- Email: [email protected]
Use of Gelidium amansii as a promising resource for bioethanol: A practical approach for continuous dilute-acid hydrolysis and fermentation
Bioresource Technology 108 (2012) 83–88
Contents lists available at SciVerse ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Use of Gelidium amansii as a promising resource for bioethanol: A practical approach for continuous dilute-acid hydrolysis and fermentation Jeong-Hoon Park a, Ji-Yeon Hong a, Hyun Chul Jang a, Seung Geun Oh a, Sang-Hyoun Kim b, Jeong-Jun Yoon a, Yong Jin Kim a,⇑ a b
Green Process Material Research Group, Korea Institute of Industrial Technology (KITECH), Chungnam 330-825, South Korea Department of Environmental Engineering, Daegu University, Gyeongbuk 712-714, South Korea
a r t i c l e
i n f o
Article history: Received 19 October 2011 Received in revised form 11 December 2011 Accepted 12 December 2011 Available online 22 December 2011 Keywords: Red seaweed Gelidium amansii Continuous process Dilute-acid hydrolysis Inhibitor
a b s t r a c t A facile continuous method for dilute-acid hydrolysis of the representative red seaweed species, Gelidium amansii was developed and its hydrolysate was subsequently evaluated for fermentability. In the hydrolysis step, the hydrolysates obtained from a batch reactor and a continuous reactor were systematically compared based on fermentable sugar yield and inhibitor formation. There are many advantages to the continuous hydrolysis process. For example, the low melting point of the agar component in G. amansii facilitates improved raw material fluidity in the continuous reactor. In addition, the hydrolysate obtained from the continuous process delivered a high sugar and low inhibitor concentration, thereby leading to both high yield and high final ethanol titer in the fermentation process. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction It is generally accepted that the dilute-acid hydrolysis of biomass produces many undesired byproducts such as 5-HMF and organic acids that act as fermentation inhibitors (Kim et al., 2011; Mathew et al., 2011; Mussatto and Roberto, 2004). To overcome this drawback, much effort has been devoted to enzymatic saccharification, (Georgieva et al., 2008) which releases sugars without the generation of inhibitors. However the production efficacy should be also considered when it comes to the mass production of ethanol for transport fuels since the rate of enzymatic saccharification process is known as slow compared with chemical transformation. Furthermore, it is a difficult challenge to process biomass using a continuous system which allows mass hydrolysate production with a good fermentable sugar concentration and few inhibitors, especially in the case of lignocellulosic biomass with its rigid crystalline structure. Therefore, it is very important to find a substrate suitable for continuous processing which produces a high sugar and low inhibitor concentration. Seaweed (macro algae) is gaining enormous interests as a potential biomass source for producing bioethanol (Kim et al., 2010a; Yun et al., 2011). Generally, seaweed can grow faster with higher CO2 fixation ability than land plants (Luning and Pang, 2003). In addition, it can be cultivated in the vast expanse of ⇑ Corresponding author. Tel.: +82 41 589 8469; fax: +82 41 589 8580. E-mail address: [email protected] (Y.J. Kim). 0960-8524/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2011.12.065
the ocean with free sunlight and no need for nitrogen-based fertilizers (Buck and Buchholz, 2004). Furthermore, they normally do not contain lignin, implying that sugars can be obtained without any expensive lignin removal process. Seaweed is usually categorized into three groups: red algae (Rhodophyta), green algae (Chlorophyta and Charophyta), and brown algae (Phaeophyceae) (Percival, 1979). Among them, the red algae are known for high carbohydrate content with Gelidium amansii as one of the most abundantly available red seaweed species, appearing along the warm and shallow coastal area of many sub-tropical countries. The G. amansii mainly consists of polysaccharide complexes of fibrin (cellulose) and agar (galactan) whose basic monomer is glucose and galactose, respectively (Jol et al., 1999). The major repeating unit of agar is agarobiose, which is a disaccharide composed of 1,3-linked-D-galactose and 1,4-linked 3,6-anhydrous-L-galactose (Quemener and Lahaye, 1998). Accordingly, Kim et al. (2010a) reported that the dilute-acid hydrolysis of G. amansii to produce sugars can be performed in a batch-type autoclave and the main products are D-galactose, 3,6-anhydro-L-galactose (3,6-AHG), and D-glucose. Among them, the galactose and glucose are classified as fermentable mono sugars and the 3,6-AHG as non-fermentable. Since the physical morphology of agar components is softer than that of cellulose, the hydrolyzed products, galactose and 3,6AHG, are to be firstly released at relatively mild hydrolysis conditions. Furthermore, the agar residue even starts to melt above 80 °C, which provides a fairly good strategy to enhance the fluidity in the continuous process resulting from the physical change of
84
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
solid into liquid. However, the 3,6-AHG, one of the main sugar component of G. amansii, is also known as so acid-labile that it is very prone to be decomposed into 5-HMF and, subsequently, into organic acids such as levulinic acid and formic acid that act as inhibitors in the fermentation process (Quemener and Lahaye, 1998). Therefore, the dilute-acid hydrolysis reaction should be designed toward increasing the concentration of fermentable sugars and decreasing that of inhibitors. It is well known that the fermentable sugar yields and the amount of inhibitor largely depends on the three major factors, reaction temperature, acid concentration, and reaction time (or residence time for continuous process (Zhang et al., 2011)). Generally, the more severe the reaction become, the more the inhibitor are generated, and thus careful tuning the reaction conditions plays pivotal role in enhancing sugar yield with suppressing inhibitor generation in the dilute-acid hydrolysis of G. amansii containing relatively large portion of acid-labile 3,6-AHG. These observations led us to develop a continuous dilute-acid hydrolysis process using the representative red seaweed G. amansii. Meanwhile, it is reported that a modified Brettanomyces custersii KCTC 18154P showed a good performance on co-fermentability using reagent grade of mixed sugar comprising galactose and glucose as substrates to produce ethanol (Yoon et al., 2011). The results revealed that the microbe showed a good ethanol yield and high final titer without any substrate or product inhibition. Accordingly, we report herein a practical approach for the dilute-acid continuous hydrolysis of G. amansii to produce a high quality hydrolysate and a subsequent study on its fermentability to generate bioethanol using Brettanomyces custersii KCTC 18154P. 2. Methods 2.1. Raw material composition and analysis The composition of G. amansii was determined according to the National Renewable Energy Laboratory (NREL, Golden, CO) analytical methods for biomass (NREL, 1996) using a two-step acid hydrolysis procedure. The same method was applied to the analysis of glucose and galactose but 1% (w/w) sulfuric acid was employed for 3,6-AHG at the first hydrolysis step due to its easy degradation mode. The sugar component (glucose, galactose, and 3,6-AHG) of the liquid phase after acid hydrolysis was analyzed by Bio-LC (ISC-3000, Dionex, USA) equipped with a ECD and a 4 mm 250 mm CarboPac PA1 column at a flow rate of 1.0 mL/ min and 30 °C with 16 mM and 2 M NaOH as the mobile phase. 5-HMF, levulinic acid, formic acid, and ethanol were analyzed by HPLC system (YL9100 series, Korea) using a refractive index detector equipped with a 300 mm 7.8 mm Aminex HPX-87H ion exclusion column (Bio-Rad, USA) at 60 °C with 5 mM of sulfuric acid as the eluent at a flow rate of 0.5 mL/min. The ethanol yield was calculated using the equation below:
