The optimization of total sugars yield was conducted based on the model generated in Eq. 9.2. The 2D contour plot (Fig. 9.2) gave the optimization result based on the quadratic model (response surface). The maximum predicted response was 32.6 % of total sugar yield (pretreated at 200 °C for 18.2 min and 20 % solid loading followed by cellulase treatment). Pretreatment examinations were conducted at the optimal condition (200 °C, 18 min, 20 %) to confirm the predicted model. Heating the reactor

Total sugars

A: Temperature

Fig. 9.2 The contour plot of RSM for optimization of total sugars yield (the maximum predicted response was 32.6 % of total sugar yield at 200 °C for 18.2 min and 20 % solid loading)

to 200 °C took about 10 min and the total sugar yield was 34 % on average, which was a little higher than model predicted value. So this model was confirmed to be reliable for determining pretreatment condition. Chemical composition analysis indicated that a complete hydrolysis would theoretically yield 75 % total sugars. However, results from these 20 experiments followed by cellulase treatment did not achieve the theoretical maximum yield. This is likely due to the low enzyme activity

Enzyme loading (% dry wood)

of the commercial cellulase used in this preliminary trial. Therefore, two industrial enzymes were chosen to conduct further enzymatic hydrolysis trials on the optimal pretreatment condition.

Entrained flow gasifiers consist of a co-current plug flow reactor and have been extensively studied and used for coal gasification. The gas and particle residence times in the reactor are very short (a few seconds), which requires very small size feedstock (100 ц-m) and very high temperature (>1,000 °C) to maximize conversion. There are two types of entrained flow gasifiers: slagging and non-slagging. In the slagging gasifier, the ash melts in the gasifier, flows down the walls of the reactor and finally leaves the reactor as a liquid slag. This type of entrained flow gasifier is preferred for biomass. In a non-slagging gasifier, the walls are free of slag, which is suitable for feedstock with low ash content [71]. Some of the advantages of entrained flow gasifiers are low tar and methane content, high carbon conversion, and ash existing as slag. Furthermore, the reactor can be operated, with a wide variety of feedstock at high pressure and temperature. Due to some problems associated with biomass molten ash and the use of very fine biomass particles, however, the use of an entrained flow biomass gasifier has been rather limited. One example of an entrained flow gasifier application for biomass is the Choren process with a maximum capacity of 45 MW h and using wood as feedstock [72].

The choice of reactor technology is critical to follow the desired kinetic pathway in pyrolysis. The emphasis of this section will be on selected reactor technologies that have been successfully demonstrated at large scale with a pilot-scale or larger unit. Several reactor concepts have been demonstrated in the scientific literature at small-scales, but they will not be considered.

The mass and energy balances are the main fundamental and semi-empirical tools to design reactors when coupled to reaction kinetics. These balances involve heat and mass transfer equations, which are dependent on the system gas/solid hydrodynam­ics. Several handbooks are dedicated to reactor design with various hydrodynamics models to represent these equipments. However, the objective of this section is not to review these models in details. Considering that heat transfer is the limiting step in pyrolysis reactors, the emphasis is put on heat transfer parameters evaluation.

Figures 13.4 and 13.5 show the refining process of the RCF facility. The facility produces less than 100 Nm3/day, making it a Class 2 producer according to the High Pressure Gas Safety Act [13]. It is also qualified by law to be a “mobile production facility”. As this mobile production facility can be moved with a crane, the fixed asset tax does not apply. Biogas purification is carried out by membrane separation inside the facility. In order to create a standardized caloric quantity required for town gas 12A, a device that adjusts the caloric content by adding propane was installed. The gas at the completion of the refining process is temporarily stored in storage units below the RCF facility. High-pressure filling of transportable cylinders using high-pressure boosters makes it possible for the biogas to be utilized outside the farm production system as fuel for general households and compressed natural gas (CNG) vehicles.

13.2 On-site Field Testing at Japanese Rural Area

13.3.1 Biogas Production from Biogas Plant

The farm where the biogas purification facility was established is located in central Hokkaido (Hokkaido: northern part of Japan). A free-stall

The total volume of the anaerobic fermentation tank is 396 m3 and the effec­tive volume is 330 m3. Operating conditions include a fermentation temperature at 55 °C and a hydraulic retention time (HRT) of 15 days. The daily material input is 22 m3/day of dairy cow slurry (density assumption: 1 kg/L dairy cow slurry=1 kg/L water). Material is added to the fermenter twice after manure removal. This is done at 9:30 in the morning and 4:30 in the evening. In order to prevent hydrogen sulfide from affecting the biogas purification membrane film, the hydrogen sulfide con­centration must be virtually 0. Desulfurization equipment that combines microbial desulfurization and dry desulfurization was used. After desulfurization, the biogas hydrogen sulfide concentration was virtually 0 ppm.

