Some supported solid acid catalysts (e. g., sulfonated carbon based solid acid, sul- fonated metal oxides and sulfonated activated-carbon) showed good activity in cellulose hydrolysis [24, 33, 34]. Supported solid acid catalysts are promising for the depolymerization of cellulose in water, since they have substantial surface acidic species (e. g., 1.5mmol/g; sulfonated activated-carbon) as compared with zeolites (e. g., 0.3 mmol/g; HSM-5B) and transition-metal oxides (e. g., 0.3 mmol/g; Nb3W7 oxide) [35, 36], and specific functional groups. The active species of protons [H+] in such catalysts are more accessible to the P-1,4-glucans in cellulose than L acid sites [37].

Solid acid catalysts, especially strong acids such as sulfated metal oxides (e. g., SO42-/Al2O3, SO42-/TiO2, SO42-/ZrO2, SO42-/SnO2 and SO42-/V2O5), are being studied extensively in an effort to replace liquid acid catalysts. Sulfated metal oxides are solid acid catalysts with the combination of B and L acids in a certain form. For­mation of proton acid centers is related to the adsorption of hydroxyl groups or H2O on SO42-. However, most acid sites are mainly formed by coordination adsorption of SO42- on the surface of metal oxides. The coordination adsorption makes strong migration of the electron cloud in metal-oxygen bond, leading to strengthening the L acid center. Many studies have proposed that water, present in biomass or produced as a reaction product, converts L acid sites to B acid sites. Therefore, in the hydrolysis of cellulose into glucose, B acid sites play an important role.

The relative activity of B acid sites depends on many factors, such as the structure of supports, the nature of reactions and the polarity of reaction media. Supported acid catalysts are the most extensively studied ones for organic synthesis [38]. As the same kind of catalysts, the number of strong B acid sites is correlated with the catalytic activity. Zhang et al. [39] studied the skeletal isomerization of я-butane and found that the catalytic activity of H4SiW12O40/SiO2 reached a maximum when the loading amount of H4SiW12O40 was 50 wt%. The density of strong B acid sites increased with loading of H4SiW12O40 on SiO2 (10-50 wt%). It was concluded that the direct interaction of H4SiW12O40 with the surface of SiO2 caused strong B sites to form in the second-layer of H4SiW12O40 exposed on the surface. However, in the previous study about the effect of support identity on B acid site density [40], it was found that the reactivity of B acid sites was not affected by the identity of supports. The turnover rate increased with the rising of density of acid sites on all supports.

The influence of supports on activity of B acid sites in cellulose hydrolysis still remains undiscovered. Sulfonic group functionalized magnetic SBA-15 cata­lyst (Fe3O4-SBA-SO3H) [41], which gave high glucose yield (98 %) in cellobiose conversion, can be recovered for reuse by an external magnetic field. Fe3O4-SBA — SO3H not only provides good access of reactants to the — SO3H groups, but also has functional characteristics that allow it to be separated and regenerated. Sulfated ZrO2 [31] has similar active acid sites (sulfonic acid group: 1.2 vs. 1.09 mmol/g) but lower glucose yield (14 % vs. 98 %) as compared with Fe3O4-SBA-SO3H. It still could not prove whether their activities were influenced by effect of supports on activity of B acid sites or by access of reactants to -SO3H groups. Suganuma et al. [42] reported that carbon materials can incorporate large amounts of hydrophilic molecules into the carbon bulk, due to the high density of the hydrophilic functional groups bound to the flexible carbon sheets. Compared with niobic acid, H-mordenite, Nafion, and Amberlyst-15 resins, the carbon catalyst had the highest catalytic activity because it can adsorb |5-1,4-glucans, which are not adsorbed by the other four solid acids.

