ABS

TRAC

T

J-FOR Journal of Science & Technology for Forest Products and Processes: VOL.1, NO.3, 201154

The research reported here has investigated consolidated bioprocessing (CBP) of purified cellulose from forest wood (PWC), hemp (PHC), flax (PFC), rice straw (PRC), and beverage-cup cellulose (BCC) compared to reagent-grade α-cellulose (AC) for ethanol and hydrogen production using Clostridium thermocellum. Net production of hydrogen and ethanol was highest for PWC (10.63 µmol/mL and 7.50 mM) compared to AC (8.54 µmol/mL and 6.57 mM respectively). Rates observed during the exponential phase of growth on PWC were compa-rable to those observed on AC, PHC, and PFC with the same pre-treatment. End-product yields were dependent on the intrinsic cellulosic content and amorphous cellulose content in test substrates. Rates of hydrogen and ethanol production were similar in the early log phase, but varied in the mid-log and late log phases. Overall, pre-treated and processed forest wood cellulose (PWC, AC, and BCC) produced higher yields than pre-treated agricultural residues (PHC, PFC, and PRC).

VALERY AGBOR, WARREN BLUNT, NAZIM CICEK, RICHARD SPARLINGALEX BERLIN, DAVID B. LEVIN*

PROCESSING OF CELLULOSIC FEEDSTOCKS FOR ETHANOL AND HYDROGEN PRODUCTION

Currently, the United States and Brazil are the leading producers of bioethanol, generating approximately 26 and 19 billion litres respectively and accounting for 89% of world ethanol production. Corn etha-nol production is currently capped at 57 billion litres in the United States, but the renewable fuel objective set by the U.S. en-ergy bill for 2022 is 136 billion litres. The remainder (79 billion litres) is expected to come from “advanced biofuels,” most of which are to be derived from lignocellu-losic biomass [1].

In North America, ethanol and hy-drogen derived from cellulosic biomass could be potential commercial energy car-riers for transportation or portable power. As biofuels plants become biorefineries, achieving a low-cost route for industrial production from cellulosic biomass is of crucial importance for investor and public endorsement. So far, separate hydrolysis and fermentation (SHF), simultaneous saccharification and fermentation (SSF), and simultaneous saccharification and co-fermentation (SSCF) have been used

INTRODUCTIONfor industrial ethanol production [2-3], while hydrogen has been produced pri-marily by electrolysis of water, gasification of oil or methane, and steam reformation of natural gas [4-5]. Current bioprocessing strategies for ethanol production have not yet become cost-competitive with natural gas [2, 6-8], and all methods for hydrogen production are energy-intensive [5].

Direct microbial conversion (DMC), also referred to as direct cellulose fer-mentation or consolidated bioprocessing (CBP), is a less energy-intensive method

*Contact: levindb@cc.umanitoba.ca

DAVID B. LEVINDepartment of Biosystems Engineering, University of Manitoba, E2-376 EITC, Winnipeg, MN Canada, R3T 5V6

WARREN BLUNTDepartment of Biosystems Engineering, University of Manitoba, E2-376 EITC, Winnipeg, MN Canada, R3T 5V6

NAZIM CICEKDepartment of Biosystems Engineering, University of Manitoba, E2-376 EITC, Winnipeg, MN Canada, R3T 5V6

RICHARD SPARLINGDepartment of Microbiology, University of Manitoba, Winnipeg, MNCanada, R3T 2N2

VALERY AGBORDepartment of Biosystems Engineering, University of Manitoba, E2-376 EITC, Winnipeg, MN Canada, R3T 5V6

ALEX BERLINNovozymes, Inc.,1445 Drew Ave, Davis, CA, USA, 95618

J-FOR Journal of Science & Technology for Forest Products and Processes: VOL.1, NO.3, 2011 55

