- Email: [email protected]
Improvement of ethanol fermentation from lignocellulosic hydrolysates by the removal of inhibitors
Accepted Manuscript Title: Improvement of ethanol fermentation from lignocellulosic hydrolysates by the removal of inhibitors Author: Hong-Joo Lee Woo-Seok Lim Jae-Won Lee PII: DOI: Reference:
S1226-086X(13)00117-2 http://dx.doi.org/doi:10.1016/j.jiec.2013.03.014 JIEC 1279
To appear in: Received date: Revised date: Accepted date:
3-12-2012 26-2-2013 12-3-2013
Please cite this article as: H.-J. Lee, W.-S. Lim, J.-W. Lee, Improvement of ethanol fermentation from lignocellulosic hydrolysates by the removal of inhibitors, Journal of Industrial and Engineering Chemistry (2013), http://dx.doi.org/10.1016/j.jiec.2013.03.014 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Improvement of ethanol fermentation from lignocellulosic hydrolysates by the removal of inhibitors
Department of Bioenergy Science and Technology, College of Agriculture & Life Sciences,
cr
a
ip t
Hong-Joo Leea, Woo-Seok Limb, Jae-Won Leeb,c*
b
us
Chonnam National University, Gwang-ju, 500-757, Republic of Korea
Department of Forest Products and Technology, College of Agriculture & Life Sciences,
c
an
Chonnam National University, Gwang-ju, 500-757, Republic of Korea
Bioenergy Research Center, Chonnam National University, Gwangju 500-757, Republic of
Ac ce p
te
d
M
Korea
*Corresponding author
E-mail address: [email protected]
1
Page 1 of 27
Abstract
In this study, the removal efficiency of fermentation inhibitors in a lignocellulosic
ip t
hydrolysate by electrodialysis (ED) and the ethanol performance of ED-treated hydrolysate were investigated. The fermentable sugars and inhibitors concentrations in the hydrolysate
cr
differed significantly depending on the kind of biomass and acid catalysts. In the mixed
us
hardwood, acetic acid and furfural in the hydrolysate was high as 8.41-8.57 g/l and 2.68-4.23 g/l, respectively, but 5-hydroxymethylfurfural (HMF) concentration was relatively low
an
compared with that of mixed softwood. The ED process showed the high effectiveness for removing acetic acid and total phenolic compounds in the hydrolysate without loss of
M
fermentable sugars. However, most of the HMF and furfural remained in the hydrolysate after ED. Ethanol fermentation was not completed in untreated and mixed hardwood ED-
d
treated hydrolysates due to the high concentration of furfural. Meanwhile, ethanol
te
fermentation was successfully performed in a mixed softwood ED-treated hydrolysate
Ac ce p
pretreated with dicarboxylic acid. The maximum ethanol concentration attained after fermentation with an initial fermentable sugar level of 27.78 g/L was 10.12 g/L after 48 h.
Keywords: Fermentation inhibitors; Lignocellulosic hydrolysate; Electrodialysis; Ethanol production
2
Page 2 of 27
1. Introduction
Lignocellulosic biomass has received attention as a resource for biofuels because it is
ip t
renewable and abundant. Moreover, lignocellulosic biomass does not compete with a food source; therefore, it has great potential as a substrate for biofuel production [1]. In particular,
cr
research on bioethanol production from lignocellulosic biomass is being carried out for its
us
direct use in converted car engines or for anhydrous ethanol to add to gasoline as a fuel enhancer [2].
an
The bioconversion process to produce ethanol from a lignocellulosic biomass consists of three steps, including pretreatment, enzymatic hydrolysis, and fermentation. Pretreatment of a
M
lignocellulosic biomass is required to improve bioconversion. Pretreatment is the process by which the biomass surface area is opened up for the subsequent enzymatic attack. The fibrous
d
structure of the biomass is destroyed as a result of pretreatment. Various pretreatment
te
processes have been suggested, including chemical, physical, physicochemical, and
Ac ce p
biological [3-5]. However, most pretreatment processes induce fermentation inhibitors in the lignocellulosic hydrolysate. These fermentation inhibitors are degradation products of sugars and lignin that may have a potential inhibitory effect on microorganisms during fermentation. Fermentation inhibitors in lignocellulosic hydrolysates are widely known as aliphatic acids (acetic, formic and levulinic acid), furaldehydes (furfural and 5-hydrolsymethylfurfural), aromatic compounds (phenolics), and extractives [6,7]. As generating inhibitors is unavoidable during acid pretreatment, a number of technologies have been employed to remove fermentation inhibitors from lignocellulosic hydrolysate. The techniques include lime treatment, ion exchange resin, organic solvent extraction, and adding adsorbents such as activated carbon and zeolite [8-10]. In particular, treating ion exchange 3
Page 3 of 27
resins effectively removes weak acidic inhibitors and can lead to an inhibitor-free hydrolysate, but it has shortcomings such as the release of exchanged ions and the periodical chemical regeneration of ion exchange resins [11].
ip t
Electrodialysis (ED) is one of ion exchange membrane process using an electrical potential as a driving force. Its system typically consists of a cell arrangement with a series of alternating
cr
anion and cation exchange membranes between an anode and a cathode to form individual
us
cells having a volume with two adjacent membranes [12]. ED has been widely applied to bioseparation processes to separate organic acids such as lactic acid, citric acid, acetic acid,
an
and their salts including conventional applications to mineralize water, desalinate saline solutions, produce table salt, and treat wastewater [13,14]. However, membrane fouling,
M
which takes place due to deposition of organics on the membrane surface, is one of the most significant considerations in ED. In this system, organics including organic acid,
d
furaldehydes, and TPC are considered as foulants. Therefore, we evaluated the removal
te
efficiency of fermentation inhibitors in a lignocellulosic hydrolysate by ED and evaluated the
Ac ce p
ethanol performance of the ED-treated hydrolysate.
2. Experimental
2.1 Biomass and pretreatment
Mixed hardwood (Quercus mongolica, Robinia pseudoacacia L, and Castanea crenata) and softwood (Pinus rigida and Pinus densiflora) chips were purchased from Poong Lim Inc. (Daejeon, Korea). The biomass was milled and screened to a size of 40-60 mesh using a JNCM Wiley mill (JISICO, Seoul, Korea) and stored at 4°C at < 10% moisture content. Oxalic, 4
Page 4 of 27
maleic, and sulfuric acid were each used as acid catalyst for pretreatment. The pretreatment was conducted in 500 mL cylindrical stainless steel reaction vessels. Each biomass portion and acid solutions were placed in stainless steel vessels that were placed into
ip t
a larger tumbling digester, heated to the reaction temperature, and then rotated to keep the liquor in contact with the material during pretreatment. Each vessel was loaded with 50 g (dry
cr
weight basis) of biomass and sufficient acid/water mixture to give a total solid/liquid ratio of
us
1:4 (w/w). Pretreatment was performed at 170°C for 60 min with each acid catalyst solution
an
of pH 1.34, corresponding to combined severity factor (CSF) of 2.5.
M
2.2 Sugar and inhibitor analysis in the hydrolysate
The concentrations of fermentable sugars in the hydrolysate were determined using an HPLC
d
(Waters 2695 system, MA, USA) outfitted with an Aminex HPX-87P column (Bio-Rad,
te
Hercules, CA, USA). The concentrations of inhibitors such as acetic acid, furfural, and 5-
Ac ce p
hydroxymethylfurfural (HMF) in the hydrolysate were determined with an Aminex HPX-87H column (Bio-Rad, Hercules, CA, USA) and a refractive index detector (Waters 2414 system, MA, USA). All samples were properly diluted and filtered through a 0.45 μm spin-filter to remove particles before analysis.
Total phenolic compounds (TPC) were estimated colorimetrically by the Folin-Ciocalteu method [15]. Inductively coupled plasma analysis of metal ions was carried out using a Nexion 300X instrument (Perkin Elmer, MA, USA). Total organic carbon (TOC) was analyzed with a TOC Muti N/C (Analytic Jena AG, Munster, Germany).
2.3. Removal of fermentation inhibitors from hydrolysate by electrodialysis 5
Page 5 of 27
ED experiments were performed using a hydrolysate obtained from following acid pretreatment. Ten pairs of cell structures were assembled in a CJ-S3 ED stack with a total
ip t
membrane effective area of 550 cm2 (Changjo Techno, Seoul, Korea). A commercial cation exchange membrane, NEOSEPTA® CMX, and an anion exchange membrane, NEOSEPTA®
cr
AMX (ASTOM Corp., Tokyo, Japan) were used to prepare of the stack. The hydrolysate was
us
fed through the membrane module in the diluate as an initial feed and the treated hydrolysate in the diluate was fermented to produce ethanol. Desalting experiments with NaCl were
an
carried out after chemical cleaning with distilled water, NaOH solution, and distilled water between the ED experiments of hydrolysate to observe the fouling effect on ED process
te
2.4 Fermentation of hydrolysate
d
M
performance. The ED-treated hydrolysate was analyzed by HPLC as described in section 2.2.