Y EtOH ¼
½EtOHmax ½Sugarini
where YEtOH = ethanol yield (g/g) (0.51 is the theoretical maximum ethanol yield), [EtOH]max = maximum ethanol titer achieved during fermentation (g/L), [Sugar]ini = total initial fermentable sugar (galactose + glucose) concentration at onset fermentation (g/L) 2.2. Continuous dilute-acid hydrolysis 190 kg of G. amansii was washed with tap-water to remove salinity and impurities, followed by a wet pulverization with water to make solid to liquid (S/L) ratio of 15% (S:L = 1:6.67). The sulfuric acid concentration was adjusted to 2% (w/w) out of the total aqueous slurry mixture. The continuous Plug and Flow reactor sys-
tem was specially designed by our research group and the structural details are described in the patent (Kim et al., 2011). A representative continuous dilute-acid hydrolysis of wet pulverized G. amansii is as follows: About 1.3 ton of wet pulverized slurry mixture containing 2% (w/w) of sulfuric acid was charged in a feeding vessel. The continuous hydrolysis reaction was carried out at 150 °C and auto-generated pressure range of 3.0–3.5 bar with a flow rate of 40 L/h which is driven by a specially modified mono pump. The product mixtures were continuously collected in a reservoir vessel and the unreacted solid residual fiber was subsequently separated by a high-capacity centrifuge, followed by neutralization of hydrolysate by CaCO3. The separated hydrolysate was concentrated in an evaporator with a negative 0.9 bars at 50 °C. Finally, the 3-fold concentrated hydrolysate was then analyzed by Bio-LC and HPLC. 2.3. Ethanol fermentation 2.3.1. Strains A wild type B. custersii KCCM11490 was irradiated with light of 245 nm at 1200 erg/mm2. The treated culture was spread over an agar plate (galactose 2% (w/v), yeast extract 1% (w/v), peptone 2% (w/v), agar 1.5% (w/v)) and cultivated for 3 days at 30 °C. Subsequently, colonies were sub-cultured to isolate the artificial mutant strain, B. custersii KTCT18154P. Also, to confirm genetic stability, the subculture was continuously carried out until the 20th generation, with galactose fermenting ability checked for each generation. B. custersii KTCT18154P cells are round to ovoid, 3–10 by 3–20 micrometers in size. They were reproduced by budding aerobically or anaerobically. The optimal growth temperature and pH were 27–30 °C and 4.8–5.5, respectively. 2.3.2. Pre-culture B. custersii KTCT18154P was inoculated on a YPD agar (yeast extract 10 g/L, bacto peptone 20 g/L, glucose 20 g/L, agar 20 g/L) and cultivated at 30 °C for 48 h. A single colony with good vegetation on the plate medium was transferred to 200 mL YSPG medium (yeast extract 10 g/L, soy peptone 20 g/L, galactose 20 g/L) and then cultured at 150 rpm and 30 °C for 24 h aerobically. Forty millilitres of the culture was centrifuged and transferred to the other 200 mL culture medium (yeast extract 10 g/L, soy peptone 20 g/L, galactose 120 g/L) and then cultured at 150 rpm and 30 °C for 16 h (OD600 18.5 at 16 h). 2.3.3. Ethanol fermentation using hydrolysate 0.68 g of yeast extract and 1.34 g of soy peptone were added to a 500-mL medium bottle. We transferred 16 mL of the pre-culture and 184 mL of the hydrolysate obtained from the continuous hydrolysis process to the medium bottle, resulting in an initial galactose and glucose concentration to 65.1 g/L and 7.1 g/L, respectively (initial OD600 1.5). Subsequently, the headspace was purged with N2 gas for 3 min. The bottle was sealed with a screw cap with a septum, and then agitated at 150 rpm and 30 °C. 2.3.4. Inhibition effect Each inhibitor was added separately to the reagent grade of mixed sugars. The series of added concentrations of 5-HMF were 5, 7.5, 10, and 12 g/L and the levulinic acid were 1, 2, 4, and 8 g/L and the formic acid were 0.1, 0.3, 0.5, 0.7, and 1.0 g/L. The fermentation experiment was performed as described in Section 2.3.2. 3. Results and discussion It is reported that G. amansii consists of polysaccharide complex of fiber and agar whose basic monomer is glucose and galactose,
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
respectively (Kim et al., 2010a). Since the physical morphology of the agar is softer than that of fiber, the optical reaction conditions for agar hydrolysis would be milder than that of fiber. Therefore, when using G. amansii as biomass, it is strongly recommended for the galactose to be firstly released from the G. amansii, followed by enzymatic saccharification using remaining cellulose-containing solid material (about 60% (w/w, dry based) of cellulose). Herein we describe a raw material composition, acid hydrolysis using batch and continuous reactor based on the formation of inhibitors, and finally comparative fermentation results using the hydrolysate from the batch and continuous method. 3.1. Raw material composition Analyses by Bio-LC and HPLC revealed that the total carbohydrate was found to be 67.3% (w/w) and among them, cellulose and agar content were found to be 14.9% (w/w) and 52.4% (w/w) based on the dried weight of G. amansii, respectively. The weight ratio of galactose to 3,6-AHG in the agar content was reported as 1:1.27 (0.44:0.56) (Kim et al., 2010a). Therefore, the maximum glucose is calculated as 0.166 [(0.149 g cellulose/g G. amansii) (180 g glucose/162 g cellulose)], galactose as 0.256 [(0.524 g agar/g G. amansii) (0.44 g galactose unit/g agar) (180 g galactose/162 g galactose unit)], and therefore the total fermentable sugar (glucose and galactose) is calculated as 0.422 (0.166 + 0.256) g/g of G. amansii, respectively. 3.2. Dilute-acid hydrolysis of G. amansii using batch reactor To investigate hydrolyzed products from agar, an isolated agar (S/L ratio = 25%, S:L = 1:4) was autoclaved in the presence of a 1.5% (w/w) sulfuric acid at 125 °C and the results depicted as a function of time are shown in Fig. 1. The isolated agar was obtained simply from the soaking of G. amansii in water charged autoclave at 121 °C for 2 h, followed by the centrifugation. The maximum galactose concentration of 93.7 g/L was calculated from (250 g agar/ L liquid) (0.375 g galactose/g agar) and 0.375 was obtained from the analytical result of the galactose portion in isolated agar (Kim et al., 2010a). Over a wide range of reaction time, the galactose concentration remains similar to give 63–65 g/L which are comparable yield of 0.68 (64.0/93.7). Conversely, the acid-labile 3,6-AHG concentration decreased very sharply, resulting in 92% loss based on the initial concentration with the concomitant increase in the concentrations
Fig. 1. Results of dilute-acid hydrolysis of agar in batch reactor as a function of time (experimental conditions: substrate = agar; S:L = 1:4; H2SO4 = 1.5% (w/w); temperature = 125 °C).
