Butanol and ethanol production from lignocellulosic feedstock: biomass pretreatment and bioconversion

Abstract

Lignocellulosic feedstock has tremendous potential to sustain the renewable production of biofuels such as ethanol and butanol. Although lignocellulosic biomass is a storehouse of energy in the form of cellulose and hemicellulose, yet lignin acts as a barrier against their hydrolysis. A dilute acid pretreatment disintegrates the biomass complex and allows cellulolytic enzymes to hydrolyze cellulose and hemicelluloses in releasing fermentable sugars. The current study investigates the effect of different H₂SO₄ doses (0–2.5%) on the three lignocellulosic feedstock material, especially pinewood, timothy grass, and wheat straw at 121°C for 1 h. Furthermore, the pretreated feedstock was subjected to enzymatic hydrolysis using cellulase, β-glucosidase, and xylanase at 45°C for 72 h. The biomass hydrolysates containing monomeric sugars (glucose and xylose) were fermented using Saccharomyces cerevisiae and Clostridium beijerinckii for ethanol and butanol production, respectively. A comparative evaluation for the concentrations of ethanol and butanol, residual sugars as well as byproducts such as acetone, acetate, and butyrate from biomass hydrolysates was performed. Pinewood hydrolysate revealed high ethanol (24.1 g/L) and butanol (11.6 g/L) concentrations due to greater sugar content. In contrast to ethanol fermentation by S. cerevisiae, butanol fermentation by C. beijerinckii demonstrated low butanol levels in the hydrolysates due to butanol toxicity toward clostridia.

Introduction

Lignocellulosic residues are inexpensive and attractive renewable resources for the production of next generation biofuels. With their vast availability, they are considered to be a suitable alternative to the diminishing fossil fuels. The inedible plant material including residues from agriculture (e.g., corn stover, wheat straw, rice husk); energy crops (e.g., switchgrass, timothy grass), and forest refuse (e.g., pinewood, spruce) are produced in abundance every year [1, 2]. Since, much of these biomasses are thrown away, turning these surplus discarded plant materials into biofuels is of great appeal and importance. The global energy consumption in 2008 being 533 EJ is projected for increase to 653 EJ by 2020 and 812 EJ by 2030 [3]. It is acceptable that a major proportion of this future energy supply (250–500 EJ per year by 2050) will be contributed by lignocellulosic biomass [4].

The major components of lignocellulosic biomass are cellulose, hemicellulose, and lignin. Cellulose is composed of glucose polymers which are largely insoluble and exist in crystalline microfibrils making the sugars difficult to extract [5]. Hemicellulose which comprises various pentose and hexose sugars is attached to the cellulose microfibrils. On the other hand, lignin is a phenyl propane polymer which forms a complex network cross-linking the cellulose and hemicellulose together. In order to break the lignocellulosic framework and extract the fermentable sugars for bioconversion to higher fuel alcohols, a pretreatment method prior to conversion is required. A number of pretreatment methods available for biomass are dilute acid, alkali, hot water, ammonia fiber explosion, carbon dioxide explosion, and organic solvent [6].

An ideal pretreatment should have the following properties: (1) disintegrate the lignin and hemicellulose complex with cellulose, (2) improve the sugar yield as a result of hydrolysis of cellulose and hemicellulose, and (3) prevent excessive degradation or loss of carbohydrate. Last but not the least, it should be cost-effective. Dilute sulfuric acid pretreatment solubilizes the hemicelluloses and thereby disrupts the lignocellulosic composite linked by covalent bonds, hydrogen bonds and van der Waals forces [7]. The pretreatment process is used to overcome the recalcitrance of lignocellulose, increase enzyme efficiency, and improve the yields of monomeric sugars.

During the past few decades, there has been a significant interest in the production of fermentation derived fuel such as ethanol from agriculture-based substrates. However, due to the controversies of food versus fuel there has been a shift in the ethanol substrates from food grains to crop residues [8]. On the other hand, there have been a few attempts to produce alternative alcohol fuels from lignocellulosic residues. One such fuel is butanol that has superior fuel properties than ethanol and can be efficiently produced from lignocellulosic feedstock. A few bacteria known to produce butanol through ABE (acetone-butanol-ethanol) fermentation are Clostridium acetobutylicum, C. beijerinckii, C. saccharoperbutylacetonicum and C. saccharobutylicum [9].

Butanol-producing Clostridium is advantageous over the long-established ethanol-producing Saccharomyces cerevisiae in efficiently metabolizing both pentose and hexose sugars. Saccharomyces cerevisiae is inefficient in metabolizing pentose (e.g., xylose); however it is known to utilize hexose (e.g., glucose) proficiently [10]. In contrast, Clostridium is capable of metabolizing both pentose and hexose [11]. Clostridium utilizes xylose by directly converting it into xylulose with the enzyme xylose isomerase. Engineered S. cerevisiae has a tendency to utilize xylulose, although it requires xylose to be reduced to xylitol and xylulose by xylose reductase and xylitol dehydrogenase, respectively [12].

As a fuel, butanol can be used in pure form or blended in gasoline in any concentration unlike ethanol that can be blended only up to 85%; however, a higher blend would necessitate motor engine modification [13]. In addition, the energy content of butanol being 29.2 MJ/L is 30% higher than that of ethanol (21.2 MJ/L) and much closer to gasoline (32.5 MJ/L). Also, butanols low vapor pressure, hydroscopic nature, less volatility and less flammability facilitates its blending and supply in existing gasoline channels and pipelines [11].