SPECIAL BIOREFINERY ISSUE

of biological hydrogen production [5,9-10] and a potential low-cost approach to cellulosic ethanol production because of the avoided cost of purchasing exogenous enzyme cocktails and using multiple reac-tors or holding vessels for the fermenta-tion process [8,11-12]. Even though CBP offers the greatest potential to become the lowest-cost pathway to biofuels, much re-mains to be elucidated with respect to (i) feedstock quality, (ii) extent of pre-treat-ment required, and (iii) the fermentability of the pre-treated biomass to produce eth-anol yields comparable to those observed from SSF or SSCF of lignocellulosics. Al-though CBP of lignocelluloses is an attrac-tive alternative to current industrial meth-ods for ethanol (i.e., SHF, SSF, SSCF) and hydrogen production, the switch to using lignocellulosic feedstocks as a substrate requires characterization and bioprocess evaluation of pre-treated feedstocks used as a source of carbon. Previous studies by the authors and other researchers have identified the major metabolic end-prod-ucts of fermentation with Clostridium ther-mocellum. The present research has assessed the fermentability of five test cellulosic feedstocks in comparison to commercial reagent grade α-cellulose used as a posi-tive control. The objective of this work was to assess the fermentability (i.e., ease of converting substrate to end-products) of various types of cellulosic biomass by Clostridium thermocellum in a single-step fermentation process using reagent-grade α-cellulose as a positive control.

MATERIALS AND METHODS

SubstratesThe test substrates included: (1) organo-solv-pre-treated hemp cellulose (PHC) originating from the Emerson hemp com-pany in Manitoba; (2) organosolv-pre-treated wood cellulose (PWC) from spruce pulp harvested in British Columbia, from Lignol Innovations in Burnaby BC; (3) organosolv-pre-treated flax shive cellulose (PFC) generously provided by Schweitzer Mauduit Canada, Inc., in Winkler MB; (4) sub-critical carbon dioxide-pre-treated rice straw cellulose (PRC) from Mitsubishi Heavy Industries, Japan; and (5) beverage-cup cellulose from Tim Horton’s Canada (BCC). Organosolv pre-treatment was conducted with 50% ethanol for 55 min at 195°C using 1.5% acid as catalyst at a 10:1 liquid:insoluble biomass ratio using the Lignol Organosolv® process [13-14]. The hydrothermal pre-treatment with hot water was conducted by Mitsubishi Heavy Industries, Ltd., Japan. These water-insol-uble substrates are referred to here as pu-rified rice cellulose (PRC), purified wood cellulose (PWC), purified hemp cellulose (PHC), purified flax cellulose (PFC), and beverage-cup cellulose (BCC).

Compositional analyses of the bio-mass substrates to determine moisture content, dry matter, acid detergent fibre (ADF), neutral detergent fibre (NDF), ash, soluble protein, cellulose, hemicellu-lose, and lignin were conducted at the Uni-versity of Saskatchewan, Feeds Innovation

Institute, and are reported on a dry-mass basis (Table 1).

A Retsch Rotor Beater mill, SR200, equipped with a 34 mesh, 0.5-mm aperture sieve was used to mill the substrates. All the milled samples were transferred into zip-lock bags and stored at room tempera-ture. Alpha (α)-cellulose (C8002) obtained from Sigma Aldrich (St. Louis, U.S.A.) was used as a positive control. All substrates were added to fermentation reactions at 2 g/L substrate loading. Negative controls were established using 1191 medium in-oculated with Clostridium thermocellum, but without any substrate.

Microorganisms and MediaClostridium thermocellum strain 1237 was obtained from Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ; Braunschweig, Germany) and maintained in the authors’ lab by culturing in 1191 medium [15]. DSMZ 1237 is equivalent to C. thermocellum strain 27405 available from the American Type Culture Collection (ATCC). For each experiment, cells used for inoculation were cultured in a 1-L Corning bottle that was used to inoculate all batch reactors used in growth experiments to ensure reproducibility. The chemicals and reagents used were obtained from Fisher Scientific unless otherwise stated. Yeast extract was obtained from Biobasic, Inc. (bacteriological grade). Serial subcultures of C. thermocellum were used to prepare and maintain fresh cultures in 1191 media containing 2 g/L α-cellulose per litre of milliQ water; 1191 medium containing KH2PO4, 1.5 g; Na2HPO4, 3.35 g; NH4Cl, 0.5 g; MgCl2.6H2O, 0.18 g; yeast extract, 2.0 g; resazurin from Sigma Aldrich, St. Louis, U.S.A. (0.25 mg/ml), 2.0 ml; 10× vitamin solution, 0.50 ml; 10× mineral solution, 1.00 ml. Vitamin supplements and mineral elixir solution were prepared according to a procedure previously described in Islam et al. [15].