The activity of solid acid catalysts for cellulose hydrolysis was not only related to the density of B acid sites even for the same support. Onda et al. [31] studied the hydrolysis of cellulose into glucose using sulfonated activated-carbon as catalyst (acid site density of 0.58 mmol/g), and a glucose yield of 41.4 % was achieved at 150 °C for 24 h. They reported a similar glucose yield of 40 % under the same reac­tion conditions using the same sulfonated activated-carbon with 1.25 mmol/g acid site density [43]. The high catalytic activity of the sulfonated carbonaceous material was attributed to (1) its ability to absorb the |5-1,4-glucans, (2) its large effective surface area in water, and (3) the presence of — SO3H groups that are tolerant to hydrolysis [44]. Vigier and Jerome [33] found that an increase in loading of sulfonic sites (i. e., the proton concentration) on poly(tetrafluoroethylene-co-perfluorovinyl ether)-gra/l-polystyrenesulfonic acid (PFA-g-PSSA) membrane surface from 28 % to 63 % enhanced the catalytic activity from 7.5 x 10-3 to 27.5 x 10-3 min-1. It was suggested that this increase in activity might be ascribed to the rise of the amount of accessible sulfonic sites and the catalyst hydrophilicity, thus improving the polymer chains mobility and therefore accessibility of the catalytic sites [24]. Similar phenomenon was obtained by Zhang and Zhao [45] who used different

H-form zeolite catalysts (i. e., HZSM-5a, HZSM-5b and H-beta) for cellulose hydrolysis with microwave-heating at 240 W. HZSM zeolites had a higher glucose yield (35 %) than that of H-beta zeolite (30 %) because of their higher acidity and more reactive sites. Therefore, disregarding the accessibility of catalytic sites to P-1,4-glucans in cellulose, hydrolysis yield is positively related to acid site density.

Sugarcane residues are generated in huge amount in the world every year and can be referred as “Green economy”. To harness its fullest potential, pretreatment is a key to unlock the green economy. Pretreatment specifically acts on either hemicellulose or lignin removal increasing the accessibility of cellulose to cellulases eventually re­leasing monomeric sugars in ready-to-use form. These sugars are further converted into value-added products by microbial mediated processes. In the past, considerable research progress has been made to develop ideal pretreatment strategy considering SB/SL as raw material. Among the pretreatment methods, alkaline and acid mediated processes have been largely explored for pretreatment of SB/SL. Biological pretreat­ment has the maximum environmental and economic benefits but their slow reaction rates make them unattractive. Pretreatment methods like auto-hydrolysis, steam ex­plosion and LHW could be effective for the removal of hemicelluloses posing less environment pollution but their efficacy toward SB/SL has limits. The need of hour is to develop the tailor made efficient pretreatment technologies aiming toward the specific, fast, economic and environmental friendly. Greater fundamental knowledge of chemicals and their possible action on SB/SL cell wall with the software-aided approach considering cheminformatics principles is required to choose the best pre­treatment strategy for the effective carbohydrate depolymerization into fermentable sugars. Such an initiative will forward the commercialization of bio-based applica­tions into new horizons which ultimately usher the overall economy of sugarcane producing countries.

Acknowledgments We are grateful to the BIOEN/FAPESP, CNPq and CAPES, Brazil for the financial assistance. We are also thankful to Dr. OmV. Singh from University of Pittsburgh, Bradford for the critical reading of chapter and valuable suggestions.

One important aspect of biomass is its potential for carbon offsetting. The energy conversion of biomass does not contribute to CO2 emission or global warming. CO2 released during biomass conversion is equivalent to the quantity it absorbed during its growth. In other words, energy production from biomass is almost carbon neutral. The veracity of this statement is, however, debatable as the calculation of the carbon balance associated with biomass conversion is complex: Many parameters must be considered, such as growth, harvest, transportation, and pretreatments. Yet the use of biomass fuels can result in the displacement of carbon dioxide emissions that are ordinarily released when using fossil fuels. This displacement will depend entirely on the efficiency with which the biomass energy can be produced and used.

Most biomass materials contain very low amounts of sulfur, chlorine, in compar­ison to most fossil fuels. Biomass conversion or its use as energy has the potential to lower pollutant emissions.

