethanol production from corn stover

https://doi.org/10.1016/j.fuel.2010.01.001Get rights and content

Abstract

This techno-economic study compares several process technologies for the production of ethanol from lignocellulosic material, based on a 5- to 8-year time frame for implementation. While several previous techno-economic studies have focused on future technology benchmarks, this study examines the short-term commercial viability of biochemical ethanol production. With that goal, yields (where possible) were based on publicly available experimental data rather than projected data. Four pretreatment technologies (dilute-acid, 2-stage dilute-acid, hot water, and ammonia fiber explosion or AFEX); and three downstream process variations (pervaporation, separate 5-carbon and 6-carbon sugars fermentation, and on-site enzyme production) were included in the analysis. Each of these scenarios was modeled and economic analysis was performed for an “nth plant” (a plant with the same technologies that have been employed in previous commercial plants) to estimate the total capital investment (TCI) and product value (PV). PV is the ethanol production cost, including a 10% return on investment. Sensitivity analysis has been performed to assess the impact of process variations and economic parameters on the PV.

The dilute-acid pretreatment process has the lowest PV among all process scenarios, which is estimated to be $1.36/l of gasoline equivalent [LGE] ($5.13/gal of gasoline equivalent [GGE]). Sensitivity analysis shows that the PV is most sensitive to feedstock cost, enzyme cost, and installed equipment costs. A significant fraction of capital costs is related to producing heat and power from lignin in the biomass.

Cellulosic ethanol production has yet to be commercialized. Hence, a pioneer plant is expected to be more costly to build and operate than an nth plant. To assess the impact of technological maturity on pioneer plant cost, a cost growth analysis was performed. The estimated value of PV for the pioneer plant is substantially larger than for the nth plant. The PV for the pioneer plant model with dilute-acid pretreatment is $2.30/LGE ($8.72/GGE) for the most probable scenario, and the estimated TCI was more than double the nth plant cost.

Introduction

There is a growing interest in liquid biofuels produced from renewable biomass resources to reduce dependence on crude oil. The US demand for gasoline is about 103 billion gal (390 million m3) annually [1]; and approximately 9 billion gal (34 million m3) per year of ethanol is being produced from corn in 2008 [2]. The production of fuels from food crops may place upward pressure on the price and availability of food, and the likelihood of such a “food versus fuel” conflict may increase as the world’s population grows. As an alternative to corn grain, lignocellulosic biomass shows promise as a feedstock for bioethanol production. It is projected that by 2030, over 38 billion gal (144 million m3) per year of renewable biofuels will be consumed in the United States, with less than half coming from conventional corn-based ethanol [1]. One study estimated that it is possible to grow enough lignocellulosic biomass – in an economically feasible and environmentally sustainable manner – to produce more than 50 billion gal (189 million m3) of biofuels annually [3].

Ethanol – along with other types of biofuels such as butanol, bio-gasoline, and dimethylfuran – can be derived from lignocellulose via different reaction pathways [4], [5]. However, bioethanol research is more advanced than many competing biofuel technologies, most of which are at the early stages of development. Ethanol can be produced from lignocellulosics following two different process pathways: (i) biochemical, where chemical or enzymatic hydrolysis and subsequent microbial fermentation is applied; and (ii) thermochemical, where gasification followed either by microbial fermentation or by catalytic upgrading are applied [6]. Unlike ethanol production from starch feedstocks, lignocellulosic biomass requires more aggressive pretreatments prior to saccharification and fermentation to increase the exposure of cellulose to enzymes during enzymatic hydrolysis. Dissolved catalysts are often used during biomass pretreatment. The effectiveness of the catalyst and pretreatment conditions contributes significantly to the yield and economics of the overall process. Both acids and bases are used as catalysts, with acids resulting in significantly different product yields than bases. Sulfuric acid, sulfur dioxide, ammonia, and lime are some of the catalysts that have been studied [7]. Hot water pretreatment [8], which relies on the reduced pH of water at elevated temperatures to hydrolyze hemicelluloses and disrupt the biomass structure, can also be used.

Following pretreatment, the biomass cellulose and hemicellulose components are hydrolyzed to monosaccharides (primarily glucose and xylose) either by acids or enzymes. The sugars are then fermented by yeast or by bacteria to produce ethanol. With separate fermentation of C5 and C6 (5-carbon and 6-carbon) sugars using selective microbes, higher ethanol yields can be achieved than with co-fermentation [9], [10]. The produced ethanol is purified in distillation columns; and advanced purification technologies such as pervaporation [11], [12], and reverse osmosis [13], [14], [15] are being developed, which may reduce operating costs.

