The solid residue from the organosolv treatment was weighed and placed in a 500 mL beaker with distilled water (5 g, 200 mL water). The mixture was treated with ultra­sound (Fisher Scientific Ultrasonic Cleaner, 100 W, 42 kHz output) for 3 h at room temperature. After that, the solid phase was separated from the liquid phase by fil­tration. The collected solid sample with (organosolv + ultrasonic) pretreatment was then used for enzymatic hydrolysis.

19.2.2.2 Organosolv Extraction Followed by NaOH Treatment

Five grams of solid residue from the organosolv treatment was placed in a 500 mL beaker with the addition of 250 mL 1N NaOH. The NaOH treatment was carried out at 70 °C in a water bath shaker at 100 rpm for 3 h. After filtration, the solid residue was collected for enzymatic hydrolysis.

19.2.2.3 Organsolv Extraction Followed by Ultrasonic and NaOH Treatment

The solid residue from the organosolv treatment was treated with ultrasound and NaOH in sequence at the same reaction conditions as described above. The solid was recovered after filtration and used in the enzymatic hydrolysis process.

9.3.1 Response Surface Model for Total Sugars Yield

Chemical composition analysis of the poplar wood was shown to consist of 49 % glucan, 21 % xylan, 1.5 % galactan, 1.0 % arabinan, 2.5 % mannan (total 75 % carbohydrate), 22 % Klason lignin, 2 % extractives, and 0.8 % ash.

In the hydrolysates, five reducing sugars (glucose, xylose, galactose, arabinose, and mannose) were measured. Neither furfural nor hydroxymethylfurfural were de­tected by HPLC in the hydrolysates and therefore not deemed in sufficient quantity to inhibit fermentation. In this preliminary trial, using a readily available cellulase enzyme, the sugars and acetic acid yields in both PL and enzymatic hydrolysates are listed in Table 9.2. In the PL, xylose was the main sugar while very little amount of other sugars could be detected. However, glucose was the major sugar released followed by xylose after enzymatic hydrolysis. Experiment 2 (200 °C, 10 min, 20 % solid loading) gave the highest total sugars yield (34 %). While experiment 11 (185 °C, 3.2 min, 30 % solid loading) gave the lowest sugars yield (14.8 %) which also had least total sugars in the PL. This result indicated that experiment 11 was not a severe pretreatment condition due to short reaction time (3.2 min).

The acetyl group is readily released from 4-o-methylglucuronoxylan as acetic acid during pretreatment [19] and therefore was quantified. Acetic acid concentrations in both PL and enzymatic hydrolysates were <1.2 mg/mL (Table 9.2) and below the level (5 mg/mL) at which it could act as an inhibitor for fermentation [8, 9, 20,

21]. Compared with an acid pretreatment, a hot-water pretreatment generates much less acetic acid during the process [2]. The pH in all experiments which was about 4 together with the low acetic acid concentrations observed will result in limited sugar degradation. The highest acetic acid levels were observed at a pretreatment of 200 °C for 30 min. This suggests that pretreatment temperature and reaction time were important factors. A hot-water wash process can help reduce acetic acid and other inhibitors levels generated from pretreatment [2, 3].

Considering further fermentation or reaction for producing PHA, total sugars yield from enzymatic hydrolysis would be a major target. Therefore, RSM used total sugars yield as response variable. Before determining the optimization pretreatment condition, an RSM model was conducted using 20 experiments (Tables 9.1 and 9.2) (total sugars yield, %). Figure 9.1 is a three-dimensional (3D) plot that modeled the pretreatment conditions for total sugars yield in a curved surface and predicted the optimal sugars yield at fixed variable (solid loading, 30 %). This is a direct view of the data generated from this experimental design. Since 30 % solid loading was the central point in the design, it was selected to be the fixed variable and the other two more significant variables (based on analysis of variance (ANOVA) results) were displayed in the response surface. From Fig. 9.1, the highest surface occurred at 200 °C and a reaction time around 20 min.

The modeling results are shown in Fig. 9.2 and Eq. 9.2.