Experimental DesignBatch experiments in Balch tubes and se-rum bottles, with working volumes of 27 mL and 127 mL respectively (Wheaton

TABLE 1 Polymer composition of all test substrates on a percentage dry matter basis. Method UsedParameter (%)

Moisture content

Crude protein

ADF

NDF

Purifi ed Wood

Purifi ed Hemp

Purifi ed Flax

Purifi ed Rice Straw

Milled Beverage Cups

Dry matter

Ash

Cellulose

Hemicellulose

Lignin

Soluble crude protein

AOAC 930.15

AOAC 930.15

AOAC 984.13ANKOM Method 5:

08-16-06ANKOM Method 6:

08-16-06AOAC 942.05

Roe et al., 1990cellulose = ADF - lignin

hemicellulose = NDF - ADF

ANKOM 08/05

4.67

95.33

0.37

93.40

95.57

0.10

0.2091.27

2.17

2.13

5.06

94.94

0.87

86.36

91.30

<0.05

78.9478.94

4.94

7.42

4.57

95.43

0.53

91.40

97.30

0.54

<0.0577.03

5.90

14.37

9.98

90.02

2.79

49.49

64.79

15.55

0.6038.61

15.30

10.88

4.38

95.62

0.27

91.47

98.01

0.50

0.0886.28

6.54

5.19

Each assay was run in duplicate and was repeated if % error was > 3 %. Roe et al. [35].

23 J-FOR Journal of Science & Technology for Forest Products and Processes: VOL.1, NO.3, 201156

Science, obtained from Fisher Scientific), were used for all tests. For all experiments, 2 g/L substrate loading was used in ves-sels containing 9 mL or 45 mL of 1191 medium. All tubes were air-sealed with butyl-rubber stoppers and crimped with aluminium seals and then gassed and de-gassed for four (1:4) cycles with 100% nitrogen to maintain an anaerobic envi-ronment. Before sterilization, each tube was reduced with 0.1 mL of sterile 2-mM Na2S reducing solution [16] and then au-toclaved. A similar procedure was used with the serum bottles, but with a start-ing volume of 45 mL 1191 medium and 0.5 mL 2-mM Na2S. A fresh culture of

C. thermocellum at exponential phase in a 1-L Corning bottle was used to inocu-late each experimental set with 10% of the volume of medium in the test vessel. All batch experiments for each substrate were conducted in three biological repli-cates, and triplicates for each experimen-tal time-point were collected for analysis according to a pre-determined interval of three hours or more. The amount of end-products carried over into test samples was determined by collecting a zero-hour time point immediately after inoculation. All growth-curve tests were conducted in Balch tubes, while end-point tests were carried out in serum bottles.

Analytical ProceduresSubstrate polymer composition for mois-ture, dry matter, crude and soluble pro-tein, lignin, acid, and neutral detergent fibre was analyzed as previously described by Mani et al. [17] (Table 1). Gas analy-sis for serum bottles was conducted us-ing an Agilent 7890 GC (gas chromato-graph) equipped with a TCD and FID using argon carrier gas, a gas sample valve with a 0.25-cc loop, a split inlet, and two PLOT columns in series, a PLOT Q 30 m x 0.53 mm ID and a PLOT molecu-lar sieve column 30 m x 0.53 mm ID. Gas production in the Balch tubes was analyzed using an SRI gas chromatograph

Fig. 1 - Unscaled super-imposed X-ray diffractograms of alpha (α)-cellulose (AC), purified wood cellulose (PWC), purified hemp cellulose (PHC), and purified flax cellulose (PFC), showing the diffraction pattern of the cellulose in test substrates and control detected at 2 theta (θ) ≈ 22.3.

J-FOR Journal of Science & Technology for Forest Products and Processes: VOL.1, NO.3, 2011 57

SPECIAL BIOREFINERY ISSUE

(SRI model instruments, Model 8610C) outfitted with a thermal conductivity de-tector, and a stainless-steel molecular sieve 13x packed column (3.2 mm x 1.8m) con-nected in series with a stainless-steel silica gel-packed column (3.2 mm x 1.8) was used to separate H2 and CO2 respective-ly using argon as a carrier gas. Total gas compositions were quantified by correct-ing for temperature, pressure, volume, and aqueous-phase concentrations estimated according to the solubility of the gases in water [18]. The amount of carbon diox-ide in equilibrium with bicarbonate was also taken into consideration [19]. Aque-ous products such as organic acids and ethanol were measured using a Waters HPLC equipped with an HPX-Aminex 87 column (Bio-Rad Laboratories, Hercules

CA) maintained at 45°C with HPLC-grade water as the mobile phase to maintain a flow of 0.6 mL/min.

