- Email: [email protected]
Enzymatic pretreatment of Chlamydomonas reinhardtii biomass for ethanol production
Bioresource Technology 101 (2010) 5330–5336
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Enzymatic pretreatment of Chlamydomonas reinhardtii biomass for ethanol production Seung Phill Choi, Minh Thu Nguyen, Sang Jun Sim * Department of Chemical Engineering, Sungkyunkwan University, Changan-gu, Suwon 440-746, Republic of Korea
a r t i c l e
i n f o
Article history: Received 18 June 2009 Received in revised form 27 December 2009 Accepted 7 February 2010 Available online 9 March 2010 Keywords: Enzymatic pretreatment Ethanol Microalgae Chlamydomonas reinhardtii
a b s t r a c t The production of ethanol from feedstock other than agriculture materials has been promoted in recent years. Some microalgae can accumulate a high starch content (about 44% of dry base) via photosynthesis. Algal biomass, Chlamydomonas reinhardtii UTEX 90, was converted into a suitable fermentable feedstock by two commercial hydrolytic enzymes. The results showed that almost all starch was released and converted into glucose without steps for the cell wall disruption. Various conditions in the liquefaction and saccharification processes, such as enzyme concentration, pH, temperature, and residence time, have been investigated to obtain an optimum combination using the orthogonal analysis. As a result, approximately 235 mg of ethanol was produced from 1.0 g of algal biomass by a separate hydrolysis and fermentation (SHF) method. The main advantages of this process include the low cost of chemicals, short residence time, and simple equipment system, all of which promote its large-scale application. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction From a greenhouse standpoint, renewable fuels such as ethanol are considered to be excellent alternative clean-burning fuels to gasoline in the future as the combustion products are environmentally safe (Kito-Borsa et al., 1998). Today, the most common renewable fuel is ethanol, derived mainly from glucose or starch sources of agricultural stock (Gray et al., 2006; Nigam and Singh, 1995). The human demand for food, however, has yet to be met. To solve both the energy and food problem, there has been increasing interest and worldwide studies in producing bioethanol from algal biomass, the alternative to agricultural stock (Baras et al., 2002; Kim and Dale, 2003; Sánchez and Cadona, 2008). C. reinhardtii, a unicellular green alga, is well-known as a photoautotrophic microorganism having a great ability to fix CO2 and accumulate a high content of stored polysaccharides, mainly starch, in complex multilayered cell walls (Hall and Rao, 1994; Hirano et al., 1997). This bears a strong structural and functional resemblance to higher plant storage starch (Libessart et al., 1995). With a high growth rate, the microalgae can be easily cultured at high yields and low costs utilizing an unlimited energy source, sunlight (Hirayama et al., 1998; Sze, 1998). These advantages allow the microalgae to be preferentially selected as a safe and prospective feedstock for bioethanol production by Saccharomyces cerevisiae. Prior to ethanol fermentation, the feedstock needs to be processed by enzymatic or acidic pretreatment technology in order * Corresponding author. Tel.: +82 31 290 7341; fax: +82 31 290 7272. E-mail address: [email protected] (S.J. Sim). 0960-8524/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2010.02.026
to release fermentable sugars. Since the first application of microbial enzyme in the food industry in the early 1960s, a great deal of effort has been made to replace traditional acid hydrolysis with enzymatic hydrolysis in almost all glucose production due to higher yields under mild conditions, less by-products, and no corrosion issues (Balat et al., 2008). Several hydrolytic enzymes possessing improved activity have been developed and used extensively in the starch processing industry for liquefaction and saccharification (Olsen, 2004; Schäfer et al., 2007). To this end, a commercial amyloglucosidase, AMG 300L, was produced by a genetically modified strain of Aspergillus. The main obstacle of enzymatic hydrolysis is that intercellular starch granules are bound within rigid cell walls (Libessart et al., 1995), thus a biomass pretreatment step is needed to break down the cell wall to release polysaccharides such as starch, structural carbohydrates, and other nutrients, prior enzymatic hydrolysis and fermentation steps. The cell wall of C. reinhardtii contains glycoproteins as the predominant constituents in its extracellular matrix (Sze, 1998). A commercial a-amylase derived from Bacillus licheniformis, Termamyl 120L, shows a protease activity particular to the degradation of glycoproteins within cell walls (Imam and Snell, 1987). Starch hydrolysis involves liquefaction and saccharification of the starch. The objective of this study was to optimize conditions to apply the starch-degrading enzymes sequentially to pretreatment of the algal biomass. The optimal conditions required for the enzymatic hydrolysis of starch, such as enzyme concentration, temperature, pH, and residence time, were selected by the orthogonal analysis method, which weighed the effects on both hydrolysis and ethanol fermentation; a maximum rate of ethanol
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
production could then be achieved. Two-step enzymatic hydrolysis by commercially available a-amylase and amyloglucosidase would yield lower energy consumption, lower content of non-glucosidic impurities, and thus, a much better suitability for ethanol production (Mojovic´ et al., 2006). To the best of the authors’ knowledge, this is the first report regarding enzymatic pretreatment of microalgal biomass optimized for ethanol fermentation. The work herein demonstrates that the enzymatic hydrolysate from the microalgae is potentially useful as a feedstock for fermentation by the ethanolproducing yeast, S. cerevisiae S288C. 2. Methods 2.1. Algal biomass The medium used for the culture of the green algae, C. reinhardtii UTEX 90, was 1.6 L of tris–acetate–phosphate (TAP) medium, which was prepared as reported in our previous work (Thu et al., 2009). A cell culture was carried out during 96 h at 23 °C and 130 rpm in a 2.5 L photo-bioreactor with 10% inoculation of seed under continuous illumination (450 lE m 2 s 1) by white fluorescent lamps. Light intensity in the photo-bioreactor was measured with an Li-250 Li-Cor quantum photometer (Lambda Instrument Corp., Lincoln, USA). The pH of the medium was automatically adjusted to the range of 7.0–7.4 by the feeding of 1.0 M acetic acid. Cell growth of the algae was monitored by measuring optical density (OD) at 600 nm. Dry cell weight (DCW) was determined by an 80 °C oven drying method using filter paper (GC/F, City, England). The cells harvested by centrifugation were characterized and used for pretreatment of the feedstock biomass. 2.2. Enzymes Two enzymes for liquefaction and saccharification, thermostable a-amylase of B. licheniformis origin (EC 3.2.1.1, Termamyl 120L) and amyloglucosidase from Aspergillus niger (EC 3.2.1. 3, AMG 300L), were purchased from Novo Nordisk (Gentafte, Denmark). The enzymatic activities of these enzymes were 120 KNU g 1 and 300 AGU mL 1, respectively, as defined by Novo Nordisk. The Kilo Novo a-amylase Unit (KNU) is defined as the amount of enzyme that can hydrolyze 5.26 g of soluble starch per hour at pH 5.6 and 37 °C. The Amyloglucosidase Novo Unit (AGU) is the amount of enzyme which cleaves 1.0 lmol of maltose/min at pH 4.3 and 25 °C. 2.3. Enzyme pretreatment of algae Enzyme pretreatment consisted of two parts, liquefaction of algal biomass by a-amylase (Termamyl 120L) and saccharification by amyloglucosidase (AMG 300L). The harvested algal biomass were washed once and slurried in water at a 5% solid to liquid ratio (w/v), mixed with Termamyl 120L at a final concentration between 0.0001% and 0.02% (v/w), and pretreated as follows. Hydrolysis reactions were performed in capped flasks in a thermostated water bath. The pH of the mixture was adjusted to pH 6.0, a value known as the optimum (Richardson et al., 2002), with 10% H2SO4. The protease activity of the Termamyl 120L, the optimum temperature of which is in the range of 50–60 °C, was activated by increasing the incubation temperature from 25 °C to different temperatures (70, 80, 90, 100 °C) for 30 min, and thenceforth the algal biomass was hydrolyzed by the a-amylase activity of Termamyl 120L by maintaining it at the optimal temperature for 10–60 min. For subsequent saccharification after liquefaction, The pH of the mixture was adjusted to a range of 4.5–5.5, the temperature was reduced to a range of 50–65 °C, and AMG 300L at concentrations of
5331