Despite many advantages, production of butanol has some drawbacks such as its low final titer levels, culture toxicity due to low butanol tolerance by clostridia, and purification issues [14]. This paper emphasizes on bio-production of ethanol and butanol from wood (pinewood) and herbaceous-based (timothy grass, wheat straw) lignocellulosic feedstock. These biomasses are least explored in terms of their butanol and ethanol-producing potentials. In addition, a dilute sulfuric acid pretreatment followed by enzymatic hydrolysis of these lignocellulosic residues was developed for higher sugar yields. Furthermore, the biomass hydrolysates with cellulosic and hemicellulosic components were fermented for the production of ethanol and butanol.

Materials and Methods

Lignocellulosic biomass

The biomass samples used in this study were pinewood, PW (Pinus banksiana), timothy grass, TG (Phleum pratense), and wheat straw, WS (Triticum aestivum). The biomasses were collected from Saskatchewan, Canada in 2011. After collection, the biomass samples were air-dried and pulverized using a Wiley mill (Wiley Mill No. 1.706.648, Arthur H. Thomas Co., Philadelphia, PA, USA) to pass through a sieve screen of 1.18 mm. The crushed biomass samples were stocked in clean glass jars at room temperature and used as necessary.

Compositional analysis

The compositional analysis (cellulose-hemicellulose-lignin) of the feedstock samples was performed using a two-step standard NREL method. In the first step, the feedstock was subjected to a Soxhlet extraction procedure using a sequential application of water, ethanol, and hexane for the extraction of any inorganic materials and nonstructural sugars from the samples such as chlorophyll, waxes, sterols, and lipids [15]. The solvents after the extraction were removed in a rotary evaporator under reduced pressure and the residual biomass was air-dried for sugar determination. The extracts after evaporation were recovered and weighed.

In the next step, the air-dried residual biomasses were hydrolyzed with 72% H₂SO₄ at 30°C for 1 h followed by 4% H₂SO₄ at 121°C for 1 h [16]. The hydrolyzed sugars (e.g., arabinose, cellobiose, glucose and xylose) were quantified through an Agilent 1100 Series HPLC (Agilent Technologies, Waldbronn, Germany) equipped with refractive index detector. The analysis was performed using an Aminex HPX 87H ion-exclusion column (BioRad, Hercules, CA, USA) with a Cation H micro-guard cartridge (BioRad). The mobile phase used was 5 mmol/L H₂SO₄ with a flow rate of 0.6 mL/min and column temperature of 55°C. After a complete conversion, the composition of glucose and cellobiose was considered as cellulose and that of xylose and arabinose as hemicellulose [17]. The sugar standards used in the experiments were purchased from Sigma-Aldrich, Oakville, Canada.

Acid and enzymatic hydrolysis

For dilute acid pretreatment, 10 g of biomass (dry basis) was mixed well with 100 mL of 0–2.5% v/v H₂SO₄ to distilled water in 250 mL conical flask. The solid loading ratio or the amount of dry feedstock loading was 1:10 w/v biomass/water. The flasks were cotton plugged and covered with aluminum foil prior to autoclaving at 121°C for 1 h in a Primus PSS5-C-MSSD steam sterilizer (Primus Sterilizer Company LLC, Omaha, NE). The flasks were weighed before and after autoclaving to account for any loss of water. The difference in water loss was supplemented by adding sterilized distilled water to the flasks. After autoclaving, the hydrolysates were cooled to room temperature and neutralized to pH 5.0 with 10 mol/L NaOH [18].

About 6 mL/L each of cellulase, β-glucosidase, and xylanase was added to the neutralized biomass hydrolysates and incubated at 45°C for 72 h with agitation at 80 rpm in an incubator shaker [18]. The enzymes cellulase (brand name: Celluclast 1.5 L; enzyme activity: ≥700 EGU/g), β-glucosidase (brand name: Novozyme 188; enzyme activity ≥250 CBU/g) and xylanase (brand name: Xylanase 1; activity: ≥40 units/mg protein) were obtained from Sigma-Aldrich, Canada. After 72 h of enzymatic hydrolysis, the mixture was centrifuged at 3300g to remove the biomass residues. The supernatant liquid containing hydrolyzed sugars were recovered and filter sterilized by passing through a 0.2 μm filter disc (Fisher Scientific, Ottawa, Canada). The sterilized biomass hydrolysates were stored in sterilized screw-capped Pyrex bottles at 4°C prior to the fermentation experiments. The sugar (e.g., glucose and xylose) levels in the hydrolysates were quantified through HPLC using an Aminex HPX 87H column with similar operating conditions mentioned in the section 'Compositional analysis'. The hydrolysate with greater sugar yields was used for ethanol and butanol fermentation.

The percent saccharification was calculated using the following equation [19].

${\mbox{Saccharification}}\left(\%\right)={\frac {{\mbox{Glucose (g/L)}}\times 0.9}{\mbox{Substrate (g/L)}}}\times 100$

(1)

Ethanol fermentation

Saccharomyces cerevisiae ATCC 96581 (Cedarlane, Burlington, Canada) was used for ethanol fermentation. The inoculum for fermentation was maintained in sterilized ATCC 1245 YPD broth medium containing yeast extract (10 g/L), peptone (20 g/L) and dextrose (20 g/L) for 24 h at 30°C, pH 5.6 with 150 rpm.