This study, which is a techno-economic analysis of biochemical ethanol production from corn stover, focuses on technologies projected to be viable within a 5- to 8-year time frame. Based on this time frame – and after considering time for design, construction, and start-up – the process would likely have to be based on experimental data available today. Initially, 35 published technologies of various liquid fuels were reviewed, and a matrix was prepared considering economics, technological maturity, environmental aspects, process performance, and technical and economic risks. Both butanol and ethanol production processes were initially included in the technology matrix. However, butanol technologies are at the lab-scale or very early pilot stage of development, and published data on butanol-producing organisms indicate low yields relative to ethanol production, so they are not included in further analysis. Seven lignocellulosic ethanol process scenarios were selected, with four involving pretreatment variations (dilute-acid, 2-stage dilute-acid, hot water, and ammonia fiber explosion or AFEX); and three involving downstream process variations (pervaporation, separate C5 and C6 fermentation, and on-site enzyme production). Fig. 1 shows a basic schematic of the cellulosic ethanol process with the model variations considered in this study listed below their respective process step.

Each of these scenarios is modeled in detail and economic analysis is performed assuming an nth plant design, meaning that the technologies used in the design have been employed in previous commercial plants and are relatively well understood. However, cellulosic ethanol production has yet to be commercialized, and a pioneer plant is expected to be significantly more expensive than an nth plant. To assess the impact of using immature technologies on the PV for a pioneer plant, the potential increase in capital cost and decreased plant performance were estimated using models developed by the RAND Corporation [16].

Section snippets

Materials and methods

A list of assumptions common to all process scenarios includes the following:

  • Plant capacity is 2000 Mg/day of dry corn stover.

  • 2007 Publicly available and experimentally validated reaction conversions and parameters are used.

  • Equipment, chemical, and labor costs indexed to 2007 dollars.

  • Process and steam generation plants depreciate in 7 and 20 years, respectively, following the modified accelerated cost recovery system (MACRS) method.

  • Project is 100% equity financed.

  • Contingency factor is 20% of total

Results and discussion

Results of the techno-economic process model and product values (PV) are presented in Table 2 and Fig. 2, respectively.

There is significant variation of ethanol yields per mass of dry feedstock among the pretreatment processes, with 2-stage dilute-acid pretreatment being the lowest (47 gal/Mg or 0.18 m3/Mg) and dilute-acid pretreatment (base case scenario) being the highest (76 gal/Mg or 0.29 m3/Mg). The installed equipment cost of the dilute-acid pretreatment scenario is $164 million, and the costs

Comparison with previous studies

The results of this study deviate considerably from a number of previous techno-economic analyses of cellulosic ethanol production. There are many contributing factors to this deviation and an explanation of the most significant of these factors is discussed here. Fig. 8 presents a plot of estimated ethanol prices from seven previous studies as a function of feedstock price. The ethanol and feedstock prices were updated to 2007 dollars using the Consumer Price Index. The solid line on the plot

Conclusions

The PV for the dilute-acid pretreatment scenario is $1.36/LGE, which is the lowest among all pretreatments and process variations. This is due primarily to the higher sugar yields – and, therefore, ethanol yields – from dilute-acid pretreatment and enzymatic hydrolysis than for the other process scenarios. The exception to this is the scenario with separate C5 and C6 sugar fermentation, which has higher ethanol yields. However, the PV is higher because of the high capital cost of extra

Acknowledgements

This project was made possible by support from ConocoPhillips Company and the National Renewable Energy Laboratory. We would like to thank the Department of Energy (DOE) Office of the Biomass Program for support and feedback. Rick Elander of NREL and Tim Eggeman of Neoterics International provided assistance with the CAFI model results. We greatly appreciate the helpful comments throughout the project from Bob Wallace of Pennsylvania State University, Ron Brown of ConocoPhillips Company, and Ed

References (39)

  • E.N. Sendich et al.

    Recent process improvements for the ammonia fiber expansion (AFEX) process and resulting reductions in minimum ethanol selling price

    Bioresour Technol

    (2008)
  • Holtberg PD, Smith KA, Mayne L, Doman L, Cymbalsky JH, Boedecker EE, et al. Annual energy outlook with projection to...
  • Renewable Fuels Association, Growing innovation: 2009 ethanol industry outlook;...
  • Perlack RD, Wright LL, Turhollow AF, Graham RL, Stokes BJ, Erbach DC. Biomass as feedstock for a bioenergy and...
  • T.C. Ezeji et al.

    Acetone–butanol–ethanol (ABE) production from concentrated substrate: reduction in substrate inhibition by fed-batch technique and product inhibition by gas stripping

    Appl Microbiol Biotechnol

    (2004)
  • G.W. Huber et al.

    Production of liquid alkanes by aqueous phase processing of biomass-derived carbohydrates

    Science

    (2005)
  • R.C. Brown
  • N. Moiser et al.

    Optimization of pH controlled hot water pretreatment of corn stover

    Bioresour Technol

    (2005)
  • Z. Kadar et al.

    Ethanol fermentation of various pretreated and hydrolyzed substrates at low initial pH

    Appl Biochem Biotechnol

    (2007)
  • Cited by (440)

    View all citing articles on Scopus

    This article is sponsored by the National Renewable Energy Laboratory and ConocoPhillips Company as part of the Supplement Techno-economic Comparison of Biomass-to-Biofuels Pathways.

    View full text