Figure 13.7 shows biogas production from an on-farm biogas plant. Table 13.1 shows the operation conditions in the anaerobic fermentation tank. In winter, the biogas produced was reduced to a low raw material supply volume, because slurry on the free-stall floor was frozen by the cold outside air.

The loading rate was 3.87 kg VS/m3/day, the average amount of biogas produced was 450 ± 23 m3/day, and the average methane concentration was 58 ± 1.8 %. The input materials and digestive slurry had a high ammonia nitrogen level compared to

Table 13.3 Average composition of raw gas, off-gas, and refined gas

the typical value, but the concentration of methane was the same as in other reported cases [14, 15].

Furthermore, lack of accumulation of propionic acid meant that the fermentation state was favorable. In Japan, the merits of methane fermentation include not only energy production, but also its role in odor mitigation countermeasures. Proprionic acid, one of the main malodorous organic acids produced in the dairy cow slurry due to the anaerobic fermentation process, was 95 % decomposed by methane fermentation processing. The odor intensity of the post-treatment digestive slurry was reduced to 1/50 of the input material. Thus, the refinement process provided highly effective odor control in neighboring villages when the digestive slurry was sprayed over grasslands.

Of all the solid acid supports, carbonaceous support seems to be the most effec­tive. It was reported that — SO3H groups are linked to amorphous structure in the form of C-O-SO3H [31]. As mentioned in Sect. 15.3.1, a carbon material can in­corporate large amounts of hydrophilic molecules, which provide good access to P-1,4glucans by — SO3H groups, giving high catalytic performance for hydrolysis [42]. The incorporation results in the decease of activation energy for the hydrolysis of cellulose.

Suganuma et al. [42] reported that carbonaceous solid acids obtained at mild carbonization temperatures (ca. 450 °C) exhibited high catalytic activity, because — SO3H groups were bonded to carbon sheets with poor cross-linking, which made it easy for access to reactants by the — SO3H groups. However, when carbonized at higher temperature (ca. 550 °C), most of the — SO3H groups bonded to the carbon sheets were not located on surface, which resulted in a poor catalytic performance. Onda et al. [31] used sulfonated activated-carbon catalyst to selectively hydrolyze cellulose at 150 °C for 24 h, and obtained 40.5 % glucose yield with 90 % glu­cose selectivity. Only few amount of SO42- was leached (<0.03 mmol/L) after the reaction.

Carbonaceous solid acid catalysts ground to nanosize (10-100 nm) could achieve high catalytic activity. Vyver et al. [79] proposed an effective conversion tech­nique for cellulose hydrolysis by using a sulfonated silica/carbon nano-composite as catalyst. Glucose formation was faster on the nano-composite as compared with ion-exchange resins (glucose yield of 50 % vs. 29 %) under reaction conditions of 150 °C for 24 h with 0.05 g of ball-milling cellulose, 0.05 g of catalysts, and 5 mL of water. The catalyst was superior in the formation rate of glucose (4.6 ^mol h-1), and turnover frequency (0.37 h-1) as compared with typical sugar catalysts (formation rate of glucose, 0.93 h-1; turnover frequency, 0.06 h-1). However, separation and re­covery of it from un-hydrolyzed cellulose residues are needed for further study. Lai et al. [41] developed a process for recovering carbonaceous solid catalysts by using a paramagnetic solid acid (Fe3O4-SBA-SO3H). When microcrystalline cellulose was pretreated with [BMIM][Cl], glucose yield reached 52 % in 3h. The incorporation of paramagnetic nanoparticles into the carbonaceous carriers not only provides good access of reactants to the — SO3H groups, but also has functional characteristics that allow it to be separated and regenerated.

Carbonaceous solid acid catalysts are considered as the most promising catalyst for cellulose hydrolysis, since they provide good access of reactants to the acidic sites of — SO3H groups. High glucose yields of up to 75 % with 80 % selectivity have been achieved at 150 °C for 24 h with carbonaceous solid acid catalysts. However, sep­aration of carbonaceous solid acid catalysts from un-hydrolyzed cellulose residues after hydrolysis needs further research since these catalysts have similar physical and chemical properties to the residues. Use of functionalized carbonaceous solid acid catalysts that contain paramagnetic groups is one method to improve carbonaceous solid acid catalysts’ separation and reuse.