X-ray diffraction analysis (XRD) was conducted using a Siemens D5000 pow-der diffractometer (Siemens AXS, Madi-son WI) using Cu radiation (Cu K-alpha lambda = 1.5418 Å) and operated at 40 kV and 40 mA. The anti-scatter and receiving slits were set to 1.0 mm. Samples were mounted in a depressed “well” on a zero-

background quartz plate and pressed into the well with a frosted glass slide. Scans were obtained from 8 to 42 degrees two-theta using a step-width of 0.05 degree two-theta and a dwell time of 1 s/step.

The contribution of amorphous

cellulose to the test substrates was deter-mined according to a method originally proposed by Segal et al. [20] for the mea-surement of cellulose crystallinity, but here used to provide empirical measurements to enable rapid comparison of cellulosic samples by making rough approximations of the contribution of amorphous cel-lulose (CAC) to the cellulose diffraction spectrum [21]:

RESULTS AND DISCUSSIONThe composition of the biomass sub-strates used in the fermentation reactions is shown in Table 1. All test substrates had > 90% dry matter content. Acid detergent

Fig. 2 - Scaled super-imposed X-ray diffractograms of alpha (α)-cellulose (AC), purified wood cellulose (PWC), purified hemp cellulose (PHC), and purified flax cellulose (PFC), showing the diffraction pattern of the cellulose in test substrates and control detected at 2 theta (θ) ≈22.3.

23 J-FOR Journal of Science & Technology for Forest Products and Processes: VOL.1, NO.3, 201158

fibre (ADF) was used to fibre (ADF) was used to determine the portions of the bio-mass consisting of cellulose and lignin, while neutral detergent fibre (NDF) was used to measure the total cell-wall com-position of the three major plant poly-mers: hemicelluloses, cellulose, and lignin. The difference between NDF and ADF represents the amount of hemicellulose, while ADF minus the lignin represents the approximate cellulose composition, expressed as percentage of the dry mass (Table 1). The results show less available cellulose following pre-treatment in PRC, PFC, and PHC than in PWC and BCC. Alpha (α-)-cellulose was used as a reagent-grade substrate with a well-defined com-position: 92.2% glucan, 3.4% xylan, and 3.2% mannan on a dry-mass basis, with negligible ash content (Sigma Aldrich, St. Louis MO, U.S.A.; product # C8002). Therefore, the cellulose composition can be assumed equal to the glucan content.

X-ray diffraction analysis was per-formed on biomass substrates pre-treated using the Lignol Organosolv® process to solubilize hemicellulose and extract lignin, resulting in a cellulose allomorph identified between 22 and 23 two-theta (θ) following the 002 peak and similar to the α-cellulose used as a positive control (Fig. 1). Hence, the celluloses in the Organosolv®-extracted test substrates can be considered to contain the same transition cellulose

allomorph or polymorph as the positive control, providing a good basis for com-parison. Previous studies have shown that a shift from cellulose I to cellulose III can be observed by following the position of the “002” peak as it shifts from 20-23 to 20-21 for Cellulose III, which is known to be a more digestible cellulose allomorph than native cellulose or cellulose I with a similar crystallinity index (CI) [21-24].

From the percentage intensity of the diffractogram shown in Fig. 2, the es-timated contribution of amorphous cel-lulose (CAC) suggests that PWC has the highest contribution of amorphous cellu-lose in its diffractogram spectrum (CAC ≈ 74.2%), followed by AC (CAC ≈ 69%), PHC (CAC ≈ 67%), and PFC (CAC = 59.8%) after pre-treatment and milling. CAC values were calculated as the ratio of the height of the 002 peak and the height of the minimum (IAM) between the 002 peak and the 101 peak, as shown in Fig. 2, using Eq. 1 [21, 24]. Amorphous cellu-losic material is more accessible to water, exogenous enzymes, and bacterial catalytic components than crystalline or less amor-phous cellulosic material and is therefore hydrolyzed at a much faster rate [25-26]. Previous studies have agreed that cellulase digestibility of pre-treated biomass is lim-ited the accessibility of the cellulose fibres [27-32].