85
of 5-HMF and organic acids, which indicates a obvious transformation of 3,6-AHG into undesirable byproducts. Similarly, the G. amansii (S/L ratio = 15%, S:L = 1:6.67) was directly subjected to the same batch type diluted-acid hydrolysis in the presence of a 1.5% (w/w) sulfuric acid at 140 °C to examine the product distribution. As can be seen in Fig. 2, a similar trends with the agar were observed but more sharp decrease in 3,6AHG concentration, resulting in 94% loss was found due to the slightly severer reaction conditions than in the case of agar, which causes increasing inhibitor concentrations as a function of time. From these results, it is concluded that when using batch reactor, it is very difficult to obtain hydrolysate consisting of high sugar and low inhibitor concentration. It is well known that the fermentation inhibitors such as 5-HMF and organic acids are generated from the chemical transformation of biomass, which can stress fermentative organisms to a point beyond where the efficient utilization of sugars is reduced and product formation decreases (Mussatto and Roberto, 2004). According to Olsson and HahnHägerdal, toxic compounds can be divided in four groups: sugar degradation products, lignin degradation products, compounds derived from lignocellulose structure, and heavy metal ions (Olsson and Hahn-Hagerdal, 1996). Since the marine seaweed do not contain any lignin or its structure, only sugar degradation products could be generated through the chemical transformation. Indeed, there are found to be only the sugar degraded products such as 5-HMF, levulinic acid, formic acid, and negligible trace amount of acetic acid without the generation of lignin degraded products in the hydrolysate from the dilute-acid hydrolysis of G. amansii in the level of more than 10–20 g/L totally (see Figs. 1 and 2) using batch reactor. Thus it should be carefully considered at this moment that the concentrations of these inhibitors should be suppressed under a certain level.
3.3. Inhibitor effects and dilute-acid continuous hydrolysis process To investigate the effect of the inhibitors on ethanolic fermentation with a modified B. custersii (KCTC18154P) and to obtain an useful information on each inhibitors’ concentration limit that allow microbe to show proper fermentability, a known amount of formic acid, levulinic acid, and 5-HMF were added separately to sugars containing reagent grade of glucose (20 g/L) and galactose (100 g/L), giving total 120 g/L of mixed sugar. The inhibitory concentration limits of 5-HMF, levulinic acid, and formic acid along with their fermentabilities are summarized in Table 1.
Fig. 2. Results of dilute-acid hydrolysis of G. amansii in batch reactor as a function of time (experimental conditions: substrate = G. amansii; S:L = 1:6.67; H2SO4 = 1.5% (w/w); temperature = 140 °C).
86
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
The concentration limits described in the left column of Table 1 show that the formic acid is found to act as critical inhibitory from the level of 0.5 g/L, and even after 56 h, a small amount of galactose still remained in the mixture solution. And the critical levulinic acid concentration limit appears to be 2 g/L level, in turn, 10 g/L level for 5-HMF, suggesting that the formic acid is identified as most inhibitory among those, which is in good agreement with trends reported in the previous study (Maiorella et al., 1983). When these results are compared with the hydrolysis results obtained from batch reactor (right column of Table 1, from batch), they do not meet the end of criteria especially in the case of levulinic acid (5.7 g/L) and formic acid (2.2 g/L), and accordingly, the fermentation yield thereof, as expected, was very low (data not shown). From these, it is likely that suppressing the inhibitor concentration below the critical limit in the dilute-acid hydrolysis of G. amansii using the batch reactor is very difficult. This is presumably as result of the long induction time until reaching high temperature and the long cooling time, thereby exposing sugars (especially, 3,6-AHG) upon the harsh reaction conditions, which are playing a critical role in generating undesirable byproducts in the batch process. To solve this problem, new method using a continuous-type reactor has to be designed (Kim et al., 2010b) for reducing the induction and cooling time, leading to suppressing the inhibitor formation dramatically. Although many efforts have been devoted to the continuous type dilute-acid hydrolysis with lignocellulosic biomass, many of them were found to be unsuccessful due to the blocking problem in the reactor channel arising from the morphological hardness of wood biomass. On the contrary, owing to much softer morphology of the seaweed biomass than wood, it would be easier for G. amansii to be processed in the continuous Plug and Flow reactor, thereby giving more chance to produce a good hydrolysate composition with minimizing inhibitor formation. To correlate this postulation, two sets of separate hydrolysis reactions of G. amansii were performed using batch and continuous reactor, and their product distribution were compared systematically (Fig. 3). As shown in Fig. 3, the concentrations fermentable sugar, galactose and glucose, obtained from the continuous reactor are found to be even slightly higher than that from batch reactor, producing almost comparable yield to the theoretical galactose maximum. It is worth mentioning that the total amount of inhibitor concentration from the continuous reactor (B) is only 3.1 g/L which is much less than those from the batch reactor (10.4 g/L) (A) (See right columns of Table 1), apparently because this continuous system prevented acid-labile 3,6-AHG from being degraded into 5-HMF and subsequent transformations into organic acids. These results strongly indicate that the continuous process has the advantage of yielding fermentable sugars in good yield as well as minimizing undesirable inhibitor by controlling the residence time very effectively. 3.4. Fermentation using hydrolysate Today, the most important microoganisms applied for ethanol production are the yeast Saccharomyces cerevisiae and the bacte-
Fig. 3. Product distributional comparison between batch and continuous reactor (experimental conditions: substrate = G. amansii; S:L = 1:10; H2SO4 = 2.0% (w/w); temperature = 121 °C for (A) and 150 °C for (B); reaction time = 59 min for (A) and 4 h for(B)).
rium Zymomonas mobilis. However, both ethanol producers have a very narrow substrate range (Horn, 2000). Although the hyrolysate from G. amansii also contains different concentration of mixed sugars, e.g., galactose as major and glucose as minor, a successful utilization of this red seaweed as bio source requires that both sugars should be efficiently converted to ethanol in high yield. This might be attained using two different organisms or two-step process, however, it would be more efficient when using a single organism that can utilize both substrates simultaneously. The authors presented that the B. custersii (KCCM11490) consumed galactose much faster than S. cerevisiae in the presence of mixed sugars, and therefore, the B. custersii will be more suitable microbe for the fermentation of galactose-rich hydrolysate (Ryu et al., 2008). For comparing the fermentability using two kinds of hydrolysates from batch and continuous reactor, each hydrolysate was concentrated by vacuum evaporation, thus the volume was reduced to about 1/3 to obtain 3-fold concentrated sugar mixture. Meanwhile, the concomitant decrease in the concentration of relatively volatile 5-HMF was observed during the evaporation process, resulting in less than 1 g/L level (Larsson et al., 1999) but the concentration of organic acids did not show diminish effect. Subsequently, the concentrated hydrolysate was subjected to the fermentative process using the modified B. custersii (KCTC18154P) and the results are shown in Fig. 4 and summarized in Table 2. As can be seen in Fig. 4 and Table 2, the hydrolysate from the batch reactor does not show good performance for producing ethanol. The maximum final ethanol titer is only 11.8 g/L (yield: 0.13) even after 56 h. In contrast, the hydrolysate from continuous reactor produces ethanol more than double compared with that from the batch, resulting in the final ethanol titer of 27.6 g/L (yield: 0.38) after 39 h. This huge difference might be ascribed to the amount of inhibitors existing in the hydrolysate and indeed, the sum up of inhibitor concentration in the concentrated hydrolysate from the batch
Table 1 Comparison of inhibitor concentration limit with the results from hydrolysates.
a b
Inhibitors
Inhibitor concentration limita, g/L/EtOH, g/L/EtOH yield, g/g
Inhibitor concentrationb, g/L From batch
From continuous
Formic acid Levulinic acid 5-HMF Total
0.5/52.1/0.44 2.0/52.0/0.43 10.0/50.5/0.44 –
2.2 5.7 2.5 10.4
0.5 1.1 1.5 3.1
120 g/L (gal:glu = 5:1), reagent grade. Hydrolysate.