Ac ce p
Pichia stipitis CBS 6054 was used for the fermentation. Yeast cells were grown in 1000 mL Erlenmeyer flasks containing 400 mL of YPD (10 g/L yeast extract, 10 g/L peptone, and 20 g/L glucose) in a shaking incubator for 24 h, after which the cell cultures were harvested and washed with sterile deionized water. The hydrolysate was adjusted to pH 6.0 with sodium hydroxide. Washed cells at a dry cell weight of 2.0 g/L were transferred to hydrolysate with 5 g/L yeast, 5 g/L urea, 0.5 g/L MgSO4 7H2O and 1 g/L KH2PO4. Fermentation was performed at 30°C in an orbital shaker at 150 rpm. Samples were taken at 24, 48, 72, and 96 h, and the amount of monosaccharides remaining and the ethanol produced were analyzed by HPLC as described in section 2.2.
6
Page 6 of 27
3. Results and discussion
ip t
3.1 Sugars and inhibitors in the hydrolysate following acid pretreatment
The compositional analysis of the hydrolysate obtained from each pretreatment is shown in
cr
Tables 1 and 2. Xylose was the most abundant sugar in the mixed hardwood hydrolysate of
us
the pretreatment conditions studied, whereas glucose released from cellulose was relatively low. The mixed softwood hydrolysate mainly contained mannose and glucose. The sugars
an
and inhibitors concentrations differed significantly depending on the kind of biomass and acid catalyst.
M
In the mixed hardwood, the highest fermentable sugar reached 35.27 g/L, which was obtained from the oxalic acid pretreatment. Fermentable sugar production was low at 23.61 g/L when
d
sulfuric acid was used as the catalyst. Dicarboxylic acids exhibited a high catalytic efficiency
te
for lignocellulosic biomass hydrolysis with low sugar degradation rate than that of sulfuric
Ac ce p
acid. This result agreed with previous reports and is explained by the typical characteristic of dicarboxylic acid pretreatment for a lignocellulosic biomass [16-18]. Acetic acid derived from hemicelluloses reached > 8 g/L under all pretreatment conditions. The major component of hardwood hemicelluloses is glucuronoxylan. Most xylose residues of glucuronoxylan contain an acetyl group at C-2 or C-3, which gives xylose/acetyl residues a ratio of 7:10 [19]. Therefore, a high concentration of acetic acid was obtained in the mixed hardwood hydrolysis. HMF and furfural concentrations in the hydrolysate were slightly higher than that of dicarboxylic acid when pretreatment was performed with sulfuric acid. This result supports the suggestion made in an earlier study that oxalic acid does not catalyze sugar degradation [20]. 7
Page 7 of 27
Dicarboxylic acid pretreatment of mixed softwood provided a high concentration of fermentable sugars in the hydrolysate, which was very similar to that of mixed hardwood. However, the acetic acid concentration was low compare to that for the mixed hardwood. The
ip t
reason is that the composition and structure of the hemicelluloses in softwood differ from those in hardwood [21]. In contrast with mixed hardwood, the HMF concentration in
cr
hydrolysate was relatively higher than the furfural concentration. This was due to differences
us
in the hydrolysis rate of producing diastereomers such as glucose, galactose, and mannose. Galactoglucomannans are the principal hemicelluloses in softwood and their monomeric
an
components are galactose, glucose, and mannan. Of these, galactose and mannose are more easily degraded to HMF than glucose due to differences in hydrolysis rate [21]. Therefore, a
M
high concentration of HMF generated from hexose was measured in the hydrolysate. The TOC concentration in a hydrolysate represents the degree of hydrolysis efficiency of the
d
lignocellulosic biomass by the acid catalyst. TOC was high in the dicarboxylic acid
Ac ce p
in the hydrolysate.
te
pretreatment regardless of biomass type. This result was similar to that of fermentable sugars
The influence of acid catalysts on the dissolution of metal ions during pretreatment was significantly different. In general, metal ions are dissolved from the reactor surface by acid catalysts during pretreatment. Furthermore, the lignocellulosic biomass provides calcium and magnesium. The Fe and Ni concentration induced from the pretreatment reactor were the highest in the hydrolysate pretreated with sulfuric acid. In particular, Fe concentration was > 500 mg/L in all biomass type, suggesting that sulfuric acid is a source of large amounts of corrosion in the reactor, which is a problem when using sulfuric acid.
(Table 1) 8
Page 8 of 27
(Table 2)
ip t
3.2. Electrodialysis experiments of hydrolysates
cr
ED experiments of the hydrolysates were carried out to remove the fermentation inhibitors.
us
The removal efficiency of acetic acid as a representative inhibitor is shown in Fig. 1. Acetic acid was removed from the mixed hardwood hydrolysates at the highest rate when oxalic acid
an
was used as the catalyst.
Table 3 summarizes the ED performance in the mixed hardwood and softwood hydrolysates
M
in terms of the power consumption to 1 g acetic acid, the flux of acetic acid through ion the exchange membrane, and the current efficiency of acetic acid transport. Among the
d
hydrolysates pretreated with acid catalysts, the oxalic acid pretreatment showed the lowest
te
power consumption and the highest acetic acid flux and current efficiency. ED performance
Ac ce p
of the mixed hardwood was higher than that of mixed softwood. The lower performance of the mixed softwood hydrolysates may be related to a higher concentration of HMF, a nonionizable inhibitor.
Fouling occurs due to the deposition of organics (foulants) on the membrane surface and can be classified as reversible or irreversible according to the interactions between the foulants and the membrane surfaces. In the case of reversible fouling, the foulant can be removed by cleaning and performance recovers. Foulant that cannot be removed by cleaning and therefore performance cannot easily be recovered are irreversible [22,23]. Of chemical species in Table 1, HMF, furfural and TPC may give fouling effect by the deposition on the membrane surface, decreasing ED performance. Table 4 shows desalting 9
Page 9 of 27
experiments with NaCl carried out after chemical cleaning. The change in ED performance was investigated in terms of transport rate, current efficiency, and power consumption. As shown in Table 4, ED performance did not decrease even after the hydrolysate experiments.
ip t
The deposition of organics in the hydrolysates did not have a significant effect on ED performance. Because HMF, furfural, and TPC are hydrophobic, they have low adsorption
cr
properties on the hydrophilic ion exchange membrane surface [22]. In addition, chemical
us
cleaning with water and NaOH removed deposited chemical on the membrane surface as
an
reversible fouling.
M
(Fig. 1)
te
Ac ce p
(Table 4)
d
(Table 3)
3.3. Removal of inhibitors from the hydrolysates
Acetic, lactic, levulinic, malonic, and formic acids are the by-products of acid pretreatment released from hemicelluloses. Delgenes (1996) suggested that > 5 g/L acetic acid inhibits fermentation because it is capable of penetrating the cell walls of microorganisms and hindering cell activity by acidifying the cytoplasm and disrupting the protein gradient across the cell membrane [24,25]. Therefore, removing the acetic acid from a mixed hardwood hydrolysat is required before ethanol fermentation due to its high concentration. In addition, most microorganisms are almost completely inhibited by furfural and HMF concentrations of 10
Page 10 of 27
2-4 g/L [6,26]. The concentration of heavy metals such as Ni and Cr above a certain level dissolved in the fermentation media might also have affected the physiology of the microorganisms and impeded cell activity.
ip t
The ED process was effective for removing the fermentation inhibitors (Figs. 2 and 3), but the fermentable sugar concentrations were unaffected. Therefore, most of the fermentable
cr
sugars remained in the hydrolysate (data not shown). This result is similar to previous reports
us
[27,28]. Most of the acetic acid is removed due to its ionic properties [29,30]. TPC was removed with an efficiency of > 50% under all pretreatment conditions. It is assumed that the
an
removal of non-ionizable hydrophobic inhibitors is related to their rejection from the membrane surface, as ion exchange membrane surfaces have hydrophilic properties [31].