0.001–0.3% (v/w) was supplied for 5–60 min. By analyzing parameters such as temperature, residence time, enzyme concentration, and pH by orthogonal arrays, two processes of the enzymatic hydrolysis were optimized. Duplicate batches were run to verify results. 2.4. Ethanol fermentation by yeast To test fermentability of the pretreated algal biomass, separate hydrolysis and fermentation (SHF) was performed using ordinary compressed baker’s yeast, S. cerevisiae S288C as reported in our previous work (Thu et al., 2009). The solid fraction of the pretreated algal biomass was removed by centrifugation (4000g, 10 min). After the aerobically pre-cultured yeast cells were inoculated at 10% of volume size to the liquid fraction of pretreated algal biomass, the yeast cells were cultured anaerobically at 30 °C for 40 h at a rotation of 160 rpm. The concentration of the produced ethanol was analyzed by HPLC under the same conditions as the analysis of monomeric sugars in the pretreatment. 2.5. Analytical methods The cells were disrupted by vortexing with glass beads in a methanol solution. The methanol was then vaporized by placing the cell lysates in the oven at 80 °C for 6 h. After the addition of 72% perchloric acid, the starch was extracted at 0 °C for 12 h and filtered through a 0.45 lm filter. The contents of the starches in the filtrate were determined by an iodo-starch reaction method (Hirokawa et al., 1982). The total cell carbohydrate was analyzed by a colorimetric method using an anthrone reagent. The algal cell pellets were mixed with 67% sulfuric acid by vortexing for 30 min and then reacted with the anthrone reagent for 5 min at 100 °C. The OD of the resulting solution was measured at 630 nm. The intracellular monomeric sugar content of the C. reinhardtii was determined by high performance anion-exchange chromatography (HPAEC; DX-300 series chromatography system, Dionex, USA). The effluent was monitored with pulsed amperometric detection detector (PAD, Dionex, CA, USA). The cell pellets were washed two times with autoclaved distilled water and freezedried. Sample of 2.5 mg dry cell was dissolved in 1 ml trifluoroacetic acid (TFA; Sigma, USA). After hydrolysis at 100 °C for 4 h, the sample was cooled to room temperature and the volatile acid was removed by centrifugal evaporation (Speed-Vac; SPD1010115, Savant Instruments, USA) under a stream of nitrogen gas. The dried sample was dissolved in 10.0 ml of water and filtered through 0.2 lm syringe filter. The standard sugars were also treated in the same way as described for the samples in order to identify and quantify. The 10 ll of filtered samples were injected into a CarboPac PA-1 anion-exchange column (0.4 250 mm, Dionex, CA, USA) that was pre-equilibrated in 18 mM NaOH. Chromatographic separation of the monomeric sugars from the samples was achieved in the isocratic mode with 18 mM NaOH at a flow rate of 1.0 ml/min in 20 min. The content of the intracellular proteins was determined by the Bradford method (1976). The cells were disrupted by sonication in PBS buffer (pH 7.3), the supernatant was colored by mixing with the Bradford reagent (Sigma Chemical Co., St. Louis, MO, USA), and the OD of it was measured at 595 nm. The concentrations of the oligosaccharide (dextrin) and monomeric sugar (glucose) in the liquid fraction of the enzymatic hydrolysate and ethanol produced by yeast were quantified by an Agilent 1200 high-pressure liquid chromatography (HPLC) system equipped with a quaternary solvent delivery system (Agilent Technologies, Palo Alto, CA, USA), an autosampler (No. 60044; Spark, Holland, Emmen, Netherlands), a refractive index detector (Acme
5332
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
9000; Younglin instrument, Seoul, Korea), and a computer software-based integration system (Agilent Chemstation; Agilent Technologies). The samples were separated on ion-moderated partition chromatography columns (Aminex HPX-87P, 300 mm 7.8 mm; Bio-Rad, Richmond, CA, USA) using deionized water (Sigma, USA) as the mobile phase at a flow rate of 0.6 ml/min. The column was maintained at 60 °C. Target compounds were detected using a refractive index detector and quantified by comparison with authentic standards: glucose (Amresco, USA); dextrin hydrate (Kanto, Japan); ethanol (J.T. Baker, USA). Each measurement was repeated at least three times and averaged. 3. Result and discussion 3.1. Algal biomass content After acetic acid fed-batch cultivation in a 2.5 L photo-bioreactor for 4 d, the Chlamydomonas biomass was harvested and concentrated at 12.4 g L 1 (dry cell weight). The content of the main components was determined by chemical analysis and presented in Table 1. It was found that the algae contained a high content of carbohydrates (59.7%) and various proteins (9.2%), useful sources of carbon and nitrogen for yeast fermentation. Therefore, it was supposed that the additional nitrogen source was unnecessary. Monomeric sugars in TFA-hydrolyzed microalgal biomass were also analyzed by HPAEC (Table 1). As a result of the substantial starch accumulation (43.6%) achieved under controlled conditions of the fed-batch culture, the most predominant monosaccharide was glucose (44.7%), which S. cerevisiae strain can readily ferment, while the total composition of the other monosaccharides like L-fucose, L-rhamnose, D-arabinose, D-galactose, and D-mannose was 7.3%. The algal biomass is cheaper and contains less intracellular glucose than the other biomass sources (sugar cane, wheat, rice, corn, and cassava) (Table 1). After quantitative determination of the cell constituents, the algal biomass was adjusted at a 5% solid liquid ratio (w/v) and ready to be used in further experiments. 3.2. Liquefaction In the enzymatic pretreatment of algal biomass, the first step is liquefaction of starch with a thermostable a-amylase (Termamyl 120L), which catalyzes the hydrolysis of the internal a-D-(1–4)glucosidic linkages in starch in a random manner (Montesinos and Navarro, 2000). The products resulting from the liquefaction
Table 1 Characterization of cell mass composition for Chlamydomonas reinhardtii UTEX 90 obtained from the pH-stat fed-batch culture by feeding 1.0 M acetic acid under continuous illumination (450 lE m 2 s 1). All analyses were conducted in triplicate and the error limits were estimated. Components
Composition based on dry cell mass (%, w/w)
Proteins
9.2 ± 0.6
Total carbohydrates
59.7 ± 0.5
(Starch)
(43.6 ± 1.4)
D-Glucose
44.7 ± 0.8
L-Fucose
0.4 ± 0.01
L-Rhamnose
0.9 ± 0.02
D-Arabinose
1.9 ± 0.04
D-Galactose
2.7 ± 0.04
D-Mannose
1.4 ± 0.03
Others
31.1
Analytical method
Colorimetric method with anthrone (Iodo-starch reaction method) Chromatographic separation by HPAEC
–
process were oligosaccharides with three or more 1 ? 4-a-linked D-glucose units, generally denoted as dextrin produced from starch (Petrova et al., 2000; Schäfer et al., 2007). Liquefaction is a process that converts a suspension of starch granules into a partially hydrolyzed starch solution of low viscosity suitable for subsequent processing. Its effectiveness depends highly on parameters such as enzyme concentration, temperature, and residence time. Therefore, the main focus was determination of how the liquefaction process is influenced by various conditions; the optimal conditions had to be determined for this process to obtain the highest yield of product. The degree of liquefaction of the algal suspension was evaluated by HPLC analysis of the amount of dextrin released from the samples. To assess the effect of enzyme concentration, liquefaction experiments were carried out at six different levels, from 1 10 4 to 2 10 2 (%, v/w), at a fixed residence time (30 min) and constant temperature (90 °C). Fig. 2 shows the concentration of dextrin obtained as a function of various operation parameters. The incubation of the algal suspension with Termamyl 120L, particularly at enzyme concentrations below 5 10 3 (%, v/w), significantly increased the concentration of the released dextrin, compared with the control of no enzyme (Fig. 1a). However, further increase in enzyme concentration appeared to not be required to establish an economic process in terms of cost. A reaction progress curve could be obtained by mixing an algal suspension with 0.001% of the Termamyl 120L enzyme and measuring the resulting dextrin generated at 90 °C over a period of time (10–60 min). During the early 10 min, the enzyme-catalyzed hydrolysis reaction exhibited linear kinetics and the highest initial rate of reaction, whereby the amount of the product dextrin increased linearly with residence time (Fig. 1b). This indicated either that the level of the enzyme was low enough for the curve to reach a plateau early, due to depletion of the substrate (algal biomass), or that it was possible to maintain the initial conditions of liquefaction over a given period of time. Only a marginal change in the dextrin was observed during the late period (30–60 min). It is certain that a longer residence time would generally be more beneficial for liquefaction. However, it was more reasonable, from an economic vantage, that 30 min of residence time should be a suitable choice for the liquefaction process. The effect of reaction temperature (70, 80, 90, 100 °C) on liquefaction performed with 0.001% enzyme for 30 min is shown in Fig. 1c. As expected, the rate of reaction, and thus the amount of produced dextrin, increased with an increase in reaction temperature. The most dramatic effect of liquefaction time was observed between 70 and 80 °C, where the released dextrin increased from 10.3 g L 1 at 70 °C to 16.8 g L 1 at 80 °C. However, temperature change seemed to be less effective above 80 °C. A further increase of 10 °C (from 80 to 90 °C) led to a slight increase of dextrin (approximately 0.5 g L 1) and furthermore, there was no significant difference between 90 and 100 °C. It was found that the optimum temperature of the liquefaction process was near 90 °C. Optimum combinations of liquefaction parameters (enzyme concentration, temperature, residence time) were obtained by analysis at three different levels using the orthogonal analysis technique (RenJie, 2008). Based on the provisional optimum values (0.005% enzyme, 30 min, 90 °C) obtained from the above experiments, the three levels for each parameter were determined. Nine of the experiments concomitant with different liquefaction parameters were collectively conducted. Table 2 shows the concentration of dextrin released in each hydrolysis experiment, the average concentration (k) of dextrin for the parameter chosen at a level, and the maximum difference (R) between the three k values. For example, k2 value for enzyme at level 2 can be calculated as the average of dextrin concentrations (16.15, 16.94, 25.21) in experiment 2, 4, 9. A higher k value and a higher R value would indicate
5333
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
30
a
b
20
Dextrin (g L-1)
Dextrin (g L-1)
25 20 15 10
15
10
5 5 0 0.000
0.005
0.010
0.015
0
0.020
0
10
20
Enzyme concentration (%, v/w)
Dextrin (g L-1)
20
30
40
50
60
Time (min)
c
15
10
5
0
70
75
80
85
90
95
100
Temperature ( oC) Fig. 1. Effects of various factors on the liquefaction process by the commercial thermostable a-amylase, Termamyl 120L. (a) Effect of enzyme concentration at a fixed residence time (30 min) and constant temperature (90 °C); (b) effect of liquefaction time with 0.001% enzyme at 90 °C; (c) effect of liquefaction temperature with 0.001% enzyme at 30 min.
