Hydrolysis of cellulose to glucose is virtually an essential step in any practical cellulosic biofuel production via a biological route. However, for a long time, the heterogeneous acidic hydrolysis of cellulose to the production of glucose took the dominant position due to the limit of cellulose solvent. However, the traditional acid hydrolysis of lignocellulose was inefficient and cost-intensive. Considering the full dissolution of cellulose in ILs, it is rational to expect that the dissolution process could break internal and external supramolecular structures among the cellulosic fibers, which will facilitate the interaction between the cellulose and external catalysts and reactants, thus a new hydrolysis behavior of cellulose will be envisioned in ILs.

In 2007, Li et al. first reported the hydrolysis behavior of cellulose in ILs in the presence of mineral acids [58]. The results showed that catalytic amounts of mineral acid were sufficient to stimulate the hydrolysis reaction. For example, when the acid/cellulose mass ratio was set to 0.46, yields of total reducing sugar (TRS) and glucose were 64 and 36%, respectively, after 42 min at 100°C. In fact, excess acid loading in the ILs system was detrimental in terms of sugar yields because side reaction tended to occur which consumed the hydrolysis products. Preliminary kinetic study indicated that the cellulose hydrolysis catalyzed by H2SO4 followed a consecutive first-order reaction sequence, where k1 for TRS formation and k2 for TRS degradation were 0.073 min-1 and 0.007 min-1, respectively. Their further study on the hydrolysis behavior of lignocellulose in ILs demonstrated that hydrochloric acid was also an effective catalyst [59]. TRS yields were up to 66, 74, 81, and 68% for hydrolysis of corn stalk, rice straw, pine wood, and bagasse, respectively, in the presence of only 7 wt% catalyst at 100°C under an atmospheric pressure within 60 min. Under those conditions, the con­stants for k1 and k2 were 0.068 and 0.007 min-1, respectively, for the hydrolysis of corn stalk. Similar work was also done by Li et al. using different woody lignocellulosic materials, and it was found that the acidic pretreatment of woody biomass species (Eucalyptus grandis, Southern pine and Norway spruce) in [Amim] Cl resulted in the near-complete hydrolysis of cellulose, hemicellulose and a significant amount of lignin [60]. Acid-catalyzed conversion of loblolly pine wood was also investigated in [Bmim] Cl and almost identical results were achieved [61].

Toward a better understanding of the acidic hydrolysis behavior of cellulose in ILs, Vanoye et al. investigated the kinetics of the acid-catalyzed hydrolysis of cellobiose in the ILs 1-ethyl-3-methylimidazolium chloride ([Emim]Cl), which was usually studied as a model for general lignocellulosic biomass hydrolysis in ILs systems. The results showed that the rates of the two competitive reactions, polysaccharide hydrolysis, and sugar decomposition, varied with acid strength, and that for acids with an aqueous pKa below approximately zero. It was found that the hydrolysis reaction was significantly faster than the degradation of glucose, thus allowing hydrolysis to be performed with a high selectivity in glucose, which was consisted with the results obtained in Li’s work [62]. It was expected that the higher the degree of polymerization (DP) value of cellulose, the longer the reaction time will be required for a satisfactory glucose yield, while more TRS will be observed with a shorter reaction time in ILs, which implies that cellulose hydro­lysis in ILs catalyzed by mineral acids most likely follows a random hydrolysis mechanism, as observed with the concentrated-acid system [58]. It was proposed that both endoglycosidic and exoglycosidic scissions occur during the hydrolysis process, but the endoglycosidic product, oligoglucoses, is the major one at the initial stage, which was usually observed in traditional heterogeneous hydrolytic systems. Since then, a lot of mineral acids, organic acids, and solid acids have been applied for the homogeneous hydrolysis of cellulose and lignocellulosic materials in ionic liquids. The results have been summarized in Table 3.1 [63, 64].