M
However, most of the HMF and furfural, which are also non-ionizable hydrophobic inhibitors, remained in the hydrolysate after ED, showing low removal efficiency for all experiments.
d
Fig. 3 shows the removal efficiency of metal ions by ED. Fe, Mg, and Ni were significantly
te
removed by ED, corresponding to a removal efficiency of > 95%. The different removal
Ac ce p
efficiencies of metal may be related to the complex formation with organic acid, their charge properties, and adsorption on the membrane surface. In particular, 10 mg/L Ni inhibits ethanol production and hence decreases specific growth rate of cells, because nickel ions bind to the cell surface and inhibits the uptake of D-glucose in yeast [32]. From these results, the hydrolysate obtained by ED could provide a suitable condition for ethanol fermentation, as it contained high concentration of fermentable sugars and low concentration of fermentation inhibitors.
(Fig. 2)
11
Page 11 of 27
(Fig. 3)
ip t
3.3. Effect of inhibitors on fermentation
Ethanol fermentation was not completed by using the untreated and ED-treated hydrolysate
cr
of mixed hardwood. In the softwood hydrolysate without ED treatment, ethanol fermentation
us
was not successfully performed due to fermentation inhibitors. Ethanol production by P. stipitis was clearly inhibited by the high concentration of inhibitors present in the untreated
an
hydrolysate. Although the hydrolysate was treated by ED to remove the fermentation inhibitors, most of the HMF and furfural remained in the treated hydrolysate. However,
M
ethanol fermentation was successfully performed in dicarboxylic acid pretreated ED-treated hydrolysate of mixed softwood (Table 5). The ethanol production and consumption of
d
fermentable sugars by P. stipitis using dicarboxylic acid pretreated ED-treated hydrolysates
te
were compared, and the results are shown in Fig. 4. The maximum ethanol concentration
Ac ce p
attained after fermentation with an initial fermentable sugar level of 27.78 g/L was 10.12 g/L after 48 h in the ED-treated hydrolysate obtained following oxalic acid pretreatment. The hydrolysate contained 2.06 g/L HMF, 0.41 g/L furfural, and 0.86 g/L TPC. The ethanol yield was relatively low, and fermentable sugars were slowly consumed by P. stipitis during fermentation in the maleic acid pretreated ED-treated hydrolysate compared with that of the oxalic acid pretreatment. The highest ethanol production was 9.36 g/L after 72 h, which corresponded to an ethanol production yield of 0.33 g/g and an ethanol volumetric productivity of 0.13 g/L.h. Compared to the oxalic acid pretreated ED-treated hydrolysate, this low ethanol production rate was caused by the synergistic effect of the inhibitors that remained in the hydrolysate [11.33]. The hydrolysate contained 1.80 g/L HMF, 0.62 g/L 12
Page 12 of 27
furfural and 1.18 g/L TPC, respectively. As a result, furfural and TPC were more significant factors for ethanol fermentation than
ip t
other inhibitors. This result was similar to previous reports [6,26].
cr
(Table 5)
us
(Fig. 4)
an
4. Conclusions
M
ED process removed acetic acid and TPC with high efficiency from hydrolysates regardless of their initial concentrations. In particular, the removal efficiency of acetic acid from the
d
hydrolysate was 100%. In contrast, neither furfural nor 5-hydroxymethylfurfural (HMF) were
te
efficiently removed. Nevertheless, the ED treatment significantly increased fermentation
Ac ce p
performance in terms of ethanol yield and ethanol production in a mixed softwood EDtreated hydrolysate pretreated with dicarboxylic acid. Additional removal of selected fermentation inhibitors such as furfural is required to improve fermentation performance, as furfural influences ethanol fermentation more than other inhibitors.
Acknowledgements
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012-0008177) and by Priority Research Centers Program through the National 13
Page 13 of 27
Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2010-0020141).
ip t
References
us
challenges. Annu. Rev. Eng. Environ. 24 (1999) 189-226.
cr
[1] C.E. Wyman, Biomass ethanol: technical progress, opportunities, and commercial
[2] J.Y. Zhu, X.J. Pan, Woody biomass pretreatment for cellulosic ethanol production:
an
Technology and energy consumption evaluation. Bioresour. Technol. 101 (2010) 49925002.
M
[3] T.A. Lloyd, C.E. Wyman, Combined sugar yields for dilute sulfuric acid pretreatment of corn stover followed by enzymatic hydrolysis of the remaining solids. Bioresour. Technol.
d
96 (2005) 1967-1977.
te
[4] X. Pan, C. Arato, N. Gilkes, D. Gregg, W. Mabee, K. Pye, Z. Xiao, X. Zhang, J.N. Saddler,
Ac ce p
Biorefining of softwoods using ethanol olganosolve pulping: reliminary evaluation of process streams for manufacture of fuel-grade ethanol and co-products. Biotechnol. Bioeng. 90 (2005) 473-481.
[5] A. Ferraz, J. Rodriguez, J. Free, J. Baeza, Biodegradation of Pinus radiate softwood by white- and brown-rot fungi. World J. Microbiol. Biotechnol. 17 (2001) 31-34.
[6] C. Bellido, S. Bolado, M. Coca, S. Lucas, G. Gonzalez-Benito, M.T. Garcia-Cubero, Effect of inhibitors formed during wheat straw pretreatment on ethanol fermentation by Pichia stipitis. Bioresour. Technol. 102 (2011) 10868-10874. [7] C. Martin, L.J. Jönsson, Comparison of the resistance of industrial and laboratory strains of Saccharomyces and Zygosaccharomyces to lignocellulosic derived fermentation 14
Page 14 of 27
inhibitors. Enzyme Microbiol. Technol. 32 (2003) 386-395. [8] L. Olsson, B. Hahn-Hägerdal, Fermentation of lignocellulosic hydrolysates for ethanol production. Enzyme Microbial. Technol. 18 (1996) 312-331.
ip t
[9] M. Ali, R. Mark, J.S. Daniel, Conditioning hemicelluloses hydrolysates for fermentation: effect of overliming pH on sugar and ethanol yields. Process Biochem. 41 (2006) 1806-
cr
1811.
us
[10] R. Ranjan, S. Thust, C.E. Gounaris, M. Woo, C.A. Floudas, M.V. Keitz, K.J. Valentas, J. Wei, M. Tsapatsis, Adsorption of fermentation inhibitors from lignocellulosic biomass
an
hydrolysates for improved ethanol yield and value-added product recovery. Micropor. Mesopor. Mater. 99 (2009) 251-260.
M
[11] S. Larsson, E. Palmqvist, B. Hahn-Hägerdal, C. Tengborg, K. Stenberg, G. Zacchi, N. Nilvebrant, The generation of fermentation inhibitors during dilute acid hydrolysis of
d
softwood. Enzyme Microbial. Technol. 24 (1999) 151-159.
te
[12] E.G. Lee, S.H. Moon, Y.K. Chang, I. Yoo, H.N. Chang, Lactic acid recovery using two-
Ac ce p
stage electrodialysis and its modeling. J. Membr. Sci. 145 (1998) 53-66. [13] C. Huang, T. Xu, Y. Zhang, Y. Xue, G. Chen, Application of electrodialysis to the production of organic acids: State-of-the-art and recent developments. J Membr. Sci. 288 (2007) 1–12.
[14] Z. Wang, Y. Luo, P. Yu, Recovery of organic acids from waste salt solutions derived from the manufacture of cyclohexanone by electrodialysis. J. Membr. Sci. 280 (2006) 134-137. [15] A. Scalbert, B. Monties, G. Janin, Tannins in wood: comparison of different estimation methods. J. Agric. Food Chem. 37 (1989) 1324-1329. [16] J.W. Lee, T.W. Jeffries, Efficiencies of acid catalysts in the hydrolysis of lignocellulosic 15
Page 15 of 27
biomass over a range of combined severity factors. Bioresour. Technol. 102 (2011) 5884-5890.
acids for cellulose hydrolysis. Biotechnol. Prog. 17 (2001) 474-480.
ip t
[17] N.S. Mosier, A. Sarikaya, C.M. Ladisch, M.R. Ladisch, Characterization of dicarboxylic
[18] Y. Lu, N.S. Mosier, Biomimetic catalysis for hemicelluloses hydrolysis in corn stover.
cr
Biotechnol. Prog. 23 (2007) 116-123.
us
[19] D. Fengel, G. Wegener, Wood: Chemistry, Ultrastructure. Walter de Gruyter, Berlin, 1984. [20] A.M. Kootstra, H.H. Beeftink, E.L. Scott, J.P.M. Sanders, Comparison of dilute mineral
an
and organic acid pretreatment for enzymatic hydrolysis of wheat straw. Biochem. Eng. J. 46 (2009) 126-131.
