Naphtha is produced as a by-product of the BTL-FT process, both straight-run from the FT reactor and as a co-product of the upgrading of the FT wax to middle distillates. BTL-FT naphtha has a low octane number and cannot be used as a gasoline blending component. The two dominant processes that have been considered for upgrading FT naphtha to high-octane gasoline are isomerization and reforming. Given that straight-run FT naphtha contains olefins and oxygenates that are not compatible with commercial reforming or isomerization technologies, a hydrotreating step is first required to convert olefins and oxygenates in the naphtha to paraffins (Gregor and Fullerton, 1989). According to a techno-economic study by Kreutz et al. (2008), the optimum BTL-FT plant configuration in order to maximize the yield of premium diesel and gasoline fuels is to isomerize a portion of the naphtha in order to convert normal paraffins to isoparaffins and boost its octane value and catalytically reform the other fraction to provide some aromatic content to (and further boost the octane value of) the final gasoline blendstock. However, it is still uncertain whether the additional gasoline blending stock value can justify the great capital and operational costs that these upgrading units impose on the BTL-FT process and explains why this option has yet to be considered in commercial operations.

Different types of biomass, whether these are primary crops or residues, are applied nowadays as the raw material for biofuels. These raw materials are used just for their caloric value. Both direct (oil) and indirect use, after conversion of the biomass into a liquid component, preferably with an increased caloric value per volume, are distinguished. Ethanol or butanol are good examples of the latter.

Other components that are present in the biomass crops or their residues are often regarded as primary residues, however, these components could have significant value, and their valorisation could well contribute to the economic feasibility of the overall biofuels production process as well (Brehmer et al, 2009).

Thermal processing of biomass has the advantage that the heterogeneous biomass components are converted towards a much more homogeneous mixture, i. e. syngas or pyrolysis oil (Manurung et al., 2009). These processes, however, will not benefit from the presence of the specific components that are not easily converted into biofuels but could create value because of their functionality on a molecular level or on a macroscopic level, e. g. as construction material or as a product to make paper.

Functionalised bulk chemicals offer great economic potential, if one wants to substitute fossil raw materials by biomass, since these chemicals need to be synthesised using, apart from oil, large quantities of energy. The production of these chemicals conventionally also requires high capital costs due to the fact that many conversion steps are necessary to convert raw oil via naphtha and ethylene to these more complex functionalised chemicals (Sanders et al., 2007).

In plant material often molecular functionality is present that can be used to make the same bulk chemicals as nowadays by petrochemical processes, but now with short synthesis routes starting from biomass (Haveren, 2007; Scott, 2007). Certainly not all components in biomass represent these high values; also biofuels, power, heat and soil improvers like fertilisers will contribute to the overall valorisation of raw biomass materials.

When more than one product is produced from a biorefinery unit, the logistics and pre-treatment become more important (Bennett, 2009). Because the transport of water and minerals is not very sustainable, the first pre-treatment and fractionation steps will be performed close to the biomass production fields. This favours small scale operations close to the fields that make intermediate products that are easy to transport and that do not deteriorate in time. Questions that still have to be solved are: (1) how can small scale processes that do not suffer from diseconomy-of-scale, and preferably can operate more economically on small scale than on large scale, be developed? and (2) how sustainable are these processes taking into account People, Planet and Profit issues?

The high molecular weight and size of triglycerides molecules, which comprise vegetable oils and animal fats, prevent their direct use as transport fuels, and hence, they must be upgraded. Hydrotreatment of triglyceride-based feedstocks (vegetable oils and animal fats) for automotive fuels has been studied in detail (Bezergianni et al., 2009; Huber et al., 2007; Lappas et al., 2009; Petri and Marker, 2006). Hydrocracking of these renewable raw materials has been also studied by several authors (Bezergianni et al, 2009; da Rocha Filho et al., 1993; Gusmao et al., 1989; Kubickova et al., 2005). However, high amounts of hydrogen are required to enhance hydrodeoxygenation processes. Such reaction pathway implies the conversion of the oxygen present in the triglyceride in form of water (Gusmao et al, 1989; Huber et al., 2007). The formed water, as well as the initial content in the feedstocks of metals (such as sodium, potassium, calcium or phosphorous), and other impurities (solid particles, water or detergents) are associated with problems related to the durability of the sensitive hydrogenation catalyst (Petri and Marker, 2006). Furthermore, there is always a problem with the operating costs related to the high hydrogen consumed along the reactions, which advise against the co-processing of renewable raw materials in refining units that work with high-pressured hydrogen.