Biological pre-treatment is a slow process which can last for few weeks. However, the process has a mild reaction, less energy demand, low chemical usage, low capital cost, less side reaction, and is environment friendly [59,60]. The process if compared to other pre-treatment processes can be an attractive alternative process due to its green technology. This pre-treatment employs microorganism and their enzymatic mechanism to break down the lignin and liberate cellulose and hemicellulose from the complex lignin. These selective microorganisms which produce oxidative enzyme to break the lignin are from type of fungi and bacteria.

Fungi, the wood rotten microorganism can be divided into three groups according to the morphology of wood decay which are soft-rot ascomycetes or detromycetes, white-and brown-rot basidomycetous. Degradation of the lignin and hemicelluloses by the action of white-rot fungi is an aerobic process but there are some bacteria like E. lignoluticus SCF1 and rumen microorganisms with lignin degrading capability under anaerobic condition [16, 17].

These fungi have shown positive effect on delignification process [61, 62]. White — and soft-rot fungi attack both cellulose and lignin while brown-rot fungi mainly attack cellulose and hemicellulose components in wood [62-64]. Some white-rot fungi can selectively delignify (lignin and hemicellulose) and leave enriched cellulose. De­pending on types of fungi and wood, the lignin lost, observed, can reach up to 44 % [64]. These microorganisms, associated with lignin-degrading enzyme, consist of mainly two major families of enzymes which are laccase and peroxidase. Bacteria such asactinomycetes, which is a filamentous bacteria belonging to the genus Strep — tomyces are well known degrader that can mineralize up to 15 % of lignin. Others nonfilamentous bacteria mineralize lignin less than 10 % and can degrade the low molecular weight part of lignin [64].

Laccase (benzenediol:oxygenoxidoreductase, EC 1.10.3.2) belongs to the small group of enzymes called the blue copper proteins or the blue copper oxidases that catalyze one-electron oxidations of aromatic amines and phenolic compounds such as phenolic structures of lignin [60, 65]. Laccase, widely distributed in higher plants and fungi, is especially found abundant from white-rot fungi and is also found in insects and bacteria [65]. Fungal laccases have higher redox potential than bacterial or plant laccases (up to +800 mV), and their action seems to be relevant in nature, also finding important applications in biotechonology [65]. Thus, fungal laccases are involved in the degradation of lignin and detoxification of phenols arising during lignin degradation which inhibit fermentation process. Vikineswary et al. studied the production of laccase from sago hampas and OPF parenchyma tissue, which gave higher laccase productivity compared to rubberwood sawdust, using Pycnoporus sanguineus [66].

Peroxidase family consists of ligninolytic enzymes such as lignin peroxidase (LiP), manganese peroxidase (MnP), and versatile peroxidases (VP) [60, 62]. Lignin peroxidase (LiP, diarylpropane peroxidase, EC 1.11.1.14), an extracellular lignin olyticperoxidases which is commonly associated with Phanerochaete chrysosporium

[67] . It has capability of catalyzing the depolymerization of the aromatic polymer lignin and a variety of non-phenolic lignin model compounds in the presence of H2O2 [60]. Others such as Phanerochaete sordida [68], Aspergilllusstrains, and bacteria such as Acinetobactercalcoaceticus, Streptomyces viridosporus and Streptomyces lividans [59] are also producing extracellular LiP.

MnP, hydrogen peroxidase oxidoreductase (EC 1.11.1.13), is able to oxidize Mn(II) to Mn(III) [60, 67]. It has also been reported isolated from Cunninghamella elegans [69], Schizophyllum sp., Ceriporiopsis subvermispora, Panus tigrinus, Lentinula edodes, Nematolomafrowardii, Bjerkandera adusta, Tinea versicolor, and Dichomitus squalens. VP (EC 1.11.1.16) oxidize Mn(II) and a high redoc-potential aromatic compound as MnP and LiP, respectively [68]. VP is also reported to be produced by fungi from the genera Pleurotus, Bjerkandera, and Lepista and maybe also by Panus and Trametes species [70].