M
[21] E. Sjöström, Wood chemistry: fundamentals and application. Academic Press, New York, 1981.
d
[22] H.J. Lee, J.H. Choi, J. Cho, S.H. Moon, Characterization of anion exchange membranes
te
fouled with humate in electrodialysis, J. Membr. Sci. 203 (2002) 115-126.
Ac ce p
[23] H.J. Lee, S.J. Oh, S.H. Moon, Removal of hardness in fermentation broth by electrodialysis. J. Chem. Technol. Biotechnol. 77 (2002) 1005-1012.
[24] J. Delgenes, R. Moletta, J. Navarro, Effects of lignocellulose degradation products on ethanol fermentations of glucose and xylose by Saccharomyces cerevisiae, Zymomonas
mobilis, Pichia stipitis, and Candida shehatae. Enzyme Microbial. Technol. 19 (1996) 220-225.
[25] C. Takahashi, D. Takahashi, M. Carvalhal, F. Alterthum, Effects of acetate on the growth and fermentation performance of Escherichia coli KO11. Appl. Biochem. Biotechnol. 81 (1999) 193-203. [26] M.J. Diaz, E. Ruiz, I. Romero, C. Cara, M. Moya, E. Castro, Inhibition of Pichia stipitis 16
Page 16 of 27
fermentation of hydrolysates from olive tree cuttings. World J. Microbiol. Biotechnol. 25 (2009) 891-899. [27] S. Walton, A.v. Heiningen, P.v. Walsum, Inhibition effects on fermentation of hardwood
ip t
extracted hemicelluloses by acetic acid and sodium. Bioresour. Technol. 101 (2010) 1935-1940.
cr
[28] Y.H. Weng, H.J. Wei, T.Y. Tsai, T.H. Lin, T.Y. Wei, G.L. Guo, C.P. Huang, Separation of
us
furans and carboxylic acids from sugars in dilute acid rice straw hydrolyzates by nanofiltration, Bioresour. Technol. 101 (2010) 4889-4894.
an
[29] K.K. Cheng, J.A. Zhang, D.H. Liu, Y. Sun, M.D. Yang, J.M. Xu, Production of 1,3propanediol by Klebsiella pneumonia from glycerol broth. Biochem. Lett. 28 (2006)
M
1817-1821.
[30] E.G. Lee, S.H. Moon, Y.K. Chang, I. Yoo, H.N. Chang, Lactic acid recovery using two-
d
stage electrodialysis and its modeling. J. Membr. Sci. 145 (1998) 53-66.
te
[31] H.J. Lee, S.H. Moon, S.P. Tsai, Effects of pulsed electric fields on membrane fouling in
Ac ce p
electrodialysis of NaCl solution containing humate, Sep. Purif. Technol. 27 (2002) 89-95. [32] N. Watson, B. Prior, P. Lategan, M. Lussi, Factors in acid treated bagasse inhibiting ethanol production from D-xylose by Pachysolen tannophilus. Enzyme Microb. Technol.
6 (1984) 451-456.
[33] E. Palmqvist, B. Hahn-Hägerdal, Fermentation of lignocellulosic hydrolysates. II: inhibitors and mechanisms of inhibition. Bioresour. Technol. 74 (2000) 25-33.
17
Page 17 of 27
ip t cr us
Table 1. Sugar, inhibitors and total organic carbon (TOC) in hydrolysate during acid pretreatment of lignocellulosic biomass Sugar (g/L)
Mixed softwood
Inhibitor (g/L)
TOC (g/L)
Glucose
Xylose
Mannose
Acetic acid
HMF
Furfural
TPC
Oxalic acid Maleic acid Sulfuric acid
10.10
25.17
NA
8.41
1.31
2.68
5.04
19.54
12.71 8.12
18.92 15.49
NA NA
8.57 8.52
1.37 1.46
3.99 4.23
7.98 5.51
25.65 17.65
Oxalic acid Maleic acid
9.31
6.38
M an
Mixed hardwood
Acid catalyst
15.10
1.84
2.24
0.47
2.80
16.82
12.41
5.22
12.18
1.60
1.82
0.90
4.40
22.11
Sulfuric acid
7.99
1.44
3.82
1.84
5.42
1.26
4.05
13.89
ed
Biomass
Ac
ce pt
*NA is data not available *Pretreatment was performed at 170°C for 60 min with each acid catalyst solution of pH 1.34, corresponding to combined severity factor (CSF) of 2.5.
18
Page 18 of 27
ip t Mixed softwood
Ca
Cr
Cu
Oxalic acid Maleic acid
0.67
NAa
130.10
1.69
135.47
Sulfuric acid
1.84
164.59 a
Oxalic acid Maleic acid
2.79 2.83
NA 48.49
Sulfuric acid
2.86
79.84
Mg
Ni
Zn
0.25
39.04
22.70
3.50
1.06
10.58
0.18
17.24
49.57
10.06
0.58
94.64
1.28
511.51
54.43
103.54
2.02
138.83 5.11
1.76 0.25
55.66 15.12
22.95 33.23
3.60 6.35
0.79 3.20
79.72
3.83
566.59
32.85
118.17
1.78
ce pt
ed
NA is data not available.
Fe
Ac
a
Al
us
Mixed hardwood
Acid catalyst
M an
Biomass
cr
Table 2. Metal ion concentration in hydrolysate during pretreatment (unit: mg/L)
19
Page 19 of 27
ip t cr
100% degradation (min) 81 90 60 40 110 60
Ac
ce pt
ed
M an
us
Table 3. Process performances of acetic acid in electrodialytic treatment of the hydrolysates Power consumption Acetic acid flux Current efficiency Biomass Acid catalyst (wh/g) (g/m2h) (%) Oxalic acid 0.27 112.0 83.6 Mixed Maleic acid 0.45 92.7 49.4 hardwood Sulfuric acid 0.31 143.8 65.8 Oxalic acid 0.40 48.9 43.1 Mixed Maleic acid 1.15 15.5 10.9 softwood Sulfuric acid 0.86 33.5 19.8
20
Page 20 of 27
ip t cr
Ac
ce pt
ed
M an
us
Table 4. Electordialysis (ED) performances of NaCl desalting experiments Before ED treatments of After ED treatments of hydrolysates hydrolysates 2 NaCl transport rate (mol/m h) 0.89 0.95 Current efficiency of NaCl (%) 94.0 96.2 Power consumption (wh/mol NaCl) 25.4 24.3
21
Page 21 of 27
ip t cr
Ac
ce pt
ed
M an
us
Table 5. Ethanol fermentation performance of electrodialysis-treated hydrolysate by P. stipitis Mixed softwood Mixed softwood Mixed softwood (oxalic acid) (maleic acid) (sulfuric acid)c a b Ethanol production (g/L) 10.12 9.36 Ethanol productivity (g/L.h) 0.21 0.13 Ethanol yield (g/g sugar) 0.36 0.33 Theoretical yield of ethanol (%) 71.43 64.15 a The highest ethanol production was after fermentation 48 h b The highest ethanol production was after fermentation 72 h c Ethanol fermentation was not successfully performed in sulfuric acid pretreated hydrolysate.
22
Page 22 of 27
Figure legends
ip t
Fig. 1. Removal of acetic acid in a lignocellulosic biomass hydrolysate by electrodialysis
cr
(top: mixed hardwood, bottom: mixed softwood)
us
Fig. 2. Removal of inhibitors in a lignocellulosic biomass hydrolysate by electrodialysis (top: mixed hardwood, bottom: mixed softwood, OA: oxalic acid, MA: maleic acid, SA:
an
sulfuric acid, ED: electrodialysis).
M
Fig. 3. Removal of metal ions in a lignocellulosic biomass hydrolysate by electrodialysis
te
SA: sulfuric acid).
d
(top: mixed hardwood, bottom: mixed softwood, OA: oxalic acid, MA: maleic acid,
Ac ce p
Fig. 4. Fermentable sugar consumption and ethanol production in electrodialysis-treated mixed softwood hydrolysate (OA: oxalic acid, MA: maleic acid).
