Of the three major biopolymers that constitute wood, lignin is distinctly different from the other macromolecular polymers [34]. Lignin is an amorphous, cross-linked, and three-dimensional polyphenolic polymer that is synthesized by enzymatic de- hydrogenative polymerization of 4-hydroxyphenyl propanoid units [35, 36]. The biosynthesis of lignin stems from the polymerization of three types of phenylpropane units as monolignols: coniferyl, sinapyl, andp-coumaryl alcohol [37, 38]. Figure 8.4 depicts these three structures. It has been identified that lignin from softwood is made up of more than 90 % of coniferyl alcohol with the remaining being mainly p-coumaryl alcohol units. Contrary to SW, lignin contained in hardwood is made up of varying ratios of coniferyl, sinapyl, and typically lesser amounts of p-coumaryl alcohol type of units.

The polymerization process is initiated by an enzyme-catalyzed oxidation of the monolignol phenolic hydroxyl groups to yield free radicals. A monolignol free radical can then couple with another monolignol to generate a dilignol. Subsequent nucle­ophilic attack by water, alcohols, or phenolic hydroxyl groups on the benzyl carbon of the quinone methide intermediate, restores the aromaticity of the benzene ring. The generated dilignols then undergo further polymerization to form protolignin.

Although the exact structure of protolignin is unknown, improvements in meth­ods for identifying lignin-degradation products and advancements in spectroscopic methods have enabled scientists to elucidate the predominant structural features of lignin. Table 8.3 showed the typical abundance of common linkages and functional groups found in softwood lignin [39, 40].

The property of polydispersity, just as with hemicellulose, characterizes lignin as well. The DP for softwood lignin is approximately 60-100 and the molecular weight is in excess of 10,000 [41, 42].

R,= OMe, R2 = H Ri= R2 = OMe R,= R2 =H

Lignin in wood behaves as an insoluble three-dimensional network. It plays an important role in the cell’s endurance and development, as it affects the transport of water, nutrients, and metabolites in the plant cell. It acts as binder between cells cre­ating a composite material that has a remarkable resistance to impact, compression, and bending [26].

Lignin is much less hydrophilic than either cellulose or hemicelluloses, and it has a general effect of inhibiting water adsorption and fiber swelling. Solvents that have been identified to significantly dissolve lignin include low molecular alcohols, dioxane, acetone, pyridine, dimethyl sulfoxide and select ionic liquids. Further­more, it has been observed that at elevated temperatures, thermal softening of lignin takes place, which allows depolymerization reactions of acidic or alkaline nature to accelerate [43].

As discussed earlier, the role and importance of VA is very essential in bio — ligninolysis. It is generally synthesized de novo from glucose via shikimate pathway at the early stage of secondary metabolism in parallel with the LiPs production. The biosynthetic pathway for VA was performed with 14 C isotope trapping exper­iments in the ligninolytic fungus P. chrysosporium (ATCC 34541); and concluded that the pathway proceeds as follows: Phenylalanine ^ cinnamic acid ^ Benzoate/ Benzaldehyde ^ VA [115, 123]. In P. chrysosporium, VA production is induced by nitrogen-limitation, whereas in Bjerkandera sp., the nitrogen element does not have any significant regulatory effect on the VA biosynthesis [124]. Furthermore, LiPs action on non-phenolic residue of lignin can be enhanced by addition of VA in bio­logical pretreatment of lignocellulose. As per Hammel et al. [115], VA protects LiPs against H2O2-mediated inactivation reaction (rate limiting step) in the LiPs catalytic cycle reaction (Fig. 1.2) and it has been proposed that VA act, in vivo as a stabilizer for the enzymes.

One of the more successful continuous high pressure steam pre-treatment processes for lignocellulosic biomass use has been the StakeTech reactor developed by Stake Technology, Ontario, Canada. The StakeTech reactor is primarily composed of a stainless steel horizontal pressure vessel which is designed to withstand operational pressures of 31 bar (i. e. 450 psig) [13]. Using an upstream screw conveyor the digester (i. e. the part of the vessel where the main stream treatment occurs) is fed continuously by the movement of the lignocellulosic biomass through the compression tube which also serves to build up the vessel pressure. The densifled biomass resulting from the feeder compression tube upon entering the reactor is transferred to a conical choke which facilitates a ‘break up’ of the dense biomass plug resulting in a scattering of undensified biomass materials onto the retention screw [13]. Once fed, the retention screw transports the biomass towards the discharge end in such a manner that the precise retention time is achieved to meet the required processing conditions. The post treated material is then conveyed to the discharge valve (which is made up of a rotary ball valve with its opening timed according to the desired production rate), using the discharge screw at the end of the digester [13]. The discharge valve further aids in providing an explosion effect on emptying the pre-treated samples while readying it for the subsequent processing stages. The StakeTech continuous technology has been widely used in research and has been reported to reach full commercialisation [13].