Table 2 Orthogonal analysis of the liquefaction process. Number
Temperature (°C)
Time (min)
Level 1 Level 2 Level 3
70 80 90
10 20 30
0.001 0.005 0.01
70 70 70 80 80 80 90 90 90
10 20 30 10 20 30 10 20 30
0.001 0.005 0.01 0.005 0.01 0.001 0.01 0.001 0.005
15.13 18.65 21.18 6.05
16.73 17.85 20.39 3.65
Experiment Experiment Experiment Experiment Experiment Experiment Experiment Experiment Experiment k1 k2 k3 R
1 2 3 4 5 6 7 8 9
a preferred level for the chosen parameters and a greater influence of that parameter, respectively. Therefore, optimum combinations of the liquefaction parameter can be determined from the maximum value of the three k values (k1, k2, k3). In this study, it could be tentatively suggested that the optimum temperature should be at level 3 (90 °C), because k3 is the highest (21.18 g L 1) of the three k values, and similarly that the optimum residence time should be 30 min and the optimum enzyme concentration 0.01% (v/w). Using such a combination, as shown in Fig. 1a, 27.09 g L 1 of dextrin could be obtained. Meanwhile, focusing upon the nine
Enzyme (%, v/w)
Dextrin (g L
1
)
10.13 16.15 19.12 16.94 22.18 16.83 23.13 15.21 25.21
14.06 19.43 21.48 7.42
experiments combining the three levels, a maximum dextrin concentration was determined, 25.21 g L 1, in the experiment performed at temperature 90 °C with 0.005% enzyme at 30 min. Though twice as much enzyme (0.01% enzyme) was introduced, increase in the dextrin concentration was only 1.88 g L 1, very small. To establish an economic pretreatment process, as mentioned above, the smallest amount of enzyme should be used as possible. Therefore, the choice of the former combination (0.005% enzyme, 90 °C, 30 min) as the optimum conditions for liquefaction was a natural one.
5334
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
The orthogonal analysis also revealed the relative importance of each parameter in influencing dextrin yield. With an R value of 7.42, the enzyme concentration seemed to be the most critical factor, followed by temperature (R value 6.05) (Table 2). With an R value of only 3.65, the variation in time appeared to have the least influence on the liquefaction process. Cell disruption, in general, is an essential, initial step regarding the biomass treatment process. Judging from the results in this liquefaction process, it was certain that the Termamyl 120L was successfully applied to interrupt the cell wall, release starch out of algal biomass having thick cell walls, and hydrolyze dextrin from the starch. This was because Termamyl 120L is a powerful complex enzyme harboring the protease activity capable of the necessary cell wall-degrading activity (Nigam and Singh, 1995; Imam and Snell, 1987), in addition to the a-amylase activity that degrades the bulk of the starch rapidly at 100 °C. This was also because the residence time of increasing temperature from room temperature to liquefaction temperatures was enough for cell wall degradation by the protease. Thus, without a separate pretreatment step, enzymatic hydrolysis rendered or disrupted the cell wall structure of the biomass and made it more accessible at reasonable rates and yields. 3.3. Saccharification and ethanol fermentation Using the starch saccharifying enzyme, amyloglucosidase (AMG 300L), which catalyzes the hydrolysis of a-D-(1–4) and a-D-(1–6)glucosidic bonds of oligosaccharides in the starch liquefied under optimal conditions, a saccharification process was subsequently performed. The degree of saccharification was evaluated by the 25
amount of glucose hydrolyzed out of the liquefied starch, which was analyzed by HPLC. The effective usefulness of the pretreated algal biomass as a medium for yeast growth was further estimated by checking ethanol production from the SHF process (Amutha and Gunaskaran, 1994). The primary advantage of SHF is that hydrolysis and fermentation occur at optimum conditions (Balat et al., 2008). Fig. 2 shows the effects of enzyme concentration, pH, residence time, and temperature on the saccharification process. The effect of enzyme concentration was firstly investigated at fixed conditions (pH 5.5, temperature 55 °C, residence time 30 min), and its effect was found to be very significant. There is no doubt that the increase in enzyme concentration would normally enhance hydrolysis and, therefore, increase ethanol yield. As demonstrated in Fig. 2a, the most dramatic effect of enzyme concentration on ethanol yield was observed between 0.001% and 0.05% (v/w). At an enzyme concentration above 0.05% (v/w), the slope of the ethanol yield line reduces gradually. Enzyme concentration is one of the factors contributing to production cost, thus proper amounts give the best results at a reduced expense. The effect of pH is presented in Fig. 2b (enzyme concentration 0.1% (v/w), temperature 55 °C, residence time 30 min). It was found that pH had a significant influence on the saccharification and fermentation processes. The maximum glucose and ethanol yield was observed at pH 4.5, a value also known as the optimum pH condition for yeast fermentation (Yong et al., 1980). Residence time seemed to maintain a positive effect over the saccharification process (Fig. 2c). Effects of small variations in time, from 0 to 60 min, were estimated at other fixed conditions (0.1% enzyme, 55 °C, pH 5.5). The slope in the initial 5 min showed the highest rate of reaction. As a result, the ethanol yield increased
a
c
20
Concentration (g L-1)
Concentration (g L-1)
20
15
10
Glucose Ethanol
5
0 0.00
15
10
Glucose Ethanol
5
0
0.05
0.10
0.15
0.20
0.25
0
0.30
10
20
b
20
40
50
60
d Glucose Ethanol
14
Concentration (g L-1)
18
Concentration (g L-1)
30
Time (min)
Enzyme concentration (%, v/w)
16 14 12 10
12
Glucose Ethanol 10
8
8 3.8
4.0
4.2
4.4
4.6
4.8
pH
5.0
5.2
5.4
5.6
48
50
52
54
56
58
60
62
64
66
Temperature ( oC)
Fig. 2. Effects of various factors on the saccharification and fermentation processes by commercial thermostable amyloglucosidase, AMG 300L. (a) Effect of enzyme concentration; (b) effect of pH; (c) effect of saccharification time; (d) effect of saccharification temperature.
5335
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
more upon extending residence time, particularly between 45 and 60 min, than glucose production did. This reveals that the saccharification process was continuously carried out at the fermentation stage. The optimal temperature of the saccharification process was tested between 50 and 65 °C (enzyme concentration of 0.1% (v/ w), pH 5.5, residence time 30 min); results are shown in Fig. 2d. It is apparent that too low or too high temperatures were not preferable. The highest yield of glucose concentration was observed at 55 °C. Therefore, maximum ethanol could also be obtained at this point. This result was compatible with the value recommended by the supplier. Finally, saccharification conditions such as enzyme concentration, temperature, pH, and the residence time, were analyzed by the orthogonal method in the same way as in the above liquefaction and optimized in relation to maximization of the ethanol yield. The above four factors were examined at three different levels (Table 3). Based on the ethanol obtained after fermentation, analysis showed that all four factors played important roles in the saccharification and fermentation processes. Of the four parameters, enzyme concentration and pH appeared to have a more dominant influence (R values of 2.47 and 1.07, respectively). With smaller R values, temperature (R value 0.48) and time (R value 0.15) seem to be relatively less critical. From this orthogonal analysis, it was found that the optimal conditions for saccharification were 55 °C, 0.2% (v/w) enzyme concentration, pH 4.5, and a 45 min residence time. But, in experiment five, combining three levels, the highest ethanol yield obtained was 11.73 g L 1 with 1.85 g L 1 of the dextrin remaining (data not shown). Therefore, it was reasonable to choose the latter combination (0.2% enzyme, 55 °C, 30 min) as the optimum conditions for liquefaction, due to the shorter residence time. This indicates that under these condi-
tions, approximately 94% of the carbohydrates were hydrolyzed into glucose and shows that algal biomass can be used for ethanol production without further pretreatment. The majority of processing schemes for hydrolyzing biomass into glucose to be used as feedstock, which have been developed over the years, utilizes either enzymes or acid of varying concentrations. We reported a study on acid treatment in our previous work (Thu et al., 2009). Compared with the ethanol production efficiency obtained by the acid treatment, the one obtained currently by enzymatic treatment is a bit lower, 29.2% and 23.5%, respectively. Nevertheless, enzymatic process would be quite promising, particularly due to higher yields under mild conditions, less byproducts, no corrosion issues, and lower utility cost (Balat et al., 2008). Really, the ethanol production using the microalgal biomass was performed at the highest level of efficiency, compared with those obtained from the different biomass sources (Table 4). Using algae as a feedstock has many advantages over other biomass sources (sugar cane, wheat, rice, corn, and cassava), because of its ability to be produced quickly and cheaply (Balat and Balat, 2009; Lynd et al., 1999). Recently, enzymes for production of fuel ethanol from biomass have been manufactured at substantially reduced cost by modern biotechnology (Schäfer et al., 2007). Therefore, it is desirable to focus on the potential of searching the mesophilic or new activity of enzymes and their application towards the SHF or the simultaneous saccharification and fermentation (SSF) process (Sánchez and Cadona, 2008), thus improving the efficiency of the process even further. 4. Conclusions This work demonstrates that commercially available a-amylase and glucoamylase were applicable to the pretreatment of algal bio-
Table 3 Orthogonal analysis of the saccharification and fermentation processes. Number
Temp. (°C)
Time (min)
Level 1 Level 2 Level 3
50 55 60
15 30 45
0.05 0.1 0.2
4.5 5 5.5
50 50 50 55 55 55 60 60 60
15 30 45 15 30 45 15 30 45
0.05 0.1 0.2 0.1 0.2 0.05 0.2 0.05 0.1
4.5 5 5.5 5.5 4.5 5 5 5.5 4.5
8.37 10.11 10.84 2.47
10.41 9.58 9.34 1.07
Experiment Experiment Experiment Experiment Experiment Experiment Experiment Experiment Experiment
1 2 3 4 5 6 7 8 9
k1 k2 k3 R
9.70 10.05 9.57 0.48
Enzyme (%, v/w)
9.71 9.75 9.86 0.15
pH
Ethanol (g L
1
)
8.87 9.81 10.42 9.88 11.73 8.54 10.38 7.71 10.63
Table 4 Comparison of ethanol yields from different biomass sources. Biomass
Conversion to sugar or starch (%)
Ethanol yield (g ethanol g
Sugar cane Sugar beet Cassava Sweet sorghum Corn Wheat Cane bagasse Corn stover Microalgae Microalgae
12.5 – 25 14 69 66 – – – 57
0.055 0.079 0.118 0.063 0.324 0.308 0.111 0.260 0.292 0.235
1
biomass)
References Moreira and Goldemberg (1999) Berg (2001) Wang (2002) Wang (2002) Wang (2002) Wang (2002) Moreira (2000) Kadam and McMillan (2003) Thu et al. (2009) This study
5336
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
mass for yeast fermentation. Due to proteases mingled with aamylase, cell disruption could be carried out without additional energy, time, and cell wall degrading enzymes. The other intracellular components released from algal cells will be beneficial in subsequent fermentations, regardless of how much is required. With the help of orthogonal analysis, the optimum conditions for the SHF process were identified as 0.005% a-amylase, 90 °C, and 30 min for liquefaction, and 0.2% glucoamylase, 55 °C, pH 4.5, and 30 min for saccharification. By yeast fermentation, 235 mg of ethanol could be obtained from the hydrolyzed starch of 1.0 g algal biomass. It has been suggested that algal biomass could be a potential feedstock for ethanol production on large-scale. Acknowledgements This research was supported by a grant (DG2-201) from the Carbon Dioxide Reduction and Sequestration Research Center, one of the 21st Century Frontier Programs that was funded by the Ministry of Science and Technology of the Korean Government. References Amutha, R., Gunaskaran, P., 1994. Simultaneous saccharification and fermentation of cassava starch using Zymomonas mobilis. J. Microbiol. Biotechnol. 9, 22–34. Balat, M., Balat, H., 2009. Recent trends in global production and utilization of bioethanol fuel. Appl. Energy 86, 2273–2282. Balat, M., Balat, H., Oz, C., 2008. Progress in bioethanol processing. Prog. Energy Combust. Sci. 34, 551–573. Baras, J., Gac´eša, S., Pejin, D., 2002. Ethanol is a strategic raw material. Chem. Ind. 56, 89–105. Berg, C., 2001. World Fuel Ethanol. Analysis and Outlook. F.O. Licht. (March 2004). Bradford, M.M., 1976. A rapid and sensitive method for the quantization of microgram quantities of protein utilizing the principle of protein dye binding. Anal. Biochem. 72, 248–254. Gray, K.A., Zhao, L., Emptage, M., 2006. Bioethanol. Curr. Opin. Chem. Biol. 10, 141– 146. Hall, D.O., Rao, K.K., 1994. Photosynthesis, fifth ed. Cambridge University Press, Cambridge. Hirano, A., Ueda, R., Hirayama, S., Ogushi, Y., 1997. CO2 fixation and ethanol production with microalgal photosynthesis and intracellular anaerobic fermentation. Energy 22, 137–142. Hirayama, S., Ueda, R., Ogushi, Y., Hirano, A., Samejima, Y., Hon-Nami, K., Kunito, S., 1998. Ethanol production from carbon dioxide by fermentative microalgae. Stud. Surf. Sci. Catal. 114, 657–660. Hirokawa, T., Hata, M., Takeda, H., 1982. Correlation between the starch level and the starch synthesis during the developmental cycle Chlorella ellipsoidea. Plant Cell Physiol. 23, 813–820.
Imam, S.H., Snell, W.J., 1987. Degradation of the framework of the Chlamydomonas cell wall by proteases present in a commercially available a-amylase preparation. Appl. Environ. Microbiol. 53, 1701–1704. Kadam, K.L., McMillan, J.D., 2003. Availability of corn stover as a sustainable feedstock for bioethanol production. Bioresour. Technol. 88, 17–25. Kim, S., Dale, B.E., 2003. Global potential bioethanol production from waste crops and crop residues. Biomass Bioenergy 26, 361–375. Kito-Borsa, T., Pacas, D.A., Selim, S., Cowley, S.W., 1998. Properties of an ethanol– diethyl ether–water fuel mixture for cold-start assistance of an ethanol-fueled vehicle. Ind. Eng. Chem. Res. 37, 3366–3374. Libessart, N., Maddelein, M.L., Koornhuyse, N., Decq, A., Delrue, B., Mouille, G., D’Hulst, C., Ball, S., 1995. Storage, photosynthesis and growth: the conditional nature of mutations affecting starch synthesis and structure in Chlamydomonas. Plant Cell 7, 1117–1127. Lynd, L.R., Wyman, C.E., Gerngross, T.U., 1999. Biocommodity engineering. Biotechnol. Progr. 15, 777–793. Mojovic´, L., Nikolic´, S., Rakin, M., Vukasinovic´, M., 2006. Production of bioethanol from corn meal hydrolyzates. Fuel 85, 1750–1755. Montesinos, T., Navarro, J.M., 2000. Production of alcohol from raw wheat flour by amyloglucosidase and Saccharomyces cerevisiae. Enzyme Microb. Technol. 27, 362–370. Moreira, J.S., 2000. Sugarcane for energy – recent results and progress in Brazil. Energy Sustain. Dev. 4, 43–54. Moreira, J., Goldemberg, J., 1999. The alcohol program. Energy Policy 27, 229–245. Nigam, P., Singh, D., 1995. Enzyme and microbial systems involved in starch processing. Enzyme Microb. Technol. 17, 770–778. Olsen, H.S., 2004. Enzyme at work. Published by Novozymes A/S.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Enzymatic pretreatment of Chlamydomonas reinhardtii biomass for ethanol production Seung Phill Choi, Minh Thu Nguyen, Sang Jun Sim * Department of Chemical Engineering, Sungkyunkwan University, Changan-gu, Suwon 440-746, Republic of Korea
a r t i c l e
i n f o
Article history: Received 18 June 2009 Received in revised form 27 December 2009 Accepted 7 February 2010 Available online 9 March 2010 Keywords: Enzymatic pretreatment Ethanol Microalgae Chlamydomonas reinhardtii
a b s t r a c t The production of ethanol from feedstock other than agriculture materials has been promoted in recent years. Some microalgae can accumulate a high starch content (about 44% of dry base) via photosynthesis. Algal biomass, Chlamydomonas reinhardtii UTEX 90, was converted into a suitable fermentable feedstock by two commercial hydrolytic enzymes. The results showed that almost all starch was released and converted into glucose without steps for the cell wall disruption. Various conditions in the liquefaction and saccharification processes, such as enzyme concentration, pH, temperature, and residence time, have been investigated to obtain an optimum combination using the orthogonal analysis. As a result, approximately 235 mg of ethanol was produced from 1.0 g of algal biomass by a separate hydrolysis and fermentation (SHF) method. The main advantages of this process include the low cost of chemicals, short residence time, and simple equipment system, all of which promote its large-scale application. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction From a greenhouse standpoint, renewable fuels such as ethanol are considered to be excellent alternative clean-burning fuels to gasoline in the future as the combustion products are environmentally safe (Kito-Borsa et al., 1998). Today, the most common renewable fuel is ethanol, derived mainly from glucose or starch sources of agricultural stock (Gray et al., 2006; Nigam and Singh, 1995). The human demand for food, however, has yet to be met. To solve both the energy and food problem, there has been increasing interest and worldwide studies in producing bioethanol from algal biomass, the alternative to agricultural stock (Baras et al., 2002; Kim and Dale, 2003; Sánchez and Cadona, 2008). C. reinhardtii, a unicellular green alga, is well-known as a photoautotrophic microorganism having a great ability to fix CO2 and accumulate a high content of stored polysaccharides, mainly starch, in complex multilayered cell walls (Hall and Rao, 1994; Hirano et al., 1997). This bears a strong structural and functional resemblance to higher plant storage starch (Libessart et al., 1995). With a high growth rate, the microalgae can be easily cultured at high yields and low costs utilizing an unlimited energy source, sunlight (Hirayama et al., 1998; Sze, 1998). These advantages allow the microalgae to be preferentially selected as a safe and prospective feedstock for bioethanol production by Saccharomyces cerevisiae. Prior to ethanol fermentation, the feedstock needs to be processed by enzymatic or acidic pretreatment technology in order * Corresponding author. Tel.: +82 31 290 7341; fax: +82 31 290 7272. E-mail address: [email protected] (S.J. Sim). 0960-8524/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2010.02.026
to release fermentable sugars. Since the first application of microbial enzyme in the food industry in the early 1960s, a great deal of effort has been made to replace traditional acid hydrolysis with enzymatic hydrolysis in almost all glucose production due to higher yields under mild conditions, less by-products, and no corrosion issues (Balat et al., 2008). Several hydrolytic enzymes possessing improved activity have been developed and used extensively in the starch processing industry for liquefaction and saccharification (Olsen, 2004; Schäfer et al., 2007). To this end, a commercial amyloglucosidase, AMG 300L, was produced by a genetically modified strain of Aspergillus. The main obstacle of enzymatic hydrolysis is that intercellular starch granules are bound within rigid cell walls (Libessart et al., 1995), thus a biomass pretreatment step is needed to break down the cell wall to release polysaccharides such as starch, structural carbohydrates, and other nutrients, prior enzymatic hydrolysis and fermentation steps. The cell wall of C. reinhardtii contains glycoproteins as the predominant constituents in its extracellular matrix (Sze, 1998). A commercial a-amylase derived from Bacillus licheniformis, Termamyl 120L, shows a protease activity particular to the degradation of glycoproteins within cell walls (Imam and Snell, 1987). Starch hydrolysis involves liquefaction and saccharification of the starch. The objective of this study was to optimize conditions to apply the starch-degrading enzymes sequentially to pretreatment of the algal biomass. The optimal conditions required for the enzymatic hydrolysis of starch, such as enzyme concentration, temperature, pH, and residence time, were selected by the orthogonal analysis method, which weighed the effects on both hydrolysis and ethanol fermentation; a maximum rate of ethanol
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
production could then be achieved. Two-step enzymatic hydrolysis by commercially available a-amylase and amyloglucosidase would yield lower energy consumption, lower content of non-glucosidic impurities, and thus, a much better suitability for ethanol production (Mojovic´ et al., 2006). To the best of the authors’ knowledge, this is the first report regarding enzymatic pretreatment of microalgal biomass optimized for ethanol fermentation. The work herein demonstrates that the enzymatic hydrolysate from the microalgae is potentially useful as a feedstock for fermentation by the ethanolproducing yeast, S. cerevisiae S288C. 2. Methods 2.1. Algal biomass The medium used for the culture of the green algae, C. reinhardtii UTEX 90, was 1.6 L of tris–acetate–phosphate (TAP) medium, which was prepared as reported in our previous work (Thu et al., 2009). A cell culture was carried out during 96 h at 23 °C and 130 rpm in a 2.5 L photo-bioreactor with 10% inoculation of seed under continuous illumination (450 lE m 2 s 1) by white fluorescent lamps. Light intensity in the photo-bioreactor was measured with an Li-250 Li-Cor quantum photometer (Lambda Instrument Corp., Lincoln, USA). The pH of the medium was automatically adjusted to the range of 7.0–7.4 by the feeding of 1.0 M acetic acid. Cell growth of the algae was monitored by measuring optical density (OD) at 600 nm. Dry cell weight (DCW) was determined by an 80 °C oven drying method using filter paper (GC/F, City, England). The cells harvested by centrifugation were characterized and used for pretreatment of the feedstock biomass. 2.2. Enzymes Two enzymes for liquefaction and saccharification, thermostable a-amylase of B. licheniformis origin (EC 3.2.1.1, Termamyl 120L) and amyloglucosidase from Aspergillus niger (EC 3.2.1. 3, AMG 300L), were purchased from Novo Nordisk (Gentafte, Denmark). The enzymatic activities of these enzymes were 120 KNU g 1 and 300 AGU mL 1, respectively, as defined by Novo Nordisk. The Kilo Novo a-amylase Unit (KNU) is defined as the amount of enzyme that can hydrolyze 5.26 g of soluble starch per hour at pH 5.6 and 37 °C. The Amyloglucosidase Novo Unit (AGU) is the amount of enzyme which cleaves 1.0 lmol of maltose/min at pH 4.3 and 25 °C. 2.3. Enzyme pretreatment of algae Enzyme pretreatment consisted of two parts, liquefaction of algal biomass by a-amylase (Termamyl 120L) and saccharification by amyloglucosidase (AMG 300L). The harvested algal biomass were washed once and slurried in water at a 5% solid to liquid ratio (w/v), mixed with Termamyl 120L at a final concentration between 0.0001% and 0.02% (v/w), and pretreated as follows. Hydrolysis reactions were performed in capped flasks in a thermostated water bath. The pH of the mixture was adjusted to pH 6.0, a value known as the optimum (Richardson et al., 2002), with 10% H2SO4. The protease activity of the Termamyl 120L, the optimum temperature of which is in the range of 50–60 °C, was activated by increasing the incubation temperature from 25 °C to different temperatures (70, 80, 90, 100 °C) for 30 min, and thenceforth the algal biomass was hydrolyzed by the a-amylase activity of Termamyl 120L by maintaining it at the optimal temperature for 10–60 min. For subsequent saccharification after liquefaction, The pH of the mixture was adjusted to a range of 4.5–5.5, the temperature was reduced to a range of 50–65 °C, and AMG 300L at concentrations of
5331