23
Page 23 of 27
(Fig. 1)
10
cr
6
4
us
Acetic acid (g/L)
8
ip t
Oxalic acid Maleic acid Sulfuric acid
an
2
0 0
2000
4000
6000
8000
10000
12000
14000
d
M
Amount of current (Coloumb)
te
2.5
Oxalic acid Maleic acid Sulfuric acid
Ac ce p
Acetic acid (g/L)
2.0
1.5
1.0
0.5
0.0 0
2000
4000
6000
8000
10000
12000
14000
Accumulated current (Coloumb)
24
Page 24 of 27
Ac ce p
te
d
M
an
us
cr
ip t
(Fig. 2)
25
Page 25 of 27
Ac ce p
te
d
M
an
us
cr
ip t
(Fig. 3)
26
Page 26 of 27
Ac ce p
te
d
M
an
us
cr
ip t
(Fig. 4)
27
Page 27 of 27
S1226-086X(13)00117-2 http://dx.doi.org/doi:10.1016/j.jiec.2013.03.014 JIEC 1279
To appear in: Received date: Revised date: Accepted date:
3-12-2012 26-2-2013 12-3-2013
Please cite this article as: H.-J. Lee, W.-S. Lim, J.-W. Lee, Improvement of ethanol fermentation from lignocellulosic hydrolysates by the removal of inhibitors, Journal of Industrial and Engineering Chemistry (2013), http://dx.doi.org/10.1016/j.jiec.2013.03.014 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Improvement of ethanol fermentation from lignocellulosic hydrolysates by the removal of inhibitors
Department of Bioenergy Science and Technology, College of Agriculture & Life Sciences,
cr
a
ip t
Hong-Joo Leea, Woo-Seok Limb, Jae-Won Leeb,c*
b
us
Chonnam National University, Gwang-ju, 500-757, Republic of Korea
Department of Forest Products and Technology, College of Agriculture & Life Sciences,
c
an
Chonnam National University, Gwang-ju, 500-757, Republic of Korea
Bioenergy Research Center, Chonnam National University, Gwangju 500-757, Republic of
Ac ce p
te
d
M
Korea
*Corresponding author
E-mail address: [email protected]
1
Page 1 of 27
Abstract
In this study, the removal efficiency of fermentation inhibitors in a lignocellulosic
ip t
hydrolysate by electrodialysis (ED) and the ethanol performance of ED-treated hydrolysate were investigated. The fermentable sugars and inhibitors concentrations in the hydrolysate
cr
differed significantly depending on the kind of biomass and acid catalysts. In the mixed
us
hardwood, acetic acid and furfural in the hydrolysate was high as 8.41-8.57 g/l and 2.68-4.23 g/l, respectively, but 5-hydroxymethylfurfural (HMF) concentration was relatively low
an
compared with that of mixed softwood. The ED process showed the high effectiveness for removing acetic acid and total phenolic compounds in the hydrolysate without loss of
M
fermentable sugars. However, most of the HMF and furfural remained in the hydrolysate after ED. Ethanol fermentation was not completed in untreated and mixed hardwood ED-
d
treated hydrolysates due to the high concentration of furfural. Meanwhile, ethanol
te
fermentation was successfully performed in a mixed softwood ED-treated hydrolysate
Ac ce p
pretreated with dicarboxylic acid. The maximum ethanol concentration attained after fermentation with an initial fermentable sugar level of 27.78 g/L was 10.12 g/L after 48 h.
Keywords: Fermentation inhibitors; Lignocellulosic hydrolysate; Electrodialysis; Ethanol production
2
Page 2 of 27
1. Introduction
Lignocellulosic biomass has received attention as a resource for biofuels because it is
ip t
renewable and abundant. Moreover, lignocellulosic biomass does not compete with a food source; therefore, it has great potential as a substrate for biofuel production [1]. In particular,
cr
research on bioethanol production from lignocellulosic biomass is being carried out for its
us
direct use in converted car engines or for anhydrous ethanol to add to gasoline as a fuel enhancer [2].
an
The bioconversion process to produce ethanol from a lignocellulosic biomass consists of three steps, including pretreatment, enzymatic hydrolysis, and fermentation. Pretreatment of a
M
lignocellulosic biomass is required to improve bioconversion. Pretreatment is the process by which the biomass surface area is opened up for the subsequent enzymatic attack. The fibrous
d
structure of the biomass is destroyed as a result of pretreatment. Various pretreatment
te
processes have been suggested, including chemical, physical, physicochemical, and
Ac ce p
biological [3-5]. However, most pretreatment processes induce fermentation inhibitors in the lignocellulosic hydrolysate. These fermentation inhibitors are degradation products of sugars and lignin that may have a potential inhibitory effect on microorganisms during fermentation. Fermentation inhibitors in lignocellulosic hydrolysates are widely known as aliphatic acids (acetic, formic and levulinic acid), furaldehydes (furfural and 5-hydrolsymethylfurfural), aromatic compounds (phenolics), and extractives [6,7]. As generating inhibitors is unavoidable during acid pretreatment, a number of technologies have been employed to remove fermentation inhibitors from lignocellulosic hydrolysate. The techniques include lime treatment, ion exchange resin, organic solvent extraction, and adding adsorbents such as activated carbon and zeolite [8-10]. In particular, treating ion exchange 3
Page 3 of 27
resins effectively removes weak acidic inhibitors and can lead to an inhibitor-free hydrolysate, but it has shortcomings such as the release of exchanged ions and the periodical chemical regeneration of ion exchange resins [11].
ip t
Electrodialysis (ED) is one of ion exchange membrane process using an electrical potential as a driving force. Its system typically consists of a cell arrangement with a series of alternating
cr
anion and cation exchange membranes between an anode and a cathode to form individual
us
cells having a volume with two adjacent membranes [12]. ED has been widely applied to bioseparation processes to separate organic acids such as lactic acid, citric acid, acetic acid,
an
and their salts including conventional applications to mineralize water, desalinate saline solutions, produce table salt, and treat wastewater [13,14]. However, membrane fouling,
M
which takes place due to deposition of organics on the membrane surface, is one of the most significant considerations in ED. In this system, organics including organic acid,
d
furaldehydes, and TPC are considered as foulants. Therefore, we evaluated the removal
te
efficiency of fermentation inhibitors in a lignocellulosic hydrolysate by ED and evaluated the
Ac ce p
ethanol performance of the ED-treated hydrolysate.
2. Experimental
2.1 Biomass and pretreatment
Mixed hardwood (Quercus mongolica, Robinia pseudoacacia L, and Castanea crenata) and softwood (Pinus rigida and Pinus densiflora) chips were purchased from Poong Lim Inc. (Daejeon, Korea). The biomass was milled and screened to a size of 40-60 mesh using a JNCM Wiley mill (JISICO, Seoul, Korea) and stored at 4°C at < 10% moisture content. Oxalic, 4
Page 4 of 27
maleic, and sulfuric acid were each used as acid catalyst for pretreatment. The pretreatment was conducted in 500 mL cylindrical stainless steel reaction vessels. Each biomass portion and acid solutions were placed in stainless steel vessels that were placed into
ip t
a larger tumbling digester, heated to the reaction temperature, and then rotated to keep the liquor in contact with the material during pretreatment. Each vessel was loaded with 50 g (dry
cr
weight basis) of biomass and sufficient acid/water mixture to give a total solid/liquid ratio of
us
1:4 (w/w). Pretreatment was performed at 170°C for 60 min with each acid catalyst solution
an
of pH 1.34, corresponding to combined severity factor (CSF) of 2.5.
M
2.2 Sugar and inhibitor analysis in the hydrolysate
The concentrations of fermentable sugars in the hydrolysate were determined using an HPLC
d
(Waters 2695 system, MA, USA) outfitted with an Aminex HPX-87P column (Bio-Rad,
te
Hercules, CA, USA). The concentrations of inhibitors such as acetic acid, furfural, and 5-
Ac ce p
hydroxymethylfurfural (HMF) in the hydrolysate were determined with an Aminex HPX-87H column (Bio-Rad, Hercules, CA, USA) and a refractive index detector (Waters 2414 system, MA, USA). All samples were properly diluted and filtered through a 0.45 μm spin-filter to remove particles before analysis.
Total phenolic compounds (TPC) were estimated colorimetrically by the Folin-Ciocalteu method [15]. Inductively coupled plasma analysis of metal ions was carried out using a Nexion 300X instrument (Perkin Elmer, MA, USA). Total organic carbon (TOC) was analyzed with a TOC Muti N/C (Analytic Jena AG, Munster, Germany).
2.3. Removal of fermentation inhibitors from hydrolysate by electrodialysis 5
Page 5 of 27
ED experiments were performed using a hydrolysate obtained from following acid pretreatment. Ten pairs of cell structures were assembled in a CJ-S3 ED stack with a total
ip t
membrane effective area of 550 cm2 (Changjo Techno, Seoul, Korea). A commercial cation exchange membrane, NEOSEPTA® CMX, and an anion exchange membrane, NEOSEPTA®
cr
AMX (ASTOM Corp., Tokyo, Japan) were used to prepare of the stack. The hydrolysate was
us
fed through the membrane module in the diluate as an initial feed and the treated hydrolysate in the diluate was fermented to produce ethanol. Desalting experiments with NaCl were
an
carried out after chemical cleaning with distilled water, NaOH solution, and distilled water between the ED experiments of hydrolysate to observe the fouling effect on ED process
te
2.4 Fermentation of hydrolysate
d
M
performance. The ED-treated hydrolysate was analyzed by HPLC as described in section 2.2.