Biomass chemical composition is obtained through ultimate (also called elemen­tal) and proximate analysis. Ultimate analysis yields the composition of biomass in terms of carbon, hydrogen, nitrogen, oxygen, and sulfur content (mass %). On the other hand, proximate analysis reports the composition in terms of volatile matter, ash, moisture, and fixed carbon. There are separate ASTM standards for measure­ments of individual components of biomass, such as volatile matter, ash, moisture, and fixed carbon.

• Volatile matter refers to the condensable and non-condensable vapor released during the early stages of biomass heating. It depends strongly on the heating rate and final temperature. There are different ASTM standards for the measurement of volatile matter; each standard is specific to different types of biomass.

• Ash is the inorganic residue composed mainly of silica, iron, calcium, magnesium, sodium, and sometimes potassium. It is the remaining material once biomass has been completely consumed in the reactor.

• Fixed carbon is an important parameter in biomass gasification, because its conversion is a limiting reaction, and it is used to size the gasifier.

• The heating value of the biomass is the energy chemically bound in the biomass with respect to a reference state. The best common ones are the lower heating value (LHV) where the reference state is water in its gas state and the higher heating value (HHV) where the reference state is water in its liquid state.

The biomass properties detailed above have significant effects on gasification condi­tions and product compositions: The basis on which they are measured and reported (wet, dry, or dry-ash-free basis) is very important. For example, low ash content improves thermal balance and reduces operating problems due to slagging and sin­tering. Water vapor is also an essential key component in gasification reactions; at high levels, it may negatively affect the process thermal balance. Biomass-containing sulfur, chlorine (may be present in low amounts), and alkali metals can also lead to the formation of corrosive components. Most biomass also contains nitrogen that can form gas-phase ammonia, which can oxidize and form NOX.

The kinetic model of the fluidized bed gasifier consists of reactor hydrodynamics, which define the transport phenomena of the gasification medium through the system and solid mixing behavior. There are several versions of fluidized bed hydrodynamic models [91-93]:

1. The two-phase model in a bubbling fluidized bed, which considers bubble and emulsion phases.

2. The three-phase model in a bubbling fluidized bed consists of bubbling, cloud, and emulsion phases.

3. The core-annulus model for circulating fluidized beds where the core is the upward flow of gas and solid in the center and the annulus is the downward flow of gas and solid close to the wall.

4. The compartment models in which the fluidized bed is divided into slices or horizontal sections.

These types of fluidized bed models avoid the complexity of gas-solid dynamics but still keep the fluid-dynamic effects by considering different regions and phases throughout the reactor, which are described by semi-empirical correlations [91, 92, 93, 94 and 95].

Gas flow through the bed can be modeled as follows:

1. Bubble phase as plug flow and emulsion phase as ideally mixed gas.

2. Both phases as ideally mixed gases.

3. Both phases as plug flow with mass transfer between two phases.

4. Upward gas in the core as plug flow and solid backflow in the annulus [95].

There are three ways to describe conversion models for single char particles:

1. Shrinking core model;

2. Shrinking particle model; and

3. Uniform conversion model [96, 97].

Shrinking core and shrinking particle models are both surface reaction models where the fast reaction takes place as soon as the reactant gas reaches the external surface of the particle. In the shrinking particle model, however, the ash peels off instan­taneously and the particle shrinks during reaction. In the shrinking core model, the particle size stays constant since the ash remains attached to the particle and becomes an additional heat and mass resistance. In the uniform conversion model the reaction takes place all over the char particle uniformly.

Different types of fluidized bed modeling have been applied for coal and biomass gasifiers [90, 98-102]. The following section presents essential equations for a one­dimensional steady-state model of a bubbling fluidized bed gasifier. The fluidized bed is divided into two regions, a dense zone and a free board. The gas flow in the dense part consists of bubble and emulsion phases, which deals with drying, pyrolysis, and gasification of biomass. The freeboard is free of solid particles and only gas phase reactions continue from the bed. The mass balance for the emulsion and bubble phases can be expressed as (10.1) and (10.2):

Emulsion phase:

— (1 — 8b)smfdz (CieUe) — Kbe(Cib — Cie) + (1 — Sb)(1 — Smf )’Lg-sRi + (10 1) (1 — Sb)Smf^g-gRi = 0 (.)

Bubble phase:

d

Sb (Cibub) + Kbe(Cib

d z

(1 — Yb)Sb^g-gRi = 0

The boundary conditions for the above equations are defined in the feeding zone, which is the gas composition, predicted using the pyrolysis kinetics.

For solid hydrodynamics, a concurrent back mixing model is considered [95]. Based on the CCBM, (10.3) could be written for the char particles.

Char balance in the ascending phase:

Char balance in the descending phase:

-Fds,0 ^ + Kw(Cas, c — Cds, c)Afas + A(1 — fas)^g-sRi = 0 dz

The bubble and emulsion equations, which give the gas concentration profile in the bed, and the char balance equation, which provides the char conversion profile in the bed, will be solved simultaneously for all gas species to obtain produced gas compositions and yields in the bed.

Based on this assumption there won’t be any solid in the freeboard and it will be considered as a plug flow reactor. The mass balance for each gas species in this region can be written as (10.5):

A Circulating Fluidized Bed (CFB) works at higher superficial gas velocities than a BFB to increase the slip velocity and heat transfer to the particles. Due to the high gas velocities, the particles are entrained outside the bed region (called the riser) and cyclones are used downstream to separate the particles from the gas and return them to the bed. During biomass pyrolysis, char and inert (generally sand) particles are present. Depending on the design of the reactor, the char/sand particles mix can be transported to a second reactor where the char is burned. In this case, the hot sand is returned to the riser and the combustion flue gases can be directly fed to the riser as an inert fluidization and heating medium (see Fig. 11.3b).

Ensyn is a well-known company that operates a commercial scale CFB biomass pyrolysis system. Their largest pilot plant can process up to 100 bone-dry tons of biomass per day. Sand is co-fed with biomass in the riser, while the residual char and sand are recovered in a separate BFB combustor where the char is burned. The hot sand is then fed into the riser.

13.4.4.1 Specification of Process Potential Material Flow and Boundaries

Figure 13.10 shows the process potential material flow of the BGUS. There are three GHGs released during the course of BGUS: carbon dioxide (CO2), methane (CH4), and nitrogen monoxide (N2O). Carbon dioxide shown by the symbolCSD in the diagram is emitted due to treatment of biomass (livestock waste in this case). It is not included in the calculation of the process potential because it is considered to be released to the atmosphere in the same amount as CO2 that was absorbed and removed from the atmosphere in the short term by plants; thus, it is considered carbon neutral according to international agreements concerning emission of GHGs made by the Intergovernmental Panel on Climate Change (IPCC)[ 16]. Chemicals indicated by the following symbols and (carbon dioxide), (methane), and C51D (nitrogen monoxide) are gases produced during crop production and biogas plant operation. Please note that when the symbol «“* indicates the use of refined biogas as an alternative to fossil fuels by equipment, it is counted toward CO2 reduction. The scope of the calculation of the process potential in terms of GHG reduction evaluation is shown in Fig. 13.10.

The main sources of GHGs produced in the processing of livestock waste by the biogas plant were divided roughly into three groups: (1) combustion of fossil fuels as power and heat sources for vehicles, (2) combustion of fossil fuels for the pro­duction of electricity when using commercial electricity, and (3) GHGs produced by the livestock waste itself by fermentation or sublimation. Further analysis shows that the process of carbon load generation can be divided into the following eight categories: (1) fuel combustion by equipment to transport livestock waste into the plant, (2) combustion of fossil fuels used by the biogas plant, (3) combustion of fossil fuels by equipment within the biogas plant, (4) deposit fermentation of solids after solid-liquid separation, (5) sublimation of digestive liquid during storage, (6) fuel combustion by equipment transporting digestive liquid and compost, (7) fuel com­bustion by equipment that spreads digestive liquid and compost, and (8) sublimation from digestive liquid and compost in fields after spreading. In addition, carbon loads were generated during (1) use of commercial electricity in the process of refining surplus biogas, (2) combustion of fossil fuels by farm trucks inside the farm pro­duction system, and (3) combustion of LPG fuel by kitchen equipment inside and outside the farm production system.

The calculation of the process potential evaluation of the BGUS modeled in Town A in terms of GHG reduction per year used the emission of carbon dioxide as its single indicator. It did not use other environmental indicators such as eutrophication of water in the region or acidification of the atmosphere in its evaluation. Below are the qualifications used in this calculation.

• Because energy involved in the delivery of biogas cylinders by the gas vendor

was almost the same as that of the previous delivery system, it was not counted

as a GHG that contributes to warming in the analysis.

<Filling refined biogas cylinders, supplying by gas vendor>

Previously existing biogas plant>

• Carbon dioxide produced when biogas was created by anaerobic fermentation of livestock waste and CO2 produced by sediment fermentation were considered the same amount as CO2 absorbed by plants during their growth, so they were considered carbon neutral and were not included in the count of GHGs.

• Besides equipment in the farm production system, there was equipment in the barn that used fossil fuels, such as boilers and tractors.

Many solid acid catalysts, such as cation-exchange resins [76], carbonaceous solid acids [79], and solid HPA catalysts [32], have shown high activity and selectivity for cellulose hydrolysis. Up to now, amorphous carbon bearing — SO3H showed the best recyclability among the reported results, and no decrease in activity was observed even after 25 cycles [42]. However, all these solid acid catalysts still suffered deactivation after a period of time, so the operation life of the catalysts still cannot meet the requirement for industrial applications. Catalyst deactivation is a problem that must be solved in the hydrolysis reaction. Fundamental understanding of the deactivation mechanisms during cellulose hydrolysis is the key to extending catalyst lifetime. The main deactivation mechanisms of solid acid catalysts are: (1) leaching of surface acid sites, (2) carbon deposition on the catalyst and poisoning by toxic substance, and (3) surface reconstruction.

Leaching of acid sites, especially sulfonic or sulfuric acid species, limits the reusability of these families of catalysts. It has been reported that all the reactants and products could cause leaching of such active species (especially H2O) even at low temperatures (e. g., 100 °C) [90]. The leaching seems unavoidable under these oper­ation conditions. Therefore, such kinds of solid acid catalysts are not proposed to be used in reactions in aqueous solutions. Moreover, it is rather complicated to regener­ate them by a pickling technique. It is well known that Al, Fe, Ni, and Pt are effective promoters to increase the stability and activity of sulfated metal oxide catalysts [91]. The promoters improve sulfate contents and acidities of solid acid catalysts.

Carbon deposition represents a significant effect on the recyclability of a viable catalyst for cellulose hydrolysis. Shi et al. [92] prepared a series of Al-promoted SO42-/ZrO2/SBA-15 catalysts and investigated the deactivation and regeneration capacities of the catalysts during the dehydration of xylose. It was found that when the catalysts were reused without regeneration, the yield of furfural decreased from

52.7 % to 19.1 %. Based on the characterization of the catalysts, the accumulation of byproducts was the main reason for the deactivation. Regeneration with H2O2 can completely recover the catalytic activity of the deactivated catalysts. After first regen­eration, the catalytic activity recovered completely. The corresponding xylose con­version rate and furfural selectivity were 98.6 % and 53.5 %, respectively, very close to those with the fresh catalysts (98.7 % and 53.4 %, respectively). During the hydrol­ysis of cellulose, the produced glucose was polymerized into carbonaceous polymers under certain hydrothermal conditions [93]. The carbonaceous polymers were further deposited on the surface of solid acids, which decreased the catalytic efficiency. The deactivated catalysts can be easily regenerated by calcination to remove the deposited coke. However, some catalysts (e. g., carbonaceous solid acid catalysts) cannot be treated by calcination at high temperature because both deposits and catalysts them­selves were combusted. As mentioned in Sect. 15.3.4.4, recovery of carbonaceous solid catalysts by incorporating paramagnetic compounds may be feasible.

Carbon deposition not only covers the active sites, but also binds to catalyst surface inducing a surface reconstruction and affecting activity. Especially for metal oxides, the diffusion of carbon into metal results in the formation of bulk metal carbide, causing the loss in both activity and selectivity [94]. The oxidative treatment was usually applied to regenerate used catalysts by removing carbonaceous phases. Saib et al. [95] studied the deactivation and regeneration of cobalt Fischer-Tropsch syn­thesis catalysts. The spent catalysts recovered their activity completely by oxidative regeneration.

The biodiesel acceptable standards around the world normally follow either the American Society for Testing and Materials (ASTM) specification D6751 [5] or the European Specification EN14214 [6]. The variable level of substituted or bound glycerol to fatty acid as tri-, di- and mono-glycerides is summed up and calculated as the total bound glycerin value with an allowable value lower than 0.24 % in biodiesel. The biodiesel specification in Europe EN14214 includes a minimum requirement for the ester content (96.5 %) and individual maximum levels for the mono-, di-, and tri­glycerides. Both specifications limit the level of total glycerin remaining in biodiesel fuel to approximately the same value (0.24 % in ASTM D 6751 and 0.25 % in EN 14214). When measuring the level of free fatty acids (FFAs), the total FFAs should be below 0.25 mg/ml.

18.2 Feedstock’s Decomposition

Virgin oil when exposed to moisture, microbial contamination, heating, and light in the presence of air such as in the case of long-term storage or cooking, undergoes a decomposition process such as oxidation and hydrolysis leading to the formation of FFAs and other low — and high-molecular weight hydrocarbons, alcohols, oxidized monomers, dimmers, and trimers. Moisture and lipases excreted by microorganisms promote the hydrolysis of vegetable oil triglycerides to form FFAs, mono-, and di- acylglycerols, which result in increasing of the refining losses directly related to the free fatty acid content of oils and fats (Fig. 18.1). Oxygen and heat cause oxidation

—OH

RCOOH

Fatly acid

-OH — OCOR’ — OH

1-Monoglyccridc 2-Monoglyccridc Fatly acid Fatly acid

of oil triglycerides which results in the formation of hydroperoxides, leading to volatiles responsible later for the unpleasant odor of spent frying oils. The heat can also initiate cyclization reactions; both intra — and inter-molecular, occurring via electrocyclic Diels-Alder and Ene reactions. Furthermore, deterioration of heated oil used for multiple frying operations, is characterized by the formation of a polymeric dark mass and other low-molecular weight compounds. During frying operations, different compounds which include fatty acids, mono-, di-, and tri-glycerides, low — and high-molecular weight polymeric dark brown materials accumulate in oil. It has been reported that oxidation of oil is mostly responsible for much more of the deterioration of fats and oils than hydrolysis [7].

The decomposed oil loses fatty acids which become detached FFAs, or “used” oil and therefore less expensive feedstock to purchase. In order to increase the overall yield of biodiesel, the high-FFAs oil requires pretreatment processing that means subjecting the oil to acidic conditions, where the FFAs are converted to fatty acid esters and thereby lowered pH to acceptable values in the oil, so that the oil will later be converted to biodiesel by conventional alkaline catalysts (reactions 1 and 3, Fig. 18.2) [8]. The FFAs can also be removed under alkaline conditions (reaction 2, Fig. 18.2) when treated with alkaline reagent such as KOH to yield fatty acid potassium salts or soap which can be removed by water-wash process [9, 10].

The conventional chemical synthesis of biodiesel is typically carried out under al­kaline conditions and to a lesser extentunder acidic conditions [11,12]. The alkaline reaction involves transesteriflcation conditions using vegetable oil (triglyceride and to a lesser extent mono — and di-glycerides), methanol, and a chemical base such as sodium hydroxide (NaOH) (reaction 3, Fig. 18.2). The reaction can also be carried out with sodium methylate (NaOCH3) as a catalyst which produces biodiesel and glycerol with improved reaction yields [13-15]. The transesteriflcation of glycerides

Reaction -1

R—COOII + KOH —————————— R—COOK + П2°

batty acid Potassium Hydroxide Soap Water

Reaction -3

— OCOR

Catalyst

—OCOR + CII3OH ————- ► 3R-COOCH3

—— CX OK

….. … Methanol Methyl ester

I nglycmdc

Fig. 18.2 Formation of fatty acid methyl esters (biodiesel) and fatty acid salts from vegetable oil and fatty acids in the presence of methanol and a catalyst and methanol can also be carried out by acid catalysis to produce fatty acid methyl esters (FAMEs) and glycerol; however, this type of reaction is unfavored due to its low rates and its strict reaction conditions with regards to high temperatures and high excess of methanol.

Although wet oxidation pretreatment is considered a promising technology for con­verting biomass into biofuels, it was rarely applied on softwood species. Palonen et al. [103] reported a 79 % cellulose conversion yield obtained from wet oxidation pretreatment of spruce. This pretreatment was performed at 200 °C for 10 min. This cellulose conversion yield was much higher than DAP and alkaline pretreatment of similar softwood species.

8.3.2 Sulfite Pretreatment to Overcome Recalcitrance of Lignocellulose (SPORL)

Recently Zhu et al. developed SPORL pretreatment for robust and efficient con­version of biomass through enzymatic saccharification [109]. During the SPORL pretreatment, the wood chips were pretreated in an aqueous sulfite solution followed by mechanical size reduction using disk refining. The terms sulfite and bisulfite are used interchangeably in the SPORL because the active reagents in the pretreatment liquor can be sulfite (SO3-), bisulfite HSO-, or a combination of two of the three reagents, sulfite (SO3-), bisulfite HSO-, and sulfur dioxide (SO2, or H2SO3), de­pending on the pH of the pretreatment liquor at a pretreatment temperature [110]. The pretreatment liquor can be prepared and recovered using existing industrial practices as described elsewhere [111]. The pH of the solution can be easily controlled by the amount of SO2 absorbed. SO2 can be substituted by other acids, such as H2SO4, HCl, oxalic acid, and acetic acid (such as the acetic acid released from acetyl groups during pretreatment of hardwood or agricultural residues).

Calcined dolomite is the most popular and most investigated material as a tar crack­ing catalyst. Dolomite is a calcium magnesium ore with a general chemical formula CaMg(CO3)2. This catalyst is relatively inexpensive, abundant, and disposable. Az — nar, Corella and their research groups, in several publications, have investigated the effect of the in-bed use of dolomite, which can decrease the tar level from 6.5 %wt to

1.3 %wt [50]. The addition of 3-10 % calcined dolomite to biomass feed decreases the tar level by 40 % and improves gas quality significantly [51, 52]. Although dolomite has proved to be effective in terms of tar reduction, it has some critical limitations. Dolomite in its naturally occurring form is not very active in tar crack­ing, and it needs to be calcined. Calcination of dolomite involves its decomposition and the elimination of CO2 to form the MgO-CaO complex. Calcination reduces the surface area of the dolomite catalyst, and makes it more friable resulting in severe catalyst attrition and fine-particle production. While using dolomite in a fluidized bed, dust entrainment due to the eroding of soft dolomite particles necessitates the continuous feeding of dolomite into the reactor by mixing it with biomass fuel [51]. This requires a major gas-cleaning operation.

Another popular catalyst in this group is olivine, which could be an alternative for dolomite. Olivine is a mineral, which contains magnesium, iron oxide, and silica. In terms of attrition, olivine has certain advantages over dolomite. However, Corella et al reported that dolomite was 1.4 times more active than olivine in biomass gasification with air, but dolomite generated ~ 4-6 times more particulates in the gasification gas than olivine [53]. Nickel has also been used with an olivine support, which led to improved catalytic activity compared to unsupported olivine catalysts [54, 55].