Y = a0 + a1 x A + a22 x B2 + a12 x A x B (9.2)

Equation 9.2 was established based on ANOVA results in Table 9.3. Temperature (A) was shown to be the most significant variable while time (B) in the second order was significant and the interaction term of temperature x time had influence on total sugars yield (95 % significant level). Since solid loading was the least effective variable based on ANOVA results, the 3D plot was given on the other two variables (temperature and time). The ^-square of the model was 0.86, which was acceptable to give a decent prediction on total sugars yield with appropriate pretreatment condition. Repeated experiments on the optimal pretreatment condition were done to support the model. Model coefficients were generated by fitting Y to the least squares of variables A, B, and C (solid loading).

Fluidized bed reactors have been used extensively for coal gasification. Fluidized bed gasifiers are characterized by excellent heat and mass transfer, which facilitates the control of the bed temperature. Furthermore, this allows the use of a wide variety of fuels, such as fluffy and fine grained materials without the need of pre-treatment. However, problems with solid feeding and fly-ash sintering in the gas channels can occur with some biomass fuels.

There are two major categories of fluidized beds for biomass gasification: bub­bling (BFB) and circulation fluidized beds (CFB). These two types of gasifiers are characterized by significantly different gas and solids hydrodynamics. CFBs are op­erated with a higher gas velocity (3-5 m/s) compared to BFB (0.4-1 m/s). In CFBs, the solid particles are entrained along a tall tubular section (called a riser), which allows long gas and solid residence times. CFB gasifiers at atmospheric pressure have proven very reliable with a variety of feedstock and are relatively easy to scale up from a few MWh up to 100 MW h [67].

The family of fluidized bed reactors include other relatively new concepts for gasifiers: chemical looping, dual gasifier, and an internally circulating fluidized bed. The main motivation driving the development of these reactors is to reduce the syngas dilution in nitrogen or carbon dioxide. In chemical looping gasification, a solid oxygen carrier is circulated continuously between two fluidized bed reactors: (1) a biomass fluidized bed gasifier and (2) a fluidized bed to oxidize the oxygen carrier. In the gasifier, the oxygen-carrier releases the oxygen for the gasification reactions before being entrained back to the second fluidized bed where it is re­oxidized [68]. The concept of circulating solids between two fluidized beds has also been extensively studied for CO2 capture [69]. There is currently an operating industrial plant in Gussing (Austria), which consists of one circulating fluidized bed combustor and a bubbling fluidized bed gasifier. This system has, however, its own drawbacks. For example, the amount of char combustion in the gasifier may not be sufficient to provide the required heat for endothermic reactions. Co-firing may therefore be necessary [70].

Fast and ultrafast pyrolyses are performed under high heating rates (600— 12,000°C/min [36]), which are many orders of magnitude higher than those of conventional pyrolysis. Thus, the peak rate of decomposition is reached at higher temperatures compared to slow pyrolysis. Under these conditions, the macromolecu­lar reorganization kinetics are slower than the volatile release kinetics. Consequently, the bio-char yield is significantly lower compared to slow pyrolysis, while the volatile yield is higher. Since the bio-oil (condensable gases) products are of higher interest for biorefineries, the emerging industrial biomass pyrolysis processes ideally target fast and ultrafast pyrolysis processes. Meanwhile, bio-char obtained at a higher tem­perature show a greater specific surface, which is another motivation for operating at very high heating rate and temperature. In addition, higher heating and reaction rates allow higher biomass process rates or the use of smaller, more compact systems: both of these aspects increase process profitability.

11.4.1 Vacuum Pyrolysis

The third category of pyrolysis process is referred to as vacuum pyrolysis. This py­rolysis process is performed under vacuum conditions independent of the heating rate (slow and fast). Under vacuum, the heavier products in the gas phase are en­trained out of the reacting environment without having time to crack into smaller molecules. For that reason, vacuum pyrolysis oils contain high molecular weight components and are consequently tarry and more viscous than common pyrolysis oils. Because of their specific molecular composition, vacuum pyrolysis oils are of great interest for specialty chemicals production in biorefineries. It is nevertheless a great challenge to produce industrial scale vacuum environments. The company Pyrovac (Laval, Quebec, Canada) attempted to operate a commercial and large-scale industrial vacuum pyrolysis plant in the 1990s and failed due to high operating costs.