Among all these significant contributions into the production of monosugars from biomass with the ILs platform, it is worthy of mention that, in 2010, Zhang et al. demonstrated that under relatively mild conditions (<140°C, 1 atm) and in the absence of acid catalysts, such as HCl, H2SO4, the dissolved cellulose in [Emim] Cl could be converted into reducing sugars in up to 97% yield. Their combined study of experimental methods and ab initio calculations demonstrated that the Kw value of water in the mixture was up to three orders of magnitude higher than that of the pure water under ambient conditions. Such high Kw values are typically achievable under high temperature or subcritical conditions, which is responsible for the remarkable performance in the absence of acid catalysts. They hypothesized that the increased [H+] was attributed to the enhanced water auto ionization by ionic liquids. This process will be affected by the electrostatic environment of the solution, the broad dielectric medium of the solvent, and the temperature. Comparative ab initio calculations based on the thermodynamic cycle shows that IL-water mixture exhibits higher concentrations of both [H+] and [OH-] than pure water, thus enabling the acid — and base-catalyzed reactions [70].

Under homogeneous conditions, the physical barriers of cellulose (such as crystallinity, morphology, surface area, and other physical features) are not pres­ent. But the recycling of the acidic catalysts is one of the main drawbacks of the conventional acid-catalyzed reaction processes. Separation processes represent more than half of the total investment in equipment for the chemical and fuel industries, while the introduction of heterogeneous catalysis made the catalyst separation easy after the reaction for industrial processes [72]. After the dissolu­tion of cellulose in ionic liquids, different solid acid catalysts have also been investigated for the hydrolysis of cellulose. In 2008, Rinaldi et al. reported that a solid acid (Amberlyst 15 DRY) catalyzed hydrolysis of cellulose and (ligno)cel — lulose in ILs [73, 74]. In these studies, depolymerized cellulose was precipitated and recovered by addition of water to the hydrolytic system, and the DP value was estimated by gel-permeation chromatography. It was found that the size of recovered cellulose fibers became successively smaller over time, resulting in a colloidal dispersion for the material recovered after 5 h. The depolymerization of cellulose proceeded progressively, resulting in the formation of soluble oligosac­charides if the reaction was carried out over a long time. For example, celloo — ligomers consisted of approximately 10 anhydroglucose units (AGU) which were seen after 5 h. The phenomena observed in these studies further supported the proposed hydrolytic pathway in ILs by Li et al. [58]. It was interesting to observe that there was an induction period for the production of glucose, and further titration results of the ILs separated from a suspension of Amberlyst 15DRY in [Bmim]Cl suggested that proton was progressively released into the bulk liquid

a Carbohydrates were hydrolyzed at 1.4—1.5 mol of HCl/g wood acid concentration b The reaction was carried out in aqueous solution c TRS selectivity d N. C not characterized

within an hour upon through an ion-exchange process involving [Bmim]+ of the ionic liquid and H+ species of the solid acid.

The design of solid catalysts, that are suitable for both heterogeneous and homogeneous conversion, is one of the most top challenges for biomass utilization [75]. It was found that the H+ species and reaction media are highly related to their catalytic activity toward the hydrolysis of cellulose. For example, Shimizu et al. developed H3PW12O40 and Sn075PW12O40 for the hydrolysis of lignocellulose, which showed higher TRS yield than conventional H2SO4 in water [66]. Other solid acids, such as Nafion® NR50, sulfonated silica/carbon nanocomposites, have also been studied for the hydrolysis of cellulose in ILs. It was found that the crystalline cellulose was partially loosened and transformed to cellulose II from cellulose I, then to glucose assisted by Nafion® NR50. Afterwards, a catalyst was recycled and the residual (hemi) cellulose solid, which could be hydrolyzed into monosugars by enzymes, was separated by adding antisolvents [67]. Due to the presence of strong, accessible Brpnsted acid sites and the hybrid surface structure of sulfonated silica/carbon nanocomposites, it was found that a 42.5% glucose yield was achieved after three recycles of this catalyst in ILs [69].

Solid acid-catalyzed hydrolysis of cellulose in ILs was greatly promoted by microwave heating. The results showed that H-form zeolites with a lower Si/Al molar ratio and a larger surface area exhibited better performance than that of the sulfated ion-exchanging resin NKC-9. The introduction of microwave irradiation at an appropriate power significantly reduced the reaction time and increased the yields of reducing sugars. A typical hydrolysis reaction with Avicel cellulose produced glucose in around 37% yield within 8 min [68].