Apart from the batch and continuous pre-treatment processes previously outlined, various patents, that is [38-41] on the use and optimisation of the catalysed and auto-hydrolytic steam pre-treatment technologies, as well as improving the purity and accessibility of the desired process outputs from a wide variety of lignocellulosic biomass inputs have been granted and are continuously being researched.

3.2 Conclusion

In this chapter, different types of steam pre-treatment have been described, and examples of commercial implementations of steam pre-treatment have been given. It is clear that for biomass to play the future role of major supplier of carbon for energy and products, full utilisation of all parts of the biomass is necessary. This necessitates technologies that go beyond high quality, cellulosic feedstocks, targeting the more abundant yet significantly more complex lignocellulosic fractions. For this purpose, steam pre-treatment offers an efficient yet environmentally benign approach to opening the biomolecular structure, and preparing the substrate for subsequent separation or conversion. Uniqueness for steam pre-treatment is that it is able to address most of the limiting factors for subsequent digestion (conversion) and at the same time turn the feed into a pumpable slurry or paste, even without the addition of catalysts or other additives, which call for subsequent separation or effluent treatment. Being a thermal process, it can be possible to design the entire plant with a significant amount of heat integration, reducing the need for external supply of heat for steam generation.

5.3.1 Drying

Drying is a crucial step in the densification process of moist material. The optimum MC for pelletization was reported between 9 % and 15 % depending on feedstock species. However, the final quality of the feedstock can be affected by different factors during drying, and these include the type of dryer, drying conditions, drying medium, and the biomass characteristics. The degree to which the material is agitated and broken up in the drying process, the residence time and temperature of the material in the dryer influence the downstream densification process. In rotary dryers, the biomass is agitated to some extent and thus a significant amount of fines is usually produced in these dryers. The prolong drying of the fine particles caused by lag time of rotary dryer may degrade the biomass that affects the final quality of the resulted pellet. In contrast, PMB dryers are designed to circulate the drying medium through biomass layers, and the biomass is relatively stable on the bed, which can act as a filtration bed to trap any fines from the exhaust, resulting in low particulate emission.

Different components and products may be emitted in gas phase during drying. These include entrained fine particulates, volatile organic components (VOCs), and products of thermal degradation of the biomass [13]. The vaporized components can be further categorized into those that remain volatile at ambient conditions and those that condense after drying the stack. The most volatile components consist of mono-terpenes, which are naturally emitted from wood at ambient temperatures and the emission rate increases with temperature, particularly above 100 °C.

The condensable category consists of extractable components such as fatty acids, resin acids, di-terpenes, and tri-terpenes. Although these have high boiling points, they have sufficient vapor pressure at high drying temperatures (180-220 °C) to be released from wood. They are responsible for the formation of blue haze, a blue-gray discoloration of the exhaust gas from a wood dryer [16].

Thermal degradation products, such as formic and acetic acids, alcohols, aldehy­des, furfurals, and carbon dioxide, are released at higher drying temperatures (200 °C or higher) when pre-pyrolysis occurs. Increasing the wood temperature rapidly in­creases the amount of thermal degradation products that have a strong smell. The formation of these degradation products (pseudo-lignin) can influence the quality of produced wood pellets and improve the binding characteristics of woody pellets [63].

The wood color change (darkening) during drying is another issue to produce high-value pellets. Chromophoric groups (carboxylates and phenol) may be produced within the lignin or extractive molecules at high temperatures and humidity. During drying furfural and some polysaccharides with low molecular weights are created from hemicelluloses degradation. As these components are dark in color (blue green), they lead to the darker color in wood appearance [64].