The enzyme activity and lignin degradation are influenced by several factors which include the type of strain and nutrient composition (i. e., in the case of delignification by using microorganism), enzyme dosage (i. e., in the case of delignification by isolated enzyme), moisture content, pH, aeration, and temperature. These factors can be manipulated to obtain the optimum pre-treatment process. Ahmad Khushairi and Zainol screened factors affecting biological delignification process of oil palm trunk using local oyster mushroom (Pleurotus ostreatus) [61]. In their study, temperature contributed the most in delignification process followed by pH. Other studied factors were fungi-to-medium ratio, moisture content, contact time, lighting, and humidity. Interesting to note that, in the study, even though biological delignification is known to be time consuming, the contact time between 2 and 10 days are among the least important compared to others.

In another study, direct delignification with a commercial biocatalyst called lac — case was performed. Taguchi method was applied to determine the optimum lignin degradation of OPF from laccase treatment. The effects of laccase dose, pH, pre­treatment temperature, and treatment time were investigated. The experiment results of the nine trial conditions with two runs per trial condition are shown in Table 17.3 where the percentage of lignin degradation is exhibited.

Table 17.3 Lignin Trial Lignin degradation (%)

degradation (%) of laccase

treated OPF ________ RR

Results in Table 17.4 indicated that factor that influenced lignin degradation is in the following order: pH value > laccase dose > treatment time > treatment tem­perature. The optimum combination of factors and levels in lignin degradation of OPF may be predicted in general as follow: pH value of 4, laccase dose of 20 mg, treatment time for 2 h, and treatment temperature at 30 °C. However, the experiment results revealed very low percentage of lignin degradation (10-11 %) as shown in Table 17.3.

Syafwina et al. studied three white-rot fungi and found that Dichomitus squalens degraded lignin most rapidly compared to Ceriporiopsis subvermispora and Pleu- rotus ostreatuson in OPEFB [71]. After 8 weeks, weight loss of the lignin and holocellulose in beech wood reached 75.9, and 49.9 %, respectively. The fungus also delignifled EFB. After 8 weeks, weight loss of lignin and holocellulose in EFB was

25.7 % and 22.8 %, respectively. Hamisan et al. performed chemical pre-treatment and compared to microbial on OPEFB [72]. Microbial pre-treatment using Phane- rochate chrysosporium was shown to significantly removed the lignin, but it is timely (7 days) compared to chemical pre-treatment (3 h) which is fast reaction but quite harsh to the substrate. Namoolnoy et al. isolated 63 fungal of white-rot fungi; 27 of them showed high activity of laccase, MnP or LiP was selected to culture on OPFs [73]. Only seven isolates could degrade more than 50 % lignin content in OPFs within 30 days of cultivation. The efficiency in lignin degradation of selected fungal isolates seemed to be related to the ability of the fungi to produce more than one ligninolytic enzyme.

A good understanding of the gasification reaction kinetics and gasifier hydrodynamics is essential for the design, operation, and optimization of gasification processes. This section will discuss the main chemical reactions that govern gasification and the different kinetic models in the scientific literature that are available.

During a gasification process, the biomass particles undergo the following reaction steps:

1. Drying: The biomass particles are heated and dried on entering the reactor (endothermic step).

2. Pyrolysis (thermal cracking): As they reach high temperatures, the biomass particles undergo pyrolysis and decompose to gas and solid char (endothermic step).

3. Gasification:

a. Combustion: The char and gases (condensable and non-condensable) react with oxygen to produce H2O, CO, CO2 (exothermic step).

b. Gasification: Where the produced gas and solid char from previous steps react with the gasifying agent and each other (endothermic step).

c. Tar cracking: The condensable gas decomposes (thermal cracking) to smaller molecular weight components (endothermic).

The level of oxygen in the gas can be set in order for the system to be autothermal. These steps are generally modeled in series, but it is widely accepted that there are no sharp boundaries between them. Table 10.1 lists the important reactions taking place during gasification.

11.2.1.1 Bark and Wood Residues from the Pulp and Paper Sector

In 2007, forest mills in the United States of America (USA) produced about

86.7 million dry tons of primary mill residues [17], which were composed mainly of bark, sawdust, wood chips and shavings. Of this amount, over 35 million dry tons of wood residues were used as combustibles and could have been used as feedstock for bioreflneries. Wood pyrolysis has been shown to generate high-value products, such as bio-char (promising activated carbon) and bio-oil. Ensyn and DynaMotive are two companies running commercial-scale pilot plants, which convert wood residue via fast pyrolysis. There are many incentives to develop in situ bioreflneries (close to pulp and paper plants) in order to avoid significant issues related to transportation and storage.