Batch growth experiments (Figs. 3a

and 3b) suggest that C. thermocellum was able to utilize the cellulose present in the test substrates to end-products during all phases of growth. More hydrogen was produced with the more amorphous and cellulose-rich PWC and AC than with PHC and PFC. The rate of hydrogen produc-tion varied during the exponential phase, with the highest rate observed in PWC, followed by AC, PHC, and PFC, as shown in Fig. 4. The variable rates observed dur-ing the exponential phase of growth re-flect the other cellular activities that come into play as the cells engage in active cell division and growth processes while utiliz-ing the available cellulose as their source of carbon. However, because amorphous cellulose is more accessible than crystalline cellulose, the highest rates were observed in PWC and AC (Fig. 4), which have great-er amounts of amorphous cellulose as in-dicated by their diffractograms (Fig. 2).

All end-products produced during growth on 2 g/L substrate loading were analyzed after a growth period of 50 hours on test samples from the authors’ laboratory. Figure 5 shows the gaseous end-products, with PWC having the high-est hydrogen production. More hydrogen production was observed with the forest-derived cellulosic products, PWC, AC, and BCC, than with the agricultural residues, PHC, PFC, and PRC. Ethanol was the soluble end-product produced in largest

Fig. 3a - Hydrogen (H2) generated by C. thermocellum from 2 g/L α-cellulose (AC), purified wood cellulose (PWC), purified hemp cellulose (PHC), and purified flax cellulose (PFC). Each data point is the average of three biological replicates, and the error bars represent the standard deviations.

Fig. 3b - Carbon dioxide production (CO2) by C. thermocellum from 2 g/L α-cellulose (AC), purified wood cellulose (PWC), purified hemp cellulose (PHC), and purified flax cellulose (PFC). Each data point is the average of three biological replicates, and the error bars represent the standard deviation.

J-FOR Journal of Science & Technology for Forest Products and Processes: VOL.1, NO.3, 2011 59

SPECIAL BIOREFINERY ISSUE

quantity. Figure 6 shows that more ethanol was produced with PWC and PHC than with AC, BCC, PFC, and PRC. The yield of gaseous and soluble end-products washigher in samples with less crystalline cel-lulose such as PWC and AC (Table 2).

The organic acid profile for all test substrates shows that acetate was the or-ganic acid produced in largest quantity in all test vessels, followed by formate and lactate. There was a concomitant decrease in pH, ranging from 6.53 to 7.04, associ-ated with organic acid production, with the pH dropping lower in test substrates showing higher amounts of organic acid production from a starting pH of 7.20. The final pH was lowest in PWC (6.53) and highest in PRC (7.04), which indicates the amount of substrate cellulose that has been converted to end-products. Little or no production was observed in 1191 me-dium inoculated with C. thermocellum. This may be due to carryover during inocula-tion or may result from the yeast extract in the 1191 medium acting as a carbon source.

The final amounts of end-products produced during growth varied based on the innate cellulose contents of the substrates. PWC had the highest yield of ethanol and hydrogen compared to BCC, PHC, PFC, PRC, and AC (Table 2). For the four test substrates pre-treat-ed with the Lignol Organosolv® process used in the growth-curve experiments (Figs. 3a and 3b), the end-production reflects the content of amorphous cel-lulose. PWC, which had the highest esti-mated contribution of amorphous cellu-

lose, also had the highest production of gaseous and soluble end-products com-pared to AC (used as a positive control) and the purified cellulose from agricultural residues such as PHC and PFC. However, the specific yield of ethanol per gram was higher for PHC than for the other test substrates. Acetate-to-ethanol ratios (Ta-ble 2) indicated that the C. thermocellum cultures used here were beyond their late exponential growth phase when the exper-iments were terminated. Acetate/ethanol ratios between 1.05 and 1.30 have been reported for the late exponential phase [33-34]. The average amounts of formate produced increased with an increase in available cellulose, but acetate/formate ra-tios were higher in PFC, PRC, and PHC (purified cellulose agricultural residues) than in PWC, AC, and BCC (Table 2), which suggests a shift in metabolism due to cellulose availability.