0.001–0.3% (v/w) was supplied for 5–60 min. By analyzing parameters such as temperature, residence time, enzyme concentration, and pH by orthogonal arrays, two processes of the enzymatic hydrolysis were optimized. Duplicate batches were run to verify results. 2.4. Ethanol fermentation by yeast To test fermentability of the pretreated algal biomass, separate hydrolysis and fermentation (SHF) was performed using ordinary compressed baker’s yeast, S. cerevisiae S288C as reported in our previous work (Thu et al., 2009). The solid fraction of the pretreated algal biomass was removed by centrifugation (4000g, 10 min). After the aerobically pre-cultured yeast cells were inoculated at 10% of volume size to the liquid fraction of pretreated algal biomass, the yeast cells were cultured anaerobically at 30 °C for 40 h at a rotation of 160 rpm. The concentration of the produced ethanol was analyzed by HPLC under the same conditions as the analysis of monomeric sugars in the pretreatment. 2.5. Analytical methods The cells were disrupted by vortexing with glass beads in a methanol solution. The methanol was then vaporized by placing the cell lysates in the oven at 80 °C for 6 h. After the addition of 72% perchloric acid, the starch was extracted at 0 °C for 12 h and filtered through a 0.45 lm filter. The contents of the starches in the filtrate were determined by an iodo-starch reaction method (Hirokawa et al., 1982). The total cell carbohydrate was analyzed by a colorimetric method using an anthrone reagent. The algal cell pellets were mixed with 67% sulfuric acid by vortexing for 30 min and then reacted with the anthrone reagent for 5 min at 100 °C. The OD of the resulting solution was measured at 630 nm. The intracellular monomeric sugar content of the C. reinhardtii was determined by high performance anion-exchange chromatography (HPAEC; DX-300 series chromatography system, Dionex, USA). The effluent was monitored with pulsed amperometric detection detector (PAD, Dionex, CA, USA). The cell pellets were washed two times with autoclaved distilled water and freezedried. Sample of 2.5 mg dry cell was dissolved in 1 ml trifluoroacetic acid (TFA; Sigma, USA). After hydrolysis at 100 °C for 4 h, the sample was cooled to room temperature and the volatile acid was removed by centrifugal evaporation (Speed-Vac; SPD1010115, Savant Instruments, USA) under a stream of nitrogen gas. The dried sample was dissolved in 10.0 ml of water and filtered through 0.2 lm syringe filter. The standard sugars were also treated in the same way as described for the samples in order to identify and quantify. The 10 ll of filtered samples were injected into a CarboPac PA-1 anion-exchange column (0.4 250 mm, Dionex, CA, USA) that was pre-equilibrated in 18 mM NaOH. Chromatographic separation of the monomeric sugars from the samples was achieved in the isocratic mode with 18 mM NaOH at a flow rate of 1.0 ml/min in 20 min. The content of the intracellular proteins was determined by the Bradford method (1976). The cells were disrupted by sonication in PBS buffer (pH 7.3), the supernatant was colored by mixing with the Bradford reagent (Sigma Chemical Co., St. Louis, MO, USA), and the OD of it was measured at 595 nm. The concentrations of the oligosaccharide (dextrin) and monomeric sugar (glucose) in the liquid fraction of the enzymatic hydrolysate and ethanol produced by yeast were quantified by an Agilent 1200 high-pressure liquid chromatography (HPLC) system equipped with a quaternary solvent delivery system (Agilent Technologies, Palo Alto, CA, USA), an autosampler (No. 60044; Spark, Holland, Emmen, Netherlands), a refractive index detector (Acme
5332
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
9000; Younglin instrument, Seoul, Korea), and a computer software-based integration system (Agilent Chemstation; Agilent Technologies). The samples were separated on ion-moderated partition chromatography columns (Aminex HPX-87P, 300 mm 7.8 mm; Bio-Rad, Richmond, CA, USA) using deionized water (Sigma, USA) as the mobile phase at a flow rate of 0.6 ml/min. The column was maintained at 60 °C. Target compounds were detected using a refractive index detector and quantified by comparison with authentic standards: glucose (Amresco, USA); dextrin hydrate (Kanto, Japan); ethanol (J.T. Baker, USA). Each measurement was repeated at least three times and averaged. 3. Result and discussion 3.1. Algal biomass content After acetic acid fed-batch cultivation in a 2.5 L photo-bioreactor for 4 d, the Chlamydomonas biomass was harvested and concentrated at 12.4 g L 1 (dry cell weight). The content of the main components was determined by chemical analysis and presented in Table 1. It was found that the algae contained a high content of carbohydrates (59.7%) and various proteins (9.2%), useful sources of carbon and nitrogen for yeast fermentation. Therefore, it was supposed that the additional nitrogen source was unnecessary. Monomeric sugars in TFA-hydrolyzed microalgal biomass were also analyzed by HPAEC (Table 1). As a result of the substantial starch accumulation (43.6%) achieved under controlled conditions of the fed-batch culture, the most predominant monosaccharide was glucose (44.7%), which S. cerevisiae strain can readily ferment, while the total composition of the other monosaccharides like L-fucose, L-rhamnose, D-arabinose, D-galactose, and D-mannose was 7.3%. The algal biomass is cheaper and contains less intracellular glucose than the other biomass sources (sugar cane, wheat, rice, corn, and cassava) (Table 1). After quantitative determination of the cell constituents, the algal biomass was adjusted at a 5% solid liquid ratio (w/v) and ready to be used in further experiments. 3.2. Liquefaction In the enzymatic pretreatment of algal biomass, the first step is liquefaction of starch with a thermostable a-amylase (Termamyl 120L), which catalyzes the hydrolysis of the internal a-D-(1–4)glucosidic linkages in starch in a random manner (Montesinos and Navarro, 2000). The products resulting from the liquefaction
Table 1 Characterization of cell mass composition for Chlamydomonas reinhardtii UTEX 90 obtained from the pH-stat fed-batch culture by feeding 1.0 M acetic acid under continuous illumination (450 lE m 2 s 1). All analyses were conducted in triplicate and the error limits were estimated. Components
Composition based on dry cell mass (%, w/w)
Proteins
9.2 ± 0.6
Total carbohydrates
59.7 ± 0.5
(Starch)
(43.6 ± 1.4)
D-Glucose
44.7 ± 0.8
L-Fucose
0.4 ± 0.01
L-Rhamnose
0.9 ± 0.02
D-Arabinose
1.9 ± 0.04
D-Galactose
2.7 ± 0.04
D-Mannose
1.4 ± 0.03
Others
31.1
Analytical method
Colorimetric method with anthrone (Iodo-starch reaction method) Chromatographic separation by HPAEC
–
process were oligosaccharides with three or more 1 ? 4-a-linked D-glucose units, generally denoted as dextrin produced from starch (Petrova et al., 2000; Schäfer et al., 2007). Liquefaction is a process that converts a suspension of starch granules into a partially hydrolyzed starch solution of low viscosity suitable for subsequent processing. Its effectiveness depends highly on parameters such as enzyme concentration, temperature, and residence time. Therefore, the main focus was determination of how the liquefaction process is influenced by various conditions; the optimal conditions had to be determined for this process to obtain the highest yield of product. The degree of liquefaction of the algal suspension was evaluated by HPLC analysis of the amount of dextrin released from the samples. To assess the effect of enzyme concentration, liquefaction experiments were carried out at six different levels, from 1 10 4 to 2 10 2 (%, v/w), at a fixed residence time (30 min) and constant temperature (90 °C). Fig. 2 shows the concentration of dextrin obtained as a function of various operation parameters. The incubation of the algal suspension with Termamyl 120L, particularly at enzyme concentrations below 5 10 3 (%, v/w), significantly increased the concentration of the released dextrin, compared with the control of no enzyme (Fig. 1a). However, further increase in enzyme concentration appeared to not be required to establish an economic process in terms of cost. A reaction progress curve could be obtained by mixing an algal suspension with 0.001% of the Termamyl 120L enzyme and measuring the resulting dextrin generated at 90 °C over a period of time (10–60 min). During the early 10 min, the enzyme-catalyzed hydrolysis reaction exhibited linear kinetics and the highest initial rate of reaction, whereby the amount of the product dextrin increased linearly with residence time (Fig. 1b). This indicated either that the level of the enzyme was low enough for the curve to reach a plateau early, due to depletion of the substrate (algal biomass), or that it was possible to maintain the initial conditions of liquefaction over a given period of time. Only a marginal change in the dextrin was observed during the late period (30–60 min). It is certain that a longer residence time would generally be more beneficial for liquefaction. However, it was more reasonable, from an economic vantage, that 30 min of residence time should be a suitable choice for the liquefaction process. The effect of reaction temperature (70, 80, 90, 100 °C) on liquefaction performed with 0.001% enzyme for 30 min is shown in Fig. 1c. As expected, the rate of reaction, and thus the amount of produced dextrin, increased with an increase in reaction temperature. The most dramatic effect of liquefaction time was observed between 70 and 80 °C, where the released dextrin increased from 10.3 g L 1 at 70 °C to 16.8 g L 1 at 80 °C. However, temperature change seemed to be less effective above 80 °C. A further increase of 10 °C (from 80 to 90 °C) led to a slight increase of dextrin (approximately 0.5 g L 1) and furthermore, there was no significant difference between 90 and 100 °C. It was found that the optimum temperature of the liquefaction process was near 90 °C. Optimum combinations of liquefaction parameters (enzyme concentration, temperature, residence time) were obtained by analysis at three different levels using the orthogonal analysis technique (RenJie, 2008). Based on the provisional optimum values (0.005% enzyme, 30 min, 90 °C) obtained from the above experiments, the three levels for each parameter were determined. Nine of the experiments concomitant with different liquefaction parameters were collectively conducted. Table 2 shows the concentration of dextrin released in each hydrolysis experiment, the average concentration (k) of dextrin for the parameter chosen at a level, and the maximum difference (R) between the three k values. For example, k2 value for enzyme at level 2 can be calculated as the average of dextrin concentrations (16.15, 16.94, 25.21) in experiment 2, 4, 9. A higher k value and a higher R value would indicate
5333
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
30
a
b
20
Dextrin (g L-1)
Dextrin (g L-1)
25 20 15 10
15
10
5 5 0 0.000
0.005
0.010
0.015
0
0.020
0
10
20
Enzyme concentration (%, v/w)
Dextrin (g L-1)
20
30
40
50
60
Time (min)
c
15
10
5
0
70
75
80
85
90
95
100
Temperature ( oC) Fig. 1. Effects of various factors on the liquefaction process by the commercial thermostable a-amylase, Termamyl 120L. (a) Effect of enzyme concentration at a fixed residence time (30 min) and constant temperature (90 °C); (b) effect of liquefaction time with 0.001% enzyme at 90 °C; (c) effect of liquefaction temperature with 0.001% enzyme at 30 min.
Table 2 Orthogonal analysis of the liquefaction process. Number
Temperature (°C)
Time (min)
Level 1 Level 2 Level 3
70 80 90
10 20 30
0.001 0.005 0.01
70 70 70 80 80 80 90 90 90
10 20 30 10 20 30 10 20 30
0.001 0.005 0.01 0.005 0.01 0.001 0.01 0.001 0.005
15.13 18.65 21.18 6.05
16.73 17.85 20.39 3.65
Experiment Experiment Experiment Experiment Experiment Experiment Experiment Experiment Experiment k1 k2 k3 R
1 2 3 4 5 6 7 8 9
a preferred level for the chosen parameters and a greater influence of that parameter, respectively. Therefore, optimum combinations of the liquefaction parameter can be determined from the maximum value of the three k values (k1, k2, k3). In this study, it could be tentatively suggested that the optimum temperature should be at level 3 (90 °C), because k3 is the highest (21.18 g L 1) of the three k values, and similarly that the optimum residence time should be 30 min and the optimum enzyme concentration 0.01% (v/w). Using such a combination, as shown in Fig. 1a, 27.09 g L 1 of dextrin could be obtained. Meanwhile, focusing upon the nine
Enzyme (%, v/w)
Dextrin (g L
1
)
10.13 16.15 19.12 16.94 22.18 16.83 23.13 15.21 25.21
14.06 19.43 21.48 7.42
experiments combining the three levels, a maximum dextrin concentration was determined, 25.21 g L 1, in the experiment performed at temperature 90 °C with 0.005% enzyme at 30 min. Though twice as much enzyme (0.01% enzyme) was introduced, increase in the dextrin concentration was only 1.88 g L 1, very small. To establish an economic pretreatment process, as mentioned above, the smallest amount of enzyme should be used as possible. Therefore, the choice of the former combination (0.005% enzyme, 90 °C, 30 min) as the optimum conditions for liquefaction was a natural one.
5334
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
The orthogonal analysis also revealed the relative importance of each parameter in influencing dextrin yield. With an R value of 7.42, the enzyme concentration seemed to be the most critical factor, followed by temperature (R value 6.05) (Table 2). With an R value of only 3.65, the variation in time appeared to have the least influence on the liquefaction process. Cell disruption, in general, is an essential, initial step regarding the biomass treatment process. Judging from the results in this liquefaction process, it was certain that the Termamyl 120L was successfully applied to interrupt the cell wall, release starch out of algal biomass having thick cell walls, and hydrolyze dextrin from the starch. This was because Termamyl 120L is a powerful complex enzyme harboring the protease activity capable of the necessary cell wall-degrading activity (Nigam and Singh, 1995; Imam and Snell, 1987), in addition to the a-amylase activity that degrades the bulk of the starch rapidly at 100 °C. This was also because the residence time of increasing temperature from room temperature to liquefaction temperatures was enough for cell wall degradation by the protease. Thus, without a separate pretreatment step, enzymatic hydrolysis rendered or disrupted the cell wall structure of the biomass and made it more accessible at reasonable rates and yields. 3.3. Saccharification and ethanol fermentation Using the starch saccharifying enzyme, amyloglucosidase (AMG 300L), which catalyzes the hydrolysis of a-D-(1–4) and a-D-(1–6)glucosidic bonds of oligosaccharides in the starch liquefied under optimal conditions, a saccharification process was subsequently performed. The degree of saccharification was evaluated by the 25
amount of glucose hydrolyzed out of the liquefied starch, which was analyzed by HPLC. The effective usefulness of the pretreated algal biomass as a medium for yeast growth was further estimated by checking ethanol production from the SHF process (Amutha and Gunaskaran, 1994). The primary advantage of SHF is that hydrolysis and fermentation occur at optimum conditions (Balat et al., 2008). Fig. 2 shows the effects of enzyme concentration, pH, residence time, and temperature on the saccharification process. The effect of enzyme concentration was firstly investigated at fixed conditions (pH 5.5, temperature 55 °C, residence time 30 min), and its effect was found to be very significant. There is no doubt that the increase in enzyme concentration would normally enhance hydrolysis and, therefore, increase ethanol yield. As demonstrated in Fig. 2a, the most dramatic effect of enzyme concentration on ethanol yield was observed between 0.001% and 0.05% (v/w). At an enzyme concentration above 0.05% (v/w), the slope of the ethanol yield line reduces gradually. Enzyme concentration is one of the factors contributing to production cost, thus proper amounts give the best results at a reduced expense. The effect of pH is presented in Fig. 2b (enzyme concentration 0.1% (v/w), temperature 55 °C, residence time 30 min). It was found that pH had a significant influence on the saccharification and fermentation processes. The maximum glucose and ethanol yield was observed at pH 4.5, a value also known as the optimum pH condition for yeast fermentation (Yong et al., 1980). Residence time seemed to maintain a positive effect over the saccharification process (Fig. 2c). Effects of small variations in time, from 0 to 60 min, were estimated at other fixed conditions (0.1% enzyme, 55 °C, pH 5.5). The slope in the initial 5 min showed the highest rate of reaction. As a result, the ethanol yield increased
a
c
20
Concentration (g L-1)
Concentration (g L-1)
20
15
10
Glucose Ethanol
5
0 0.00
15
10
Glucose Ethanol
5
0
0.05
0.10
0.15
0.20
0.25
0
0.30
10
20
b
20
40
50
60
d Glucose Ethanol
14
Concentration (g L-1)
18
Concentration (g L-1)
30
Time (min)
Enzyme concentration (%, v/w)
16 14 12 10
12
Glucose Ethanol 10
8
8 3.8
4.0
4.2
4.4
4.6
4.8
pH
5.0
5.2
5.4
5.6
48
50
52
54
56
58
60
62
64
66
Temperature ( oC)
Fig. 2. Effects of various factors on the saccharification and fermentation processes by commercial thermostable amyloglucosidase, AMG 300L. (a) Effect of enzyme concentration; (b) effect of pH; (c) effect of saccharification time; (d) effect of saccharification temperature.
