Acid impregnation and steam explosion of corn stover in batch processes

https://doi.org/10.1016/j.indcrop.2007.03.005Get rights and content

Abstract

The synergistic effect of pre-impregnation by sulphuric acid and steam explosion has been investigated. Sugar recovery by water extraction and cellulose digestibility by enzymes have been considered. The acid diffusion inside a representative particle, having thickness of 0.7 mm, has been modelled on the basis of available models and diffusion coefficients. The experiments of acid impregnation have pointed out a plateau of the acid uptake after 10 min corresponding to an acid adsorption of 22 g/kg of dry stover. An experiment of acid desorption from the pre-impregnated biomass particles to a new bulk of pure water (reverse diffusion), has confirmed that only a small fraction of the uptaken acid remained free, while most has been consumed by the substrate. Nine conditions have been tested for the steam explosion treatment selecting the temperature of 180, 190, 200 °C and sulphuric acid loadings of 0, 1.5, 3 wt.%. The maximum sugar recovery by water extraction is produced by a SE treatment of 190 °C for 5 min and an acid loading of 1.5 wt.%; at these conditions 25% of the sugars in the feedstock can be recovered as monomers or oligomers. The acid has greatly improved the sugar solubility, e.g. at 180 °C the sugar recovery has been only 1.5% without acid, while by using 3 wt.% of acid the sugar recovery has increased to 16.8%. Also the cellulose digestibility has been improved by the acid pre-impregnation, after 48 h of digestion the yield of glucose reached 93% of the theoretical by using the substrate that was pre-impregnated with 3 wt.% of acid an treated at 190 °C. The high acid loading has also been required to achieve the best recovery of glucose (85% of the initial glucan) as sum of water extraction and 48 h hydrolysis.