On the other hand, there is the possibility of cracking triglyceride-based feedstocks in refining units without the presence of hydrogen. These possible units are thermal cracking units such as visbreaker or coker and the FCC unit. Thermal cracking units are used for the breakdown of heavy crude oil into smaller molecules in the absence of catalyst and hydrogen. Several vegetable oils have been thermally cracked and the results reported in literature: tung oil (Chang and Wan, 1947), soybean oil (Demirbas and Kara, 2006; Lima et al., 2004; Schwab et al., 1988), high-oleic safflower oil (Schwab et al, 1988), palm oil (PO) (Chew and Bhatia, 2008; Lima et al., 2004), castor oil (Lima et al., 2004), canola oil (Idem et al., 1996; Sadrameli and Green, 2007), several tropical vegetable oils (Alencar et al., 1983) and oleaginous waste feedstocks such as waste cooking oil (Dandik and Aksoy, 1998), oils from non-edible fruits (such as Macauba fruit; Fortes and Baugh, 1999, 2004) and non-edible animal fats (Adebanjo et al., 2005; Demirbas, 2009). Moreover, Padmaja et al. (2009) have recently reported the thermal cracking of a biocrude extracted from Calotropis procera (laticiferous arid plant from India) under conditions similar to those found in visbreaking and delayed coking. All of the above-mentioned reactions have been usually performed in batch reactors, although fixed bed reactors (usually under the presence of inert materials) and fluidized bed have been also reported. Temperature ranges usually between 300°C and 500°C and the operating pressure is always close to the atmospheric. From this work, a high amount of oxygenated hydrocarbons is found in the final reaction products independently of reaction temperature. Although the thermal decomposition of triglyceride molecules and their associated heavy oxygenated hydrocarbons is always initiated at temperatures of 240-300°C (without the presence of oxygen) (Adebanjo et al, 2005; Crossley et al, 1962), the presence of a catalyst is necessary to remove oxygen from oxygenated hydrocarbons such as carboxylic acids, esters, aldehydes or ketenes and to obtain an organic liquid fraction suitable for gasoline and diesel formulation. Thus, the co-feeding of this renewable feedstock to an FCC unit would be more feasible. This unit is the most widely used process for the conversion of heavy fraction of

crude oil into high-value products (e. g. diesel, gasoline). This unit operates under high temperatures (> 500°C) and pressure close to the atmospheric in the absence of hydrogen and the presence of an acid catalyst.

In this section, we will discuss the chemistry involved in the catalytic cracking of triglyceride molecules as well as the work dealing with the processing of this biomass feedstock under FCC realistic conditions.

The major obstacle for the large-scale implementation of biomass gasification is the ‘tar problem’. The circulating fluidised bed gasifier typically produces ~10 g/ M03. Tars in the synthesis gas give rise to fouling, as they deposit on the walls of the system when the tars, due to cooling, condensate. There are numerous purification processes, but few of them are suitable for our application. The OLGA3 concept seems to be most promising for this process. This concept reaches highest efficiency, has been tested thoroughly and is commercially available. Besides that, this system is designed especially for this purpose. For thermal tar cracking, high temperatures will have to be reached, which leads to extra loss of energy and therefore lower efficiency. The catalytic tar cracking also seems to be promising, but isn’t commercially obtainable yet. Besides that, one of the boundary conditions to the plant is that as few as possible additives are to be used. The catalyst in this concept needs extra care especially as deactivation can be a problem and the higher investment costs.

17.3.3 Fermentation

After gasification and gas purification the next process step to produce ethanol is fermentation. Fermentation occurs when bacteria metabolise a material into another one. In this case the cleaned synthesis gas can be converted anaerobically into ethanol. Biological production of chemicals from synthesis gas offers several advantages over catalytic techniques:

• Biological conversion occurs under mild temperatures and pressures, whereas catalytic reactors are operated at high temperatures and pressures.

• The reaction specificity of enzymes is typically higher than that of inorganic catalysts.

• Most biological catalysts are tolerant to sulphur gases, reducing the cost of gas cleanup prior to the conversion step.

• In the fermenter the waste gas shift takes place biologically so preventing the use of a separate shift reactor for adjusting the CO/H2 ratio.