Monosugars are intermediates linking the sustainable biomass and clean ener­gies, such as bioethanol and microbial biodiesel. In 2010, Binder et al. first investigated the fermentation potential of sugars produced from cellulose in ILs after separation of ILs by ion-exclusion chromatography. The results showed that adding water gradually to a chloride ionic liquid-containing catalytic HCl led to a nearly 90% yield of glucose from cellulose and 70-80% yield of sugars from untreated corn stover. Ion-exclusion chromatography allowed the recovery of the ILs and delivered sugar feedstocks that support the vigorous growth of ethanol — ogenic microbes. This simple chemical process presents a full pathway from biomass to bio-energy based on the ionic liquids platform, although the devel­opment of more economic technologies for the recovery and separation of the ILs and sugars is still in high demand [65].

Recent work has demonstrated that the recovery of sugars from ILs could be fulfilled by extraction based on the chemical affinity of sugars to boronates such as phenyl boronic acid and naphthalene-2-boronic acid [71]. 90% of mono — and di-saccharides could be extracted up by boronate complexes from aqueous ILs solutions, pure ILs systems, or hydrolysates of corn stove-containing ILs.

After cellullose dissolution in ILs and its regeneration with an anti-solvent, the IL is usually filtered (or centrifuged) and washed with ethanol, acetone, ethyl ether, or water several times to remove by-products of wood degradation. Due to the IL low vapor pressure [60], the excess can be removed with a rotary evaporator, possibly at high temperatures, before the IL reuse [25, 31, 46, 55, 89]. It can also be separated with ethyl ether, then dissolved in acetonitrile/ethyl acetate and frozen for 24 h. The IL is then placed in a vacuum over at 90°C for 8 h before reuse [16].

Using these recycling procedures, the reuse of ILs for the pretreatment of native biomass through multiple cycles was reported. [BAIM][Cl] and [MAIM][Cl] could dissolve Metasequoia glyptostroboides wood sawdust without any efficiency loss after five cycles [16]. After five cycles, [EMIM][OAc] only lost 10% of its effi­ciency to dissolve maple wood flour [25]. The dissolution of rice straw in [EMIM][OAc] was repeated for 20 cycles with no reported efficiency loss. The cellulose recovery even increased over time due to the accumulation of dissolved cellulose residues that can be recovered in later cycles [46]. ILs, such as [BMIM][PF6] and [BMIM][BF4], were successfully recycled through multiple reaction cycles. Their recyclability was attributed mainly to their low solubility in some organic solvents or water. They can thus be extracted with an organic solvent or washed with water [60].

The IL recyclability is limited by the formation and accumulation of by­products or impurities. The degradation of cellulose was reported in reactions conducted at high temperatures [22, 36] or with acid catalysts [6, 32, 53]. The dehydration of free monosaccharides could lead to the formation of 5-hydrox — ymethylfurfural and furfural [53]. After the IL use for the dissolution of native biomass, 31P NMR spectra revealed signatures from 5-hydroxymethylfurfural, acetol, 2-methoxy-4-methylphenol, catechol, and acetic acid [32]. Even if these by-products can be avoided, the lignin extracted from the biomass accumulated in the recycled ILs with the increasing number of cycles [25, 32]. There was also accumulation of hemicelluloses, which are polar and have good affinity with polar ILs, such as 1-allyl-3-methylimidazolium chloride [32].

Wood naturally contains acid groups that can become free by hydrolysis and generate acids in the IL solution, such as acetic acid (pKa = 4.76) and glucuronic acid (pKa = 3.18) [31]. The generation of strong acids can protonate the acetate anion in [EMIM][OAc], for example, reduce the IL dissolution efficiency and complicate its recovery for reuse [31, 36]. Recycling is further limited by the high viscosity of ILs, which complicates handling and purification steps [36]. There­fore, efficient methods to separate the different dissolved products are necessary for the recycling of ILs [53].

Recycling efficiency depends on the anti-solvent used for the regeneration of wood after dissolution. Using water as the anti-solvent resulted in a higher yield of regenerated wood than using methanol [31]. The glucose yield after enzymatic hydrolysis after four cycles was also higher using water as an anti-solvent. In the case of E. grandis, the lower yield with methanol was explained by the larger amount of extractives dissolved in the methanol/IL mixture. However, after four recycling cycles, the recycled IL yield was 96% with methanol as the anti-solvent and 91% with water. At an industrial scale, water is preferable to methanol, since it is cheaper and more benign environmentally [31].