Introduction

Corn grain is widely used to produce ethanol, it accounts for about 46% of the whole corn plant, the remaining 54% of biomass is stalk, leaf, cob and husk (Pordersimo et al., 2005, Pordersimo et al., 2004). While the grain can be processed with high yield on a commercial scale, the same it is not possible for the plant residues. The reason is that while the convertible part of the grain is starch, the rest of the plant is mainly cellulose, lignin and hemicellulose. Starch can be efficiently and cheaply hydrolysed with amylases, but cellulose and hemicellulose are more difficult to treat. In fact, not only are the carbohydrate chains poorly accessible in the lignocellulose domains, but the cellulolytic and hemicellulolytic enzymes that are currently available have an lower efficiency and higher cost than amylases (Lynd et al., 2005). Increasing the availability of all the carbohydrates would increase the yield of ethanol per unit of cultivated land. Moreover, the production of ethanol fuel from residues could avoid the diversion of human food resources (Hahn-Hägerdal et al., 2006, Pimentel, 2003). Even considering that a part of the residues should be left on field to avoid soil erosion, the economic advantage would be great (Fenton et al., 2005, Sheehan et al., 2003). The same considerations apply for any lignocellulose material, both energy crop or residue; Kim and Dale (2004) estimated that the 491 GL year−1 of ethanol that could be produced in this way could replace 353 GL of gasoline (32% of the global gasoline consumption in the world). In the case of corn stover the benefit would be immediate and of deep impact because of its world-wide cultivation. The world production of corn stover is estimated as high as 324 Mtonnes year−1 and the ethanol theoretically available with the current best available technology would be about 100 Mtonnes (FAO, 2004, Hamelinck et al., 2003). Plenty of resources have been invested in developing new enzymes and significant targets have been achieved in cost reduction and efficiency (Cherry, 2005, Hong et al., 2003, Sun and Cheng, 2002, Bruins et al., 2001). These improvements have to be joined with suitable pretreatments of the biomass because otherwise the enzymes, hampered by their high molecular weight, have difficulty reaching the carbohydrate chains that are embedded in the lignin matrix (Mosier et al., 2005). Moreover, cellulose itself consists of crystalline domains difficult to permeate and attack. Thermal treatments with aqueous solutions or saturated steam are reported to break down the solid matrix by partially hydrolysing chemical bonds. Depending on the treatment severity (i.e. temperature and time) hemicellulose can be solubilised and cellulose domains can be accessed by cellulolytic enzymes, while lignin is depolymerised and partially removed from the fibers (Sun et al., 2005a, Sun et al., 2005b, Sun et al., 2005c). Steam explosion (SE) provides a further destructing effect consisting in the defiberization caused by the sudden decompression. The high severity improves sugar solubility and enzyme digestibility, but sugar degradation can also be increased. Several chemicals have been tested to improve the yield under mild conditions, among them acids, alkalis, gases, and liquids (Ramos, 2003, Lu and Mosier, 2007). Beside the carbohydrate hydrolysis, in which they act as catalyst, the chemicals are involved in several side reactions, such as the neutralization of minerals (Lloyd and Wyman, 2004). As a result, at least 1–5 wt.% of chemicals have to be used in order to achieve significant results. Acid neutralization by ash has been hypothesised by Kim et al. (2001) to explain why, under extremely low sulphuric acid concentration of 0.07% (corresponding to 0.7 wt.% on biomass basis), the yield of cellulose hydrolysis has been much lower from poplar sawdust than from α-cellulose, while by increasing the acid loading to 1.7 wt.% they have obtained yields higher than 90%. At low severities the use of sulphuric acid provides an effective edge in recovering thermo labile sugars and improving the cellulose digestibility (Neureiter et al., 2002, Neuter et al., 2004). Dilute sulphuric acid pretreatment of corn stover has been investigated at lab and pilot scale. Acid loadings up to 16 wt.% and temperature of 170–180 °C are required to obtain xylose yields of 80%; the process also produces cellulose that can be converted into ethanol by SSF with yields up to 87% (Esteghlalian et al., 1997, Shell et al., 2003). Lloyd and Wyman (2005) reported an overall recovery up to 92.5% of the total sugars originally available in the corn stover by coupling the dilute acid pretreatment with enzymatic hydrolysis. An alternative to the dilute sulphuric acid process is the SE of pre-impregnated biomass. Steam explosion provides both energetic and chemical consumption advantages because, at the same treatment temperature, SE can be carried out with 1.5 kg of water per kg of biomass (1 kg of impregnation water plus 0.5 kg of saturated steam to heat the mixture) while dilute acid process usually require 5–10 kg of water. The low water content in a SE reactor allows the use of the same concentration of acid that is used in dilute acid process, but with proportional lower consumption. The efficiency of the pre-impregnation procedure remains to be demonstrated and some differences have been pointed out between full immersion of the biomass in chemicals or the spraying on of a liquid mixture (Martin et al., 2002). A complete particle penetration should be attained for the acid to be most effective. Sulphuric acid-catalysed SE has been investigated for woods and good results are reported (Emmel et al., 2003, Browell et al., 1985). Large particle size hinders the wood impregnation with chemicals and overnight impregnation is usually used in the experiments. Using X-ray analysis, Kazi et al. (1998) evaluated that a concentrated NaOH solution impregnated only 20% of a wood particle having radius 15 mm after 60 min at 75 °C and 0.79 MPa, while in the case of straw 10 min impregnation at 25 °C and 0.2 MPa ensured a steady state chemical concentration distribution within the straw pieces. The use of SO2 overcomes the diffusion problem and good sugar yields and enzyme digestibility have been reported in batch SE (Söderström et al., 2004, Söderström et al., 2002, Shevchenko et al., 2000, Bura et al., 2002). Practical problems of working with and storing SO2 would instead suggest the use of sulphuric acid on a production scale. However, the enhanced concentration of degradation products and sulphate salts produced after neutralization have negative effects on the hydrolysate fermentation (Söderström et al., 2005, Larsson et al., 1999); Agbogbo and Wenger (2006) reported that (NH4)2SO4 enhanced the xylose consumption and ethanol production by Pichia stipitis, while CaSO4 enhanced growth and xylitol production but produced the lowest ethanol concentration and yield. The aim of this work has been to investigate the kinetics of sulphuric acid uptake during the impregnation of corn stover and to point out the synergistic effect of the pre-impregnation and SE on the solubilisation of sugars and enzymatic hydrolysis.