11.2.1.2 Black Liquor

Black-liquor pyrolysis has been the subject of several studies, but the main efforts have been invested towards gasification. This has been motivated by the fact that black-liquor pyrolysis generates too much solid char [18], which would need to be burned to release the inorganics. The advantage of gasification is that it includes the char combustion process. Thus far, pyrolysis has been mostly considered in the scientific literature as a precursor step to gasification. Consequently, it will not be considered as a potential feedstock for pyrolysis aiming at biorefineries.

Biomass and MSW contain oxygen, nitrogen, sulfur, halogens, metals and other elements whose concentrations in the pyrolysis products must be controlled as per existing standards.

Sulfur and nitrogen have been shown to mostly cluster in the bio-oil phase [21]. Sulfur has been successfully avoided in pyrolysis oil from coal in the past [67] by the use of lime (CaO). Similarly, the same experiments showed that the oxygen weight fraction in oil could be as well significantly decreased. The main disadvantage of this technology is that the char will remain charged with calcium and that the gas phase will mostly receive this excess of sulfur [68]. However, the gas phase can be post-treated efficiently (scrubbing) and considering the high-level of inorganics and contaminants in the solid phase, it will not likely be used as bio-char or as activated carbon.

Adding CaO in the pyrolysis environment can also significantly inhibit the for­mation of liquid chlorinated organics in the bio-oil [69]. The co-feeding of CaO in MSW pyrolysis could eventually become widespread as this additive is relatively cheap and available.

Post-processing of bio-oil to remove oxygen, nitrogen and sulfur is also possible through hydrodeoxygenation using metal catalysts (and an H2 stream) [70]. For biofuels production, this post-treatment process is desirable, as the oxygen weight fraction can be reduced below 1 %. Furthermore, the effect of this post-treatment on the bio-oil chemical composition has not been investigated such that there is a possibility that it also generates higher value chemicals.

Most metals and metal oxides cluster in the solid phase due to the low tempera­ture associated with pyrolysis, which remains below metal sublimation or melting temperatures. Metals can also be absorbed by activated carbon produced from bio-char [56].

Nomenclature

Abbreviations

A = Pre-exponential factor [s-1] for a 1st order reaction d = Diameter

E = Activation energy [Jmol-1]

Fr = Froude number g = Standard gravity [m s-2] h = Heat transfer constant [Wm-2K-1] k = Thermal conductivity [Wm-1K-1] m = Dimensional & non-dimensional weight n = Order of reaction, non-dimensional Nu = Nusselt number Pr = Prandtl number

R = Universal gas constant OR Radius [Jmol-1K-1] OR m Re = Reynolds number

t = Time

T = Temperature

U = Velocity [m s-1]

Symbols

ц, = Viscosity [Pa s]

p = Density [kgm-3] rn = Angular velocity [s-1]

Subscripts

bed = Fluidized bed

bp = Bed of particles (in rotary drums)

g = Gas phase

p = Particle

t = Terminal (velocity)

14.2.3.1 General Process of Organosolv Pretreatment

A general process of organosolv pretreatment is described in Fig. 14.2. Pure or aque­ous organic solvents, such as methanol, ethanol, acetone, ethylene glycol, triethylene glycol, and phenol, were typically used as the working medium. Biomass was ther­mochemically treated in these working medium by mixing with (or without) the addition of acid, alkaline, or neutral catalysts, such as H2SO4, NaOH, or MgCl2, at a

relatively lower (< 180 °C) or higher temperature (> 180 °C) [23,24,55,56]. Organo­solv pretreatment also dictated a unique pathway of downstream refinery process: Af­ter treating the substrate for several minutes or days, the cellulose-rich solid fraction was separated by filtration and washed by the same solvent and, subsequently, water, and then was ready for hydrolysis. The liquid stream, which contained the solvent, alone with lignin and sugars from hemicellulose, as well as their derived products (e. g., furfural, 5-hydroxymethylfurfural(5-HMF),depolymerizedlignin), was recov­ered by distillation. After distillation, the residual was washed by water to precipitate the organosolv lignin. Chemicals recovered from the water-soluble fraction include xylose, glucose, oligomeric sugars, organic acids, 5-HMF, and other lipophilic extracts. They can be further utilized after separation and concentration [24].