The upgrading process is basically a separation of CH4 and CO2 in the biogas, in order to obtain town gas quality with regard to the caloric value, Wobbe index (WI: the value was obtained by dividing the amount of heat released by the square root of the gas density), and relative density. In Japan, town gas is defined as gas fuel supplied by companies that conduct "General Gas Utility Business" under the Gas Business Act. Specifications regard the gas’s heat capacity; the gas produced in the developed purification facility corresponds to town gas 12A specifications. Town gas 12A specifications are the common classification of town gas in Japan. The WI for town gas 12A is in the range of 49.2-53.8 and the combustion speed is in the range of 34-47 m/s.

Figure 13.3 illustrates how the WI increases and the relative density decreases as the methane content of the upgraded gas increases. A raw biogas composition of approximately 60 % methane and 40 % carbon dioxide is assumed.

For specific methane purity, the methane yield can be improved by recirculation of a part of the permeated CO2-enriched gas. In the case of several modules connected

in series, the best result is obtained with recirculation of only the permeated gas from the last module. Another way to maximize the methane yield and still obtain pipeline quality gas is to upgrade the biogas to a lower quality than required and then add propane in order to meet the specifications.

Cation-exchange resins have been used commercially as solid acid catalysts in es­terification, alkylation, hydration/dehydration, isomerization, oligomerization, and condensation reaction [71]. Amberlyst® 15 and Nafion® resins are the mostly applied catalysts in organic catalysis. These resins are both chemically and thermally stable (up to 280 °C). Amberlyst-15 is an effective catalyst for the selective conversion of cellulose to glucose. More than 25 % glucose yield was obtained under reaction conditions of 150 °C for 24 h with Amberlyst® 15 (50 mg) from milled cellulose (45mg), and distilled water (5.0mL) [72]. Nafion® NR50has similar acidic char­acter but better thermal stability as compared with Amberlyst-15. However, such resins have obvious drawbacks such as low surface area and activity in both aque­ous solvent and gas phases [72]. Kim et al. [73] proposed a strategy to enhance the overall process efficiency by combining of 1-n-butyl-3-methylimidazolium chloride {[BMIM][Cl]} and Nafion® NR50, which resulted in 35 % glucose yield.

These resin catalysts have high surface area (up to 539 m2/g) and controllable hydrophobicity, but their acidic concentration is still blocked by the limitation of sulfonation in phenolic rings due to the strong steric hindrance [74]. The hy­drophobic feature of resin catalysts limited their efficiency in cellulose hydrolysis. Therefore, special solvent like, ILs is needed to dissolve cellulose. Qi et al. [75] using a strong acidic cation-exchange resin to achieve high glucose yield of 83 % from cellulose hydrolysis in 1-ethyl-3-methyl imidazolium chloride [EMIM][Cl] with gradual addition of water.

Many efforts focused on the design and synthesis of nanosized resins. Via an in-situ sol-gel route [76], Nafion/SiO2 with high specific surface area was synthesized, and further exposure of more acid sites is obtained. The feasibility of using Lewatit FO36 nano ion-exchange resin was investigated as adsorption for removal Cr (VI) from aqueous solutions using batch technique under various conditions [77]. Nanosized resins have advantages of allowing the в-1,4 glycosidic bonds in cellulose to become more accessible to catalytic sites. The other method to improve catalytic efficiency is the modification of cation-exchange resins using metallic salt. Chen et al. [78] demon­strated that gallium sulfate modified strong acid cation ion-exchange resin was highly active for the synthesis of butyl lactate with conversion rate of 92 % in 70 min. More­over, a cation-exchange resin modified by incorporation of magnetic components (e. g., iron, cobalt, and nickel) can be used to separate resin catalysts easily.

Torrefaction is the heating of biomass in the absence of oxygen at temperature lower than 300 °C, or in other words, partial pyrolysis of biomass, to produce uniform product with lower moisture contents and higher energy contents. Prins et al. in their study on torrefaction of dry willow at two different temperatures of 250 °C and 300 °C, observed that more volatiles are obtained at 350 °C with mass yield of 67% [4].

The mass and energy from the biomass is predominantly conserved in the solid product (torrefied biomass) [1]. Besides, this process is able to depolymerize the long polysaccharide chains, producing a hydrophobic solid product with an increased energy density and consequently increase grindability [3]. Much better fuel quality with higher CV for combustion and gasification applications is obtained through torrefaction [4, 37]. Torrefied materials are commonly used as solid fuel in industries and residences, similar to that of charcoal. However, when performed at temperature higher than 300 °C, the process is no longer referred to as torrefaction, but as flash or rapid pyrolysis that occurs at faster rates.

In this study, the torrefaction was performed in a horizontal tubular furnace flowed with nitrogen; N2 being inert was used as torrefaction furnace. Four factors that in­clude temperature, residence time, particle size, and inert gas flow rate were studied as listed in Table 17.2. Each was varied at three levels to perform nine sets of ex­periments following L9 orthogonal array following Taguchi Method. This method was used to analyze the energy density of torrefied products in order to get the op­timum results. From the study, temperature contributes the most. Since the process is endothermic and temperature increase will also increase the minimum amount of energy needed to induce the reaction. Residence time and particle size are moderate contributors to the high heating values resulting in energy density and mass loss,

while contribution of nitrogen flow rate is insignificant. For oil palm kernel shell, the optimum condition was found to be at 280 °C and 60 min residence time, while particle sizes have insignificant effect in the process. Torrefaction process is found to be able to overcome some of the above limitations [3, 37, 38]. Torrefaction is able to remove moisture and low weight organic volatile component (carbon monoxide (CO) and carbon dioxide (CO2)) and increase grindability as well as energy density. O/C ratio of torrefied biomass is lowered through the release of moisture, CO, and CO2, which remove oxygen from the biomass, resulting in a more efficient gasification

[4] .

In general, the decomposition of lignocellulosic biomass under inert atmosphere proceeds in three steps; decomposition of hemicellulose (180-300 °C), cellulose (240-400 °C), and lignin (280-550 °C). Torrefaction is the first step, at which hemicellulose decomposes [38]. On the contrary, only a small portion of cellulose decomposes due to its crystallinity compared to hemicellulose. The third component which is lignin decomposes only a little due to its cross-linked structure. During torrefaction, major low molecular weight products include acetic acid, water, formic acid, methanol, lactic acid, furfural, hydroxyacetone, phenol, carbon dioxide and carbon monoxide [39]. Traces of hydrogen and methane are also detected [39]. Water may come from mainly moisture of biomass, and partly from dehydration of holocellulose. Acetic acid and carbon dioxide may come from acetyl group on hemicellulose. Change in the elementary composition during torrefaction may be exhibited using the Van Krevelen plot. In Fig. 17.2, the results extracted from some torrefaction studies are plotted; the H/C atomic ratio against the O/C ratio [40]. The raw biomass was shown by a key with a circle. As torrefaction proceeds, both the ratios, H/C and O/C, decrease in parallel regardless of the type of raw material biomass including beechwood, eucalyptus, canary grass, wheat straw, willow, oil palm empty fruit bunches (OPEFB), oil palm mesocarp fiber, and oil palm kernel

shell. This change is mainly due to removal of water and carbon dioxide [39] as discussed previously.

Relationship Between Carbon Content and Calorific Value

The CV of lignocellulosic biomass and lignocellulosic biomass-derived fuels is one of the important fuel properties, which defines the energy density of those fuels. Since the estimation of this value from the elementary composition has been recognized as an important step for performance modeling of biomass and biomass-derived fuels, many correlations have been reported [31-48]. Nonetheless, only one report has been found that correlates the CV of torrefied biomass with its elementary composition

[49] . This paper proposed the following equation, by which the CV high heating value (HHV) (MJ/kg) is well correlated with the elementary composition. HHV also known as gross calorific value (GCV) is defined as the amount of heat released by the unit mass or volume of fuel (initially at 25 °C) once it is combusted [35], as shown inEq. 17.1 [38]:

HHV = 0.4373 x C — 1.6701 (17.1)

where C denotes the weight percent of carbon in the torrefied biomass. The correlation plot is shown in Fig. 17.3.

Effects of Parameters Temperature

Torrefaction consists of a variety of chemical reactions that lignocellulose undergoes, as described above. The fact, mass yield decreases with torrefaction temperature
clearly shows most of the activation energies are positive. Although the activation energy for each reaction has not been reported, the activation energy of the overall reaction has been reported in a few cases: willow biomass to be 76.0 kJ/mol [50] and oil palm empty fruit bunch (OPEFB) to be 37.3 kJ/mol [51]. The difference between these two may attribute to different type of biomass in this case due to different properties of wood and oil palm waste. The second factor is the experimental technique.

Residence Time

In previous papers for torrefaction, the residence time was selected to be between

0. 5 and 5h. The mass yield decreases fast in the first 30 min, and becomes almost constant, specifically below torrefaction temperature of 360 °C. At 280 °C, after 30 min a considerable decrease in mass yield was observed by 3 h [52] and 5 h [53].

Particle Size

Torrefaction is an endothermic reaction. The effect of heat transfer was discussed by Bergman et al. [54]. They proposed an index number, called the Pyrolysis number (py) and Biot number (Bi), as defined in Eqs. 17.2 [39] and 17.3 [40]:

Biot number:

arp

У

where a is the external heat transfer coefficient in W/(m2 K), k is the reaction rate constant in s-1, p is the density in kg/m3, Cp is the heat capacity in J/(kg K), rp is the radius in m, and X is the thermal conductivity in W/(m K) of the biomass particle.

Using the criterion, Jalan and Srivastava estimated that biomass pyrolysis up to 873 K is controlled by the intrinsic pyrolysis reaction when particle sizes are less than 1 mm [55]. Biomass torrefaction is controlled by the intrinsic torrefaction reaction when biomass particle sizes are less than 2 mm, based on the criterion [4, 52]. Larger particle sizes induced secondary cracking reaction due to increase in the resistance towards primary pyrolysis product [35].

Sweep Gas Composition

In academic studies, nitrogen is commonly used as the sweep gas. Industrially, nitrogen is not a practical option due to its cost. In an industrial concept, re-circulation use of torrefaction gas is proposed [54]. From an economic point of view, use of boiler flue gas for torrefaction of oil palm residue is proposed [51]. They investigated the effects of oxygen concentration in torrefaction gas on the mass yield of torrefied biomass.

Advantages of Torrefaction for Solid Fuel Use and Gasification The advantages of torrefaction are:

1. Higher CV due to partial decomposition and moisture removal [40, 53].

2. Longer shelf life due to moisture removal.

3. Moisture resistant pellets and briquettes due to higher hydrophobicity [56].

4. Enhanced grindability due to less toughness of torrefied biomass [52, 57].

All these advantages are related to solid use of torrefied biomass. At 1,400 °C, the two types of torrefied wood produced more hydrogen (7 %) and carbon monoxide (20 %) than the untorrefied wood [58]. This higher gas yields were attributed to a higher carbon and hydrogen content of the torrefied biomass.

Gasification is the conversion of solid fuels into gas fuels in an oxygen-lean at­mosphere. Pyrolysis, gasification, and combustion are classified as three separate thermo-chemical conversion processes. However, both pyrolysis and combustion take place during gasification. In gasification, at least four different stages of reac­tions are involved: (1) pyrolysis, (2) combustion of solid char and other gases, (3) gasification of char, and (4) tar cracking.