Section snippets

Feedstock

Corn stover was harvested in Vicenza (North Italy) and delivered as a 400 kg bale. The feedstock consisted of a mix of stalk, leaf, cob, and husk. The stover has been divided in bulks of 20–30 kg, put in bags of synthetic fabric and stored indoors. The material was sampled following the UNI/CEN 14778 procedure and coarsely chopped into 2–3 cm pieces with industrial machinery before steam explosion. To carry out the chemical analysis the stover was ground with a mill to 0.2 mm. The ash content of

Particle size analysis

The composition of corn stover is reported in Table 1. Bulk density and particle density were, respectively, 61 and 1115 kg m−3. The sieving data of the coarsely chopped stover are reported in Fig. 1a. The longer filamentous particles were sorted on the first sieve (8 mm) because they twisted and formed coils. Coils have been found only in this first fraction, the other particles having been well separated by the sieving procedure. The sieving data was complemented with the image analysis to sort

Conclusions

The impregnation of corn stover with sulphuric acid has to be described in terms of diffusion and reaction owing to the basicity of the feedstock. The procedure of acid desorption from the pre-impregnated biomass particles to a new bulk of pure water, pointed out that only a fraction of acid was free after soaking. By fitting theoretical models it has been estimated diffusion coefficients of 1.6–2.1 × 10−5 cm2 s−1, i.e. almost one order of magnitude higher than those reported in literature. The

Acknowledgements

The work has been cofinanced by the EU (contract No. ENK6-CT-2002-00604). Mr. G. Cardinale, Mr. A. Villone, Mr. V. Valerio, Mr. F. Liuzzi, Dr. I. De Bari are acknowledged for the assistance in the experiments. Dr. Eng. G. Braccio is acknowledged for the general management.

References (60)

  • S.B. Kim et al.

    Diffusion of sulfuric acid within lignocellulosic biomass particles an its impact on dilute-acid pretreatment

    Bioresour. Technol.

    (2002)
  • T.H. Kim et al.

    Pretreatment of corn stover by aqueous ammonia

    Bioresour. Technol.

    (2003)
  • S. Larsson et al.

    The generation of fermentation inhibitors during dilute acid hydrolysis of softwood

    Enzyme Microb. Technol.

    (1999)
  • T.A. Lloyd et al.

    Combined sugar yields for dilute sulphuric acid pretreatment of corn stover followed by enzymatic hydrolysis of the remaining solids

    Bioresour. Technol.

    (2005)
  • L.R. Lynd et al.

    Consolidated bioprocessing of cellulosic biomass: an update

    Curr. Opin. Biotechnol.

    (2005)
  • N. Mosier et al.

    Features of promising technologies for pretreatment of lignocellulosic biomass

    Bioresour. Technol.

    (2005)
  • K. Öhgren et al.

    Fuel ethanol production from steam-exploded corn stover using SSF at higher dry matter content

    Biomass Bioenergy

    (2006)
  • S.M. Shevchenko et al.

    Optimisation of monosaccharide recovery by post-hydrolysis of the water soluble hemicellulose component after steam explosion of softwood chips

    Bioresour. Technol.

    (2000)
  • R.C. Sun et al.

    Identification and quantization of lipophilic extractives from whet straw

    Ind. Crops Prod.

    (2001)
  • X.F. Sun et al.

    Characteristics of degraded cellulose obtained from steam exploded wheat straw

    Carbohydrate Res.

    (2005)
  • X.F. Sun et al.

    Characteristics of degraded hemicellulosic polymers obtained from steam exploded wheat straw

    Carbohydrate Polym.

    (2005)
  • Y. Sun et al.

    Hydrolysis of lignocellulosic materials for ethanol production: a review

    Bioresour. Technol.

    (2002)
  • F. Teymouri et al.

    Optimization of the ammonia fiber explosion (AFEX) treatment parameters for enzymatic hydrolysis of corn stover

    Bioresour. Technol.

    (2005)
  • C.H. Wyman et al.

    Comparative sugar recovery data from laboratory scale application of leading pretreatment technologies to corn stover

    Bioresour. Technol.

    (2005)
  • F.K. Agbogbo et al.

    Effect of pre-treatment chemicals on xylose fermentation by Pichia stipitis

    Biotechnol. Lett.

    (2006)
  • K. Belkacemi et al.

    Phenomenological Kinetics of complex systems: Mechanistic considerations in the solubilization of hemicellulose following aqueous/steam treatments

    Ind. Eng. Chem. Res.

    (1991)
  • H.H. Browell et al.

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

    Biotechnol. Bioeng.

    (1985)
  • M.E. Bruins et al.

    Thermozymes and their applications

    Appl. Biochem. Biotechnol.

    (2001)
  • R. Bura et al.

    SO2-Catalysed steam explosion of corn fiber for ethanol production

    Appl. Biochem. Biotechnol.

    (2002)
  • J.R. Cherry

    Progress on enzymes for biomass utillization and prospects for the future

  • Cited by (0)

    View full text