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
87
Fig. 4. Fermentability comparison using hydrolysates from batch and continuous (experimental conditions: B. custersii seeding = 8% (v/v); C-source = hydrolysate 90 g/L for (A) and hydrolysate 72.2 g/L for (B); N-source = yeast extract 0.68 g and soy peptone 1.34 g; agitate = 150rpm; temperature = 30 °C working volume = 200 mL).
Table 2 Fermentability comparison data of concentrated hydrolysates. Inhibitors
Formic acid Levulinic acid 5-HMF Total a b
Inhibitor concentration limit, g/L
Inhibitor concentration after evaporation, g/L/EtOH, g/L/EtOH yield, g/g From batcha
From continuousb
0.5 2.0 10.0 –
14.5/11.8/0.13 27.1/11.8/0.13 0.2/11.8/0.13 41.8
0.6/27.6/0.38 2.3/27.6/0.38 0.8/27.6/0.38 3.7
90 g/L of fermentable hydrolysate. 72.2 g/L of fermentable hydrolysate.
Fig. 5. Product distributional comparison between batch and continuous reactor (experimental conditions: B. custersii seeding = 8% (v/v); C-source = hydrolysate 90 g/L for (A) and hydrolysate 72.2 g/L for (B); N-source = yeast extract 0.68 g and soy peptone 1.34 g; agitate = 150 rpm; temperature = 30 °C working volume = 200 mL.
reactor is 41.8 g/L whereas only 3.7 g/L from the continuous reactor, indicating 1/10 times diminish effect in the amount of the inhibitor by using this continuous method (Gurram et al., 2011). Interestingly, in the fermentive process using the hydrolysate from the batch reactor, both the levulinic and formic acid concentration decreased very sharply after 32 h (Fig. 5A), which well corresponds to the moment when the glucose is consumed completely (see Fig. 4(A)). Accordingly, this result strongly suggests that the organic acids are also consumed by the microbes (Larsson et al., 1999). However, this phenomenon was less pronounced in the fermentive process of using the hydrolysate from the continuous reactor (Fig. 5B) since the consumption of fermentable sugar is more predominant than that of organic acids.
Further studies on the effective inhibitor removal methodologies for enhancing ethanol yield as well as the study on enzymatic saccharification using a residual fiber and on the anaerobic digestion using bioethanol residue from this continuous process are under investigation.
4. Conclusions There are many advantages to the continuous hydrolysis process for producing bioethanol, especially when using seaweeds as biomass. In addition to a low melting property of the agar component in the G. amansii, which is making improved raw material
88
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
fluidity in the continuous reactor, the hydrolysate obtained from this process includes high fermentable sugar and low inhibitor concentration. In the fermentive process, the hydrolysate from the continuous reactor delivered much more ethanol than that from batch reactor due to the dramatic reduce in the formation of inhibitor. Acknowledgement This work was supported by a grant (2008-N-BI08-P-01) from Korea Institute of Energy Technology Evaluation and Planning, Ministry of Knowledge Economy, Republic of Korea. References Buck, B.H., Buchholz, C.M., 2004. The offshore-ring: a new system design for the open ocean aquaculture of macroalgae. Journal of Applied Phycology 16, 355– 368. Georgieva, T.I., Hou, X., Hilstrom, T., Ahring, B.K., 2008. Enzymatic hydrolysis and ethanol fermentation of high dry matter wet-exploded wheat straw at low enzyme loading. Biotechnology for Fuels and Chemicals, 553–562. Gurram, R.N., Datta, S., Lin, Y.J., Snyder, S.W., Menkhaus, T.J., 2011. Removal of enzymatic and fermentation inhibitory compounds from biomass slurries for enhanced biorefinery process efficiencies. Bioresource Technology. 102, 7850– 7859. Horn, S.J., 2000. Bioenergy from brown seaweeds. Doctor Thesis, Department of Biotechnology, Norwegian University of Science and Technology NTNU, Trondheim. Jol, C.N., Neiss, T.G., Penninkhof, B., Rudolph, B., De Ruiter, G.A., 1999. A novel highperformance anion-exchange chromatographic method for the analysis of carrageenans and agars containing 3,6-anhydrogalactose. Analytical Biochemistry 268, 213–222. Kim, C., Ryu, H.J., Kim, S.H., Yoon, J.J., Kim, H.S., Kim, Y.J., 2010a. Acidity tunable ionic liquids as catalysts for conversion of agar into mixed sugars. Bulletin of the Korean Chemical Society Notes 31, 511–514.
Kim, Y.J., Yoon, J.J., Kim, S.H., 2010b. Korea Patent Number 10-1096183. Kim, J.W., Kim, K.S., Lee, J.S., Park, S.M., Cho, H.Y., Park, J.C., Kim, J.S., 2011. Two-stage pretreatment of rice straw using aqueous ammonia and dilute acid. Bioresource Technology. 102, 8992–8999. Larsson, S., Palmqvist, E., Hahn-Hagerdal, B., Tengborg, C., Stenberg, K., Zacchi, G., Nilvebrant, N.O., 1999. The generation of fermentation inhibitors during dilute acid hydrolysis of softwood. Enzyme and Microbial Technology 24, 151–159. Luning, K., Pang, S., 2003. Mass cultivation of seaweeds: current aspects and approaches. Journal of Applied Phycology 15, 115–119. Maiorella, B., Blanch, H.W., Wilke, C.R., 1983. By product inhibition effects on ethanolic fermentation by Saccharomyces cerevisiae. Biotechnology and Bioengineering 25, 103–121. Mathew, A.K., Chaney, K., Crook, M., Humphries, A.C., 2011. Dilute acid pretreatment of oilseed rape straw for bioethanol production. Renewable Energy 36, 2424–2432. Mussatto, S.I., Roberto, I.C., 2004. Alternatives for detoxification of diluted-acid lignocellulosic hydrolyzates for use in fermentative processes: a review. Bioresource Technology 93, 1–10. NREL, 1996. National Renewable Energy Laboratory (NREL). Chemical analysis and testing laboratory analytical procedures. NREL, Golden, CO, USA. Available from:. Olsson, L., Hahn-Hagerdal, B., 1996. Fermentation of lignocellulosic hydrolysates for ethanol production. Enzyme Microbial Technology 18, 312–331. Percival, E., 1979. The polysaccharides of green, red and brown seaweeds: their basic structure, biosynthesis and function. British Phycology Journal 14, 103– 117. Quemener, B., Lahaye, M., 1998. Comparative analysis of sulfated galactans from red algae by reductive hydrolysis and mild methanolysis coupled to two different HPLC techniques. Journal of Applied Phycology 10, 75–81. Ryu, H.-J., Kim, C., Lee, S.-G., Kim Yong Jin, Y.J.-T., Kim Kyung-Soo, Shin Myung-Kyo, 2008. Mixed sugars fermentation from hydrolyzed red seaweed with yeasts. In: Proceedings of 2008 KSIEC Spring Meeting. Yoon, J.J, Kim, S.H., Kim, Y.J., Kim, J.S., 2011. Korea Patent Number 10-1075602. Yun, E.J., Shin, M.H., Yoon, J.J., Kim, Y.J., Choi, I.G., Kim, K.H., 2011. Production of 3,6anhydro-L-galactose from agarose by agarolytic enzymes of Saccharophagus degradans 2–40. Process Biochemistry 46, 88–93. Zhang, B., Wang, L., Shahbazi, A., Diallo, O., Whitmore, A., 2011. Dilute-sulfuric acid pretreatment of cattails for cellulose conversion. Bioresource Technology 102, 9308–9312.