Syrups and molasses have long been used as substrates for fermentative production of commercial polysaccharides such as pullulan [46-49], xanthan [57], dextran [38], scleroglucan [32], levan [33, 41, 42], and gellan [39] due to their many advantages like high sucrose and other nutrient contents, low cost and ready availability, and ease of storage. Using molasses in crude form resulted in low pullulan [46] and scleroglucan [32] yields which in turn pointed out the need for pretreatments. Pre­treatment of sugar beet molasses with sulfuric acid has been reported for pullulan

[47] and levan [33] production but, when acid treatment was combined with activated carbon treatment, significant improvements in pullulan [48,49] and levan [33] yields

were obtained, most probably due to the removal of heavy metals and colored sub­stances. Activated carbon is particularly known for its efficiency in removing heavy metal pollutants [64]. However, after a systematic study on the effect of different pretreatments on the heavy metal distribution of starch and beet molasses samples, Kuquka§ik et al. [33] reported a drastic increase in the dissolved iron (Fe2+) content after the activated carbon treatment. This has been attributed to the reduction of iron from its impregnated Fe3+ form to its soluble form since this increase in soluble iron was more profound when acid treated samples were subjected to activated carbon treatment [33]. Same authors suggested tricalcium phosphate (Ca3(PO4)2,TCP) treat­ment as an effective method for selective removal of iron and zinc from molasses or other mixtures of comparable composition. Heavy metals like iron, zinc, and nickel are known to enter the apatite crystal structure of TCP by replacing the Ca atom

[65] . Goksungur et al. [49] applied potassium ferrocyanide (K3[Fe(CN)6]) treatment to precipitate heavy metals. For the microbial levan production with Paenibacillus polymyxa NRRL B-18475, clarified sugar cane syrup and crude sugar beet molasses resulted in very low yields. Therefore, peptone was added to the cane syrup and beet molasses was subjected to various expensive pretreatments like passing it through gel filtration and anion exchange columns in order to increase the levan yields to levels comparable with sucrose [41]. To produce levan from Zymomonas mobilis, both sug­arcane molasses and sugarcane syrup were clarified by centrifugation followed by filtration and then used at 250 g/L carbohydrate concentration [42]. For levan produc­tion by halophilic Halomonas sp. cultures, sugar beet molasses, and starch molasses were subjected to five different physical and chemical pretreatment methods and their combinations, that is, clarification, pH adjustment, sulfuric acid, TCP and activated carbon treatment [33]. In both molasses types, pretreatments like clarification, pH adjustment were not adequate as also reflected by the low EPS production yields due to the retained undesirable constituents (e. g., heavy metals, impurities) which influence the growth of microorganism and associated polysaccharide production [47, 66, 67]. Highest levan yields were obtained with sugar beet molasses pretreated with TCP followed by acidification with sulfuric acid and then subjected to activated carbon pretreatment [33]. On the other hand, Kalogiannis et al. [57] applied various treatment methods to sugar beet molasses including aeration, acid, activated carbon, K3[Fe(CN)6] treatments, and ion exchange chromatography, however, none of the pretreatments improved the xanthan yield of X. campestris ATCC 1395 cultures and the highest production was obtained with the untreated crude molasses. Crude beet molasses is also used to produce dextran by L. mesenteroides bacterial cultures and the yields were comparable to those of media containing pure sucrose [38]. Banik et al. [39] used Response surface methodology to optimize the production of gellan gum by S. paucimobilis ATCC-31461 using crude sugarcane molasses and reported a maximum yield of 13.81 g/L gellan. Survase et al. [32] used various dilutions of coconut water, sugarcane molasses, and sugarcane juice for scleroglucan production by filamentous fungi S. rolfsii MTCC 2156 and obtained the highest yields (23.87 g/L in 72 h) from sugarcane juice that was obtained from a local market and hence did not require any pretreatments before use. Coconut water and sugarcane juice were also used for EPS production by Lactobacillus confusus cultures [68].

In the process of steam explosion, more than 50 % hemicellulose degrades into pentose. Therefore, it is possible to fractionate hemicellulose by washing materials pretreated with steam explosion. As per the analysis in Sect. 4.1, materials pretreated with steam explosion formed porous structure, which enhance the accessibility of a substrate for solvent or other chemicals. So, steam exploded materials are suitable for component fractionation.

Research reveals that lignin could be separated by alkali oxygenation [36]. There­fore, stalk is pretreated with steam explosion-washing-alkali extraction technology to realize fractionation in the component level.

Fig. 4.3 Steam explosion-washing-alkali integrated technology chart

Organic solvent is applied in the process of clean fractionation technology from National Renewable Energy Laboratory [37]. By this way, cellulose, hemicellulose, and lignin in lignocelluloses are also separated. However, compared with organic solvent steam and alkali are cost-competitive. Moreover, alkali solvent could be recycled by nanofiltration.