Another possibility is to replace the anti-solvent by an aqueous solution of phosphate, carbonate, or sulfate. The addition of a K3PO4 solution to the biomass solution led to the precipitation of the dissolved biomass and the appearance of a biphasic system with an IL-rich phase and a salt-rich phase. The extracted IL can then be dried and reused [43]. Phenylboronic acid and naphthalene-2-boronic acid were used to extract more glucose, xylose, cello- biose from IL/corn stover solutions after enzymatic hydrolysis in order to improve recyclability [45].

Significant activity of clostridia toward consuming lignocellulosic biomass uncovers the space of cheaper feedstock for ABE fermentation. However, efficient techniques for removing the inhibitors, generated during hydrolysis of lignocel — lulosic materials, can make it a more effective feedstock. From the economic point of view, the integrated system of hydrolysis, fermentation, and recovery process also opens vital ways to reduce the capital and operational cost of butanol syn­thesis. More developments in the recovery techniques such as gas stripping will boost up to this integrated fermentation process for improving the productivity. Additionally, completion of genome sequencing of two clostridial species pro­vided the crucial opportunity to genetic engineers to engineer the genome of butanol-producing species to improve its capability toward high yield and butanol resistance.

Contrary to E. coli, Z. mobilis is an ethanologenic bacterium and lacks the ability to metabolize hemi-celluloses derived monosaccharides, except glucose. There­fore, most of the engineering strategies applied to this bacterium intended to increase their substrate utilization range. In an earlier study, the strain CP4 has been shown to be the best ethanol producer from glucose. It was first engineered toward xylose utilization by the expression, on a plasmid, of the E. coli genes encoding for xylose isomerase (XI), xylulokinase (XK), transaldolase (TAL), and transketolase (TKL) under the control of strong constitutive promoters [206]. Ethanol yield from xylose fermentation attained 86% of the theoretical. The same approach was used to engineer the strain ATCC 39676 toward arabinose fer­mentation [35]. The genes from the E. coli operon araBAD, encoding L-arabinose isomerase (AI), L-ribulokinase (RK), L-ribulose-5P 4-epimerase (L-RPE), toge­ther with TAL and TKL allowed L-arabinose fermentation at high yield (96%) but at a low rate. This was ascribed to very low affinity of the glucose facilitator to L — arabinose. The same ATCC 39676 strain was used to express the xylose pathway, followed by successful long-term (149 d) adaptation in continuous fermentation of hemicellulose hydrolysates containing xylose, glucose, and acetic acid [106]. Finally, co-fermentation of glucose, xylose, and arabinose was obtained by genomic DNA integration (AX101 strain) of the xylose and arabinose pathways

[124] . The co-fermentation process yield was about 84%, with preferential order in sugar utilization: glucose first, then xylose, and arabinose last.

Life on earth is believed to have started with algae. Algae are photosynthetic organisms capable of converting solar energy into chemical energy and in the process, consume CO2 and release O2. With the advent of fossil fuels, the focus shifted from tapping solar energy via photosynthesis toward burning the fossil fuel to generate energy. The predominant use of fossil fuels over several years has made us realize the dangers that the GHGs released from burning these fuels, pose to the environment. The rapid depletion of these fuels and the millennia required for their renewal have now forced us to look for safer and renewable alternatives to the fossil fuels. Thus, we have now come a full circle toward again reverting to biomass, the unlimited solar energy and photosynthesis, to meet our ever increasing energy demands. Algae are considered to be the most photosyntheti­cally efficient plants on the earth. There is a large variety of algae ranging from small unicellular organisms to fairly complex and differentiated forms of multi­cellular organisms. They thrive on land as well as in water, using sunlight, CO2, and water for growth. Like other plants, algae use photosynthesis to convert solar energy into chemical energy and store it in the form of high energy substances such as oils, carbohydrates, and proteins. In other plants, the predominant store­house of this chemical energy is carbohydrates, whereas in algae, this energy is stored in the form of oils. A one-hectare algae farm on wasteland can produce over 10-100 times the oil produced by any other oil crop known till date [30]. The lipid content in different species of microalgae was found to vary from as low as 12-14% of dry weight in Scenedesmus obliquus, to as high as 80% of dry weight in Botyococcus braunii [31]. Algal energy is becoming increasingly popular because algae can be grown on wastelands and unarable lands, thus enabling all agri­cultural land completely available for growing food crops. Thus, third-generation biofuels, free from the food versus fuel controversy, can be obtained in abundance by efficient cultivation and harvesting of algae.