Ac ce p
Pichia stipitis CBS 6054 was used for the fermentation. Yeast cells were grown in 1000 mL Erlenmeyer flasks containing 400 mL of YPD (10 g/L yeast extract, 10 g/L peptone, and 20 g/L glucose) in a shaking incubator for 24 h, after which the cell cultures were harvested and washed with sterile deionized water. The hydrolysate was adjusted to pH 6.0 with sodium hydroxide. Washed cells at a dry cell weight of 2.0 g/L were transferred to hydrolysate with 5 g/L yeast, 5 g/L urea, 0.5 g/L MgSO4 7H2O and 1 g/L KH2PO4. Fermentation was performed at 30°C in an orbital shaker at 150 rpm. Samples were taken at 24, 48, 72, and 96 h, and the amount of monosaccharides remaining and the ethanol produced were analyzed by HPLC as described in section 2.2.
6
Page 6 of 27
3. Results and discussion
ip t
3.1 Sugars and inhibitors in the hydrolysate following acid pretreatment
The compositional analysis of the hydrolysate obtained from each pretreatment is shown in
cr
Tables 1 and 2. Xylose was the most abundant sugar in the mixed hardwood hydrolysate of
us
the pretreatment conditions studied, whereas glucose released from cellulose was relatively low. The mixed softwood hydrolysate mainly contained mannose and glucose. The sugars
an
and inhibitors concentrations differed significantly depending on the kind of biomass and acid catalyst.
M
In the mixed hardwood, the highest fermentable sugar reached 35.27 g/L, which was obtained from the oxalic acid pretreatment. Fermentable sugar production was low at 23.61 g/L when
d
sulfuric acid was used as the catalyst. Dicarboxylic acids exhibited a high catalytic efficiency
te
for lignocellulosic biomass hydrolysis with low sugar degradation rate than that of sulfuric
Ac ce p
acid. This result agreed with previous reports and is explained by the typical characteristic of dicarboxylic acid pretreatment for a lignocellulosic biomass [16-18]. Acetic acid derived from hemicelluloses reached > 8 g/L under all pretreatment conditions. The major component of hardwood hemicelluloses is glucuronoxylan. Most xylose residues of glucuronoxylan contain an acetyl group at C-2 or C-3, which gives xylose/acetyl residues a ratio of 7:10 [19]. Therefore, a high concentration of acetic acid was obtained in the mixed hardwood hydrolysis. HMF and furfural concentrations in the hydrolysate were slightly higher than that of dicarboxylic acid when pretreatment was performed with sulfuric acid. This result supports the suggestion made in an earlier study that oxalic acid does not catalyze sugar degradation [20]. 7
Page 7 of 27
Dicarboxylic acid pretreatment of mixed softwood provided a high concentration of fermentable sugars in the hydrolysate, which was very similar to that of mixed hardwood. However, the acetic acid concentration was low compare to that for the mixed hardwood. The
ip t
reason is that the composition and structure of the hemicelluloses in softwood differ from those in hardwood [21]. In contrast with mixed hardwood, the HMF concentration in
cr
hydrolysate was relatively higher than the furfural concentration. This was due to differences
us
in the hydrolysis rate of producing diastereomers such as glucose, galactose, and mannose. Galactoglucomannans are the principal hemicelluloses in softwood and their monomeric
an
components are galactose, glucose, and mannan. Of these, galactose and mannose are more easily degraded to HMF than glucose due to differences in hydrolysis rate [21]. Therefore, a
M
high concentration of HMF generated from hexose was measured in the hydrolysate. The TOC concentration in a hydrolysate represents the degree of hydrolysis efficiency of the
d
lignocellulosic biomass by the acid catalyst. TOC was high in the dicarboxylic acid
Ac ce p
in the hydrolysate.
te
pretreatment regardless of biomass type. This result was similar to that of fermentable sugars
The influence of acid catalysts on the dissolution of metal ions during pretreatment was significantly different. In general, metal ions are dissolved from the reactor surface by acid catalysts during pretreatment. Furthermore, the lignocellulosic biomass provides calcium and magnesium. The Fe and Ni concentration induced from the pretreatment reactor were the highest in the hydrolysate pretreated with sulfuric acid. In particular, Fe concentration was > 500 mg/L in all biomass type, suggesting that sulfuric acid is a source of large amounts of corrosion in the reactor, which is a problem when using sulfuric acid.
(Table 1) 8
Page 8 of 27
(Table 2)
ip t
3.2. Electrodialysis experiments of hydrolysates
cr
ED experiments of the hydrolysates were carried out to remove the fermentation inhibitors.
us
The removal efficiency of acetic acid as a representative inhibitor is shown in Fig. 1. Acetic acid was removed from the mixed hardwood hydrolysates at the highest rate when oxalic acid
an
was used as the catalyst.
Table 3 summarizes the ED performance in the mixed hardwood and softwood hydrolysates
M
in terms of the power consumption to 1 g acetic acid, the flux of acetic acid through ion the exchange membrane, and the current efficiency of acetic acid transport. Among the
d
hydrolysates pretreated with acid catalysts, the oxalic acid pretreatment showed the lowest
te
power consumption and the highest acetic acid flux and current efficiency. ED performance
Ac ce p
of the mixed hardwood was higher than that of mixed softwood. The lower performance of the mixed softwood hydrolysates may be related to a higher concentration of HMF, a nonionizable inhibitor.
Fouling occurs due to the deposition of organics (foulants) on the membrane surface and can be classified as reversible or irreversible according to the interactions between the foulants and the membrane surfaces. In the case of reversible fouling, the foulant can be removed by cleaning and performance recovers. Foulant that cannot be removed by cleaning and therefore performance cannot easily be recovered are irreversible [22,23]. Of chemical species in Table 1, HMF, furfural and TPC may give fouling effect by the deposition on the membrane surface, decreasing ED performance. Table 4 shows desalting 9
Page 9 of 27
experiments with NaCl carried out after chemical cleaning. The change in ED performance was investigated in terms of transport rate, current efficiency, and power consumption. As shown in Table 4, ED performance did not decrease even after the hydrolysate experiments.
ip t
The deposition of organics in the hydrolysates did not have a significant effect on ED performance. Because HMF, furfural, and TPC are hydrophobic, they have low adsorption
cr
properties on the hydrophilic ion exchange membrane surface [22]. In addition, chemical
us
cleaning with water and NaOH removed deposited chemical on the membrane surface as
an
reversible fouling.
M
(Fig. 1)
te
Ac ce p
(Table 4)
d
(Table 3)
3.3. Removal of inhibitors from the hydrolysates
Acetic, lactic, levulinic, malonic, and formic acids are the by-products of acid pretreatment released from hemicelluloses. Delgenes (1996) suggested that > 5 g/L acetic acid inhibits fermentation because it is capable of penetrating the cell walls of microorganisms and hindering cell activity by acidifying the cytoplasm and disrupting the protein gradient across the cell membrane [24,25]. Therefore, removing the acetic acid from a mixed hardwood hydrolysat is required before ethanol fermentation due to its high concentration. In addition, most microorganisms are almost completely inhibited by furfural and HMF concentrations of 10
Page 10 of 27
2-4 g/L [6,26]. The concentration of heavy metals such as Ni and Cr above a certain level dissolved in the fermentation media might also have affected the physiology of the microorganisms and impeded cell activity.
ip t
The ED process was effective for removing the fermentation inhibitors (Figs. 2 and 3), but the fermentable sugar concentrations were unaffected. Therefore, most of the fermentable
cr
sugars remained in the hydrolysate (data not shown). This result is similar to previous reports
us
[27,28]. Most of the acetic acid is removed due to its ionic properties [29,30]. TPC was removed with an efficiency of > 50% under all pretreatment conditions. It is assumed that the
an
removal of non-ionizable hydrophobic inhibitors is related to their rejection from the membrane surface, as ion exchange membrane surfaces have hydrophilic properties [31].
M
However, most of the HMF and furfural, which are also non-ionizable hydrophobic inhibitors, remained in the hydrolysate after ED, showing low removal efficiency for all experiments.
d
Fig. 3 shows the removal efficiency of metal ions by ED. Fe, Mg, and Ni were significantly
te
removed by ED, corresponding to a removal efficiency of > 95%. The different removal
Ac ce p
efficiencies of metal may be related to the complex formation with organic acid, their charge properties, and adsorption on the membrane surface. In particular, 10 mg/L Ni inhibits ethanol production and hence decreases specific growth rate of cells, because nickel ions bind to the cell surface and inhibits the uptake of D-glucose in yeast [32]. From these results, the hydrolysate obtained by ED could provide a suitable condition for ethanol fermentation, as it contained high concentration of fermentable sugars and low concentration of fermentation inhibitors.
