Available online at www.sciencedirect.

com

Bioresource Technology 100 (2009) 12381245

Comparative hydrolysis and fermentation of sugarcane and agave bagasse

ndez-Salas, M.S. Villa-Ram rez, J.S. Veloz-Rendo n, K.N. Rivera-Herna ndez, J.M. Herna * lez-Ce sar, M.A. Plascencia-Espinosa, S.R. Trejo-Estrada R.A. Gonza
n en Biotecnolog a Aplicada del Instituto Polite cnico Nacional (CIBA-IPN), Exhacienda San Juan Molino, Km 1.5 Carretera Estatal Centro de Investigacio bal, Tlaxcala, Mexico Tecuexcomac-Tepetitla, Tepetitla de Lardiza Received 2 June 2003; received in revised form 12 August 2006; accepted 12 September 2006

Abstract Sugarcane and agave bagasse samples were hydrolyzed with either mineral acids (HCl), commercial glucanases or a combined treatment consisting of alkaline delignication followed by enzymatic hydrolysis. Acid hydrolysis of sugar cane bagasse yielded a higher level of reducing sugars (37.21% for depithed bagasse and 35.37% for pith bagasse), when compared to metzal or metzontete (agave pinecone and leaves, 5.02% and 9.91%, respectively). An optimized enzyme formulation was used to process sugar cane bagasse, which contained Celluclast, Novozyme and Viscozyme L. From alkalineenzymatic hydrolysis of sugarcane bagasse samples, a reduced level of reducing sugar yield was obtained (1120%) compared to agave bagasse (1258%). Selected hydrolyzates were fermented with a non-recombinant strain of Saccharomyces cerevisiae. Maximum alcohol yield by fermentation (32.6%) was obtained from the hydrolyzate of sugarcane depithed bagasse. Hydrolyzed agave waste residues provide an increased glucose decreased xylose product useful for biotechnological conversion. 2007 Elsevier Ltd. All rights reserved.
Keywords: Sugarcane; Agave; Bagasse; Fermentation; Hydrolysis

1. Introduction There is an increased interest in producing bioethanol as an octane booster or as a liquid fuel. Lignocellulosic materials from dierent crop residues have been used for conversion to ethanol (Lynd et al., 1991; Wiselogel et al., 1996; Rabinovich, 2006). One of the most extensively used agricultural residues is sugarcane bagasse. In Latin America, sugarcane is widely produced and provides the main source of fermentable carbohydrates for alcohol production. In 199798 Brazil produced more than 15,000,000 m3 of ethanol obtained from alcoholic fermentation of sugarcane juice, providing
Corresponding author. Address: Bogota No. 5, Colonia America Norte, Puebla, CP 72340, Mexico. Tel.: +52 555 7296000x87804; fax: +52 248 4870762. E-mail addresses: sr_tres@hotmail.com, sertre@hotmail.com (S.R. Trejo-Estrada). 0960-8524/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2006.09.062
*

all factory power from burning the residual bagasse (almost 100 million tons), (Boddey, 1993; CFCISO GEPLACEA, 1999). Using simultaneous saccharication and fermentation (SSF), conversion of lignocellulosic residues such as bagasse to ethanol is technically and economically feasible (Philippidis et al., 1992; Hinman et al., 1992). In Mexico, sugar cane (47 million tons produced in 1997) is used entirely for sugar production. The byproduct blackstrap molasses (1.8 million tons produced in 1997) is fermented and distilled to produce alcohol. Molasses is also used as a feed supplement for cattle production or sold on to the international markets as a fermentation raw material. Bagasse, an important residue from sugarcane processing (13.6 million tons per year), could become an important biomass source for saccharication and fermentation to produce bioethanol, but only 3% is processed lez-Ce sar, in Mexicos pulp and paper industry (Gonza 2002).

 ndez-Salas et al. / Bioresource Technology 100 (2009) 12381245 J.M. Herna

1239

Four million vehicles in Mexico City have an estimated daily consumption of 17.2 million liters of gasoline. Assuming that an alcohol gasoline blend (containing 8% anhydrous ethanol in the mixture) was used to alleviate air pollution, then all the molasses produced in the country would need to be fermented to anhydrous alcohol. However, distilleries in Mexico produce only 53 million liters of alcohol per year, a volume equivalent to the potential biofuel consumption of 5 weeks in Mexico City. In Mexico there are ve major cities that will require improved motor lez-Ce sar, 2002). vehicle emissions (Gonza Considerable attention has been given to crop residues derived from sugarcane, corn, wheat, rice and other cereals, as well as residues from the pulp and paper industry (Wiselogel et al., 1996). Other sources have been considered for converting lignocellulose to fermentable sugars, such as legume shrub (Sericea lespedeza), sorghum (Sorghum bicolor) (Wiselogel et al., 1996), horse gram (Dolichos biorus) (Reddy and Reddy, 2005), switchgrass (Panicum virgatum) (Pimentel and Patzek, 2005), and peat moss (Forsberg et al., 1986). The use of sugarcane bagasse in chemistry and biotechnology has been recently reviewed (Pandey et al., 2000). Sugarcane bagasse has a complex structure, and is primarily composed of 25% lignin, 25% hemicellulose and 4050% cellulose (Pandey et al., 2000; Neureiter et al., 2002). Conversion of sugarcane bagasse into fermentable sugars is possible through thermal, chemical, or enzymatic hydrolysis (Beguin and Aubert, 1994; Neureiter et al., 2002; Martin et al., 2002; Laser et al., 2002). Treating lignocellulosic agricultural residues with mineral acids can be used to release fermentable sugars. The level and composition of the sugars released depends on the type of acid and its concentration in the hydrolysis mixture (Israilides et al., 1978). The hemicellulose fraction can be easily extracted and hydrolyzed by dilute acid treatment. Using an HCl solution, 85% of the xylose was recovered in the hydrolyzate (Du Toit et al., 1984), whereas treatment with dilute sulfuric acid released mainly xylose, which could in turn be converted to ethanol by fermentation with a selected strain of Pichia stipitis (Roberto et al., 1991). Variations of hydrolysis conditions provided a method for acid hydrolysis of sugarcane bagasse, which proved to be useful to produce 23 g of xylose per 100 g of bagasse, a near theoretical yield (Neureiter et al., 2002). Alkaline treatment of sugarcane bagasse digests of the lignin matrix and makes cellulose and hemicellulose available to enzyme degradation (Pandey et al., 2000; Rodriguez-Vazquez and Diaz-Cervantes, 1994; Aiello et al., 1996). Similar treatment of sugarcane leaves enhanced subsequent hydrolysis by a cellulolytic enzyme complex (Krishna et al., 1998; Bhat, 2000). Alternatively, biological delignication of bagasse is possible using selected strains of Panus tigrinus, a white rot fungus (Costa et al., 2002; Goncalves et al., 2002). Commercial enzyme preparations have been used to convert sugarcane bagasse to fermentable sugars. Krishna

et al. (1998) used Trichoderma reesei cellulase and cellobiase to hydrolyze sugarcane leaves after alkaline delignication. More recently, Martin et al. (2002) used a mixture of endo-glucanases and cellobiases to saccharify steam pretreated sugarcane bagasse. The resultant hydrolyzate had a sugar composition similar to that reported from chemically treated bagasse, when analyzed by a modied TAPPI standard method (Neureiter et al., 2002). To provide a more diversied group of locally produced raw materials for bioethanol production, we initiated a study to understand the conversion of lignocellulosic materials derived from sugarcane and agave bagasse. Vast areas of Mexico have agave species, wild perennial succulent xerophytes that grow in dry habitats. Agave n) production in the plants are used for ber (heneque n, 1963); for tequila nchez-Marroqu Yucatan peninsula (Sa and mezcal production in southern, west-central and western Mexican states; and also for producing a traditional fermented beverage called pulque, in the highlands of Mexicos central plateau, including agricultural areas within the nchez-Marroqu n, 1963; Idarraga limits of Mexico City (Sa et al., 1999). The agave residues examined in the present study are by-products from pulque manufacturing and are derived from Agave atrovirens (which is called maguey in Mexico), a species widely distributed in the Mexicos central states: Tlaxcala, Puebla and Hidalgo nchez-Marroqu n, 1963). (Sa Pulque production starts by extracting sap or aguamiel, the fermentation substrate, which is fermented by mixed populations of yeast, fermentative and lactic acid bacteria. When the sap extraction is done, a cellulosic residue called metzal is obtained. Metzontete is the whole agave plant as it ends its productive life, providing an excellent cellulosic material for bioconversion to fermentable sugars. Although the composition of A. atrovirens bagasse has not been reported, ber from other agave species has been recently studied. Ceden o-Cruz and Alvares-Jacobs (1999) reported that the composition of bagasse from Agave tequilana (var. Weber, from the tequila industry) is 43% cellulose; 19% hemicellulose and 15% lignin. Its potential use for pulp and paper production and for animal feeding has been recently demonstrated (Idarraga et al., 1999; In iguezCovarrubias et al., 2001). It is possible that the increased glucan and decreased lignin content of agave ber may provide a good source of fermenting sugars produced by chemical and enzymatic hydrolysis and saccharication. Fermentation of hydrolyzate from lignocellulosic residues is necessary to produce alcohol (Wyman, 1996). For that purpose, both pentose-utilizing yeast strains (P. stipitis; Roberto et al., 1991), and non-pentose-utilizing yeast strains (Saccharomyces cerevisiae; Krishna et al., 1998) have been used. Recombinant strains of Escherichia coli, Zymomonas mobilis and S. cerevisiae capable of hexose and pentose catabolism and high ethanol production have also been constructed. Their use has made the conversion of lignocellulose to ethanol economically feasible (Mohagheghi et al., 2002; Moniruzzaman and Ingram, 1998; Martin et al., 2002).

1240

ndez-Salas et al. / Bioresource Technology 100 (2009) 12381245 J.M. Herna

The aim of the present study was to examine the production of free sugars through chemical or enzymatic hydrolysis of agave residues obtained from the pulque industry, and to compare that process to the better understood process of saccharication of sugarcane bagasse. Evidence is presented that supports the potential use of fermentable sugars thus obtained for ethanol production by fermentation. 2. Methods 2.1. Sources of lignocellulose Sugarcane bagasse samples were kindly provided by Ingenio Central Motzorongo, a sugar factory based in Motzorongo, Veracruz, Mexico. Agricultural agave residues (from A. atrovirens), metzal and metzontete were kindly provided by Rancho San Isidro, a pulque producer from Nanacamilpa Tlaxcala, Mexico (see Table 1). Mezontete was thoroughly washed with distilled water, and dried in an oven at 70 C for 72 h. Metzal was only dried, using the same procedure. Agave and sugarcane residues were ground separately in a small disc mill (Bauer Bros. Co. Springeld, Ohio, USA. SS mod. 148-2, 3540 RPM, 300 mm ID). The cellulosic materials were then sieved through stainless steel sieves. The fractions between 0.149 and 1.68 mm were collected for further processing. 2.2. Steam pretreatment A steam treatment was applied to agave and sugarcane bagasse. Two grams of either metzal, metzontete or sugarcane bagasses were mixed with distilled water. The mixtures were autoclaved at 121 C and 1.1 kg/cm2 for 4 h. 2.3. Acid hydrolysis Lignocellulosic materials were also hydrolyzed with dilute acid and steam treatment. HCl (1.2% v/v) was added to bagasse samples in a 5:1, 10:1 or 15:1 ratio (mL of solution/g of bagasse by weight) which accounted for concentrations equivalent to 30, 60 and 90 mg HCl/g of dry matter. Each suspension was autoclaved at 121 C and 1.1 kg/cm2 for 4 h. After hydrolysis 0.1 N NaOH was
Table 1 Lignocellulosic materials from sugarcane and agave Substrates Metzal (Mz): The cellulosic material obtained from scrapping agave pinecone before and during aguamiel production Mezontete (Mt): The whole agave biomass (leaves and pinecone) left after aguamiel production has ceased Pith bagasse (B): Short bers from the vascular bundle of sugarcane stalk. Low density, porous material Depithed bagasse (Bc): Long bers from the cortex of sugarcane stalk. High density

added and the pH adjusted to 5.0. The mixtures were then diluted with distilled water to a nal dilution of 30:1 (30 mL of liquid phase/g of bagasse). The homogenized mixture was then ltered and the liquid saved for analysis. 2.4. Alkaline and enzymatic hydrolysis Bagasse samples (2 g) from agave and sugarcane were delignied with alkali before enzymatic hydrolysis. This treatment was carried by the use of a dilute NaOH solution (2% w/v) applied at a ratio of 5 mL of solution/g of bagasse. Final concentration of NaOH was 50 mg/g of bagasse. The mixture was then autoclaved at 121 C and 1.1 kg/cm2 for 4 h. After this treatment, the pH was adjusted with NaOH (0.25 M), to 5.07.5, depending on the optimum pH value reported by the enzyme manufacturer. The mixture was then blended with the corresponding enzyme preparation and diluted with distilled water to a nal dilution of 15:1 (mL of liquid phase/g of bagasse). For enzymatic treatment, delignied bagasse was treated with 0.40 g of the corresponding enzyme. Final concentrations were 6% bagasse and 1.33% of enzyme (20% (w/w) of enzyme per bagasse dry weight). When several enzymes were used, each one was added at the same concentration. Enzymatic hydrolysis was performed in a water bath set at 55 C for 4 h. All the enzyme preparations were kindly provided by Novozymes (Mexico). The preparations used were Viscozyme, Celluclast, Novozyme, Cellubrix and Pulpzyme. Pulpzyme HC: Is a liquid preparation produced by submerged fermentation of modied genetically strain of Bacillus sp. with an activity of 1000 AXU/g (xylanase units per gram). Cellubrix L: Is a liquid preparation of cellulase and cellobiase produced by separated fermentations of a strain of Thichoderma longibrachiatum and a strain of Aspergillus niger. Novozyme: Is a liquid preparation produced by submerged fermentation of monocomponent endo-glucanase by modied genetically strain of Aspergillus spp. with an activity of 5000 ECU/g (endo-cellulase units per gram). Celluclast: Is a liquid preparation of cellulase produced by submerged fermentation of strain of T. reesei with an activity of 1.5 l Celluclast 700 EGU/g (endo-glucanase units per gram). Viscozyme: Is a liquid multienzyme complex preparation of arabinase, b-glucanase, hemicellulase, cellulase and xylanase produced by a strain of Aspergillus aculeatus with an activity of 100 FBG/g (fungal beta-glucanase, units per gram). 2.5. Reducing sugars The reducing sugar concentration was determined by the 3,5-dinitro salicylic acid (DNS) method (Miller, 1959), with a spectrophotometer (Hewlet Packard, Palo Alto, CA)

 ndez-Salas et al. / Bioresource Technology 100 (2009) 12381245 J.M. Herna

1241

Model 8453. Hydrolysis by either chemical (acid or alkali) or enzymatic methods, was expressed as the yield for sugars by the quantication of the reducing sugars concentration in the ltrate, as determined by the DNS method. Yields for sugars were calculated using the formula referred by Neureiter et al. (2002) as described below. Y CV =W 100 C is the concentration of reducing sugars (g/L), V is total volume of the liquid phase (L) and W is dry weight of the corresponding lignocellulosic material (g). Y is the yield of sugars expressed as percent of reducing sugars in dry weight. 2.6. Fermentation The hydrolyzates were adjusted to pH 5 and amended with mineral nutrients (NH4)2SO4 (6 g/L), KH2PO4 (7.5 g/L), K2HPO4 (2.4 g/L), MgSO4 7H2 O (0.6 g/L), CaCl2 (0.001 g/L), CuSO4 5H2 O (0.001 g/L), MnCl (0.001 g/L), ZnSO4 (0.001 g/L), CoCl2 (0.001 g/L) and sodium molybdate (0.001 g/L). The liquid media was sterilized and distributed in Erlenmeyer asks (450 mL in 1 L asks), inoculated with a commercial strain of S. cerevisiae (SAFMEX, Mexico 1 107 cfu/mL) and incubated at 30 C for 48 h. The asks were sampled (50 mL) to determine total reducing sugars and alcohol concentration. 3. Analytical methods Samples were analyzed for sugars by HPLC. One milliliters of each sample was diluted 15-fold and 10 ml of this mix was passed through 3 cm3 Sep-Pak C-18 cartridges (Waters, Milford, MA); (preequilibrated with 5 mL methanol and 10 mL of deionized water), in order to eliminate hydrophobic substances. The claried samples were then ltered through 0.22-lm Acrodisc membrane lters (Waters, Milford, MA); (Wright, 1995). The ltered samples were injected (5 lL) into a Shodex SC1011 (Shoko Co., Tokyo, Japan) column and eluted with water at a ow rate of 1.0 mL/min and a constant temperature of 80 C. Refractive index detection was used (HewlettPackard, D112), for HPLC analysis (HewlettPackard series 1100, USA), to determine monosaccharide composition. Individual solutions of true glucose and xylose (SigmaAldrich, USA) were used to prepare calibration curves. 3.1. Ethanol analysis Fermentation musts were distilled and the distillate analyzed by gas chromatography using an HP6890 chromatograph (HewlettPackard, USA) with an FID detector. The inlet temperature was 200 C, the mode was split ow (40:1) and the carrier gas was nitrogen. The HP Innowax column was 30.0 m long with a diameter of 340 lm, with a lm thickness of 0.15 lm. The oven was warmed to an initial temperature of 40 C, held for 3 min, and then raised

to 70 C at a rate of 3 C/min. The oven temperature was then raised to 180 C at a rate of 10 C/min. 4. Results and discussion 4.1. Acid hydrolysis Acid hydrolysis of bagasse was performed using HCl. The greatest post-treatment sugar concentration was detected in ryegrass straw using this acid, when compared to other mineral acids (Israilides et al., 1978). Fig. 1 shows the results of acid hydrolysis of agave waste residues. Both metzal and metzontete were treated with HCl at a concentration of 1.2% (v/v). An increase in the amount of acid solution from 5 to 15 (volume of acid solution (mL)/weight of metzal (g)) correlated with an increased saccharication. When compared to a control treated only with water and autoclaved (1.2% hydrolysis), the concentration of 15 mL of acid solution per gram of metzal yielded 10% hydrolysis. In contrast, only 4.8% by weigh of metzontete was converted to reducing sugars using hydrochloric acid (Fig. 1). Sugarcane waste by-products, namely pith bagasse and depithed bagasses, were also hydrolyzed by dilute acid (Fig. 2). Even from the lowest (5:1) ratio of acid solution to bagasse, more than 30% by weight was converted to reducing sugars, and at the highest ratio (15:1), a saccharication level of more than 35% was obtained. Acid hydrolysis of sugarcane bagasse yielded a higher level of reducing sugars (37.21% for sugarcane depithed bagasse and 35.37% for sugarcane pith bagasse), when compared to either metzal or metzontete (5.02% and 9.91%, respectively). The analysis of variance was performed using the Student t test. When the control (no treatment) was compared to the acid treated bagasse samples (90 mg HCl per gram of bagasse), statistically signicant dierences were found (F = 6.15 > F0.10 = 5.54) for every kind of bagasse
14 12 10 8 6 4 2 0 Control 30 60 90 control 30 60 90 Metzontete

Reducing Sugars (% w/w dry matter)

Metzal

(mg of HCl/g of dry matter)

Fig. 1. Acid hydrolysis of agave bagasse. Eect of HCl (mg of HCl/g of bagasse), on the hydrolysis of agave bagasse (yield of reducing sugars).

1242
45

ndez-Salas et al. / Bioresource Technology 100 (2009) 12381245 J.M. Herna

80 Reducing Sugars (% w/w dry matter)

Depithed Bagasse
40 Reducing Sugars (% w/w dry matter) 35 30 25 20 15 10 5 0 Control 30 60 90

Pith Bagasse

70 60 50 40 30 20 10 0

Agave Bagasse(Metzal)

Agave Bagasse(Metzontete)

viscozyme

Control

30

60

90

(mg of HCl/g of dry matter)

Fig. 3. Eect of dierent enzyme preparations on the release of reducing sugars from hydrolysis of alkaline-treated agave bagasse.

Fig. 2. Eect of acid concentration (mg HCl/g of dry matter) on the release of reducing sugars from the hydrolysis of sugarcane bagasse (yield of reducing sugars).

tested. But when the dierent bagasse materials were compared with respect to their hydrolysis yield under dilute HCl treatment, no signicant dierences were found (F0 = 0.95 < F0.10 = 5.39). The four kinds of bagasse tested, despite their dierent origins, responded similarly to the dilute acid treatment. 4.2. Enzymatic hydrolysis Agave and sugarcane bagasse samples were also treated with commercial enzyme preparations after autoclave pretreatment. At optimal pH and appropriate concentrations, only from treatments with Novozyme and Viscozyme L, 57% hydrolysis was obtained (data not shown). An additional step of delignication using dilute NaOH was then tested. This process renders cellulose more available for enzymatic hydrolysis (Ghosh and Singh, 1993; Rodriguez-Vazquez and Diaz-Cervantes, 1994; Aiello et al., 1996; Pandey et al., 2000). Alkaline delignication was performed using a dilute NaOH solution (2% w/v) mixed with 5 mL of solution per gram of bagasse by weight and autoclaved for 4 h. After the pH was adjusted to 7, the partially delignied materials were treated with enzyme preparations, either alone or in combinations. In contrast with results obtained from the acid treatment, the use of an alkalineenzymatic process on agave bagasse (metzal and metzontete) resulted in a higher concentration of reducing sugars (Fig. 3). From the treatment of metzal and metzontete with dilute alkali and autoclave (controls in Fig. 3), only 5% of these materials were transformed to reducing sugars. Subsequent enzymatic hydrolysis of metzal generated from 22% to 58% saccharication, depending on the enzyme preparation, whereas the metzontete values ranged from 12% to 36%. In both cases, treating of delignied samples with Viscozyme L and Celluclast showed the highest hydrolysis rates.

The dierences observed in saccharication of alkaline treated bagasse samples were clearly due to the specic enzyme preparation used. Statistically signicant dierences were obtained by ANOVA analysis using the t test (F0 = 2.99 > F0.10 = 2.27). From alkalineenzymatic hydrolysis of sugarcane bagasse samples (Fig. 4), the reducing sugar yield was lower compared to agave bagasse. For depithed bagasse, 1318% reducing sugars were generated, whereas 1120% saccharication was obtained from pith bagasse. In this case, Cellubrix and Celluclast resulted in the highest hydrolytic activity of all enzyme preparations tested. Statistically signicant dierences were determined by the ANOVA analysis when the alkalineenzymatic saccharication (the reducing sugar yield), was studied as an eect of the kind of bagasse used (F0 = 12.33 > F0.10 = 2.40). The yield of saccharication products from the alkaline enzymatic treatment of agave bagasse almost doubles the yield from sugarcane bagasse. This sharp dierence can be explained by the dierent lignin content of the corresponding tissues. Paturau (1969) and other authors (Pandey et al., 2000; Neureiter et al., 2002) report an average lignin content of 2025% for both pith and depithed sugarcane bagasse, whereas Ceden o-Cruz and Alvares-Jacobs (1999)

Depithed bagasse Reducing Sugars (% w/w dry matter) 20

Pith bagasse

15

10

0 viscozyme viscozyme novozyme novozyme pulpzyme pulpzyme celluclast celluclast cellubrix cellubrix Control Control

Fig. 4. Eect of dierent enzyme preparations on the release of reducing sugars from hydrolysis of alkaline-treated sugarcane bagasse.

viscozyme

novozyme

novozyme

pulpzyme

pulpzyme

celluclast

celluclast

cellubrix

cellubrix

Control

Control

 ndez-Salas et al. / Bioresource Technology 100 (2009) 12381245 J.M. Herna

1243

report 15% for tequila agave, a similar level described for other agave species (Idarraga et al., 1999). An optimized enzyme formulation was developed for pith and depithed bagasse which contained Celluclast, Novozyme and Viscozyme L in equal proportions. After the alkaline treatment described previously, each enzyme concentrate was added at a rate of 0.19 mL per gram of bagasse. The same formulation was used for all four substrates, and the sugars released by alkaline and enzymatic treatment were characterized by HPLC. Interestingly, the combination of enzymes used for full hydrolysis did not include a pure cellobiase preparation like Novozyme 188, which was previously used for sugarcane hydrolysis (Martin et al., 2002). It is possible that cellobiase is present in some of the commercial enzyme preparations tested. Another explanation comes from the dierence between steam-enzymatic treatments reported in other studies (Martin et al., 2002) and the alkalineenzymatic approach presented here. Alkaline treatment delignies more eciently than regular steam explosion (Rodriguez-Vazquez and Diaz-Cervantes, 1994). Fig. 5 shows the results of bagasse saccharication by alkalineenzymatic treatment for each class of lignocellulose material. HPLC analysis was performed for acid hydrolyzed samples (data not shown). The only carbohydrates analyzed quantitatively in these samples were glucose and xylose. Other non-characterized monosaccharides (mostly galactose and diverse pentose sugars) were separated and detected but not quantied. Figs. 6 and 7 refer to the amount of glucose and xylose released by acid or alkalineenzymatic treatments from sugarcane or agave bagasse, measured by HPLC-IR. Glucose concentration was clearly higher in hydrolyzates derived from the alkalineenzymatic treatment for each bagasse sample (F0 = 26.57 > F0.10 = 4.06) As expected, the amount of glucose liberated from the hydrolysis of sugarcane pith and

70 Acid hydrolysis 60 50 40 30 20 10 0

Fig. 6. Glucose content of sugarcane and agave hydrolyzates determined by HPLC-IR. Released by optimized enzyme preparation.

Glucose (g/l)

Mz Mt B Bc Mz Mt B Bc

Alkaline-enzymatic hydrolysis

40 35 30 Xylose (g/l) 25 20 15 10 5 0 0

Acid hydrolysis

Enzymatic hydrolysis

Mz Mt B Bc Mz Mt B Bc

10

Fig. 7. Xylose from samples of sugarcane and agave bagasse hydrolyzates determined by HPLC-IR. Released by optimized enzyme preparation.

Agave bagasse (Metzal) Agave bagasse (Metzontete) Sugar Cane bagasse (Depithed) Sugar Cane bagasse (Pith)

70 Reducing Sugars (%w/w dry matter) 60 50 40 30 20 10 0

Fig. 5. Reducing sugars released from hydrolysis of alkaline-treated sugarcane and agave bagasse. An optimized mixture was used, containing three dierent enzyme preparations. Celluclast, Novozyme and Viscozyme at a concentration of 0.57 g of enzyme/g dry bagasse.

partially dephited bagasse was three times the amount of xylose liberated. When saccharication is compared within either the acid or the alkalineenzymatic treated samples, no statistically signicant dierences can be attributed to the kind of bagasse used (F0 = 0.4 < F0.10 = 3.62). This results are consistent with the data reported by Paturau (1969), Wiselogel et al. (1996), Neureiter et al. (2002) and Martin et al. (2002). Agave waste residues showed dierent hydrolysis products depending on the type of tissue. Metzal, the residue from pinecone extraction, released more sugars compared to the whole agave residue (metzontete). The concentration of glucose released from metzal is 2.5 times higher than the concentration of xylose released by the alkalineenzymatic treatment. Acid hydrolysis released less glucose compared to xylose from sugarcane tissues and metzontete. This trend was previously reported from acid hydrolysis (Du Toit et al., 1984) and uncatalyzed conversion (Jacobsen and Wyman, 2002) of sugarcane bagasse. In contrast, acid hydrolysis of metzal rendered twice as much glucose as xylose, and almost half of the amount obtained from alkalineenzymatic hydroly-

1244

ndez-Salas et al. / Bioresource Technology 100 (2009) 12381245 J.M. Herna

Table 2 Enzyme preparations used in hydrolysis of sugarcane and agave bagasse Enzymes from Novo Nordisk Pulpzyme HC: Is a liquid preparation produced by submerged fermentation of modied genetically strain of Bacillus sp. With an activity of 1000 AXU/g (Xilanase units per gram) Cellubrix L: Is a liquid preparation of cellulase and cellobiase produced by separated fermentations of a strain of Thichoderma longibrachiatum and a strain of Aspergillus niger Novozyme: Is a liquid preparation produced by submerged fermentation of mono-component endo-glucanase by modied genetically strain of Aspergillus spp. With an activity of 5000 ECU/g (endo-cellulase units per gram) Celluclast: Is a liquid preparation of cellulase produced by submerged fermentation of strain of Thichoderma reesei: With an activity of 1.5 l Celluclast 700 EGU/g (endo-glucanase units per gram) Viscozyme: Is a liquid multi-enzyme complex preparation of arabinase, b-glucanase, hemicellulase, cellulase and xylanase produced by a strain of Aspergillus aculeatus. With an activity of 100 FBG/g (fungal betaglucanase, units per gram)

6. Conclusions The data presented here supports the use of agave bagasse for saccharication processes to obtain valuable fermentation products. In particular the use of metzal and pinecone waste residues from the pulque industry would provide an excellent source of a cellulosic material easily hydrolyzed to a glucose rich, xylose sparse product, comparable to other biotechnologically important bioresources, like corn straw, or sugarcane bagasse. In the present study, experimental data demonstrates a yield of 56% of sugars from the conversion of agave metzal, by a combined alkalineenzymatic treatment. A similar treatment for sugarcane depithed bagasse allowed a similar sugar yield, but a much higher alcohol yield by fermentation (32.6% alcohol by weight of sugars present in the hydrolyzate). Both agave and sugarcane are important raw materials xfor manufacturing sugar and fermented beverages in Me ico. The distribution of the corresponding cultivars is geographically distinctive, but their combined distribution covers an important proportion of Mexican land. Lignocellulosic by-products derived from the industrial processing of those bioresources may provide raw materials for producing bioethanol, using chemical and biotechnological processes. Bioethanol, a renewable source of liquid fuel, has proved to be an octane booster and an excellent alternative to decrease air pollution. Bioethanol may serve as an alternative suitable for cities such as Mexico City and Guadalajara, the most densely populated areas in Mexico. Acknowledgements

sis with the same material. A higher glucan content or the type of hemicellulose present (Type A or B, Du Toit et al., 1984) may explain the dierence of monosaccharides released from metzal, a material present only in agave pinecone (see Table 2). The origin of metzal is structurally similar to the A. tequilana bagasse, but metzal does not undergo a thermal treatment during processing. Hemicellulose should still be present in the ber, but the hydrolyzate should have a lower concentration of toxic products (like hydroxy methyl furfural), when compared to bagasse from the tequila industry (Idarraga et al., 1999; Martin et al., 2002). 5. Fermentation Fermentation of sugarcane and agave bagasse hydrolyzates was performed with a non-recombinant distillers strain of S. cerevisiae. Table 3 shows data on the content of reducing sugars in the hydrolyzates and the ethanol yield after fermentation. The ethanol yield is much lower than that obtained by SSF as reported by Philippidis et al. (1992). As reported for sugarcane leaves (Krishna et al., 1998), the fermentation of hydrolyzates yielded an ethanol concentration that correlates with the bioconversion of glucose. No glucose, only xylose was detected by HPLC analysis of hydrolyzates after fermentation ceased (data not shown).

n Gonza lez Romo from The authors thank M.Sc. Adria Colegio de Posgraduados Campus Puebla, Dr. Sonia Silva mez and Dr. Ricardo Pe rez Avile s from Posgrado de Go noma de Puebla Ciencias Ambientales, Universidad Auto for their assistance. Thanks to Dr. Don L. Crawford, University of Idaho, for helpful review of the manuscript. This work was mainly supported by Sistema Regional Zaragoza a (SIZA) from Consejo Nacional de Ciencia y Tecnolog xico) grant no. 980502014 and Fundacion (CONACYT-Me PRODUCE Puebla. This work was partially funded by PIFI, COFAA and Programa de Becas Institucionales cnico Nacional (IPN-Me xico). (SIP) from Instituto Polite

Table 3 Sugars and alcohol yield of the fermentation of hydrolyzates from sugarcane and agave bagasse Raw material and treatment Raw material Agave metzal (Mz) Agave mezontete (Mt) Sugarcane depithed (B) Sugarcane pith (Bc) Agave metzal (Mz) Agave metzontete (Mt) Sugarcane depithed (B) Sugarcane pith (Bc) Hydrolysis Acid Acid Acid Acid Alkalineenzymatic Alkalineenzymatic Alkalineenzymatic Alkalineenzymatic Sugar yield Reducing sugars (RS), g/L 24.82 19.45 35.43 29.9 56.37 28.44 38.38 50.08 Ethanol yield g/L (48 h) 7.4 6.5 5.0 4.7 6.6 6.3 12.5 12.9 Yield (% w/w initial RS) 29.81 33.42 14.11 15.72 11.71 22.23 32.57 25.76

 ndez-Salas et al. / Bioresource Technology 100 (2009) 12381245 J.M. Herna

1245

References
Aiello, C., Ferrer, A., Ledesma, A., 1996. Eect of alkaline treatments at various temperatures on cellulase and biomass production using submerged sugarcane bagasse fermentation with Trichoderma reesei QM 9414. Bioresource Technology 57 (1), 1318. Beguin, P., Aubert, J.P., 1994. The biological degradation of cellulose. FEMS Microbiology Reviews 13, 2558. Bhat, M.K., 2000. Cellulases and related enzymes in biotechnology. Biotechnology Advances 18, 355383. Boddey, R., 1993. Green energy from sugarcane. Chemistry and Industry, 355358. Ceden o-Cruz, M., Alvares-Jacobs, J., 1999. Production of tequila from agave. In: Murtagh, J. (Ed.), The Alcohol Textbook, second ed. Nottingham University Press, UK. CFCISOGEPLACEA, 1999. Project for promoting carburant alcohol. Final report. Edited by: Common Fund for Commodities, International Sugar Organization, Group of Latin America and Caribbean sugar exporting countries, Sao Paulo, Brazil. Costa, S.M., Goncalves, A., Esposito, E., 2002. Action of White-Rot Fungus Panus tigrinus on Sugarcane Bagasse: evaluation of selectivity. Applied Biochemistry and Biotechnology 98/100 (1/3), 357364. Du Toit, P., Oliver, S., Van Bijon, P., 1984. Sugarcane bagasse as a possible source of fermentable carbohydrates. I. Characterization of bagasse with regard to monosaccharide, hemicellulose, and amino acid composition. Biotechnology and Bioengineering 26, 10711078. Forsberg, C.W., Schellhorn, H., Gibbins, L., Maine, F., Mason, E., 1986. The release of fermentable carbohydrate from peat by steam explosion and its use in the microbial production of solvents. Journal of Biotechnology and Bioengineering 28, 176184. Ghosh, P., Singh, A., 1993. Physicochemical and biological treatment for enzymatic/microbial conversion of lignocellulosic biomass. Advances in Applied Microbiology 39, 295333. Goncalves, A., Costa, S., Esposito, E., 2002. Panus tigrinus strains used in delignication of sugarcane bagasse prior to kraft pulping. Applied Biochemistry and Biotechnology 98/100 (1/3), 373382. lez-Ce sar, R., 2002. Sustentabilidad de la produccio n e introGonza n de alcohol como carburante para uso en automotores en la duccio xico. Masters thesis. Beneme rita Universidad Auto noma Ciudad de Me xico. de Puebla (BUAP), Me Hinman, N., Schel, D., Riley, C., Bergeron, P., Walter, P., 1992. Preliminary estimate of the cost of ethanol production for SSF Technology. Applied Biochemistry and Biotechnology 3435, 639649. n Idarraga, G., Ramos, J., Zu iga, V., Sahin, T., Young, R.A., 1999. Pulp and paper from blue agave waste from tequila agave. Journal of Agricultural and Food Chemistry 47, 44504455. In iguez-Covarrubias, G., Lange, S., Rowell, R., 2001. Utilization of byproducts from the tequila industry. Part 1: Agave bagasse as a raw material for animal feeding and berboard production. Bioresource Technology 77 (ER1), 2532. Israilides, C., Grant, G., Han, Y., 1978. Sugar level, fermentability, and acceptability of straw treated with dierent acids. Applied and Environmental Microbiology 36, 4346. Jacobsen, S., Wyman, C., 2002. Xylose monomer and oligomer yields for uncatalyzed hydrolysis of sugarcane bagasse hemicellulose at varying solids concentration. Industrial and Engineering Chemistry Research 41 (6), 14541461. Krishna, S.H., Prasanthi, K., Chowdary, G.V., Ayyanna, C., 1998. Simultaneous saccharication and fermentation of pretreated sugarcane leaves to ethanol. Process Biochemistry 33, 825830. Laser, M., Schulman, D., Allen, S., Lichwa, J., Antal, M., Lynd, L., 2002. A comparison of liquid hot water and steam pretreatment of sugarcane

bagasse for bioconversion to ethanol. Bioresource Technology 81 (ER1), 3344. Lynd, L., Cushman, J., Nichols, R., Wyman, C., 1991. Fuel ethanol from cellulosic biomass. Science 51, 13181323. Martin, C., Galbe, M., Wahlbom, C., Hahn-Hagerdal, B., Jonsson, L., 2002. Ethanol production from enzymatic hydrolysates of sugarcane bagasse using recombinant xylose-utilizing Saccharomyces cerevisiae. Enzyme and Microbial Technology 31 (3), 274282. Miller, G., 1959. Use of dinitrisalicylic acid reagent for determination of reducing sugars. Analytical Chemistry 31, 426429. Mohagheghi, A., Evans, K., Chou, Y., Zhang, M., 2002. Cofermentation of glucose, xylose, and arabinose by genomic dna-integrated xylose/ arabinose fermenting strain of Zymomonas mobilis AX101. Applied Biochemistry and Biotechnology 98 (13), 885898. Moniruzzaman, M., Ingram, L., 1998. Ethanol production from dilute acid hydrolysate of rice hulls using genetically engineered Escherichia coli. Biotechnology Letters 20 (10), 943947. Neureiter, M., Danner, H., Thomasser, C., Saidi, B., Braun, R., 2002. Dilute-acid hydrolysis of sugarcane bagasse at varying conditions. Applied Biochemistry and Biotechnology 98/100 (1/3), 4958. Pandey, A., Soccol, C., Soccol, V., 2000. Biotechnological potential of agro-industrial residues. I: Sugarcane bagasse. Bioresource Technology 74 (2), 69. Paturau, J., 1969. By-Products of the Cane Sugar Industry, An Introduction to their Industrial Utilization. Elsevier Publishing Company, pp. 01115. Philippidis, G., Smith, T., Wyman, C., 1992. Study of the enzymatic hydrolysis of cellulose for production of fuel ethanol by the simultaneous saccharication and fermentation process. Biotechnology and Bioengineering 41, 846853. Pimentel, D., Patzek, T., 2005. Ethanol production using corn, switchgrass, and wood; biodiesel production using soybean and sunower. Natural Resources Research 14 (1), 6576. Rabinovich, M., 2006. Ethanol production from materials containing cellulose: the potential of Russian research and development. Applied Biochemistry and Microbiology 42 (1), 126. Reddy, L., Reddy, O., 2005. Improvement of ethanol production in very high gravity fermentation by horse gram (Dolichos biorus) our supplementation. Letters in Applied Microbiology 41 (5), 440444. Roberto, I., Lacis, L., Barbosa, M., De Mancilha, I., 1991. Utilization of sugarcane bagasse hemicellulosisc hydrolysate by Pichia stipitis for the production of ethanol. Process Biochemistry 26, 1521. Rodriguez-Vazquez, R., Diaz-Cervantes, D., 1994. Eect of chemical solutions sprayed on sugarcane bagasse pith to produce single cell protein: physical and chemical analysis of pith. Bioresource Technology 47 (2), 159. nchez-Marroqu n, A., 1963. Industrializacio n del Agave. Revista de la Sa mica. Universidad Nacional Autonoma de Me xico Sociedad Qu (UNAM), Mexico. Wiselogel, A., Tyson, S., Johnson, D., 1996. Biomass feedstock resources and composition. In: Wyman, C.E. (Ed.), Handbook of Bioethanol: Production and Utilization. Taylor & Francis, Washington, DC, USA, ISBN 1-56032-553-4 (Chapter 6). Wright, S.A., 1995. Distillery quality control. In: Kelsall, D.R., Lyons, T.P., Murtagh, J.E. (Eds.), The Alcohol Textbook: Ethanol Production by Fermentation and Distillation. Nottingham University Press, UK (Chapter 19). Wyman, C.E., 1996. Ethanol production from lignocellulosic biomass: overview. In: Wyman, C.E. (Ed.), Handbook of Bioethanol: Production and Utilization. Taylor & Francis, Washington, DC, USA, ISBN 1-56032-553-4 (Chapter 1).