CONCLUSIONSWhen different lignocellulosic feedstocks are subjected to the same type of physio-chemical pre-treatment, the rate of hydro-lysis and subsequent utilization depends on the content of amorphous cellulose in the biomass. Lignocellulosic biomass

that has been pre-treated physically and chemically, or physio-chemically, contains unique cellulose allomorphs or poly-morphs with unique properties that vary in their digestibility. In this research, three types of lignocellulosic biomass were sub-jected to the same type of physio-chemi-cal pre-treatment (in this case, the Lignol Organosolv® pre-treatment) to generate the same allomorph. The results obtained indicate that the amorphous cellulose con-tent of the pre-treated feedstocks affected subsequent hydrolysis and fermentation of the biomass to the desired end-products. The purified cellulose from forest biomass (PWC, AC, and BCC) and agricultural resi-dues (PHC, PFC, and PRC) was subjected to a series of pre-treatments that extracted lignin, solubilized hemicelluloses, reduced crystallinity, and reduced the degree of po-lymerization (in this case, by both physical pre-treatment (hammer milling) and phys-io-chemical pre-treatment). This provided the highly reactive cellulose needed for the production of cellulosic biofuels (i.e., hy-drogen and ethanol). Rendering cellulosic feedstocks more amorphous (porous, dis-ordered, and accessible) is a good strategy to increase yields of bio-conversion prod-ucts through CBP.

Fig. 4 - Rate of hydrogen production by C. thermocellum from 2 g/L α-cellulose (AC), purified wood cellulose (PWC), purified hemp cellulose (PHC), and purified flax cellulose (PFC) during exponential phase of growth.

Fig. 5 - Hydrogen and carbon dioxide production after 50 hours of C. thermocellum growth on test substrates: AC (α-cellulose), purified wood cellulose (PWC), purified hemp cellulose (PHC), purified flax cellulose (PFC), beverage cup cellulose (BCC), and purified rice cellulose (PRC). Each bar is the average of three biological replicates, and the error bars represent the standard deviation.

23 J-FOR Journal of Science & Technology for Forest Products and Processes: VOL.1, NO.3, 201160

ACKNOWLEDGEMENTSThis research was supported by the Natu-ral Sciences and Engineering Research Council of Canada (NSERC) through the Hydrogen Canada (H2CAN) Network and by Genome Canada.

REFERENCESSannigrahi, P., Pu, Y., and Ragaus-kas, A., “Cellulosic Biorefineries—Unleashing Lignin Opportunities”.

1.

Current Opinion in Environmental Sustainability, 2(5): J383–393 (2010).Lynd, L.R.,”Overview and Evalua-tion of Fuel Ethanol from Cellulosic Biomass: Technology, Economics, the Environment, and Policy”. Annual Review of Energy Environment, 21: J403-465 (1996). Sassner, P., Galbe, M., and Zacchi, G. “Techno-Economic Evaluation of Bioethanol Production from Three Different Lignocellulosic Materials”.

2.

3.

Biomass and Bioenergy, 32(5):J422-430 (2008).Nath, K. and Das, D., “Improvement of Fermentative Hydrogen Produc-tion: Various Approaches”. Applied Microbiology and Biotechnology 65(5):J520-529 (2004). Levin, D.B., Pitt, L., and Love, M., “Biohydrogen Production: Prospects and Limitations to Practical Applica-tion”. International Journal of Hydro-gen Energy, 29(2):J173-185 (2004). Galbe, M. and Zacchi, G., “Pretreat-ment of Lignocellulosic Materials for Efficient Bioethanol Production”. Biofuels108:J41-65 (2007). Lynd, L.R., Elander, R.T., and Wy-man, C.E., “Likely Features and Costs of Mature Biomass Ethanol Technol-ogy”. Applied Biochemistry and Bio-technology, 57-58(1):J741-761 (1996). Lynd, L.R., Laser, M.S., Bransby, D., Dale, B.E., Davison, B., Hamilton, R., Himmel, M., Keller, M., Mc-Millan, J.D., Sheehan, J., and Wyman, C.E., “How Biotech Can Transform Biofuels”. Nature Biotechnology, 26(2):J169-172 (2008). Carere, C.R., Sparling, R., Cicek, N., and Levin, D.B., “Third-Generation Biofuels via Direct Cellulose Fer-mentation”. International Journal of Molecular Sciences, 9(7): J1342–1360 (2008). Das, D. and Veziroglu, T.N., “Hydro-gen Production by Biological Process-es: A Survey of the Literature”. Inter-national Journal of Hydrogen Energy, 26(1):J13-28 (2001). Weimer, P.J., Russell, J.B., and Muck, R.E., “Lessons From The Cow: What the Ruminant Animal Can Teach Us About Consolidated Bioprocessing of Cellulosic Biomass”. Bioresource Technology, 100(21):J5323-5331 (2009). Xu, Q., Singh, A., and Himmel, M.E., “Perspectives and New Directions for the Production of Bioethanol Us-ing Consolidated Bioprocessing of Lignocellulose”. Current Opinion in Biotechnology, 20(3):J364-371 (2009). Pan, X., Arato, C., Gilkes, N., Gregg, D., Mabee, W., Pye, K., Xiao, Z., Zhang, X., and Saddler, J.,