(Fig. 2)
11
Page 11 of 27
(Fig. 3)
ip t
3.3. Effect of inhibitors on fermentation
Ethanol fermentation was not completed by using the untreated and ED-treated hydrolysate
cr
of mixed hardwood. In the softwood hydrolysate without ED treatment, ethanol fermentation
us
was not successfully performed due to fermentation inhibitors. Ethanol production by P. stipitis was clearly inhibited by the high concentration of inhibitors present in the untreated
an
hydrolysate. Although the hydrolysate was treated by ED to remove the fermentation inhibitors, most of the HMF and furfural remained in the treated hydrolysate. However,
M
ethanol fermentation was successfully performed in dicarboxylic acid pretreated ED-treated hydrolysate of mixed softwood (Table 5). The ethanol production and consumption of
d
fermentable sugars by P. stipitis using dicarboxylic acid pretreated ED-treated hydrolysates
te
were compared, and the results are shown in Fig. 4. The maximum ethanol concentration
Ac ce p
attained after fermentation with an initial fermentable sugar level of 27.78 g/L was 10.12 g/L after 48 h in the ED-treated hydrolysate obtained following oxalic acid pretreatment. The hydrolysate contained 2.06 g/L HMF, 0.41 g/L furfural, and 0.86 g/L TPC. The ethanol yield was relatively low, and fermentable sugars were slowly consumed by P. stipitis during fermentation in the maleic acid pretreated ED-treated hydrolysate compared with that of the oxalic acid pretreatment. The highest ethanol production was 9.36 g/L after 72 h, which corresponded to an ethanol production yield of 0.33 g/g and an ethanol volumetric productivity of 0.13 g/L.h. Compared to the oxalic acid pretreated ED-treated hydrolysate, this low ethanol production rate was caused by the synergistic effect of the inhibitors that remained in the hydrolysate [11.33]. The hydrolysate contained 1.80 g/L HMF, 0.62 g/L 12
Page 12 of 27
furfural and 1.18 g/L TPC, respectively. As a result, furfural and TPC were more significant factors for ethanol fermentation than
ip t
other inhibitors. This result was similar to previous reports [6,26].
cr
(Table 5)
us
(Fig. 4)
an
4. Conclusions
M
ED process removed acetic acid and TPC with high efficiency from hydrolysates regardless of their initial concentrations. In particular, the removal efficiency of acetic acid from the
d
hydrolysate was 100%. In contrast, neither furfural nor 5-hydroxymethylfurfural (HMF) were
te
efficiently removed. Nevertheless, the ED treatment significantly increased fermentation
Ac ce p
performance in terms of ethanol yield and ethanol production in a mixed softwood EDtreated hydrolysate pretreated with dicarboxylic acid. Additional removal of selected fermentation inhibitors such as furfural is required to improve fermentation performance, as furfural influences ethanol fermentation more than other inhibitors.
Acknowledgements
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012-0008177) and by Priority Research Centers Program through the National 13
Page 13 of 27
Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2010-0020141).
ip t
References
us
challenges. Annu. Rev. Eng. Environ. 24 (1999) 189-226.
cr
[1] C.E. Wyman, Biomass ethanol: technical progress, opportunities, and commercial
[2] J.Y. Zhu, X.J. Pan, Woody biomass pretreatment for cellulosic ethanol production:
an
Technology and energy consumption evaluation. Bioresour. Technol. 101 (2010) 49925002.
M
[3] T.A. Lloyd, C.E. Wyman, Combined sugar yields for dilute sulfuric acid pretreatment of corn stover followed by enzymatic hydrolysis of the remaining solids. Bioresour. Technol.
d
96 (2005) 1967-1977.
te
[4] X. Pan, C. Arato, N. Gilkes, D. Gregg, W. Mabee, K. Pye, Z. Xiao, X. Zhang, J.N. Saddler,
Ac ce p
Biorefining of softwoods using ethanol olganosolve pulping: reliminary evaluation of process streams for manufacture of fuel-grade ethanol and co-products. Biotechnol. Bioeng. 90 (2005) 473-481.
[5] A. Ferraz, J. Rodriguez, J. Free, J. Baeza, Biodegradation of Pinus radiate softwood by white- and brown-rot fungi. World J. Microbiol. Biotechnol. 17 (2001) 31-34.
[6] C. Bellido, S. Bolado, M. Coca, S. Lucas, G. Gonzalez-Benito, M.T. Garcia-Cubero, Effect of inhibitors formed during wheat straw pretreatment on ethanol fermentation by Pichia stipitis. Bioresour. Technol. 102 (2011) 10868-10874. [7] C. Martin, L.J. Jönsson, Comparison of the resistance of industrial and laboratory strains of Saccharomyces and Zygosaccharomyces to lignocellulosic derived fermentation 14
Page 14 of 27
inhibitors. Enzyme Microbiol. Technol. 32 (2003) 386-395. [8] L. Olsson, B. Hahn-Hägerdal, Fermentation of lignocellulosic hydrolysates for ethanol production. Enzyme Microbial. Technol. 18 (1996) 312-331.
ip t
[9] M. Ali, R. Mark, J.S. Daniel, Conditioning hemicelluloses hydrolysates for fermentation: effect of overliming pH on sugar and ethanol yields. Process Biochem. 41 (2006) 1806-
cr
1811.
us
[10] R. Ranjan, S. Thust, C.E. Gounaris, M. Woo, C.A. Floudas, M.V. Keitz, K.J. Valentas, J. Wei, M. Tsapatsis, Adsorption of fermentation inhibitors from lignocellulosic biomass
an
hydrolysates for improved ethanol yield and value-added product recovery. Micropor. Mesopor. Mater. 99 (2009) 251-260.
M
[11] S. Larsson, E. Palmqvist, B. Hahn-Hägerdal, C. Tengborg, K. Stenberg, G. Zacchi, N. Nilvebrant, The generation of fermentation inhibitors during dilute acid hydrolysis of
d
softwood. Enzyme Microbial. Technol. 24 (1999) 151-159.
te
[12] E.G. Lee, S.H. Moon, Y.K. Chang, I. Yoo, H.N. Chang, Lactic acid recovery using two-
Ac ce p
stage electrodialysis and its modeling. J. Membr. Sci. 145 (1998) 53-66. [13] C. Huang, T. Xu, Y. Zhang, Y. Xue, G. Chen, Application of electrodialysis to the production of organic acids: State-of-the-art and recent developments. J Membr. Sci. 288 (2007) 1–12.
[14] Z. Wang, Y. Luo, P. Yu, Recovery of organic acids from waste salt solutions derived from the manufacture of cyclohexanone by electrodialysis. J. Membr. Sci. 280 (2006) 134-137. [15] A. Scalbert, B. Monties, G. Janin, Tannins in wood: comparison of different estimation methods. J. Agric. Food Chem. 37 (1989) 1324-1329. [16] J.W. Lee, T.W. Jeffries, Efficiencies of acid catalysts in the hydrolysis of lignocellulosic 15
Page 15 of 27
biomass over a range of combined severity factors. Bioresour. Technol. 102 (2011) 5884-5890.
acids for cellulose hydrolysis. Biotechnol. Prog. 17 (2001) 474-480.
ip t
[17] N.S. Mosier, A. Sarikaya, C.M. Ladisch, M.R. Ladisch, Characterization of dicarboxylic
[18] Y. Lu, N.S. Mosier, Biomimetic catalysis for hemicelluloses hydrolysis in corn stover.
cr
Biotechnol. Prog. 23 (2007) 116-123.
us
[19] D. Fengel, G. Wegener, Wood: Chemistry, Ultrastructure. Walter de Gruyter, Berlin, 1984. [20] A.M. Kootstra, H.H. Beeftink, E.L. Scott, J.P.M. Sanders, Comparison of dilute mineral
an
and organic acid pretreatment for enzymatic hydrolysis of wheat straw. Biochem. Eng. J. 46 (2009) 126-131.
M
[21] E. Sjöström, Wood chemistry: fundamentals and application. Academic Press, New York, 1981.
d
[22] H.J. Lee, J.H. Choi, J. Cho, S.H. Moon, Characterization of anion exchange membranes
te
fouled with humate in electrodialysis, J. Membr. Sci. 203 (2002) 115-126.