Algae can be classified on the basis of their fundamental cellular structure, life cycle, and pigmentation. According to the cellular structure, algae may be either unicellular or multicellular. The multicellular algae growing mainly in saltwater or freshwater are called macroalgae or ‘‘seaweeds’’. There are three different types of pigmentations seen in macroalgae: (1) green seaweed (Chlorophyceae), (2) red seaweed (Rhodophyceae), and (3) brown seaweed (Phaeophyceae). Microalgae which are microscopic photosynthetic organisms growing in both marine and freshwater environments are called microalgae (Cyanophyceae or blue-green algae).

In order to produce good quality briquettes, feed preparation is very important. For densification of biomass, it is important to know the feed parameters that influence the extrusion process. For different briquetting machines, the required parameters of raw materials like their particle size, moisture content, temperature are different. These are discussed below.

2.6.2.1 Effect of Particle Size

Particle size and shape are of great importance for densification. It is generally agreed that biomass material of 6-8 mm size with 10-20% powdery component (<4 mesh) gives the best results. Although the screw extruder which employs high pressure (1,000-1,500 bar), is capable of briquetting material of oversized parti­cles, the briquetting will not be smooth and clogging might take place at the entrance of the die resulting in jamming of the machine. The larger particles which are not conveyed through the screw start accumulating at the entry point and the steam produced due to high temperature (due to rotation of screw, heat conducted from the die and also if the material is preheated) inside the barrel of the machine starts condensing afresh. Cold feed results in the formation of lumps and leads to jamming. That is why the processing conditions should be changed to suit the requirements of each particular biomass. Therefore, it is desirable to crush larger particles to get a random distribution of particle size so that an adequate amount of sufficiently small particles is present for embedding into the larger particles.

The presence of different size particles improves the packing dynamics and also contributes to high static strength. Only fine and powdered particles of size <1 mm are not suitable for a screw extruder because they are less dense, more cohesive, non-free flowing entities.

In addition to acids, other catalysts such as Li salts (LiCl, LiBr, LiAc, LiNO3, or LiClO4) were added to enhance the dissolution of cellulose in [EMIM][OAc]. It was believed that the lithium cation can disrupt the hydrogen bonding network in cellulose [54].

Two polyoxometalates, an acidic form H5[PV2Mo10O4o] and an [EMIM][OAc] compatible form [1-ethyl-3-methylimidazolium]4H[PV2Mo10O40], were prepared and used as catalysts for the dissolution of southern yellow pine (particle size <0.125 mm, 5 wt%) in [EMIM][OAc] at 110°C [55]. The addition of 0.5 wt% acidic polyoxometalate reduced the time for complete dissolution of pine from 46 to 15 h. The regenerated cellulose contained significantly less lignin, with limited losses in cellulose. The [EMIM]-compatible form improved delignification, but with greater cellulose losses in the regenerated cellulose [55].

Bartosz Rozmystowicz, Paivi Maki-Arvela and Dmitry Yu. Murzin

6.1 Introduction

Constant decrease of fossil fuels reserves creates a great need for development of the new technologies for production of liquid transportation fuels based on renewable sources. World crude oil reserves, according to OPEC [1], are at the level of 1,337.2 billion barrels. In year 2010 the daily world consumption reached

86.6 million barrels per day (mb/d) [2] with the forecast of increase by 1.4 and

1.6 mb/d in the following 2 years [2]. Even with an assumption that the world fuel consumption will be maintained at the same level, reserves of oil should run out in approximately 40 years. This threat of oil pools depletion leads to an increase of interest in biofuels, both by governments and industries.

Recently in numerous countries legislation measures were taken to increase biofuels share in transportation fuels. European Union will increase the usage of biofuels both in gasoline and diesel to 10% by 2020 [3]. Analogous regulations were proposed in China, Brazil, India, and USA which indicates that biofuels will have significant share of liquid transportation fuels market.

Renewable fuels can be named as first or second generation biofuels depending on the origin. The first-generation biofuels are made mainly from crops. To produce bioethanol cereals, maize or sugar beet is used, whereas biodiesel feed­stock consists of canola, soybean, or palm oil. There is a great concern that the production of those fuels in large scale could in a significant way decrease food cropland. Therefore, the second generation of biofuels was introduced using non-food crops source such us lignocellulosic residues, tall oil, or algae. Another advantage of those fuels is lower emissions of CO2 per unit of energy content

B. Rozmystowicz • P. Maki-Arvela • D. Yu. Murzin (H)

Laboratory of Industrial Chemistry and Reaction Engineering, Process Chemistry Centre, Abo Akademi University, 20500 Turku/Abo, Finland e-mail: dmurzin@abo. fi

C. Baskar et al. (eds.), Biomass Conversion,

DOI: 10.1007/978-3-642-28418-2_6, © Springer-Verlag Berlin Heidelberg 2012

(Fig. 6.1) by LCA WTW assessment (life cycle assessment, well-to wheel), which is the product life cycle starting from extraction to waste disposal [4, 5].

While in first — and second-generation bioethanol there are no differences in composition of the fuel, significant differences could be observed in the case of biodiesel. Originally biodiesel was the name connected to fatty acids methyl ester (FAME), but recently new technologies emerged for production of diesel fuels that originates from biomass. Deoxygenation of fatty acids is a process involving hydrodeoxygenation (HDO) or decarboxylation/decarbonylation of carboxylic group that leads to formation of diesel-like hydrocarbons (Green diesel). Process of HDO is already applied on industrial scale by Neste Oil (NExBTL oil). There are three units already operating (Singapore and two in Finland) and one in construction (Rotterdam, which should be ready in the end of the year 2011) with combined capacity of around 2 million tons per year [6]. In this process fatty acids are converted to aliphatic hydrocarbons, which is advantageous compared to transesterification method, where products contain significant amount of oxygen.

The other option besides HDO is decarboxylation/decarbonylation. Pioneering work was performed recently with participation of the authors [7-9]. It was found that it is possible to remove carboxylic group using heterogeneous catalysts, with less hydrogen consumption than in HDO process.

In this chapter comparison of different routes of deoxygenation will be described as well as recent research in this field.

Sugarcane bagasse is the wastes from the sugar factory. It is obtained as a left-over material after the juice is extracted from the sugarcane. Sugarcane bagasse (SCB) was analyzed for its composition, structure, and surface properties (Table 9.4).

Because of its lower ash content, 1.9% [111], bagasse offers numerous advantages compared to other agro-based residues such as paddy straw, 16% [4], rice straw, 14.5% [60], and wheat straw, 9.2% [210]. In another study, SCB was obtained from a small sugarcane juice factory and milled for analysis of different types of fibers. It is important to note that most developments in SCB transformation into sugars and ethanol have a common scientific base with other lignocellulosic materials, due to considerable similarity in composition and structure.

A variable volume fed-batch culture was adopted (incremental feeding of same concentration solution to that of initial medium resulting in an increase in volume). All the fermentations were performed in a fed-batch mode in a 5-l bioreactor controlled by a computer having advanced fermentation soft water. The fermenter was equipped with temperature, agitation, and aeration systems with precise control for these parameters. Aeration was measured in terms of dissolved oxygen. The parameters were measured and automonitored against the set values. The pH was, however, controlled manually by adding acid or alkali as the case may be. The volume of incremental feeding was adjusted in such a way that the final volume in the fermenter reached to about 4.75-5.00 l. The samples were drawn using sampling port at a 2-h interval during fermentation using injection syringe under aseptic conditions. Incremental feeding was started after 1 h of actual start of fermentation (called as activation period) and stopped before 1 h of actual completion of fermentation (called as terminal cell maturity step). For incremental feeding additional accessories were attached to the fermenter. Generally, 7-8 h were taken by incremental feeding at this rate. The fermentation parameters were kept arbitrary but constant, except that used for standardization during the parameter optimization experiments.