5335
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
more upon extending residence time, particularly between 45 and 60 min, than glucose production did. This reveals that the saccharification process was continuously carried out at the fermentation stage. The optimal temperature of the saccharification process was tested between 50 and 65 °C (enzyme concentration of 0.1% (v/ w), pH 5.5, residence time 30 min); results are shown in Fig. 2d. It is apparent that too low or too high temperatures were not preferable. The highest yield of glucose concentration was observed at 55 °C. Therefore, maximum ethanol could also be obtained at this point. This result was compatible with the value recommended by the supplier. Finally, saccharification conditions such as enzyme concentration, temperature, pH, and the residence time, were analyzed by the orthogonal method in the same way as in the above liquefaction and optimized in relation to maximization of the ethanol yield. The above four factors were examined at three different levels (Table 3). Based on the ethanol obtained after fermentation, analysis showed that all four factors played important roles in the saccharification and fermentation processes. Of the four parameters, enzyme concentration and pH appeared to have a more dominant influence (R values of 2.47 and 1.07, respectively). With smaller R values, temperature (R value 0.48) and time (R value 0.15) seem to be relatively less critical. From this orthogonal analysis, it was found that the optimal conditions for saccharification were 55 °C, 0.2% (v/w) enzyme concentration, pH 4.5, and a 45 min residence time. But, in experiment five, combining three levels, the highest ethanol yield obtained was 11.73 g L 1 with 1.85 g L 1 of the dextrin remaining (data not shown). Therefore, it was reasonable to choose the latter combination (0.2% enzyme, 55 °C, 30 min) as the optimum conditions for liquefaction, due to the shorter residence time. This indicates that under these condi-
tions, approximately 94% of the carbohydrates were hydrolyzed into glucose and shows that algal biomass can be used for ethanol production without further pretreatment. The majority of processing schemes for hydrolyzing biomass into glucose to be used as feedstock, which have been developed over the years, utilizes either enzymes or acid of varying concentrations. We reported a study on acid treatment in our previous work (Thu et al., 2009). Compared with the ethanol production efficiency obtained by the acid treatment, the one obtained currently by enzymatic treatment is a bit lower, 29.2% and 23.5%, respectively. Nevertheless, enzymatic process would be quite promising, particularly due to higher yields under mild conditions, less byproducts, no corrosion issues, and lower utility cost (Balat et al., 2008). Really, the ethanol production using the microalgal biomass was performed at the highest level of efficiency, compared with those obtained from the different biomass sources (Table 4). Using algae as a feedstock has many advantages over other biomass sources (sugar cane, wheat, rice, corn, and cassava), because of its ability to be produced quickly and cheaply (Balat and Balat, 2009; Lynd et al., 1999). Recently, enzymes for production of fuel ethanol from biomass have been manufactured at substantially reduced cost by modern biotechnology (Schäfer et al., 2007). Therefore, it is desirable to focus on the potential of searching the mesophilic or new activity of enzymes and their application towards the SHF or the simultaneous saccharification and fermentation (SSF) process (Sánchez and Cadona, 2008), thus improving the efficiency of the process even further. 4. Conclusions This work demonstrates that commercially available a-amylase and glucoamylase were applicable to the pretreatment of algal bio-
Table 3 Orthogonal analysis of the saccharification and fermentation processes. Number
Temp. (°C)
Time (min)
Level 1 Level 2 Level 3
50 55 60
15 30 45
0.05 0.1 0.2
4.5 5 5.5
50 50 50 55 55 55 60 60 60
15 30 45 15 30 45 15 30 45
0.05 0.1 0.2 0.1 0.2 0.05 0.2 0.05 0.1
4.5 5 5.5 5.5 4.5 5 5 5.5 4.5
8.37 10.11 10.84 2.47
10.41 9.58 9.34 1.07
Experiment Experiment Experiment Experiment Experiment Experiment Experiment Experiment Experiment
1 2 3 4 5 6 7 8 9
k1 k2 k3 R
9.70 10.05 9.57 0.48
Enzyme (%, v/w)
9.71 9.75 9.86 0.15
pH
Ethanol (g L
1
)
8.87 9.81 10.42 9.88 11.73 8.54 10.38 7.71 10.63
Table 4 Comparison of ethanol yields from different biomass sources. Biomass
Conversion to sugar or starch (%)
Ethanol yield (g ethanol g
Sugar cane Sugar beet Cassava Sweet sorghum Corn Wheat Cane bagasse Corn stover Microalgae Microalgae
12.5 – 25 14 69 66 – – – 57
0.055 0.079 0.118 0.063 0.324 0.308 0.111 0.260 0.292 0.235
1
biomass)
References Moreira and Goldemberg (1999) Berg (2001) Wang (2002) Wang (2002) Wang (2002) Wang (2002) Moreira (2000) Kadam and McMillan (2003) Thu et al. (2009) This study
5336
S.P. Choi et al. / Bioresource Technology 101 (2010) 5330–5336
mass for yeast fermentation. Due to proteases mingled with aamylase, cell disruption could be carried out without additional energy, time, and cell wall degrading enzymes. The other intracellular components released from algal cells will be beneficial in subsequent fermentations, regardless of how much is required. With the help of orthogonal analysis, the optimum conditions for the SHF process were identified as 0.005% a-amylase, 90 °C, and 30 min for liquefaction, and 0.2% glucoamylase, 55 °C, pH 4.5, and 30 min for saccharification. By yeast fermentation, 235 mg of ethanol could be obtained from the hydrolyzed starch of 1.0 g algal biomass. It has been suggested that algal biomass could be a potential feedstock for ethanol production on large-scale. Acknowledgements This research was supported by a grant (DG2-201) from the Carbon Dioxide Reduction and Sequestration Research Center, one of the 21st Century Frontier Programs that was funded by the Ministry of Science and Technology of the Korean Government. References Amutha, R., Gunaskaran, P., 1994. Simultaneous saccharification and fermentation of cassava starch using Zymomonas mobilis. J. Microbiol. Biotechnol. 9, 22–34. Balat, M., Balat, H., 2009. Recent trends in global production and utilization of bioethanol fuel. Appl. Energy 86, 2273–2282. Balat, M., Balat, H., Oz, C., 2008. Progress in bioethanol processing. Prog. Energy Combust. Sci. 34, 551–573. Baras, J., Gac´eša, S., Pejin, D., 2002. Ethanol is a strategic raw material. Chem. Ind. 56, 89–105. Berg, C., 2001. World Fuel Ethanol. Analysis and Outlook. F.O. Licht.
Imam, S.H., Snell, W.J., 1987. Degradation of the framework of the Chlamydomonas cell wall by proteases present in a commercially available a-amylase preparation. Appl. Environ. Microbiol. 53, 1701–1704. Kadam, K.L., McMillan, J.D., 2003. Availability of corn stover as a sustainable feedstock for bioethanol production. Bioresour. Technol. 88, 17–25. Kim, S., Dale, B.E., 2003. Global potential bioethanol production from waste crops and crop residues. Biomass Bioenergy 26, 361–375. Kito-Borsa, T., Pacas, D.A., Selim, S., Cowley, S.W., 1998. Properties of an ethanol– diethyl ether–water fuel mixture for cold-start assistance of an ethanol-fueled vehicle. Ind. Eng. Chem. Res. 37, 3366–3374. Libessart, N., Maddelein, M.L., Koornhuyse, N., Decq, A., Delrue, B., Mouille, G., D’Hulst, C., Ball, S., 1995. Storage, photosynthesis and growth: the conditional nature of mutations affecting starch synthesis and structure in Chlamydomonas. Plant Cell 7, 1117–1127. Lynd, L.R., Wyman, C.E., Gerngross, T.U., 1999. Biocommodity engineering. Biotechnol. Progr. 15, 777–793. Mojovic´, L., Nikolic´, S., Rakin, M., Vukasinovic´, M., 2006. Production of bioethanol from corn meal hydrolyzates. Fuel 85, 1750–1755. Montesinos, T., Navarro, J.M., 2000. Production of alcohol from raw wheat flour by amyloglucosidase and Saccharomyces cerevisiae. Enzyme Microb. Technol. 27, 362–370. Moreira, J.S., 2000. Sugarcane for energy – recent results and progress in Brazil. Energy Sustain. Dev. 4, 43–54. Moreira, J., Goldemberg, J., 1999. The alcohol program. Energy Policy 27, 229–245. Nigam, P., Singh, D., 1995. Enzyme and microbial systems involved in starch processing. Enzyme Microb. Technol. 17, 770–778. Olsen, H.S., 2004. Enzyme at work. Published by Novozymes A/S.