Contents lists available at SciVerse ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Use of Gelidium amansii as a promising resource for bioethanol: A practical approach for continuous dilute-acid hydrolysis and fermentation Jeong-Hoon Park a, Ji-Yeon Hong a, Hyun Chul Jang a, Seung Geun Oh a, Sang-Hyoun Kim b, Jeong-Jun Yoon a, Yong Jin Kim a,⇑ a b
Green Process Material Research Group, Korea Institute of Industrial Technology (KITECH), Chungnam 330-825, South Korea Department of Environmental Engineering, Daegu University, Gyeongbuk 712-714, South Korea
a r t i c l e
i n f o
Article history: Received 19 October 2011 Received in revised form 11 December 2011 Accepted 12 December 2011 Available online 22 December 2011 Keywords: Red seaweed Gelidium amansii Continuous process Dilute-acid hydrolysis Inhibitor
a b s t r a c t A facile continuous method for dilute-acid hydrolysis of the representative red seaweed species, Gelidium amansii was developed and its hydrolysate was subsequently evaluated for fermentability. In the hydrolysis step, the hydrolysates obtained from a batch reactor and a continuous reactor were systematically compared based on fermentable sugar yield and inhibitor formation. There are many advantages to the continuous hydrolysis process. For example, the low melting point of the agar component in G. amansii facilitates improved raw material fluidity in the continuous reactor. In addition, the hydrolysate obtained from the continuous process delivered a high sugar and low inhibitor concentration, thereby leading to both high yield and high final ethanol titer in the fermentation process. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction It is generally accepted that the dilute-acid hydrolysis of biomass produces many undesired byproducts such as 5-HMF and organic acids that act as fermentation inhibitors (Kim et al., 2011; Mathew et al., 2011; Mussatto and Roberto, 2004). To overcome this drawback, much effort has been devoted to enzymatic saccharification, (Georgieva et al., 2008) which releases sugars without the generation of inhibitors. However the production efficacy should be also considered when it comes to the mass production of ethanol for transport fuels since the rate of enzymatic saccharification process is known as slow compared with chemical transformation. Furthermore, it is a difficult challenge to process biomass using a continuous system which allows mass hydrolysate production with a good fermentable sugar concentration and few inhibitors, especially in the case of lignocellulosic biomass with its rigid crystalline structure. Therefore, it is very important to find a substrate suitable for continuous processing which produces a high sugar and low inhibitor concentration. Seaweed (macro algae) is gaining enormous interests as a potential biomass source for producing bioethanol (Kim et al., 2010a; Yun et al., 2011). Generally, seaweed can grow faster with higher CO2 fixation ability than land plants (Luning and Pang, 2003). In addition, it can be cultivated in the vast expanse of ⇑ Corresponding author. Tel.: +82 41 589 8469; fax: +82 41 589 8580. E-mail address: [email protected] (Y.J. Kim). 0960-8524/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2011.12.065
the ocean with free sunlight and no need for nitrogen-based fertilizers (Buck and Buchholz, 2004). Furthermore, they normally do not contain lignin, implying that sugars can be obtained without any expensive lignin removal process. Seaweed is usually categorized into three groups: red algae (Rhodophyta), green algae (Chlorophyta and Charophyta), and brown algae (Phaeophyceae) (Percival, 1979). Among them, the red algae are known for high carbohydrate content with Gelidium amansii as one of the most abundantly available red seaweed species, appearing along the warm and shallow coastal area of many sub-tropical countries. The G. amansii mainly consists of polysaccharide complexes of fibrin (cellulose) and agar (galactan) whose basic monomer is glucose and galactose, respectively (Jol et al., 1999). The major repeating unit of agar is agarobiose, which is a disaccharide composed of 1,3-linked-D-galactose and 1,4-linked 3,6-anhydrous-L-galactose (Quemener and Lahaye, 1998). Accordingly, Kim et al. (2010a) reported that the dilute-acid hydrolysis of G. amansii to produce sugars can be performed in a batch-type autoclave and the main products are D-galactose, 3,6-anhydro-L-galactose (3,6-AHG), and D-glucose. Among them, the galactose and glucose are classified as fermentable mono sugars and the 3,6-AHG as non-fermentable. Since the physical morphology of agar components is softer than that of cellulose, the hydrolyzed products, galactose and 3,6AHG, are to be firstly released at relatively mild hydrolysis conditions. Furthermore, the agar residue even starts to melt above 80 °C, which provides a fairly good strategy to enhance the fluidity in the continuous process resulting from the physical change of
84
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
solid into liquid. However, the 3,6-AHG, one of the main sugar component of G. amansii, is also known as so acid-labile that it is very prone to be decomposed into 5-HMF and, subsequently, into organic acids such as levulinic acid and formic acid that act as inhibitors in the fermentation process (Quemener and Lahaye, 1998). Therefore, the dilute-acid hydrolysis reaction should be designed toward increasing the concentration of fermentable sugars and decreasing that of inhibitors. It is well known that the fermentable sugar yields and the amount of inhibitor largely depends on the three major factors, reaction temperature, acid concentration, and reaction time (or residence time for continuous process (Zhang et al., 2011)). Generally, the more severe the reaction become, the more the inhibitor are generated, and thus careful tuning the reaction conditions plays pivotal role in enhancing sugar yield with suppressing inhibitor generation in the dilute-acid hydrolysis of G. amansii containing relatively large portion of acid-labile 3,6-AHG. These observations led us to develop a continuous dilute-acid hydrolysis process using the representative red seaweed G. amansii. Meanwhile, it is reported that a modified Brettanomyces custersii KCTC 18154P showed a good performance on co-fermentability using reagent grade of mixed sugar comprising galactose and glucose as substrates to produce ethanol (Yoon et al., 2011). The results revealed that the microbe showed a good ethanol yield and high final titer without any substrate or product inhibition. Accordingly, we report herein a practical approach for the dilute-acid continuous hydrolysis of G. amansii to produce a high quality hydrolysate and a subsequent study on its fermentability to generate bioethanol using Brettanomyces custersii KCTC 18154P. 2. Methods 2.1. Raw material composition and analysis The composition of G. amansii was determined according to the National Renewable Energy Laboratory (NREL, Golden, CO) analytical methods for biomass (NREL, 1996) using a two-step acid hydrolysis procedure. The same method was applied to the analysis of glucose and galactose but 1% (w/w) sulfuric acid was employed for 3,6-AHG at the first hydrolysis step due to its easy degradation mode. The sugar component (glucose, galactose, and 3,6-AHG) of the liquid phase after acid hydrolysis was analyzed by Bio-LC (ISC-3000, Dionex, USA) equipped with a ECD and a 4 mm 250 mm CarboPac PA1 column at a flow rate of 1.0 mL/ min and 30 °C with 16 mM and 2 M NaOH as the mobile phase. 5-HMF, levulinic acid, formic acid, and ethanol were analyzed by HPLC system (YL9100 series, Korea) using a refractive index detector equipped with a 300 mm 7.8 mm Aminex HPX-87H ion exclusion column (Bio-Rad, USA) at 60 °C with 5 mM of sulfuric acid as the eluent at a flow rate of 0.5 mL/min. The ethanol yield was calculated using the equation below:
Y EtOH ¼
½EtOHmax ½Sugarini
where YEtOH = ethanol yield (g/g) (0.51 is the theoretical maximum ethanol yield), [EtOH]max = maximum ethanol titer achieved during fermentation (g/L), [Sugar]ini = total initial fermentable sugar (galactose + glucose) concentration at onset fermentation (g/L) 2.2. Continuous dilute-acid hydrolysis 190 kg of G. amansii was washed with tap-water to remove salinity and impurities, followed by a wet pulverization with water to make solid to liquid (S/L) ratio of 15% (S:L = 1:6.67). The sulfuric acid concentration was adjusted to 2% (w/w) out of the total aqueous slurry mixture. The continuous Plug and Flow reactor sys-
tem was specially designed by our research group and the structural details are described in the patent (Kim et al., 2011). A representative continuous dilute-acid hydrolysis of wet pulverized G. amansii is as follows: About 1.3 ton of wet pulverized slurry mixture containing 2% (w/w) of sulfuric acid was charged in a feeding vessel. The continuous hydrolysis reaction was carried out at 150 °C and auto-generated pressure range of 3.0–3.5 bar with a flow rate of 40 L/h which is driven by a specially modified mono pump. The product mixtures were continuously collected in a reservoir vessel and the unreacted solid residual fiber was subsequently separated by a high-capacity centrifuge, followed by neutralization of hydrolysate by CaCO3. The separated hydrolysate was concentrated in an evaporator with a negative 0.9 bars at 50 °C. Finally, the 3-fold concentrated hydrolysate was then analyzed by Bio-LC and HPLC. 2.3. Ethanol fermentation 2.3.1. Strains A wild type B. custersii KCCM11490 was irradiated with light of 245 nm at 1200 erg/mm2. The treated culture was spread over an agar plate (galactose 2% (w/v), yeast extract 1% (w/v), peptone 2% (w/v), agar 1.5% (w/v)) and cultivated for 3 days at 30 °C. Subsequently, colonies were sub-cultured to isolate the artificial mutant strain, B. custersii KTCT18154P. Also, to confirm genetic stability, the subculture was continuously carried out until the 20th generation, with galactose fermenting ability checked for each generation. B. custersii KTCT18154P cells are round to ovoid, 3–10 by 3–20 micrometers in size. They were reproduced by budding aerobically or anaerobically. The optimal growth temperature and pH were 27–30 °C and 4.8–5.5, respectively. 2.3.2. Pre-culture B. custersii KTCT18154P was inoculated on a YPD agar (yeast extract 10 g/L, bacto peptone 20 g/L, glucose 20 g/L, agar 20 g/L) and cultivated at 30 °C for 48 h. A single colony with good vegetation on the plate medium was transferred to 200 mL YSPG medium (yeast extract 10 g/L, soy peptone 20 g/L, galactose 20 g/L) and then cultured at 150 rpm and 30 °C for 24 h aerobically. Forty millilitres of the culture was centrifuged and transferred to the other 200 mL culture medium (yeast extract 10 g/L, soy peptone 20 g/L, galactose 120 g/L) and then cultured at 150 rpm and 30 °C for 16 h (OD600 18.5 at 16 h). 2.3.3. Ethanol fermentation using hydrolysate 0.68 g of yeast extract and 1.34 g of soy peptone were added to a 500-mL medium bottle. We transferred 16 mL of the pre-culture and 184 mL of the hydrolysate obtained from the continuous hydrolysis process to the medium bottle, resulting in an initial galactose and glucose concentration to 65.1 g/L and 7.1 g/L, respectively (initial OD600 1.5). Subsequently, the headspace was purged with N2 gas for 3 min. The bottle was sealed with a screw cap with a septum, and then agitated at 150 rpm and 30 °C. 2.3.4. Inhibition effect Each inhibitor was added separately to the reagent grade of mixed sugars. The series of added concentrations of 5-HMF were 5, 7.5, 10, and 12 g/L and the levulinic acid were 1, 2, 4, and 8 g/L and the formic acid were 0.1, 0.3, 0.5, 0.7, and 1.0 g/L. The fermentation experiment was performed as described in Section 2.3.2. 3. Results and discussion It is reported that G. amansii consists of polysaccharide complex of fiber and agar whose basic monomer is glucose and galactose,
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
respectively (Kim et al., 2010a). Since the physical morphology of the agar is softer than that of fiber, the optical reaction conditions for agar hydrolysis would be milder than that of fiber. Therefore, when using G. amansii as biomass, it is strongly recommended for the galactose to be firstly released from the G. amansii, followed by enzymatic saccharification using remaining cellulose-containing solid material (about 60% (w/w, dry based) of cellulose). Herein we describe a raw material composition, acid hydrolysis using batch and continuous reactor based on the formation of inhibitors, and finally comparative fermentation results using the hydrolysate from the batch and continuous method. 3.1. Raw material composition Analyses by Bio-LC and HPLC revealed that the total carbohydrate was found to be 67.3% (w/w) and among them, cellulose and agar content were found to be 14.9% (w/w) and 52.4% (w/w) based on the dried weight of G. amansii, respectively. The weight ratio of galactose to 3,6-AHG in the agar content was reported as 1:1.27 (0.44:0.56) (Kim et al., 2010a). Therefore, the maximum glucose is calculated as 0.166 [(0.149 g cellulose/g G. amansii) (180 g glucose/162 g cellulose)], galactose as 0.256 [(0.524 g agar/g G. amansii) (0.44 g galactose unit/g agar) (180 g galactose/162 g galactose unit)], and therefore the total fermentable sugar (glucose and galactose) is calculated as 0.422 (0.166 + 0.256) g/g of G. amansii, respectively. 3.2. Dilute-acid hydrolysis of G. amansii using batch reactor To investigate hydrolyzed products from agar, an isolated agar (S/L ratio = 25%, S:L = 1:4) was autoclaved in the presence of a 1.5% (w/w) sulfuric acid at 125 °C and the results depicted as a function of time are shown in Fig. 1. The isolated agar was obtained simply from the soaking of G. amansii in water charged autoclave at 121 °C for 2 h, followed by the centrifugation. The maximum galactose concentration of 93.7 g/L was calculated from (250 g agar/ L liquid) (0.375 g galactose/g agar) and 0.375 was obtained from the analytical result of the galactose portion in isolated agar (Kim et al., 2010a). Over a wide range of reaction time, the galactose concentration remains similar to give 63–65 g/L which are comparable yield of 0.68 (64.0/93.7). Conversely, the acid-labile 3,6-AHG concentration decreased very sharply, resulting in 92% loss based on the initial concentration with the concomitant increase in the concentrations
Fig. 1. Results of dilute-acid hydrolysis of agar in batch reactor as a function of time (experimental conditions: substrate = agar; S:L = 1:4; H2SO4 = 1.5% (w/w); temperature = 125 °C).