Different types of gasifiers must deal with these reactions. The contribution of these reactions upon the final gaseous product as well as the specific conversion of each reaction depends on the operating conditions (temperature, pressure, etc.), the biomass characteristics (chemical composition, particle size, etc.), and gasifier type. The fundamental gasification reactions (reduction reactions) are endothermic. The necessary energy is supplied by the exothermic combustion reactions. Compared to the original solid fuel, the produced gaseous fuel is easy to clean, to transport, and if cleaned properly it can be used in fuel cells as well as burned in gas turbines, furnaces, boilers, and reciprocating engines [10].

As mentioned earlier, syngas is one of the products of gasification, which is an important source for valuable chemicals and energy, such as hydrogen, diesel (through Fischer-Tropsh synthesis), electricity (through combustion), fertilizer (through ammonia production), and methanol.

In the gasification process, heavier hydrocarbons, called tar, are produced along with syngas. Tar consists of high-molecular weight components, usually rich poly aromatic hydrocarbons (PAH). The presence of tar in syngas can cause serious prob­lems in syngas applications. Condensation, fouling, and the polymerization of tar are general problems that can arise. The amount of undesirable products in the gas depends mainly on the design of the gasifier, feedstock characteristics as well as the gasifier operating conditions. Once the raw syngas leaves the gasifier, it goes through treatment processes, such as gas cleaning of dust or particulates (by cyclone, fabric or electrostatic filters, or solvent scrubber), conditioning (with the use of shift reac­tion to adjust the molar ratio of CO and H2), and separation (to refine the syngas stream from tar and other catalyst poisons).

Most of the biomass feedstocks can be classified into three families as defined by the US Department of Energy [8]: forestry, agriculture and municipal. Table 11.1 summarizes key chemical and physical characteristics of the main feedstocks that are considered for biorefineries.

Through the conservation of mass, the biomass chemical composition determines the chemical elements present in the three pyrolysis products: non-condensable gas, condensable gas and char. The presence of specific chemical elements in each product fraction is determined by the pyrolysis conditions.

Environmental and purity standards restrict the presence of oxygen, nitrogen, sulfur and inorganics in the pyrolysis products. During pyrolysis, the tendency of producing an aqueous phase generally increases with increasing biomass-oxygen fraction (dry weight) since water is produced [16]. The presence of oxygen may also lead to the production of acids, which are detrimental to the oil stability. On

Note 1: Wood energy crops as per bark & wood residue Note 2: Agricultural crops as per perennial crops

the other hand, sulfur and nitrogen are not present in biomass in large amounts as shown in Table 11.1, but they will, nonetheless, be present in the products. In this case, the pyrolysis products may need post-treatment since sulfurous compounds are corrosive, while nitrogen affects reactivity as well as pollutant emissions (e. g. fuel — bound NOX). Moreover, these species are also problematic when performing bio-oil upgrade. Finally, inorganics in the biomass and pyrolysis products may represent a risk of slagging and sintering.

Furthermore, the biomass physical properties strongly influence the gas/solid hydrodynamics as well as the heat/mass transfer in the pyrolysis reactor such that it affects the pyrolysis products (respective yield of the three pyrolysis products and their composition). The important physical properties include the shape of the biomass feedstock, its particle size and moisture fraction. These properties will determine the required biomass physical transformations or pre-treatments.

Char is sometimes used as fuel for pyrolysis reactors. However, the market value of bio-char must be considered before taking decisions. Activated carbon sells on the market at around 1 USD per kg, which is comparable to the value of some chemicals found in bio-oil. Bio-oil and bio-char both need post-processing transformation in order to yield their valuable products.

Activated carbon production implies carbonization and chemical activation. Typi­cally, carbonization consists of a very slow pyrolysis process, which yields very high amounts of char. The principal characteristic of activated carbon is a very high spe­cific surface, generally over 500 m2/g. To be competitive with the actual commercial carbons, this specific surface must be attained and it has recently been demonstrated as feasible [7]. Since 2006, the number of publications on bio-char activity has increased considerably [56].