Ac ce p
[23] H.J. Lee, S.J. Oh, S.H. Moon, Removal of hardness in fermentation broth by electrodialysis. J. Chem. Technol. Biotechnol. 77 (2002) 1005-1012.
[24] J. Delgenes, R. Moletta, J. Navarro, Effects of lignocellulose degradation products on ethanol fermentations of glucose and xylose by Saccharomyces cerevisiae, Zymomonas
mobilis, Pichia stipitis, and Candida shehatae. Enzyme Microbial. Technol. 19 (1996) 220-225.
[25] C. Takahashi, D. Takahashi, M. Carvalhal, F. Alterthum, Effects of acetate on the growth and fermentation performance of Escherichia coli KO11. Appl. Biochem. Biotechnol. 81 (1999) 193-203. [26] M.J. Diaz, E. Ruiz, I. Romero, C. Cara, M. Moya, E. Castro, Inhibition of Pichia stipitis 16
Page 16 of 27
fermentation of hydrolysates from olive tree cuttings. World J. Microbiol. Biotechnol. 25 (2009) 891-899. [27] S. Walton, A.v. Heiningen, P.v. Walsum, Inhibition effects on fermentation of hardwood
ip t
extracted hemicelluloses by acetic acid and sodium. Bioresour. Technol. 101 (2010) 1935-1940.
cr
[28] Y.H. Weng, H.J. Wei, T.Y. Tsai, T.H. Lin, T.Y. Wei, G.L. Guo, C.P. Huang, Separation of
us
furans and carboxylic acids from sugars in dilute acid rice straw hydrolyzates by nanofiltration, Bioresour. Technol. 101 (2010) 4889-4894.
an
[29] K.K. Cheng, J.A. Zhang, D.H. Liu, Y. Sun, M.D. Yang, J.M. Xu, Production of 1,3propanediol by Klebsiella pneumonia from glycerol broth. Biochem. Lett. 28 (2006)
M
1817-1821.
[30] E.G. Lee, S.H. Moon, Y.K. Chang, I. Yoo, H.N. Chang, Lactic acid recovery using two-
d
stage electrodialysis and its modeling. J. Membr. Sci. 145 (1998) 53-66.
te
[31] H.J. Lee, S.H. Moon, S.P. Tsai, Effects of pulsed electric fields on membrane fouling in
Ac ce p
electrodialysis of NaCl solution containing humate, Sep. Purif. Technol. 27 (2002) 89-95. [32] N. Watson, B. Prior, P. Lategan, M. Lussi, Factors in acid treated bagasse inhibiting ethanol production from D-xylose by Pachysolen tannophilus. Enzyme Microb. Technol.
6 (1984) 451-456.
[33] E. Palmqvist, B. Hahn-Hägerdal, Fermentation of lignocellulosic hydrolysates. II: inhibitors and mechanisms of inhibition. Bioresour. Technol. 74 (2000) 25-33.
17
Page 17 of 27
ip t cr us
Table 1. Sugar, inhibitors and total organic carbon (TOC) in hydrolysate during acid pretreatment of lignocellulosic biomass Sugar (g/L)
Mixed softwood
Inhibitor (g/L)
TOC (g/L)
Glucose
Xylose
Mannose
Acetic acid
HMF
Furfural
TPC
Oxalic acid Maleic acid Sulfuric acid
10.10
25.17
NA
8.41
1.31
2.68
5.04
19.54
12.71 8.12
18.92 15.49
NA NA
8.57 8.52
1.37 1.46
3.99 4.23
7.98 5.51
25.65 17.65
Oxalic acid Maleic acid
9.31
6.38
M an
Mixed hardwood
Acid catalyst
15.10
1.84
2.24
0.47
2.80
16.82
12.41
5.22
12.18
1.60
1.82
0.90
4.40
22.11
Sulfuric acid
7.99
1.44
3.82
1.84
5.42
1.26
4.05
13.89
ed
Biomass
Ac
ce pt
*NA is data not available *Pretreatment was performed at 170°C for 60 min with each acid catalyst solution of pH 1.34, corresponding to combined severity factor (CSF) of 2.5.
18
Page 18 of 27
ip t Mixed softwood
Ca
Cr
Cu
Oxalic acid Maleic acid
0.67
NAa
130.10
1.69
135.47
Sulfuric acid
1.84
164.59 a
Oxalic acid Maleic acid
2.79 2.83
NA 48.49
Sulfuric acid
2.86
79.84
Mg
Ni
Zn
0.25
39.04
22.70
3.50
1.06
10.58
0.18
17.24
49.57
10.06
0.58
94.64
1.28
511.51
54.43
103.54
2.02
138.83 5.11
1.76 0.25
55.66 15.12
22.95 33.23
3.60 6.35
0.79 3.20
79.72
3.83
566.59
32.85
118.17
1.78
ce pt
ed
NA is data not available.
Fe
Ac
a
Al
us
Mixed hardwood
Acid catalyst
M an
Biomass
cr
Table 2. Metal ion concentration in hydrolysate during pretreatment (unit: mg/L)
19
Page 19 of 27
ip t cr
100% degradation (min) 81 90 60 40 110 60
Ac
ce pt
ed
M an
us
Table 3. Process performances of acetic acid in electrodialytic treatment of the hydrolysates Power consumption Acetic acid flux Current efficiency Biomass Acid catalyst (wh/g) (g/m2h) (%) Oxalic acid 0.27 112.0 83.6 Mixed Maleic acid 0.45 92.7 49.4 hardwood Sulfuric acid 0.31 143.8 65.8 Oxalic acid 0.40 48.9 43.1 Mixed Maleic acid 1.15 15.5 10.9 softwood Sulfuric acid 0.86 33.5 19.8
20
Page 20 of 27
ip t cr
Ac
ce pt
ed
M an
us
Table 4. Electordialysis (ED) performances of NaCl desalting experiments Before ED treatments of After ED treatments of hydrolysates hydrolysates 2 NaCl transport rate (mol/m h) 0.89 0.95 Current efficiency of NaCl (%) 94.0 96.2 Power consumption (wh/mol NaCl) 25.4 24.3
21
Page 21 of 27
ip t cr
Ac
ce pt
ed
M an
us
Table 5. Ethanol fermentation performance of electrodialysis-treated hydrolysate by P. stipitis Mixed softwood Mixed softwood Mixed softwood (oxalic acid) (maleic acid) (sulfuric acid)c a b Ethanol production (g/L) 10.12 9.36 Ethanol productivity (g/L.h) 0.21 0.13 Ethanol yield (g/g sugar) 0.36 0.33 Theoretical yield of ethanol (%) 71.43 64.15 a The highest ethanol production was after fermentation 48 h b The highest ethanol production was after fermentation 72 h c Ethanol fermentation was not successfully performed in sulfuric acid pretreated hydrolysate.
22
Page 22 of 27
Figure legends
ip t
Fig. 1. Removal of acetic acid in a lignocellulosic biomass hydrolysate by electrodialysis
cr
(top: mixed hardwood, bottom: mixed softwood)
us
Fig. 2. Removal of inhibitors in a lignocellulosic biomass hydrolysate by electrodialysis (top: mixed hardwood, bottom: mixed softwood, OA: oxalic acid, MA: maleic acid, SA:
an
sulfuric acid, ED: electrodialysis).
M
Fig. 3. Removal of metal ions in a lignocellulosic biomass hydrolysate by electrodialysis
te
SA: sulfuric acid).
d
(top: mixed hardwood, bottom: mixed softwood, OA: oxalic acid, MA: maleic acid,
Ac ce p
Fig. 4. Fermentable sugar consumption and ethanol production in electrodialysis-treated mixed softwood hydrolysate (OA: oxalic acid, MA: maleic acid).
23
Page 23 of 27
(Fig. 1)
10
cr
6
4
us
Acetic acid (g/L)
8
ip t
Oxalic acid Maleic acid Sulfuric acid
an
2
0 0
2000
4000
6000
8000
10000
12000
14000
d
M
Amount of current (Coloumb)
te
2.5
Oxalic acid Maleic acid Sulfuric acid
Ac ce p
Acetic acid (g/L)
2.0
1.5
1.0
0.5
0.0 0
2000
4000
6000
8000
10000
12000
14000
Accumulated current (Coloumb)
24
Page 24 of 27
Ac ce p
te
d
M
an
us
cr
ip t
(Fig. 2)
25
Page 25 of 27
Ac ce p
te
d
M
an
us
cr
ip t
(Fig. 3)
26
Page 26 of 27
Ac ce p
te
d
M
an
us
cr
ip t
(Fig. 4)
27
Page 27 of 27