85
of 5-HMF and organic acids, which indicates a obvious transformation of 3,6-AHG into undesirable byproducts. Similarly, the G. amansii (S/L ratio = 15%, S:L = 1:6.67) was directly subjected to the same batch type diluted-acid hydrolysis in the presence of a 1.5% (w/w) sulfuric acid at 140 °C to examine the product distribution. As can be seen in Fig. 2, a similar trends with the agar were observed but more sharp decrease in 3,6AHG concentration, resulting in 94% loss was found due to the slightly severer reaction conditions than in the case of agar, which causes increasing inhibitor concentrations as a function of time. From these results, it is concluded that when using batch reactor, it is very difficult to obtain hydrolysate consisting of high sugar and low inhibitor concentration. It is well known that the fermentation inhibitors such as 5-HMF and organic acids are generated from the chemical transformation of biomass, which can stress fermentative organisms to a point beyond where the efficient utilization of sugars is reduced and product formation decreases (Mussatto and Roberto, 2004). According to Olsson and HahnHägerdal, toxic compounds can be divided in four groups: sugar degradation products, lignin degradation products, compounds derived from lignocellulose structure, and heavy metal ions (Olsson and Hahn-Hagerdal, 1996). Since the marine seaweed do not contain any lignin or its structure, only sugar degradation products could be generated through the chemical transformation. Indeed, there are found to be only the sugar degraded products such as 5-HMF, levulinic acid, formic acid, and negligible trace amount of acetic acid without the generation of lignin degraded products in the hydrolysate from the dilute-acid hydrolysis of G. amansii in the level of more than 10–20 g/L totally (see Figs. 1 and 2) using batch reactor. Thus it should be carefully considered at this moment that the concentrations of these inhibitors should be suppressed under a certain level.
3.3. Inhibitor effects and dilute-acid continuous hydrolysis process To investigate the effect of the inhibitors on ethanolic fermentation with a modified B. custersii (KCTC18154P) and to obtain an useful information on each inhibitors’ concentration limit that allow microbe to show proper fermentability, a known amount of formic acid, levulinic acid, and 5-HMF were added separately to sugars containing reagent grade of glucose (20 g/L) and galactose (100 g/L), giving total 120 g/L of mixed sugar. The inhibitory concentration limits of 5-HMF, levulinic acid, and formic acid along with their fermentabilities are summarized in Table 1.
Fig. 2. Results of dilute-acid hydrolysis of G. amansii in batch reactor as a function of time (experimental conditions: substrate = G. amansii; S:L = 1:6.67; H2SO4 = 1.5% (w/w); temperature = 140 °C).
86
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
The concentration limits described in the left column of Table 1 show that the formic acid is found to act as critical inhibitory from the level of 0.5 g/L, and even after 56 h, a small amount of galactose still remained in the mixture solution. And the critical levulinic acid concentration limit appears to be 2 g/L level, in turn, 10 g/L level for 5-HMF, suggesting that the formic acid is identified as most inhibitory among those, which is in good agreement with trends reported in the previous study (Maiorella et al., 1983). When these results are compared with the hydrolysis results obtained from batch reactor (right column of Table 1, from batch), they do not meet the end of criteria especially in the case of levulinic acid (5.7 g/L) and formic acid (2.2 g/L), and accordingly, the fermentation yield thereof, as expected, was very low (data not shown). From these, it is likely that suppressing the inhibitor concentration below the critical limit in the dilute-acid hydrolysis of G. amansii using the batch reactor is very difficult. This is presumably as result of the long induction time until reaching high temperature and the long cooling time, thereby exposing sugars (especially, 3,6-AHG) upon the harsh reaction conditions, which are playing a critical role in generating undesirable byproducts in the batch process. To solve this problem, new method using a continuous-type reactor has to be designed (Kim et al., 2010b) for reducing the induction and cooling time, leading to suppressing the inhibitor formation dramatically. Although many efforts have been devoted to the continuous type dilute-acid hydrolysis with lignocellulosic biomass, many of them were found to be unsuccessful due to the blocking problem in the reactor channel arising from the morphological hardness of wood biomass. On the contrary, owing to much softer morphology of the seaweed biomass than wood, it would be easier for G. amansii to be processed in the continuous Plug and Flow reactor, thereby giving more chance to produce a good hydrolysate composition with minimizing inhibitor formation. To correlate this postulation, two sets of separate hydrolysis reactions of G. amansii were performed using batch and continuous reactor, and their product distribution were compared systematically (Fig. 3). As shown in Fig. 3, the concentrations fermentable sugar, galactose and glucose, obtained from the continuous reactor are found to be even slightly higher than that from batch reactor, producing almost comparable yield to the theoretical galactose maximum. It is worth mentioning that the total amount of inhibitor concentration from the continuous reactor (B) is only 3.1 g/L which is much less than those from the batch reactor (10.4 g/L) (A) (See right columns of Table 1), apparently because this continuous system prevented acid-labile 3,6-AHG from being degraded into 5-HMF and subsequent transformations into organic acids. These results strongly indicate that the continuous process has the advantage of yielding fermentable sugars in good yield as well as minimizing undesirable inhibitor by controlling the residence time very effectively. 3.4. Fermentation using hydrolysate Today, the most important microoganisms applied for ethanol production are the yeast Saccharomyces cerevisiae and the bacte-
Fig. 3. Product distributional comparison between batch and continuous reactor (experimental conditions: substrate = G. amansii; S:L = 1:10; H2SO4 = 2.0% (w/w); temperature = 121 °C for (A) and 150 °C for (B); reaction time = 59 min for (A) and 4 h for(B)).
rium Zymomonas mobilis. However, both ethanol producers have a very narrow substrate range (Horn, 2000). Although the hyrolysate from G. amansii also contains different concentration of mixed sugars, e.g., galactose as major and glucose as minor, a successful utilization of this red seaweed as bio source requires that both sugars should be efficiently converted to ethanol in high yield. This might be attained using two different organisms or two-step process, however, it would be more efficient when using a single organism that can utilize both substrates simultaneously. The authors presented that the B. custersii (KCCM11490) consumed galactose much faster than S. cerevisiae in the presence of mixed sugars, and therefore, the B. custersii will be more suitable microbe for the fermentation of galactose-rich hydrolysate (Ryu et al., 2008). For comparing the fermentability using two kinds of hydrolysates from batch and continuous reactor, each hydrolysate was concentrated by vacuum evaporation, thus the volume was reduced to about 1/3 to obtain 3-fold concentrated sugar mixture. Meanwhile, the concomitant decrease in the concentration of relatively volatile 5-HMF was observed during the evaporation process, resulting in less than 1 g/L level (Larsson et al., 1999) but the concentration of organic acids did not show diminish effect. Subsequently, the concentrated hydrolysate was subjected to the fermentative process using the modified B. custersii (KCTC18154P) and the results are shown in Fig. 4 and summarized in Table 2. As can be seen in Fig. 4 and Table 2, the hydrolysate from the batch reactor does not show good performance for producing ethanol. The maximum final ethanol titer is only 11.8 g/L (yield: 0.13) even after 56 h. In contrast, the hydrolysate from continuous reactor produces ethanol more than double compared with that from the batch, resulting in the final ethanol titer of 27.6 g/L (yield: 0.38) after 39 h. This huge difference might be ascribed to the amount of inhibitors existing in the hydrolysate and indeed, the sum up of inhibitor concentration in the concentrated hydrolysate from the batch
Table 1 Comparison of inhibitor concentration limit with the results from hydrolysates.