To improve the yield of hemicellulose [38], corn stalk exploded with steam at severity of logioR0 [min] = 3.05 is pretreated with 0.3-0.5 % dilute sulfuric acid at 110-120 °C for 0.5-1 h. The ratio of dilute acid to solid is 1:5-1:7, and then liquid and solid are separated with spiral extruder. As a result, hemicellulose yield reaches to 30.7 %. Lignin in solid is fractionated with 0.5-2 % NaOH for 2-3 h at 150­160 °C. The ratio of solid to liquid is 1:5-1:7 (based on the hydrolysis residues). The extract liquid is filtrated sequentially with precise filter, ultrafiltration membrane, and nanofiltration membrane. Filtrate from the nanofiltration membrane is a NaOH so­lution that could be reused. Concentrated solution from the ultrafiltration membrane is a lignin solution. The lignin solution is sent to neutralization tank to get lignin precipitation that is washed to get lignin product. The whole chart is as Fig. 4.3.As a result, lignin yields are 16.80 %, and cellulose content in cellulose fraction reaches 63 % from 25.4 %.

Wood is essentially composed of cellulose, hemicelluloses, lignin, and extractives. Table 7.1 presents major chemical compositions of some wood species.

7.2 The Biological Approach for Biomass Hydrolysis into Sugars

The biological approach for hydrolysis of biomass is composed of a pretreatment phase, to make the lignocellulosic material such as wood open to hydrolysis, fol­lowed by cellulose and hemicelluloses enzymatic hydrolysis to break them down into sugars; finally, separation of the sugar solution from the residual materials, mostly lignin and also some enzymes adsorbed on the lignin [3].

Fig. 7.1 Main structure of galactoglucomannans in softwood hemicellulose (R = CH3CO or H)

In addition to this, yields of enzymatic hydrolysis are ca. 80-85 %, the obtained sugar concentration in the medium is low as <12.5 %, and the energy costs for both pretreatment and excess water removal is relatively high, and ca. 20 % of the sugar product is required to make the processing enzymes. Also there is no past experience with such process at industrial scale.

Due to all the above-mentioned withdrawals in using a biological process, which might be the technology of choice in other cases, the use of chemical process seems
to be an optimal route for hydrolysis of biomass into sugars. This approach is further presented in this chapter.

Blanchette [86] has described two kinds of soft-rot: type I consisting of biconical or cylindrical cavities that are formed within secondary walls and type II refers to an erosion form of degradation. For example, Daldinia concentrica is the most efficient fungus of type II group, which primarily affect hardwood. Nilsson et al. [94] found 53% weight loss in birch wood within 2 months. During early stage of classification of different wood-rotting fungi, Xylariaceous ascomycetes from genera such as Daldinia, Hypoxylon, and Xylaria have often been regarded as white — rot fungi, but today these fungi are categorized to soft-rot fungi as they cause typical type II soft-rot in wood. In coniferous wood (e. g., pine wood), the weight loss was very low and it has been thought that these type of woods have more guaiacyl units in middle lamella, which inhibit the growth of soft-rot fungi [49].

Some microfungi (Penicillium chrysogenum, Fusarium oxysporum, and Fusarium solani) identified in a forest soil sample are able to mineralize grass lignins upto 27 % [49]. However, most of the soft-rot and microfungi consume readily economically important carbohydrates during invading and have very less applications in biological pretreatment.

Also referred to as auto-hydrolysis, the use of steam for the pre-treatment of lignocel — lulosic biomass is one of the most widely demonstrated and implemented methods in research and commercial facilities [7, 8]. This method employs the use of high — pressured saturated steam for the treatment of the lignocellulosic biomass. Steam pre-treatment is usually conducted at reaction temperatures of 160-260 °C using high pressures of 0.69-4.83 MPa with the period in which the biomass is exposed to this conditions ranging from several seconds to a few minutes [9]. A further clas­sification of the steam pre-treatment can be made depending on if the use of high pressured steam in the process is followed by a sudden reduction of the process pressures. The process is thus regarded as ‘Steam explosion’ where such abrupt de — pressurisation and cooling of the lignocellulosic biomass after a specified reaction period is implemented. This is since the abrupt pressure reduction results in the ex­plosive decompression of the biomass materials, thus facilitating a disruption of the lignocellulosic biomass cell walls and enhancing the accessibility of the biomass macromolecular contents. The conditions facilitated by the high-pressured steam (for both explosive and non-explosive methods) facilitates a disruption of the lignin sheath and enhances a solubilisation of the biomass hemicellulosic component (via hydrolysis), thus aiding the accessibility of cellulose to further conversion methods, that is, enzymatic hydrolysis [10]. For example, a 90 % enzymatic hydrolysis of Poplar biomass chips was achieved in 24 h after it was subjected to a steam explo­sion pre-treatment compared to only a 15 % hydrolysis obtained for the untreated chips [11].