For ethanol fermentation, 100 mL of biomass hydrolysate was supplemented with yeast extract (1.5 g/L), peptone (1 g/L), ammonium sulfate (1 g/L), dipotassium phosphate (0.5 g/L), magnesium sulfate (0.5 g/L), and manganese(II) sulfate (0.5 g/L) at pH 5.6 and autoclaved at 121°C for 30 min [20]. After cooling to room temperature, about 6 mL of actively growing S. cerevisiae was added to the sterilized media and incubated for 60 h at 30°C and 150 rpm. Ethanol production from glucose as the control substrate was studied at levels varying from 50–150 g/L. The specified concentrations of glucose solutions were supplemented with the above-mentioned nutrients. All the chemicals were purchased from Sigma-Aldrich, Canada.

About 1.5 mL of samples was drawn every 12 h for ethanol and residual sugar determinations in an HPLC with conditions previously mentioned. Prior to all measurements, the liquid samples were filtered through 0.2 μm filter disks to remove any sediment.

Butanol fermentation

Clostridium beijerinckii B-592 (NRRL, Peoria, IL, USA) was used for butanol fermentation. The dried clostridia spores were rejuvenated in a sterilized ATCC 2107 modified reinforced clostridial broth medium (Cedarlane, Burlington, Canada) and incubated in an anaerobic chamber (New Brunswick Galaxy 170R CO₂ incubator; Eppendorf, Mississauga, Canada). About 6 mL of actively growing C. beijerinckii culture was transferred to 100 mL of freshly prepared pre-sterilized ATCC 2107 media in 125 mL screw-capped Pyrex bottle for inoculum development. The ATCC 2107 modified reinforced clostridial broth contained peptone (10 g/L), beef extract (10 g/L), yeast extract (3 g/L), dextrose (5 g/L), NaCl (5 g/L), starch (1 g/L), L-cysteine HCl (0.5 g/L), sodium acetate (3 g/L), and 0.025% resazurin (4 mL/L). The inoculum was allowed to grow for 16–18 h at 35°C prior to inoculation into the fermentation media.

Butanol production was studied with glucose as the control substrate at levels varying from 50–150 g/L. To 100 mL of the specified concentration of glucose, 2.5 mL of 40 g/L yeast extract was added and sterilized at 121°C for 15 min followed by cooling to room temperature. After cooling, 6 mL of C. beijerinckii (grown in ATCC 2107) culture was added to the medium and incubated in the anaerobic chamber for 72 h at 35°C.

Similarly, 100 mL of filter sterilized biomass hydrolysate (pH 6.5) was transferred to 250 mL pre-sterilized screw-capped Pyrex bottle. To the hydrolysate, 2.5 mL of 40 g/L sterilized yeast extract solution was added followed by inoculation with 6 mL of the actively growing C. beijerinckii (in ATCC 2107 broth) as described above. During all fermentations, 1.5 mL of liquid samples was taken every 12 h and filtered with 0.2 μm filter disks. The filtered samples were quantified for acetone, butanol, ethanol, acetic acid, butyric acid, and residual sugar in an HPLC (Aminex HPX 87H column) with conditions previously mentioned.

All the analyses mentioned above such as compositional analysis, acid/enzymatic hydrolysis, and ethanol and butanol fermentations were performed in triplicates with standard deviation less than 5%. The product yields (g/g) were calculated as gram quantities of ethanol or butanol produced per gram of glucose or sugar utilized. The productivity (g/L/h) was calculated as the maximum ABE concentration (g/L) divided by the particular fermentation time (h).

Results and Discussion

The lignocellulosic biomasses were pretreated using 0–2.5% H₂SO₄ followed by the addition of a mixture of cellulolytic enzymes. The dilute acid and enzymatic hydrolysis resulted in the breakdown of lignocellulosic network for the release of pentose and hexose sugars. Prior to this pretreatment, the compositional analysis of PW, TG and WS was performed to quantify the lignocellulosic components along with extractives and ash.

About 90% of dry matter in lignocellulosic biomass comprises cellulose, hemicellulose, and lignin, whereas the remaining consists of extractives and ash [21]. The exhaustive extraction process of feedstock samples using water, ethanol, and hexane led to the removal of extractives. Table 1 gives the compositional analysis of the three lignocellulosic feedstock. The yield of extractives in PW, TG, and WS was 15.7, 16.5, and 19.2 wt%, respectively. The high amount of total extractives in WS indicated higher amount of water, ethanol, and hexane soluble components such as terpenes, terpenoids, tannins, resins, fats, waxes, lipids, proteins, and organic acids. PW, TG and WS represented 1.5, 1.3, and 1.1 wt% of ash content, respectively.

Table 1. Compositional analysis (in weight percent) of lignocellulosic feedstock.

Biomass	Cellulose	Hemicellulose	Lignin	Extractives	Ash
Pinewood	38.8 ± 1.4	23.6 ± 0.8	20.4 ± 1.0	15.7 ± 0.6	1.5 ± 0.2
Timothy grass	34.2 ± 1.2	30.1 ± 1.0	18.1 ± 0.7	16.5 ± 0.8	1.1 ± 0.4
Wheat straw	39.1 ± 0.8	24.1 ± 0.6	16.3 ± 1.2	19.2 ± 0.8	1.3 ± 0.1

On the other hand, the cellulose concentration in the three feedstock components ranged between 34.2 and 39.1 wt%, with WS demonstrating highest levels. The hemicellulose levels were highest (30.1 wt%) in TG and lowest (23.6 wt%) in PW. Similar results on high hemicellulose contents (26–32 wt%) have been reported from reed canary grass and switchgrass [22]. This was due to the fact that fast-growing plants are abundant in hemicellulose that aids in conducting and concentrating tissue for mineralized solutions rich in sulphates, chlorides, nitrates and silicic acid in plants [23]. The holocellulosic (cellulose and hemicellulose) composition in the three feedstock samples was in the range of 62.4–64.3 wt% indicating higher structural carbohydrate (sugar) levels.

PW showed utmost levels of lignin (20.4 wt%) compared to those of herbaceous biomasses (16.3 and 18.1 wt%), which is in accordance with various authors [24, 25]. As lignin concentrates between primary and secondary cell walls, it generally tends to occur in high amounts in woods due to their tightly bound fibrous texture compared to the loosely bound fibers in herbaceous plants [26]. Lignin acts as an organic polymer to bind the cellulose and hemicellulose together forming a complex network that makes the biomass recalcitrance to acids and enzymes for releasing the fermentable sugars. For this purpose, the biomasses were subjected to dilute H₂SO₄ pretreatment and enzymatic (i.e., cellulase, β-glucosidase and xylanase) hydrolysis.

Figure 1 illustrates the sugar yields from PW, TG, and WS as a result of different H₂SO₄ concentrations ranging from 0% to 2.5% and enzymatic hydrolysis (denoted as “E” in the figure). Enzymatic hydrolysis resulted in higher sugar yields within all feedstocks than the dilute acid pretreatment. Respectively, enzymatic hydrolysis resulted in an increase of 66.5%, 65.7%, and 60% total sugar (glucose and xylose) yield in PW, TG, and WS. Among all feedstock components, PW showed maximum fermentable sugar release of 68.5 g/L at 2% H₂SO₄ and enzymatic hydrolysis. On the other hand, 1.5% H₂SO₄ and enzymatic hydrolysis resulted in 57.4 and 63.6 g/L of sugars from TG and WS, respectively. In terms of the highest percent saccharification (Fig. 2), the following sequence was observed: PW (29.6%) > TG (24%) > WS (23.4%).

Figure 1. Sugar release from (A) pinewood, (B) timothy grass, and (C) wheat straw at 0–2.5% H2SO4 concentrations and enzymatic hydrolysis.

Figure 2. Percent saccharification of biomass after H2SO4 and enzymatic hydrolysis.

As a result of dilute H₂SO₄ pretreatment for feedstock, there was a significant increase in the xylose yields compared to that of glucose. In contrast to glucose, xylose yield was amplified by 37.6%, 60.3%, and 54.7% in dilute acid hydrolysates of PW, TG, and WS, respectively. This was due to the fact that hemicelluloses (primarily xylose) are easily hydrolysable in dilute acid than cellulose (glucose) because of their amorphous nature and lower degree of polymerization [5]. Compared to dilute acid pretreatment, enzymatic hydrolysis resulted in an increase in glucose concentration (g/L) by 72.3%, 78.8%, and 68.9% in PW, TG, and WS, respectively. Correspondingly, xylose levels (g/L) increased by 61.4%, 54.7%, and 54% in the enzymatic hydrolysates of PW, TG, and WS.

This escalation of glucose and xylose levels in enzymatic hydrolysis was due to the activity of cellulase, β-glucosidase, and xylanase. Cellulases (e.g., endoglucanases and cellobiohydrolases) hydrolyze the β-(1,4)-glycosidic linkages of cellulose releasing cellobiose molecules [27]. Cellobiose comprises cellulose chains with repeat units of D-glucose established through β-(1,4)-glycosidic linkages. Further, β-glucosidases break down cellobiose molecules releasing two glucose subunits. On the other hand, xylanases hydrolyze the β-(1,4) bond in xylan backbone, producing short xylo-oligomers and xylose [28]. Xylan is a polysaccharide carbohydrate made from xylose units.

However, dilute H₂SO₄ pretreatment is necessary prior to enzymatic hydrolysis as it could lead to high reaction rates and significantly improve cellulose hydrolysis [29]. The removal of majority of hemicelluloses (xylose) enhances cellulose (glucose) recovery by exposing the cellulose fibers to enzymes for catalysis. The highest total sugar yield from enzymatic hydrolysis of PW (68.5 g/L) was found at 2% H₂SO₄, whereas for TG (57.4 g/L) and WS (63.6 g/L) it was 1.5% H₂SO₄. A comparative analysis by Wyman et al. [30] suggests that high glucose (41.4 g/L) and xylose (22.3 g/L) fractions were recovered from poplar wood at 2% H₂SO₄ concentration. The digestibility of cellulose in biomass is influenced by the physicochemical, structural, and compositional factors [6]. The woody and fibrous nature of PW along with its high amount of lignin (20.4 wt%) led to the requirement of a relatively higher H₂SO₄ concentration than that of TG and WS. Unlike PW, herbaceous biomasses such as TG and WS were found to be porous and fragile with lesser amount of lignin (16.3–18.1 wt%) necessitating lower H₂SO₄ concentration for hydrolysis.

A gradual decrease in glucose and xylose yields was found with an increase in H₂SO₄ concentrations. There was a notable decrease in the total sugar yields of 21.5%, 61.7%, and 60.8% in the enzymatic hydrolysis of PW, TG and WS, respectively at 2.5% H₂SO₄ concentration. It has been reported that high-severity in acid levels result in the undesired conversion of sugars to furfurals and hydroxymethylfurfural [6, 31]. Recent investigations reveal that xylose degrades into insoluble compounds called pseudo-lignin at high dilute acid levels which can significantly retard cellulose digestibility [32, 33]. The deposition of pseudo-lignin on cellulose fibers (in biomass) could block the surface binding sites for acids/enzymes and may result in lower cellulose hydrolysis. A similar effect is presumed to occur in this study at high H₂SO₄ levels.

The effectiveness of biomass pretreatment could be improved by avoiding undesired sugar degradation that may lead to pseudo-lignin formation [34]. Furthermore, these sugar degradation components tend to inhibit the fermentation. Overliming is a conditioning method to reduce the toxicity of pretreated hydrolysates caused by furans (e.g., furfurals and hydroxymethylfurfural) and other phenolics [35]. It is an additional step after enzymatic hydrolysis that uses Ca(OH)₂ to increase the alkalinity of the hydrolysate followed by heating and neutralization [18, 36].

The evaluation of ethanol yields from S. cerevisiae ATCC 96581 was performed on glucose medium at different substrate levels varying from 50 to 150 g/L (Fig. 3). The final ethanol concentrations of 19.3, 31.5 and 47.1 g/L were obtained from S. cerevisiae at 50, 100 and 150 g/L glucose levels, respectively in 60 h of fermentation. In other words, the final ethanol yields were 0.39, 0.32 and 0.31 g/g in 50, 100 and 150 g glucose media (Table 2). Table 2 shows the yield and productivity of ethanol from different glucose media and biomass hydrolysates. Among the three glucose doses, maximum ethanol concentration of 48.3 g/L (productivity: 1.34 g/L/h) was obtained from 150 g/L glucose medium in 36 h. However, the highest ethanol levels in 50 and 100 g/L glucose levels were 20.3 and 32.9 g/L. High sugar and ethanol contents in the fermentation broth inhibit growth and the rate of product formation by yeast. The specific growth rate (per hour) for yeast is found to decrease from 0.61 at 50 g/L glucose to 0.51 at 140 g/L glucose [37]. The trend of glucose conversion increased from Glu-50 g/L (~74%) to Glu-100 g/L (~80%), and decreased in Glu-150 g/L (~68%). This was due to the fact that fermentation process by yeast is inhibited at glucose concentrations above 100 g/L [37, 38]. As a result of substrate inhibition, the residual glucose was relatively high (41.4 g/L) in 150 g/L glucose medium at 60 h of fermentation. In the same way, the unutilized glucose content in 50 and 100 g/L glucose media were 12.6 and 18.1 g/L, respectively.

Table 2. Yield and productivity of ethanol from glucose media and biomass hydrolysates.

Fermentation media	Ethanol yield (g/g)					Productivity (g/L per hour)
	12 h	24 h	36 h	48 h	60 h
The yield and productivity values presented are average of triplicate measurements with the standard error less than ±5%.
Glucose (50 g/L)	0.07	0.16	0.41	0.4	0.39	0.56
Glucose (100 g/L)	0.08	0.18	0.33	0.33	0.32	0.91
Glucose (150 g/L)	0.1	0.21	0.32	0.32	0.31	1.34
Pinewood	0.07	0.15	0.35	0.32	0.31	0.67
Timothy grass	0.05	0.16	0.39	0.35	0.33	0.63
Wheat straw	0.05	0.13	0.36	0.34	0.33	0.64

Figure 3. Ethanol production by Saccharomyces cerevisiae ATCC 96581 from 50, 100, and 150 g/L glucose substrates.

Similarly, ethanol at a titer level of 90 g/L is completely inhibitory (no growth) to S. cerevisiae, although the inhibition is initiated at 24 g/L ethanol in a batch fermentation experiment. Ethanol at higher concentrations can alter the composition, structure, and function of the microbial cell membranes as well as inhibit cell division and decrease cell viability, thus reducing metabolic activity [39]. However, cell organelles such as mitochondria and vacuoles along with cellular metabolism such as sugar transport systems are found to play vital roles in ethanol tolerance mechanism for yeasts.

Following the glucose control fermentations, the biomass hydrolysates were fermented using S. cerevisiae to evaluate their ethanol yields. The bioconversion experiments were performed on PW, TG and WS hydrolysates with initial sugar (glucose and xylose) concentrations of 68.5, 57.4, and 63.6 g/L, respectively (Fig. 4). The ethanol levels in all hydrolysates were found to be high (22.6–24.1 g/L) at 36 h of fermentation at 30°C. The highest ethanol yields were in the order: TG (0.39 g/g) > WS (0.36 g/g) > PW (0.35 g/g). The residual total sugar levels at 60 h of fermentation were found to be 13.6, 11.5 and 14.4 g/L for PW, TG, and WS, respectively.

Figure 4. Ethanol production by Saccharomyces cerevisiae ATCC 96581 from (A) pinewood, (B) timothy grass, and (C) wheat straw hydrolysates.

Two significant phases were observed in the ethanol fermentation for glucose and biomass hydrolysates. The first phase was characterized with a marked increase in ethanol levels between 12 and 36 h, whereas the second phase showed a significant decrease in the sugar levels within the same period. This could be explained through the yeast metabolism. Saccharomyces cerevisiae started metabolizing the sugars for ethanol production at 12–36 h of fermentation and as the level reached up to 24.1 g/L in hydrolysates and 25.5 g/L in glucose media, an inhibition mechanism restricted the yeast to further multiply and produce ethanol.

Figure 5 illustrates the trend of butanol production from glucose substrates with concentrations ranging from 50–150 g/L. Butanol fermentation by Clostridium spp. is characterized by the production of two major classes of products, namely solvents (acetone, ethanol and butanol) and organic acids (acetic acid and butyric acid) [14]. However, some fractions of gas components with CO₂ and H₂ are also produced. The yield of acetone, butanol and ethanol along with acetic and butyric acid was recorded up to 72 h of fermentation. Maximum butanol concentration of 11.9 g/L (yield: 0.12 g/g) was found in 100 g/L glucose media at 60 h of fermentation (Fig. 5B). Similarly, the levels of acetone (6.1 g/L), ethanol (1.9 g/L) were high in 100 g/L glucose media at 60 h. Acetic acid (4.5 g/L) and butyric acid (2.5 g/L) levels were greater in 100 g/L glucose media at 72 h of the fermentation. The total amount of ABE produced was high (17.9 g/L) in 100 g/L glucose medium compared to those of 50 and 150 g/L glucose media.

Figure 5. Butanol production by Clostridium beijerinckii B-592 from (A) 50, (B) 100, and (C) 150 g/L glucose substrates.

With an increase in the substrate (glucose) level, there was an increase in the residual glucose concentration. In particular, the residual glucose contents in 50, 100 and 150 g/L glucose media were 18.2, 28.3 and 53.8 g/L, respectively at 72 h (Fig. 5). This large amount of remaining glucose in the fermentation media was due to the substrate inhibition and butanol toxicity. Glucose concentration of 161.7 g/L has been found to cause substrate inhibition for clostridia in butanol fermentation, resulting in a lag phase of ~40 h [40].

A typical ABE batch fermentation by clostridia occurs in two phases, especially acidogenic phase and solventogenic phase [9]. The acidogenic phase is the initial growth phase that results in H₂, CO₂, acetic acid and butyric acid. Due to the formation of acids, the pH of the medium lowers and the bacteria enters stationary growth stage. This shifts the bacterial metabolism toward the production of solvents (ABE) in the second fermentation phase that is, solventogenic phase. These two phases of clostridial metabolism were evident from the trend analysis of ABE fermentation (Figs. 5 and 6). As the acidogenic phase occurs early in the fermentation, producing acids, the levels of acetic and butyric acid were high at 24 h followed by a sharp decrease at 36 h. On the other hand, there was a noticeable increase in the acetone and butanol concentrations in the later fermentation hours, particularly at 60 h.

Figure 6. Butanol production by Clostridium beijerinckii B-592 from (A) pinewood, (B) timothy grass, and (C) wheat straw hydrolysates.

The fermentation of biomass hydrolysates was performed by C. beijerinckii B-592 to have a comparative yield analysis for ABE and organic acid (Fig. 6). The total amount of ABE in the three hydrolysates were in the order: PW (18.5 g/L) > WS (17.9 g/L) > TG (17.4 g/L) at 60 h of fermentation. The productivity (g/L/h) also showed a parallel trend: PW (0.31) > WS (0.3) > TG (0.29) (see Table 3). Among all feedstock hydrolysates, PW exhibited the highest butanol (11.6 g/L) and ethanol (1.7 g/L) levels, whereas highest acetone concentration (5.4 g/L) was found in case of TG. The average recorded yield (from 12 to 72 h) of solvents in the three feedstock components was in the order: butanol (5.7–6.6 g/L) > acetone (3.3–3.7 g/L) > ethanol (0.6–1.0 g/L). This was due to the fact that a typical ABE fermentation by Clostridium spp. yields acetone, butanol, and ethanol in the ratio of 3:6:1 [14].

Table 3. Yield and productivity of acetone-butanol-ethanol (ABE) from glucose media and biomass hydrolysates.

Fermentation media	ABE yield (g/g)						Productivity (g/L per hour)
	12 h	24 h	36 h	48 h	60 h	72 h
The yield and productivity values presented are average of triplicate measurements with the standard error less than ±5%.
Glucose (50 g/L)	0.02	0.11	0.19	0.26	0.36	0.33	0.3
Glucose (100 g/L)	0.02	0.07	0.11	0.16	0.2	0.18	0.33
Glucose (150 g/L)	0.02	0.03	0.05	0.07	0.1	0.09	0.24
Pinewood	0.02	0.08	0.13	0.21	0.27	0.26	0.31
Timothy grass	0.02	0.06	0.14	0.21	0.3	0.28	0.29
Wheat straw	0.03	0.08	0.15	0.24	0.28	0.26	0.3

The average productivity of ethanol was high compared to that of ABE, both in glucose media and biomass hydrolysates (Tables 2 and 3). Compared to ethanol fermentation, ABE fermentation resulted in lower final butanol levels, which was because of butanol toxicity. Butanol at a level of 12 g/L is inhibitory to C. beijerinkii [41]. Moreover, a maximum butanol production of 19.6 g/L has been reported by C. beijerinckii BA101 [39]. Butanol as a solvent enters in to the bacterial cytoplasmic membranes and changes the membrane structures resulting in membrane fluidity [42]. This significantly interferes with the normal metabolic functions of the bacteria. For instance, a 20–30% increase in the membrane fluidity was noticed in C. acetobutylicum in response to 1% butanol exposure during fermentation.

A few advancements such as fed-batch fermentation, gas stripping, perstraction, pervaporation, and membrane separation have shown to reduce the culture toxicity caused by excessive butanol accumulation in the media [14]. However, these features considerably add to the overall economics of the bioconversion process. Furthermore, substantial research is underway towards the development of genetically modified butanol-producing microorganisms. These microorganisms are desired to have better and modified stress responsive proteins (e.g., heat shock proteins) and membrane fatty acid composition and structure to resist membrane fluidity that is involved in alcohol tolerances [14, 39].

Conclusions

Three varieties of lignocellulosic feedstock, namely pinewood, timothy grass, and wheat straw were pretreated using dilute H₂SO₄ at varying doses (0–2.5%) followed by hydrolysis using cellulolytic enzyme mixture (cellulase, β-glucosidase and xylanase) for their bioconversion to the next generation alcohol-based fuels such as ethanol and butanol. The compositional analysis of pinewood, timothy grass, and wheat straw showed the presence of 38.8, 34.2 and 39.1 wt% cellulose; 23.6, 30.1 and 24.1 wt% hemicellulose; and 20.4, 18.1 and 16.3 wt% lignin, respectively. On the other hand, the total (water, ethanol and hexane-soluble) extractives in biomass samples was in the range of 15.7–19.2 wt%. The utmost levels of glucose (32.9 g/L) and xylose (35.6 g/L) were recovered from pinewood with 2% H₂SO₄ and enzymatic hydrolysis. In contrast, 1.5% H₂SO₄ pretreatment with enzymatic hydrolysis resulted in 26.7 and 26 g/L glucose and 30.7 and 37.6 g/L xylose yields from timothy grass and wheat straw hydrolysates, respectively. Pinewood (29.6%) showed high levels of saccharification, followed by timothy grass (24%) and wheat straw (23.4%).

Maximum ethanol concentrations of 20.3, 32.9 and 48.3 g/L were obtained using S. cerevisiae ATCC 96581 from 50, 100 and 150 g/L glucose levels, respectively in 36 h of fermentation. Among the feedstock hydrolysates, pinewood demonstrated high ethanol levels (24.1 g/L) followed by wheat straw (23.2 g/L) and timothy grass (22.6 g/L). ABE fermentation using C. beijerinckii B-592 led to uppermost butanol concentrations of 11.2, 11.9, and 9.3 g/L from 50, 100, and 150 g/L glucose substrate in 60 h. The butanol levels from the biomass hydrolysates decreased as: pinewood (11.6 g/L) > wheat straw (11.2 g/L) > timothy grass (10.8 g/L). Furthermore, the total ABE levels from pinewood, timothy grass, and wheat straw hydrolysates were found to be 18.5, 17.4, and 17.9 g/L, respectively for 60 h of fermentation.

Acknowledgment

The authors thank Natural Sciences and Engineering Research Council of Canada (NSERC) and Canada Research Chair (CRC) program for the financial support toward this biomass conversion research.

Conflict of Interest

None declared.

References

1. Sims, R. E. H., W. Mabee, J. N. Saddler, and M. Taylor. 2010. An overview of second generation biofuel technologies. Bioresour. Technol.101:1570–1580.
2. Lynd, L. R., C. E. Wyman, and T. U. Gerngross. 1999. Biocommodity engineering. Biotechnol. Prog.15:777–793.
3. United States Energy Information Administration, USEIA. 2011. International energy outlook 2011. Available at: http://www.eia.gov/forecasts/ieo/pdf/0484(2011).pdf (accessed 03 January 2012).
4. Berndes, G., M. Hoogwijk, and R. van den Broek. 2003. The contribution of biomass in the future global energy system: a review of 17 studies. Biomass Bioenergy25:1–28.
5. Hu, F., and A. Ragauskas. 2012. Pretreatment and lignocellulosic chemistry. Bioenergy Res.5:1043–1066.
6. Kumar, P., D. M. Barrett, M. J. Delwiche, and P. Stroeve. 2009. Methods for pretreatment of lignocellulosic biomass for efficient hydrolysis and biofuel production. Ind. Eng. Chem. Res.48:3713–3729.
7. Karimi, K., M. Shafiei, and R. Kumar. 2013. Progress in physical and chemical pretreatment of lignocellulosic biomass. Pp. 53–96inV. K. Gupta and M. G. Tuohy, eds. Biofuel technologies. Springer, Berlin, Heidelberg.
8. Graham-Rowe, D.2011. Beyond food versus fuel. Nature474:S6–S8.
9. Jones, D. T., and D. R. Woods. 1986. Acetone-butanol fermentation revisited. Microbiol. Rev.50:484–524.
10. Ha, S. J., J. M. Galazka, S. R. Kim, J. H. Choi, X. Yang, J. H. Seo, et al. 2011. Engineered Saccharomyces cerevisiae capable of simultaneous cellobiose and xylose fermentation. PNAS108:504–509.
11. Qureshi, N., and T. C. Ezeji. 2008. Butanol, ‘a superior biofuel’ production from agricultural residues (renewable biomass): recent progress in technology. Biofuels Bioprod. Bioref.2:319–330.
12. Walfridsson, M., J. Hallborn, M. Penttila, S. Keranen, and B. Hahn-Hagerdal. 1995. Xylose-metabolizing Saccharomyces cerevisiae strains overexpressing the TKL1 and TAL1 genes encoding the pentose phosphate pathway enzymes transketolase and transaldolase. Appl. Environ. Microbiol.61:4184–4190.
13. Durre, P.2007. Biobutanol: an attractive biofuel. Biotechnol. J.2:1525–1534.
14. Zheng, Y. N., L. Z. Li, M. Xian, Y. J. Ma, J. M. Yang, X. Xu, et al. 2009. Problems with the microbial production of butanol. J. Ind. Microbiol. Biotechnol.36:1127–1138.
15. Sluiter, A., R. Ruiz, C. Scarlata, J. Sluiter, and D. Templeton. 2008. Determination of extractives in biomass. Technical report NREL/TP-510-42619. National Renewable Energy Laboratory (NREL), Colorado.
16. Sluiter, A., B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, and D. Templeton. 2008. Determination of sugars, byproducts, and degradation products in liquid fraction process samples. Technical report NREL/TP-510-42623. National Renewable Energy Laboratory (NREL), Colorado.
17. Lenihan, P., A. Orozco, E. O'Neill, M. N. M. Ahmad, D. W. Rooney, and G. M. Walker. 2010. Dilute acid hydrolysis of lignocellulosic biomass. Chem. Eng. J.156:395–403.
18. Qureshi, N., B. C. Saha, B. Dien, R. E. Hector, and M. A. Cotta. 2010. Production of butanol (a biofuel) from agricultural residues. Part I – use of barley straw hydrolysate. Biomass Bioenergy34:559–565.
19. Araujo, A., and J. D'Souza. 1986. Enzymatic saccharification of pretreated rice straw and biomass production. Biotechnol. Bioeng.18:1503–1509.
20. Govumoni, S. P., S. Koti, S. Y. Kothagouni, S. Venkateshwar, and V. R. Linga. 2013. Evaluation of pretreatment methods for enzymatic saccharification of wheat straw for bioethanol production. Carbohydr. Polym.91:646–650.
21. Balat, M.2011. Production of bioethanol from lignocellulosic materials via the biochemical pathway: a review. Energy Convers. Manage.52:858–875.
22. Bridgeman, T. G., L. I. Darvell, J. M. Jones, P. T. Williams, R. Fahmi, A. V. Bridgwater, et al. 2007. Influence of particle size on the analytical and chemical properties of two energy crops. Fuel86:60–72.
23. Vassilev, S. V., D. Baxter, L. K. Andersen, C. G. Vassileva, and T. J. Morgan. 2012. An overview of the organic and inorganic phase composition of biomass. Fuel94:1–33.
24. Naik, S., V. V. Goud, P. K. Rout, K. Jacobson, and A. K. Dalai. 2010. Characterization of Canadian biomass for alternative renewable biofuel. Renew. Energy35:1624–1631.
25. Shen, D. K., S. Gu, K. H. Luo, A. V. Bridgwater, and M. X. Fang. 2009. Kinetic study on thermal decomposition of woods in oxidative environment. Fuel88:1024–1030.
26. McKendry, P.2002. Energy production from biomass (Part 1): overview of biomass. Bioresour. Technol.83:37–46.
27. Perez, J., J. Munoz-Dorado, T. de la Rubia, and J. Martinez. 2002. Biodegradation and biological treatments of cellulose, hemicellulose and lignin: an overview. Int. Microbiol.5:53–63.
28. Shallom, D., and Y. Shoham. 2003. Microbial hemicellulases. Curr. Opin. Microbiol.6:219–228.
29. Nanda, S., J. Mohammad, S. N. Reddy, J. A. Kozinski, and A. K. Dalai. 2014. Pathways of lignocellulosic biomass conversion to renewable fuels. Biomass Convers. Bioref.4:157–191.
30. Wyman, C. E., B. E. Dale, R. T. Elander, M. Holtzapple, M. R. Ladisch, Y. Y. Lee, et al. 2009. Comparative sugar recovery and fermentation data following pretreatment of poplar wood by leading technologies. Biotechnol. Prog.25:333–339.
31. Rosatell, A. A., S. P. Simeonov, R. F. M. Frade, and C. A. M. Afonso. 2011. 5-Hydroxymethylfurfural (HMF) as a building block platform: biological properties, synthesis and synthetic applications. Green Chem.13:754–793.
32. Kumar, R., F. Hu, P. Sannigrahi, S. Jung, A. J. Ragauskas, and C. E. Wyman. 2013. Carbohydrate derived-pseudo-lignin can retard cellulose biological conversion. Biotechnol. Bioeng.110:737–753.
33. Hu, F., S. Jung, and A. Ragauskas. 2012. Pseudo-lignin formation and its impact on enzymatic hydrolysis. Bioresour. Technol.117:7–12.
34. Hu, F., and A. Ragauskas. 2014. Suppression of pseudo-lignin formation under dilute acid pretreatment conditions. RSC Adv.4:4317–4323.
35. Chi, Z., M. Rover, E. Jun, M. Deaton, P. Johnston, R. C. Brown, et al. 2013. Overliming detoxification of pyrolytic sugar syrup for direct fermentation of levoglucosan to ethanol. Bioresour. Technol.150:220–227.
36. Mohagheghi, A., M. Ruth, and D. J. Schell. 2006. Conditioning hemicellulose hydrolysates for fermentation: effects of overliming pH on sugar and ethanol yields. Process Biochem.41:1806–1811.
37. Ghose, T. K., and T. D. Tyagi. 1979. Rapid ethanol fermentation of cellulose hydrolysate. II. Product and substrate inhibition and optimization of fermentor design. Biotechnol. Bioeng.21:1401–1420.
38. Moulin, G., H. Boze, and P. Galzy. 1980. Inhibition of alcoholic fermentation by substrate and ethanol. Biotechnol. Bioeng.22:2375–2381.
39. Liu, S., and N. Qureshi. 2009. How microbes tolerate ethanol and butanol. New Biotechnol.26:117–121.
40. Ezeji, T. C., N. Qureshi, and H. P. Blaschek. 2004. Acetone butanol ethanol (ABE) production from concentrated substrate: reduction in substrate inhibition by fed-batch technique and product inhibition by gas stripping. Appl. Microbiol. Biotechnol.63:653–658.
41. Westhuizen, A. V. D., D. T. Jones, and D. R. Woods. 1982. Autolytic activity and butanol tolerance of Clostridium acetobutylicum . Appl. Environ. Microbiol.44:1277–1281.
42. Durre, P.2008. Fermentative butanol production bulk chemical and biofuel. Ann. N. Y. Acad. Sci.1125:353–362.