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

Fig. 6 - Ethanol and organic acid production after 50 hours of C. thermocellum growth on test substrates: AC (α-cellulose), purified wood cellulose (PWC), purified hemp cellulose (PHC), purified flax cellulose (PFC), beverage cup cellulose (BCC), and purified rice cellulose (PRC). Each bar is the average of three biological replicates, and the error bars represent the standard deviation.

TABLE 2

Summary of hydrogen (H2) and ethanol (C2H5OH), yields, average of fi nal pH recorded, acetate:ethanol ratios, acetate:formate ratios, and specifi c yields of H2 and C2H5OH per gram cellulose for C. thermocellum grown on 2 g/L of test substrates AC (α-cellulose), purifi ed wood cellulose (PWC), purifi ed hemp cellulose (PHC), purifi ed fl ax cellulose (PFC), purifi ed rice cellulose, and milled Tim Horton cups (BCC).

SubstrateParameter

H2 Yield(µmol/mL)C2H5OH

Yield(mM)Final pH

Acetate/FormateSpecifi c Yield of H2 (µmol/mL/g cellulose)

Specifi c yield of C2H5OH

(mMol/g cellulose)

Acetate/Ethanol

AC

8.54

6.57

6.63±0

0.972.32

92.59

1.55

10.63

7.50

6.53±0.01

0.812.02

116.39

1.79

7.64

7.22

6.73±0.02

0.662.55

96.76

1.99

5.76

5.19

6.90±0.01

0.763.17

74.83

1.47

3.07

2.74

7.04±0.01

0.923.10

79.47

1.54

9.62

5.24

6.63±0.01

1.002.47

111.46

1.32

PWC PHC PFC PRC BCC

J-FOR Journal of Science & Technology for Forest Products and Processes: VOL.1, NO.3, 2011 61

SPECIAL BIOREFINERY ISSUE

“Biorefining of Softwoods Using Ethanol Organosolv Pulping: Prelim-inary Evaluation of Process Streams for Manufacture of Fuel-Grade Etha-nol and Co-Products”. Biotechnology and Bioengineering, 90(4):J473-481 (2005). Pan, X., Gilkes, N., Kadla, J., Pye, K., Saka, S., Gregg, D., Ehara, K., Xie, D., Lam, D., and Saddler, J., “Bioconver-sion of Hybrid Poplar to Ethanol and Co-Products Using an Organosolv Fractionation Process: Optimization of Process Yields”. Biotechnology and Bioengineering, 94(5):J851-861 (2006). Islam, R., Cicek, N., Sparling, R., and Levin, D.B., “Effects of Substrate Loading on Hydrogen Production During Anaerobic Fermentation by Clostridium thermocellum 27405”. Ap-plied Microbiology Biotechnology, 72(3):J576-583 (2006). Daniels, L., Rajagopal, B.S., and Belay, N., “Assimilatory Reduction of Sul-fate and Sulfite by Methanogenic Bac-teria”. Applied and Environmental Microbiology, 51(4):J703-709 (1986). Mani, S., Tabil, L.G., and Sokhansanj, S., “Effects of Compressive Force, Particle and Moisture Content on Me-chanical Properties of Biomass Pellets from Grasses”. Biomass and Bioener-gy, 30(7):J648-654 (2006). Sander, R., “Compilation of Henry’s Law Constants for Inorganic and Or-ganic Species of Potential Importance in Environmental Chemistry”. Mainz, Germany: Max-Planck Institute of Chemistry, 20(1): P1 (1999).Darrett, R.H. and Drisham, C.M. (eds.), Biochemistry. New York: Saun-ders College Publishing (1995).Segal, L., Creely, J.J., Martin, A.E.J., and Conrad, C.M., “An Empirical Method for Estimating Degree of Crystallinity of Native Cellulose Us-

14.

15.

16.

17.

18.

19.

20.

ing the X-Ray Diffractometer”. Tex-tile Research 29(10):J786-794 (1959). Park, S., Baker, J.O., Himmel, M., Pa-rilla, P.A., and Johnson, D.K., “Cellu-lose Crystallinity Index: Measurement Techniques and their Impact on Inter-preting Cellulase Performance”. Bio-technology for Biofuels, 3:10, (2010). Mittal, A., Katahira, R., Himmel, M.E., and Johnson, D.K., “Effects of Alka-line Pretreatment or Liquid Ammonia Treatment on Crystalline Cellulose: Changes in Crystalline Structures and Effects”. Biotechnology for Biofuels, 4:41, (2011) O’Sullivan, A.C., “Cellulose: The Structure Slowly Unravels”. Cellulose, 4(3):J173-207 (1997). Park, S., Johnson, D.K., Ishizawa, C.I., Parilla, P.A., and Davis, M.F., “Measuring the Crystallinity Index of Cellulose by Solid-state 13C Nuclear Magnetic Resonance”. Cellulose, 16(4):J641-647(2009). Hall, M., Bansal, P., Realff, J.H., and Bommarius, A.S., “Cellulose Crys-tallinity—A Key Predictor of Enzy-matic Hydrolysis Rate”. FEBS Journal 277(6):J1571-1582 (2010). Fan, L.T., Lee, Y.H., and Beardmore, D.H., “Mechanism of the Enzymatic Hydrolysis of Cellulose: Effect of Major Structural Features of Cel-lulose on Enzymatic Hydrolysis”. Biotechnology and Bioengineering, 22(1):J177-199 (1980). Converse, A.O., “Substrate Factors Limiting Enzymatic Hydrolysis. In: J.N. Saddler (ed.), Bioconversion of Forest and Agricultural Plant Residues, pp. 93-106. JCAB International (1993). Desai, S.G. and Converse, A.O., “Sub-strate Reactivity as a Function of the Extent of Reaction in the Enzy-matic Hydrolysis of Lignocellulose”. Biotechnology and Bioengineering, 56(6):J650-655 (1997).

21.

22.

23.

24.

25.

26.

27.

28.

Hendricks, A.T. and Zeeman, G., “Pre-treatments to Enhance the Digest-ibility of Lignocellulosic Biomass”. Bioresource Technology 100(1):J10-18 (2009). Jeoh, T., Ishizawa, C.I., Davis, M.F., and Himmel, M.E., Adney, W.S., John-son, D.K., “Cellulase Digestibility of Pretreated Biomass is Limited by Cel-lulose Accessibility”. Biotechnology and Bioengineering 98(1):J112-122 (2007). Laureano-Perez, L., Teymouri, F., Al-izadeh, H., and Dale, B.E., “Under-standing Factors that Limit Enzymatic Hydrolysis of Biomass: Characteriza-tion of Pretreated Corn Stover”. Ap-plied Biochemistry and Biotechnology 121-124:J1081-1099 (2005). Silverstein, R.A., Chen, Y., Sharma-Shivappaa, R.R., Boyettea, M.D., and Osborne, J., “A Comparison of Chemical Pretreatment Methods for Improving Saccharification of Cot-ton Stalks”. Bioresource Technology, 98(16):J3000-3011(2007). Islam, R., Cicek, N., Sparling, R., and Levin, D., “Influence of Initial Cel-lulose Concentration on the Carbon Flow Distribution During Batch Fer-mentation by Clostridium thermocellum ATCC 27405”. Applied Microbiology and Biotechnology, 82(1):J141-148 (2009). Lynd, L.R., Grethlein, H.G., and Wol-kin, R.H., “Fermentation of Cellulose Substrates in Batch Culture and Con-tinuous Culture Using Clostridium ther-mocellum “. Applied and Environmen-tal Microbiology 55(12):J3131-3139 (1989). Roe, M.B., Sniffen, C.J., and Chase, L.E., “Techniques for Measuring Protein Fractions in Feedstuffs”. Pro-ceedings, Cornell Nutrition Confer-ence, Ithaca NY, J81 (1990).

29.

30.

31.

32.

33.

34.

35.