a b
Inhibitors
Inhibitor concentration limita, g/L/EtOH, g/L/EtOH yield, g/g
Inhibitor concentrationb, g/L From batch
From continuous
Formic acid Levulinic acid 5-HMF Total
0.5/52.1/0.44 2.0/52.0/0.43 10.0/50.5/0.44 –
2.2 5.7 2.5 10.4
0.5 1.1 1.5 3.1
120 g/L (gal:glu = 5:1), reagent grade. Hydrolysate.
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
87
Fig. 4. Fermentability comparison using hydrolysates from batch and continuous (experimental conditions: B. custersii seeding = 8% (v/v); C-source = hydrolysate 90 g/L for (A) and hydrolysate 72.2 g/L for (B); N-source = yeast extract 0.68 g and soy peptone 1.34 g; agitate = 150rpm; temperature = 30 °C working volume = 200 mL).
Table 2 Fermentability comparison data of concentrated hydrolysates. Inhibitors
Formic acid Levulinic acid 5-HMF Total a b
Inhibitor concentration limit, g/L
Inhibitor concentration after evaporation, g/L/EtOH, g/L/EtOH yield, g/g From batcha
From continuousb
0.5 2.0 10.0 –
14.5/11.8/0.13 27.1/11.8/0.13 0.2/11.8/0.13 41.8
0.6/27.6/0.38 2.3/27.6/0.38 0.8/27.6/0.38 3.7
90 g/L of fermentable hydrolysate. 72.2 g/L of fermentable hydrolysate.
Fig. 5. Product distributional comparison between batch and continuous reactor (experimental conditions: B. custersii seeding = 8% (v/v); C-source = hydrolysate 90 g/L for (A) and hydrolysate 72.2 g/L for (B); N-source = yeast extract 0.68 g and soy peptone 1.34 g; agitate = 150 rpm; temperature = 30 °C working volume = 200 mL.
reactor is 41.8 g/L whereas only 3.7 g/L from the continuous reactor, indicating 1/10 times diminish effect in the amount of the inhibitor by using this continuous method (Gurram et al., 2011). Interestingly, in the fermentive process using the hydrolysate from the batch reactor, both the levulinic and formic acid concentration decreased very sharply after 32 h (Fig. 5A), which well corresponds to the moment when the glucose is consumed completely (see Fig. 4(A)). Accordingly, this result strongly suggests that the organic acids are also consumed by the microbes (Larsson et al., 1999). However, this phenomenon was less pronounced in the fermentive process of using the hydrolysate from the continuous reactor (Fig. 5B) since the consumption of fermentable sugar is more predominant than that of organic acids.
Further studies on the effective inhibitor removal methodologies for enhancing ethanol yield as well as the study on enzymatic saccharification using a residual fiber and on the anaerobic digestion using bioethanol residue from this continuous process are under investigation.
4. Conclusions There are many advantages to the continuous hydrolysis process for producing bioethanol, especially when using seaweeds as biomass. In addition to a low melting property of the agar component in the G. amansii, which is making improved raw material
88
J.-H. Park et al. / Bioresource Technology 108 (2012) 83–88
fluidity in the continuous reactor, the hydrolysate obtained from this process includes high fermentable sugar and low inhibitor concentration. In the fermentive process, the hydrolysate from the continuous reactor delivered much more ethanol than that from batch reactor due to the dramatic reduce in the formation of inhibitor. Acknowledgement This work was supported by a grant (2008-N-BI08-P-01) from Korea Institute of Energy Technology Evaluation and Planning, Ministry of Knowledge Economy, Republic of Korea. References Buck, B.H., Buchholz, C.M., 2004. The offshore-ring: a new system design for the open ocean aquaculture of macroalgae. Journal of Applied Phycology 16, 355– 368. Georgieva, T.I., Hou, X., Hilstrom, T., Ahring, B.K., 2008. Enzymatic hydrolysis and ethanol fermentation of high dry matter wet-exploded wheat straw at low enzyme loading. Biotechnology for Fuels and Chemicals, 553–562. Gurram, R.N., Datta, S., Lin, Y.J., Snyder, S.W., Menkhaus, T.J., 2011. Removal of enzymatic and fermentation inhibitory compounds from biomass slurries for enhanced biorefinery process efficiencies. Bioresource Technology. 102, 7850– 7859. Horn, S.J., 2000. Bioenergy from brown seaweeds. Doctor Thesis, Department of Biotechnology, Norwegian University of Science and Technology NTNU, Trondheim. Jol, C.N., Neiss, T.G., Penninkhof, B., Rudolph, B., De Ruiter, G.A., 1999. A novel highperformance anion-exchange chromatographic method for the analysis of carrageenans and agars containing 3,6-anhydrogalactose. Analytical Biochemistry 268, 213–222. Kim, C., Ryu, H.J., Kim, S.H., Yoon, J.J., Kim, H.S., Kim, Y.J., 2010a. Acidity tunable ionic liquids as catalysts for conversion of agar into mixed sugars. Bulletin of the Korean Chemical Society Notes 31, 511–514.
Kim, Y.J., Yoon, J.J., Kim, S.H., 2010b. Korea Patent Number 10-1096183. Kim, J.W., Kim, K.S., Lee, J.S., Park, S.M., Cho, H.Y., Park, J.C., Kim, J.S., 2011. Two-stage pretreatment of rice straw using aqueous ammonia and dilute acid. Bioresource Technology. 102, 8992–8999. Larsson, S., Palmqvist, E., Hahn-Hagerdal, B., Tengborg, C., Stenberg, K., Zacchi, G., Nilvebrant, N.O., 1999. The generation of fermentation inhibitors during dilute acid hydrolysis of softwood. Enzyme and Microbial Technology 24, 151–159. Luning, K., Pang, S., 2003. Mass cultivation of seaweeds: current aspects and approaches. Journal of Applied Phycology 15, 115–119. Maiorella, B., Blanch, H.W., Wilke, C.R., 1983. By product inhibition effects on ethanolic fermentation by Saccharomyces cerevisiae. Biotechnology and Bioengineering 25, 103–121. Mathew, A.K., Chaney, K., Crook, M., Humphries, A.C., 2011. Dilute acid pretreatment of oilseed rape straw for bioethanol production. Renewable Energy 36, 2424–2432. Mussatto, S.I., Roberto, I.C., 2004. Alternatives for detoxification of diluted-acid lignocellulosic hydrolyzates for use in fermentative processes: a review. Bioresource Technology 93, 1–10. NREL, 1996. National Renewable Energy Laboratory (NREL). Chemical analysis and testing laboratory analytical procedures. NREL, Golden, CO, USA. Available from: