ETHANOL PRODUCTION FROM SUGAR LIBERATED FROM PINUS SP. AND EUCALYPTUS SP. BIOMASS PRETREATED BY IONIC LIQUIDS

Pretreatment of lignocellulosic biomass using ionic liquids (ILs) has been widely studied and is considered one of the most promising methods to obtain fermentable sugars. However, few data exist on the fermentation of reducing sugars (RS) obtained by enzymatic hydrolysis of biomass pretreated with ionic liquids for the production of ethanol. Therefore, this study evaluated the production of ethanol from sugars liberated from sawdust of Pinus sp. and Eucalyptus sp. pretreated with ionic liquid [C4mim][OAc] and [C2mim] [OAc], hydrolyzed with enzymes of Penicillium echinulatum employing Saccharomyces cerevisiae and Schizosaccharomyces pombe Y698 yeasts. The data indicate that when the biomass is pre-treated by [C2mim] [OAc] there is higher production of ethanol that, when treated by [C4mim][OAc] for both evaluated yeasts, even though the pretreatment with [C2mim] [OAc] caused the highest losses of cellulose and hemicellulose. Under the conditions analyzed, it is possible to produce approximately 40 and 47 L of ethanol per ton of Pinus sp. and Eucalyptus sp, respectively. In addition to contributing to knowledge about the physiology of the yeasts in the sugars liberated from biomass pretreated in presence of ionic liquid, this data is also relevant to the development of processes for the production of lignocellulosic ethanol.


INTRODUCTION
To develop alternative methods of energy generation, enzymatic hydrolysis of sawdust has become increasingly attractive due to the large quantities generated annually (Camassola & Dillon, 2009;Idrees et al., 2014).Moreover, it constitutes a possible source of energy and chemical supplies, and it contributes significantly to the mitigation of serious trinomial pollution problems of soil, water, and air (McKendry, 2002;Reis et al., 2015).In Brazil, it is estimated that 6.4×10 6 hectares correspond to the area of planted forests, with 4.8×10 6 ha of Eucalyptus sp. and Pinus sp.The annual productivity of Pinus sp. is about 35 m 3 of wood per hectare and Eucalyptus sp. is 41 m 3 of wood per hectare (BRACELPA, 2014;Camassola & Dillon, 2007).However, sawdust, originated from the operation of saws, can reach 12% of the total volume of raw material (Cassilha et al., 2004).In Brazil, 620,000 tons of sawdust are generated annually and most of this is burnt in the open, discarded in the environment, or removed to inappropriate landfills, causing damage to the environment, especially to streams, rivers, and springs (Cardoso, 2004).

Brazilian Journal of Chemical Engineering
The process of converting lignocellulosic biomass involves five main stages: selection of suitable biomass, effective pretreatment, production of enzymes such as cellulase and hemicellulase, efficient enzymatic hydrolysis, and fermentation of hexoses and pentoses (Menon & Rao, 2012;Padilha et al., 2015).
Pretreatment with ionic 1-ethyl-3-methylimidazolium acetate and 1-butyl-3-methylimidazolium is very effective in dissolving lignocellulosic materials, such as grasses (Barr et al., 2012), rice husk (Poornejad et al., 2013), cane sugar bagasse (Sant'Ana da Silva et al., 2011), maple (Lee et al., 2009), pine (Brandt et al., 2011) and eucalyptus (Uju et al., 2012).However, the removal of excess ionic liquid in the lignocellulosic biomass prior to the enzymatic hydrolysis is necessary in order to avoid a negative effect on the cellulase activity and the consequent reduction in final concentrations of total reducing sugars (Alvira et al., 2010;Aver et al., 2014).However, the main reason is the price of ionic liquids (ILs), even the commercial grades.Because of the high price of ILs, the processes are not economically feasible without recovery and recycling of more than 99% of the ILs.Currently, numerous studies are available that employ ILs to perform pretreatment of lignocellulosic biomass, but there is little information about the fermentation processes of the reducing sugars (RS) obtained by enzymatic hydrolysis of biomass pretreated with ionic liquids to obtain ethanol.
The most widely used microorganism for alcoholic fermentation is Saccharomyces cerevisiae (Alves Jr et al., 2014), due to its ability to hydrolyze sucrose into glucose and fructose, two easily assimilated hexoses.Other yeasts, such as Schizosaccharomyces pombe, have the additional advantage of tolerating high osmotic pressure (large amounts of salts) and high percentage of solids (Sánchez & Cardona, 2008).
Therefore, the aim of this study was to evaluate the production of ethanol from sugars liberated from the sawdust of Pinus sp. and Eucalyptus sp.pre-treated with the ionic liquids [C 4 mim][OAc] and [C 2 mim] [OAc], hydrolyzed with enzymes of Penicillium echinulatum (Schneider et al., 2016) using S. cerevisiae and S. pombe Y-698 yeasts for alcoholic fermentation.

Reagents
[C 2 mim][OAc] and [C 4 mim][OAc] were purchased from Sigma-Aldrich.All other chemicals were reagent grade or better.

Raw material
The Eucalyptus sp. and Pinus sp.sawdust were obtained from the Miotto sawmill in the city of Caxias do Sul, RS, Brazil.They had an average particle size of about 5 mm × 3 mm.

Enzymes and microorganisms
Cellulases and xylanases produced from Penicillium echinulatum S1M29 (Dillon et al., 2011) were used for enzymatic hydrolysis in solid-state cultivation, using 50% wheat bran and 50% sugarcane bagasse (Camassola & Dillon, 2010) The strain used belongs to the microorganism collection of the Enzyme and Biomass Laboratory at the Institute of Biotechnology, University of Caxias do Sul, Brazil.A commercial S. cerevisiae culture and S. pombe Y 698, kindly donated by the United States Department of Agriculture, were used in the fermentation process.

Pretreatment of biomass
[C 2 mim][OAc] and [C 4 mim][OAc] were separately added to glass tubes containing Eucalyptus sp. or Pinus sp.sawdust (1:4 w/w) and maintained at 120 ºC for 24 h.After pretreatment, the lignocellulosic biomass was washed with distilled water (1:10) and then centrifuged at 800 g and 10 ºC for 15 min.The supernatant was removed, and this washing step was repeated five times (Brandt et al., 2010).Weight loss was determined gravimetrically.

Enzymatic Hydrolysis and Fermentation
The enzymatic hydrolysis of pretreated Eucalyptus sp. or Pinus sp.sawdust was performed with an enzyme loading of 15 U Filter Paper Activity (cellulase) and 108.5 U of b-glucosidases per gram of biomass.The amount of b-glucosidases used was the amount present in the volume of enzyme solution to obtain 15 FPU/g.P. echinulatum enzymes were used and these enzymes were not concentrated.Biomass (2%) and sodium citrate buffer 50 mmol.L -1 (pH 4.8) to complete 50 mL were used.The hydrolysis was carried out at 50 ºC for 24 h (Camassola et al., 2004) and 100 rpm.After hydrolysis, the broth not sterilized containing the sugars was subjected to fermentation at a concentration of 10 6 cells/mL of S. cerevisiae or S. pombe, for 48 h at 28 ºC.All experiments were performed in triplicate.Samples were collected at 0, 12, 24, and 48 h.

Determination of Reducing Sugars (RS)
RS present in the solutions from enzymatic hydrolysis were measured by the DNS method of Miller (1959).A standard curve of glucose to convert absorbance values to concentration of sugars was produced.

High Performance Liquid Chromatography (HPLC)
The dosage of sugars and ethanol by HPLC was performed on a Bio-Rad Aminex HPX-87H column (Shimadzu) at 60 ºC using H 2 SO 4 (5 mmol.L -1 ) as eluent at a flow rate of 0.6 mL min -1 and a refractive index detector.Samples were pre-filtered in polyethersulfone, 0.20 micrometers.

Characterisation of Biomass
The characterisation of the chemical composition of Pinus sp. and Eucalyptus sp. was performed according to the methodology proposed by the National Renewable Energy Laboratory (NREL-TP-510-42618, NREL-TP-510-42619, NREL-TP-510-42621), with adaptations described by Menegol et al. (2014).

Pretreatment
The weight loss after pretreatment with the ILs and subsequent washing, was 33±1  2013) with [C 2 mim] [OAc], a loss of mass of 37.2% for Pine and 40.1% for Eucalyptus sp. were verified.These higher losses verified by these authors in relation to those obtained in this work may be related to the higher concentrations of ILs and temperature used.According to Shafiei et al. (2013), about 85-97% of softwood powder with a size between 295 and 833 µm and 89-100% of the wood chips with size less than 2 cm can be recovered after pretreatment with the same ILs used in this work.Other pretreatments of wood chips, such as steam explosion and dilute acid, may have a weight loss ranging from 20-50% (Shafiei et al., 2013).

Characterisation of Biomass
In relation to the chemical composition (Figure 1) changes were verified in the composition, especially in the percentage of cellulose, hemicellulose and soluble lignin.For cellulose and hemicellulose, pretreatment with [C 2 mim] [OAc] had a more pronounced impact on Eucalyptus sp.than for Pinus sp.sawdust.As for soluble lignin, there was an increase for the two ILs and for the two evaluated biomasses, although the values of the increments were reduced.For insoluble lignin there was no difference between pretreatments and controls.Li et al. (2013) obtained greater removals of insoluble lignin than those obtained in this work, which is probably associated with pre-treatment using higher concentrations of ILs and higher temperatures.

Enzymatic hydrolysis and fermentation
Figures 1 and 2 show that liberation of RS after enzymatic hydrolysis caused a significant increase in biomass pretreated with ILs compared to non-pretreated biomass.This increased 59.61% for Eucalyptus sp.sawdust (Figure 3) and 60.67 % for Pinus sp.sawdust (Figure 2 1), it was verified that the sugar release was not proportional to the availability of cellulose and hemicellulose present in the biomasses, since there was a greater availability of polysaccharides in the Eucalyptus sp. but higher amount of RA was observed in the hydrolysed samples of Pinus sp.Yamashita et al. (2010) obtained 462 mg/g of RS from wood chips pretreated by steam explosion (25 atm) and IL [C 4 mim][OAc], and a lower concentration of AR when the pretreatment used only the IL 1-butyl-3 methylimidazolium chloride, 69.7 mg/g, with both results obtained for enzymatic hydrolysis of 48 h.An interesting result for S. pombe was an increase in the amount of reducing sugars 12 h after the start of the fermentation for both Pinus sp. and Eucalyptus sp.sawdust.S. cerevisiae was found to increase the amount of reducing sugars in the presence of [C 2 mim][OAc] in Eucalyptus sp.These increases in the concentration of sugars in the beginning of the fermentation process are due to the remaining non-hydrolyzed biomass, and the presence of enzymes that possibly are derepressed after the initial consumption of sugars by yeasts.

Brazilian Journal of Chemical Engineering
Greater consumption of RS was found during fermentation within the first 24 h (Figures 2 and 3).After this period, the levels of RS were maintained.After 48 h of fermentation, the residual amount of RS for the biomass fermented with both S. cerevisiae and S. pombe remained steady.The data suggest that both tested yeasts consumed glucose and little metabolized xylose (Figure 4).
S. cerevisiae was more efficient in consuming reducing sugar for both sawdusts (Pinus sp. and Eucalyptus sp.Although yields were reduced, it would be possible to obtain about 47 L of ethanol per ton of Pinus sp.sawdust.However, it is still necessary to increase these yields to make the process economically viable to use Pinus sp.sawdust.
Figure 5 illustrates that Eucalyptus sp.sawdust pretreated with ILs [C 2 mim][OAc] and [C 4 mim][OAc] enabled the yeast to consume higher glucose content.The increased glucose consumption, about 98%, was observed for the yeast S. cerevisiae.The xylose consumption was greater when sawdust was pretreated with [C 4 mim][OAc], consumption reaching 62% for S. cerevisiae and 59% for S. pombe.
The best ethanol yield, considering only the glucose consumption, was obtained with S. cerevisiae, with 35% for both ILs.The highest yield of ethanol, about 37%, considering only the glucose consumption, was obtained with the Pinus sp.sawdust pretreated with [C 2 mim][OAc] and fermented with S. cerevisiae and 35% for the Eucalyptus sp.sawdust pretreated with both ILs.
The work of Shafiei et al. (2013) with woodchips and sawdust pretreated with [C 2 mim][OAc] obtained yields of 66.8% and 81.5% ethanol, respectively.As for their pretreatment with [C 4 mim][OAc ], the yield was 51.8 % for the woodchips and 81% for the sawdust.Although the pretreatment was done with the same ionic liquids and the fermentation was performed with yeast of the same species, the pretreatment time, hydrolysis, and fermentation are different, which may have contributed to the different yields.Yamashita et al. (2010) obtained 30.1 g/L of ethanol using pretreatment with steam explosion (45 atm) and 73.3 g/L for the pretreatment with organic solvents, using an initial substrate concentration of 200 g/L and fermentation time of 48 h with S. cerevisiae.Haykir et al. (2013), using cotton rods pretreated with [C 2 mim][OAc], and alkali obtained ethanol yields of 22.9 mg/g and 19.8 mg/g, respectively, demonstrating a greater efficiency in the pretreatment with ILs when compared to alkali, although the ionic liquids used are also alkaline.With ethanol yields obtained for the Eucalyptus sp.sawdust, it would be possible to produce approximately 40 L of ethanol per ton of waste.
For sawdust of Pinus sp. and Eucalyptus sp., shown in Figure 6, the concentrations of arabinose and xylitol remained constant during fermentation, with the exception of Eucalyptus sp.sawdust pretreated with [C 4 mim][OAc] (Figure 5D), where there was a reduction in the concentration of arabinose by S. cerevisiae.This suggests that, when the biomass is pretreated with [C 4 mim][OAc], this interferes in the metabolism of this yeast, enabling the consumption of pentose arabinose.In the samples where cellobiose was present, it was consumed by both S. cerevisiae and S. pombe, but this consumption was possibly related to the presence of β-glucosidase in the medium.This caused the hydrolysis of the disaccharide to glucose.Cellobiose was not detected in the Eucalyptus samples pretreated with [C 4 mim][OAc], indicating that this disaccharide was converted to glucose during the hydrolysis step.

Brazilian Journal of Chemical Engineering
Comparing the yields of ethanol produced by the S. cerevisiae and S. pombe yeasts, although the differences in yields were small, S. cerevisiae had the highest yield.
Regarding the measured assessed biomass, the highest yields of ethanol were from Pinus sp., although this biomass did not contain the highest amount of carbohydrates.Pinus sp.contained 38% cellulose and 16.14% hemicellulose, while Eucalyptus sp. had 40.05% cellulose and 16.55% hemicellulose.In the quantity of lignin, the quantity of the two evaluated biomasses were very similar, 28.7% and 28% for Pinus sp. and Eucalyptus sp., respectively.These data indicate that the concentrations of lignin and carbohydrates are not the most relevant characteristics for the release of sugars from biomass pretreated by ionic liquids.
Figure 6 also shows that all the hydrolyzed samples have acetic acid emanating from the pretreatment process, but the yeasts evaluated consumed all of this acid content.Another aspect seen in Figure 6B and D is the production of glycerol, which was higher for S. pombe in both biomasses pretreated with [C 4 mim] [OAc].This may have caused a lower ethanol yield, because the formation of glycerol reduces the

Figure 2 .
Figure 2. Consumption of reducing sugars by S. cerevisiae (A) and S. pombe (B) during fermentation of monosaccharides released from Pinus sp.sawdust pretreated with ILs and without pretreatment (control).

Figure 3 .
Figure 3. Consumption of reducing sugars by S. cerevisiae (A) and S. pombe (B) during fermentation of monosaccharides released from Eucalyptus sp.sawdust pretreated with ILs and without pretreatment (control).

Figure 4 .
Figure 4. Concentration of glucose, xylose, and ethanol determined by HPLC for Pinus sp.sawdust pretreated with [C 2 mim][OAc] (A) and [C 4 mim][OAc] (B) before fermentation (Start -0 h) and after (48 h) for S. cerevisiae and S. pombe.The concentration of monosaccharides and ethanol is represented in mg released per g of Pinus sp.sawdust, considering the pretreatment losses.

Figure 5 .
Figure 5. Concentration of glucose, xylose, and ethanol determined by HPLC for Eucalyptus sp.sawdust biomass pretreated with [C 2 mim][OAc] (A) and [C 4 mim][OAc] (B) before (Start -0 h) and after (48 h) fermentation with S. cerevisiae and S. pombe.The concentration of monosaccharides and ethanol is represented in mg released per g of Eucalyptus sp.sawdust, considering the pretreatment losses.

Figure 6 .
Figure 6.Concentration of cellobiose, acetic acid, glycerol, xylitol, and arabinose for Pinus sp.(A and B) and Eucalyptus sp.(C and D) sawdust biomass pretreated with [C 2 mim][OAc] and [C 4 mim][OAc] before (Start -0 h) and after (48 h) fermentation by S. cerevisiae and S. pombe.The concentration of substances is represented in mg per g of Pinus sp. and Eucalyptus sp.sawdust considering the pretreatment losses.

Table 1 .
Ethanol yield from the amount of glucose released from the hydrolysis of Pinus sp. and Eucalyptus sp.sawdust pretreated with the ILs [C 2 mim] [OAc] and [C 4 mim][OAc] in fermentations employing different yeasts.