During the steam pre-treatment, some of the biomass lignin is solubilised (which however re-polymerises on cooling and forms a part of the acid-soluble lignin frac­tion), with the biomass hemicellulose solubilised after the application of the steam pre-treatment and is subsequently recovered in the aqueous fraction (or is further degraded to other compounds, i. e. furfural) [ 12]. Most of the cellulose content of the biomass is preserved in the solid fraction, however, its hydrolysis to glucose could be obtained under high steam pre-treatment temperature conditions (i. e. >200 °C) [12]. The hydrolysis of the hemicelluloses is proposed to be brought about mainly by the action of acetic acid formed from the acetyl groups released during the steam pre­treatment [9]. In addition to this, other acids, that is, formic, levulinic and pyromucic acids, produced during the steam pre-treatment process as described [13], may also play an important role in the acid catalysed breakdown of the hemicellulosic glyco — sidic bonds [13]. Furthermore, water at high temperatures has been demonstrated to possess some acid properties, which could also enhance the hemicelluloses hydrol­ysis [14]. The acidic conditions provided by the steam pre-treatment could thus also lead to the degradation of available sugars in the biomass materials [15].

The potential benefits accruable with the use of steam pre-treatment over other pre-treatment systems have been widely demonstrated in the literature. Compared to the use of commonly applied chemical pre-treatment techniques, that is, sodium hy­droxide, calcium hydroxide (lime) and dilute sulphuric acid hydrolysis, higher treated products recovery as well as improved substrate availability for further processing has been obtained with the application of auto-hydrolysis [16-19]. The implementa­tion of mechanical biomass pre-treatment routes such as milling and novel methods like microwave radiation have also been demonstrated to less effective than the use of high-pressure steam for the pre-treatment of lignocellulosic biomass [18, 19]. Furthermore, the application of non catalytic steam pre-treatment methods was dis­cussed in to have a lower process energy requirements compared to mechanical comminution methods (70 % less, to achieve similar size reductions), with no or a low recycling or environmental costs attached [9].

The non-catalysed steam pre-treatment method (including steam explosion tech­niques) has been widely demonstrated both in the literature and in practice to be utilisable for a variety of lignocellulosic biomass, that is, from forest products and residues (including the use of shortrotation woody biomass) [17,18,20-22], purpose grown energy crops [9], as well as from agricultural residues [16, 19]. Regarding the use of woody biomass, it was seen that the use of younger biomass materials were easier to fractionate during the steam pre-treatment process and thus better substrates production for subsequent enzymatic hydrolysis, when compared to the use of older materials [13, 23]. This is, therefore, promising for the integration of steam pre-treatment with ongoing and proposed large-scale short rotation forest and fast growing purpose grown lignocellulosic biomass projects. The use of this method is however less effective for the pre-treatment of softwood (i. e. pine) where the acid catalysed route as described in Sect. 3.2.2.1 is better suited.

Grinding unit is the second largest electricity consumer in pellet production (pel­let mill is the maximum electricity consumer) [34]. The whole process of pellet production is divided into: raw material, general investment, drying, grinding, pel­letization, cooling, storage, peripheral equipment, and personnel. If the raw material is wet, grinding cost is 3 % of the whole cost, and if the raw material is dry, the grinding cost is 2 % of the whole cost. It should be considered that sawdust is the base raw material [34]. A hammer mill cost ranges between 62,000 € (with a ca­pacity of about 2.5-3 t (d. b.)/h) and 168,000 € (two large hammer mills with a total capacity of about 9 t(d. b.)/h) depending on the plant size and the equipment used. The maintenance cost is 18 % of the investment costs per year. Scale of operation can impact overall costs, for example the overall cost of pellet production is lower in Sweden because of their larger capacity plants and lower electricity price.

One step of grinding is sufficient for material preparation for the traditional pellet production using sawdust and shaving as feedstock. When the feedstock sources change from sawdust and shaving to other mill residues (e. g., logging residues, thinning materials, agricultural residues, and short rotational trees), two or more steps of size reduction is needed to prepare the material for pelletization. Two stages of size reduction using a coarse grinder and a hammer mill were used for the pellet production of short rotation trees [35].