Type of bacteria

Several acetogenic microbes are capable of metabolising synthesis gas into ethanol. Two of the more promising strains are described below. Both are gram­positive bacteria, this means they are characterised by having as part of their cell wall structure peptidoglycan as well as polysaccharides and/or teichoic acids. The peptidoglycans are heteropolymers of glycan strands, which are cross-linked through short peptides.4

• Butyribacterium methylotrophicum (Bredwell et al., 1999): It is a gram­positive, motile, rod-shaped anaerobic bacterium, which grows on a wide variety of substrates, including glucose, formate and methanol, H2 and CO2, and CO. The latter two are of interest to us. The products achieved are acetic acid, butyric acid, ethanol and butanol. Production of ethanol and butanol is usually low. When production of ethanol and butanol is increased, butanol is dominant.

• Clostridium species: In the case of clostridial fermentation it has been proposed that acetyl-CoA is the central intermediate (Roger, 1986). The first of the clostridium species was isolated from chicken waste and grew well on the components of synthesis gas to produce acetate and ethanol. The first optimisations of this species resulted in an approximate 1:1 ratio of ethanol and acetate under optimal conditions (Vega, 1989). After that the developments of the clostridium species have been many.

— Clostridium ljungdahlii: It is a gram-positive, motile, rod-shaped anaerobic bacterium, which converts CO, H2 and CO2 into a mixture of acetate and ethanol. The ratio of these products can be adjusted by pH. When the pH is lowered to 4 the ratio ethanol: acetate becomes 3: 1 (Gaddy and Fayetteville, 1995). Further medium adjustment has reportedly nearly eliminated acetate production and led to an ethanol concentration of 48 g/L (approximately 1 mol/l) on day 25 when using an optimised medium (Phillips et al, 1993). This means the separation of ethanol from water is feasible.

— Clostridium carboxidovorans (P7) (Rajagopalan et al., 2002): P7 is a gram­positive, motile, rod-shaped anaerobic bacterium, which converts CO, H and CO2 into a mixture of acetate, butanol and ethanol. The ratio ethanol: butanol:acetate is 6:3:1 in absence of hydrogen. Further developments are expected to include hydrogen and inhibition of the butanol step.

The clostridium species has the most potential and is best suited for the development of ethanol. From the clostridium species P7 has a lot of potential, but because of the early stages of the development of this bacterium and therefore lack of data, Clostridium ljungdahlii is most promising for our goal to produce ethanol from synthesis gas at this moment.

Bacterial fermentation of CO, CO2 and H2 to ethanol using Clostridium ljungdahlii gives the following equations (Klasson et al., 1993):

(1) 6 CO + 3 H2O ^ C2H5OH + 4 CO2 AG = 216 kJ/mol

(2) 6 H2 + 2 CO2 ^ C2H5OH + 3 H2O AG = -97 kJ/mol

The distinctive feature of the followed pathway of these microorganisms seems to involve the reduction of carbon dioxide to a methyl group and then its combination with a molecule of carbon monoxide and CoA to form acetyl-CoA (Ljungdahl, 1986). This combination of reactions has been designated as the acetyl-CoA pathway (Wood et al., 1986).

After cleaning the synthesis gas, the next part of equipment needed is for fermentation. In this equipment the bacteria must be able to convert synthesis gas into ethanol and after this the product (ethanol) has to be removed from the rest of the stream. For energetic, environmental and economic reasons the rest stream (consisting of nutrients, water and a small amount of ethanol) should be recycled into the reactor. This is schematically shown in Fig. 17.3. Here a gas-sampling vessel disengages the gas from the liquid. At the recycle-bottle port, fresh nutrients are added, the liquid effluent is withdrawn, and the pH is adjusted. Experimental studies have shown that the rate-limiting step in synthesis gas fermentation is typically gas-to-liquid mass transfer. This means when the gas gets easier in the liquid the reaction-rate will increase. A common approach to enhance gas-to — liquid mass transfer in stirred tanks is to increase the agitator’s power-to-volume ratio. Increasing the power input increases bubble break-up, thereby increasing the interfacial area available for mass transfer.

However, this approach is not economically feasible for the very large reactors being considered for commercial synthesis gas fermentation, due to excessive power costs. Consequently, alternative bioreactor configurations that may provide more energy-efficient mass transfer are needed. The most interesting option is a monolith reactor.

Monolith reactor

A monolith reactor is columnar and does not require mechanical agitation and thus offers the potential for lower power costs than stirred tanks. A monolith reactor (Fig. 17.4) is a packed bed of channels where the liquid and gaseous phase flow in co/current downwards. High gas and low liquid flow rates are typically used, giving relative low-pressure drops. The cells are immobilised on the wall. This development is promising, but requires a lot of knowledge and equipment and is still in the experimental stage for fermenters (Salim et al., 2008).

Pyrolysis oil contains a mixture of a lot of chemicals of which certain fractions can be interesting for dedicated applications. An example of such a route is given in Figure 20.5 as proposed by the NREL.16 Here, pyrolysis oil is phase separated via water addition into an aqueous soluble phase and aqueous insoluble phase (here called pyrolitic lignin) which can be used for the production of phenolic resins. The remaining aqueous rich phase is then steam reformed (gas phase) to produce hydrogen. This phase could alternatively also be reformed at higher pressures in hot compressed water.

Co-feeding biomass to existing refineries

Feeding pyrolysis oil to an existing refinery could facilitate large scale implementation of the use of second generation bio-fuels.17 The pyrolysis oil would then be fed to specific sections of a refinery (like FCC and hydrotreating). To allow this, pyrolysis oil would need an upgrading step via (mild) hydrogenation. As a side product of this upgrading step, water-rich organic side streams are being produced which would then, via steam reforming or reforming in hot compressed water, be a source for hydrogen for the refinery and the upgrading itself.

Shimada et al. (2000) converted free sterols in SODD to sterol esters and completely hydrolyzed acylglycerols by applying an enzymatic reaction to the purification of tocopherols and sterols. Fractionation of these two compounds of interest was carried out by short-path distillation. It was found that C. rugosa lipase recognized sterols as substrates but not tocopherols, and that esterification of sterols with FFA could be effected with negligible influence on the water content. High boiling point substances, including steryl esters, were removed from the SODD by distillation, and the resulting distillate (SODDTC) was used as a starting material for tocopherol purification. Several factors affecting esterification of sterols were investigated. It was observed that approximately 80% of sterols were esterified when tocopherols were unmodified. After the reaction, tocopherols and FFA were recovered as a distillate by molecular distillation of the oil layer. To enhance further removal of the remaining sterols, the lipase-catalyzed reaction was repeated on the distillate. As a result, more than 95% of the sterols were esterified in total. The resulting reaction mixture was fractionated to four distillates and one residue. The main distillate fraction contained 65% tocopherols with low contents of FFA and sterols, and the residue fraction contained high-purity steryl esters. It was suggested that due to the fact that the process presented in this study included only an organic solvent-free enzymatic reaction and a molecular distillation, it could be feasible as an industrial purification method for tocopherols. However, the process had the drawback that FFA and tocopherols were not efficiently fractionated since the boiling points of the two compounds were close.

Watanabe et al. (2004) introduced the chemical modification of the DD to convert FFA to their methyl or ethyl esters, followed by short-path distillation of the reaction mixture for the elimination of fatty acid esters and thus the purification of tocopherols and sterols esters.

Tocopherols and sterols in the SODD were first recovered by short-path distillation, which was named SODD tocopherol/sterol concentrate (SODDTSC). The SODDTSC which contained MAG, DAG, FFA and unidentified hydrocarbons in addition to the two compounds of interest was then treated with a lipase to convert the free sterols to fatty acid steryl esters (FASEs), acylglycerols to FFA and FFA to FAME. It was observed that methanol inhibited the esterification of the sterols. Hence, a two-step in situ reaction was conducted and a conversion of 80% of the initial sterols to FASEs, complete hydrolysis of the acylglycerols and a 78% decrease in the initial FFA content by methyl esterification, was achieved. To enhance the degree of steryl and methyl esterification, the reaction products (FASEs and FAME) were removed by short-path distillation, and the resulting fraction containing tocopherol, sterols and FFA was again treated with the lipase. Distillation of the reaction mixture purified the tocopherols to 76% (recovery, 89%) and sterols to 97% as FASEs (recovery, 86%).

Nagao et al. (2005) described a process where SODD was first distilled and the sterols and tocopherols were enriched. The obtained fraction was SODDTSC. In this study, esterification of sterols was improved by removing water with a degree of esterification of 95%. The second-step reaction was then conducted in which

95% FFAs were converted into FAME. Finally, tocopherols and steryl esters were purified from the reaction mixture by short-path distillation. Tocopherols were purified to 72% (88% yield) and steryl esters were purified to 97% (97% yield).

Purification on a larger scale was performed with 1.5 kg SODDTSC and the procedure is shown in Fig. 22.5.

Albiez et al. (2004) described a process for concentrating and isolating sterols and/or tocopherols from physically refined DD that consisted in hydrolysis to split the glycerides present into FFA and glycerol, followed by the glycerol- containing hydrolysis water removal and distillation of the FFA and readily volatile unsaponifiable components. The distillation residue was additionally hydrolyzed to split the sterol esters into FFA and free sterols, followed by the distillation of the later one.

Accordingly, the problem addressed by the present patented invention was to provide a process for the simultaneous production of tocopherol and sterol which would be applicable to many different starting mixtures, which would not involve the use of toxicologically and ecologically unsafe solvents, which would use even low-concentration starting materials sparingly and which would still give high yields without the use of metal-containing catalysts. In addition, the process would be economically workable on an industrial scale.

Top et al. (1993) described a process for the production of tocopherols and tocotrienols where the PFAD was first modified and the resulting fractions were further purified applying different steps. The process includes the conversion of FFA and glycerides in PFAD into alkyl esters by esterification and transesterification, followed by distillation of the resulting product under reduced pressure to remove a major part of the alkyl esters and leave the tocopherols and tocotrienols and other higher boiling point substance in the residue. The residue was cooled to induce crystallization of higher melting substances and other impurities and the crystalline material was filtered off to leave the tocopherols and tocotrienols in the filtrate. The filtrate was further treated by an ion-exchange procedure with a high selectivity anionic resin to produce a concentrated fraction of tocopherols and tocotrienols, the solvent was removed by evaporation. The tocopherols and tocotrienols fraction was washed dried and then subjected to molecular distillation and deodorization to produce a further concentrated product of tocopherols and tocotrienols. After evaporation step, the concentrations of tocopherols and tocotrienols were 83% and 87%, respectively.

In the same invention, an alternative process was described, where the PFAD was pretreated to remove the majority of the FFA by distillation before sending it to the process described above.

Martins et al. (2005) described a process where the SODD was chemically modified, submitted to molecular distillation for fatty acids elimination and the product obtained was submitted to an ethanolic extraction for tocopherols and concentrations of phytosterols. Chemical modification of SODD was conducted by a saponification at 65°C, followed by an acidulation step. With this procedure it was possible to release conjugated fatty acids of acylglycerols molecules. Therefore, not only FFA can be removed from the mixture by molecular distillation, but conjugated fatty acids of acylglycerols also lead to a higher tocopherol concentration. The applied molecular distillation was characterized by using high vacuum, reduced temperature and low residence time. SODD, containing about 75% of FFA, was submitted to four steps of molecular distillation to remove the FFA from the mixture. The separation of tocopherols from sterols was difficult because they have similar molecular weights, boiling points and vapor pressure, and consequently, they are distilling together. Therefore, the resulting product of molecular distillation was submitted to an ethanolic extraction at 0°C to separate the tocopherols from the sterols. As tocopherols are soluble in ethanol, it was possible to separate and to concentrate phytosterols and tocopherols. This process obtained a purity of 26% of tocopherols and 52% of sterols.

A recent patent (Zima et al., 2009) describes a process for preparing a phytolipid composition containing squalene, phytosterols, tocopherols and vegetable wax that consist two steps of distillation at different temperatures and pressure, extraction and precipitation (Fig. 22.6). The final phytolipid product may contain ca. 10% to about 40% squalene, 2% to about 20% phytosterols, 1% to about 10% of mixed tocols and 40% to about 80% vegetable wax with possible applications in the cosmetic, nutraceutical or food industry.

M. A. MORRIS, University College Cork, Ireland

Abstract: This chapter provides a review of catalytic pyrolysis summarising the potential of the methodology. Catalytic pyrolysis is centred on the use of catalysts in the production of bio-oils and related oils by pyrolysis of biomass and various waste materials. The subject is detailed against growing requirements for development of sustainable energy and fuel sources as pressures on fossil fuels increase as well as increasing fears on climate change and potential fuel shortages. The economics of pyrolysis derived bio-oil production against competing established and emerging technologies is provided. The review summarises the current state-of-the-art with particular reference to the challenges of catalysing reactions in the harsh environments of pyrolysis reactors. The types of active solid materials that can be used to generate oil are detailed so as to indicate the flexibility of the methodology. The outlook for commercialising of the technology is also summarised. A brief review of the potential use of pyrolysis products is given and barriers to uptake of this emerging technology are explained.

Key words: pyrolysis, catalytic pyrolysis, zeolites, mesoporous silicates, transition metal catalysts, bio-oil, pyrolysis-oil.

14.1 Introduction

There is little doubt that the world is facing an uncertain future around the continued use of fossil fuels as has been outlined previously in this book and elsewhere.1 Fossil fuel related climate change due to anthropogenic emissions of carbon dioxide is well known with 98% of carbon emissions arising from fossil fuel combustion.2 Further, depletion of fossil fuel reserves is expected within a few generations3 and energy security has become a major issue.4 Whilst the major uses of fossil fuels are in domestic energy production5 and transportation,6 petroleum, gas and coal have very significant other uses. For example, oil is used to prepare in excess of 70 million tonnes of polyolefins per year.7 Perhaps most importantly, very significant amounts of gas and oil are used in the production of fertilisers.8 Fertiliser is wholly necessary for maintaining food supplies and feeding the growing world population. Fertiliser is prepared via the fixation of nitrogen by reaction of atmospheric nitrogen with hydrogen over transition metal catalysts, usually using iron based catalysts as were originally developed by Haber and Bosch.11 The product of the reaction, ammonia, can be subsequently oxidised to nitric acid and the direct reaction of further ammonia yielding ammonia nitrate and this has been the basis of the modern fertiliser industry for almost 100 years.12 The hydrogen needed in ammonia synthesis is derived from nickel catalysed reactions of gas or oil with water (steam reforming13) followed by a copper-zinc oxide catalysed reaction of carbon monoxide with water (water gas shift14). These reactions are outlined further below. Because of the hydrogen sources used, the fertiliser price is closely related to the oil price.9 It should be noted that alternatives to steam reforming exist and hydrogen can also be obtained from methane by decomposition15 and aromatisation.16 Because hydrogen is essentially derived from fossil fuels (either directly, as detailed here, or indirectly via electrolysis of water using convention fossil fuel energy sources), it seems appropriate that, for the purposes of this review, hydrogen is considered as a petroleum product.

The over-reliance on fossil fuels derives from the convenience of these energy sources as a means of transporting and delivering energy.17 Alternative sources of energy (wind, solar, nuclear, etc.) are unlikely to provide a convenient source of energy consistent with industry requirements and not precipitate large-scale industry changes and the capital required to replace current large scale chemical technologies based on oil processing that supply polymers, fertiliser, fine chemicals, etc. An alternative strategy to drastic modification of the chemical economy and the use of oil as a form of transporting energy is to find an effective means to generate petroleum-like products from renewable or waste material sources. The potential of pyrolysis is one means to affect the delivery of both petroleum and hydrogen allowing maintenance of current technologies.

Solid fuel in the presence of a gasifying agent (air, oxygen, steam) under thermal action undergoes chemical decomposition to produce the useful gas. According to the type of gasifying agent used, the heating value of the product gas obtained will also be different.

The conversion of gasification feedstocks can be divided into several gross stages: (1) decomposition of the original feedstock into volatile matter and char,

(2) conversion of the volatile matter by secondary reactions (combustion and reforming), and (3) conversion of the char by ‘char gasification’ reactions with H2O and CO2 to produce fuel gases (CO, H2, CH4), in addition to char combustion when oxygen is present. Devolatilization produces a broad spectrum of products, ranging from light gases to tars. The products are strongly dependent on the identity of the feedstock and process conditions, such as heating rate. These products may contain valuable species. Partial reforming of these products by contact with components of the char bed may result in improved gas quality. For example, if fuel gas is the desired product, such conversion could preserve methane while reforming undesirable tars. The progress of such reforming reactions is dependent on the nature of the char, including the inorganic (ash) components, and the type of reactor. The conversion of the entire feedstock to fuel gases by gasification reactions is generally endothermic, and air or oxygen is typically added to heat balance the process. In general, the solid fuel, during gasification, undergoes the following four processes that are more distinct in the case of the moving bed gasifier (such as an up and down draft gasifier) than in the case of fluidized bed gasification. The mechanism of gasification is shown in Fig. 16.3 and explained in detail below.

1 Drying: In this pricess, the moisture in the feedstock is vaporized. The feedstock does is not decompose because the temperature is not high enough to cause any chemical reaction.

2 Pyrolysis: During pyrolysis or devolatization, the volatile content of the matter is released from the feedstock and char is left. This reaction occurs in the absence of oxygen and at a temperature around 300-500°C. The reaction occurring in this process is endothermic in nature, thus the heat required is provided by the combustion of the feedstock during the oxidation process.

Feedstock = char + volatiles + energy (kJ/kg). [16.1]

3 Oxidations: In this process, the feedstock is combusted with the air supplied. As gasification is an endothermic process, the overall heat required is produced during this process. To maintain a favorable temperature in the gasifier and also avoid the excess dilution of the product gas, an equivalent ratio (actual air supply/stoichiometric air required for complete combustion) is maintained between 0.2-0.4. The reactions taking place in this process are:

C + O2 ^ CO2, [16.2]

2H2 + O2 ^ 2H2O. [16.3]

4 Reduction: In this process, several reactions take place. The product from this process is mainly the gas, consisting of carbon dioxide, hydrogen, methane and carbon monoxide. The following reactions take place:

The aforementioned reactions are the major gasification reactions. Depending upon the operating conditions, one reaction dominates over another and, thus, the product composition changes accordingly. For example, if steam gasification is used then the reactions [16.5] and [16.7] would be major ones and it can be seen that increasing temperature increases reaction [16.5] while decreasing reaction [16.7]. Similarly, if it is air or oxygen gasification, then reaction [16.4] will be the major one that increases with rise in temperature. Thus, air/oxygen gasification will have a higher concentration of CO.

The syngas purification step is the most expensive part of an FT complex. It accounts for 60-70% of the total cost in the case of natural gas (simplest option). This cost rises up to 50% more in the case of coal-based FT process, with additional 50% cost increase in the case of biomass feedstock (Zhang, in press). Syngas cleaning is, therefore, considered the biggest challenge to the commercialization of the BTL process.

The presence of impurities in the syngas produced by the gasification step is inevitable. Syngas contains different kinds of contaminants such as particulates, condensable tars, BTX (benzene, toluene and xylenes), alkali compounds, H2S, HCl, NH3 and HCN. The catalysts employed in the FT reactor for the synthesis of the liquid fuels are notoriously sensitive to such impurities, and especially sulphur and nitrogen compounds, which irreversibly poison FT catalysts. Alkaline metals and tars deposit on catalysts and contaminate the products, while particles cause fouling of the reactor. Therefore, extensive cleaning of the syngas is required prior to entering the FT reactor. Moreover, the concentration of inert gases (i. e. CO2, N2, CH4, etc.) must be approximately less than 15 vol.% (Boerrigter et al., 2004). Indicative syngas specifications for FT synthesis are shown in Table 19.1 (Boerrigter et al., 2004).

The first step in all syngas cleaning configurations considered so far is the removal of BTX and larger hydrocarbons, the tars. BTX should be removed upstream the active carbon filters in the syngas cleaning train, as active carbon adsorbs BTX and would therefore require frequent regeneration, reducing process reliability. Tars normally condensate at the typical FT reactor conditions and foul downstream equipment, coat surfaces and enter pores in filter and sorbents. Therefore, tars should be removed to a concentration below condensation point at the operating pressure of the FT reactor (Hamelinck et al., 2004). Three processes can be used for tar removal. Thermal cracking of tars involves high temperatures, 1000-1200°C, and tars are cracked in the absence of a catalyst with the use of

steam or oxygen. However, thermal cracking has low thermal efficiency, requires expensive materials and results in the production of large amounts of soot. Catalytic cracking/reforming of tars in the presence of dolomite/olivine, nickel — based catalysts or alkalis (Wang et al., 2008) overcomes these limitations. Still, this technology is not yet proven and costs are increased due to catalyst consumption (Milne et al., 1998). Alternatively, tars can be removed at a low temperature by advanced scrubbing, using a special organic washing liquid (‘oil’). Such a system has been developed by ECN, who have patented the OLGA tar removal technology (Boerrigter et al., 2004). It should be mentioned that the use of entrained flow gasifiers removes the need for a tar cracking/removal step as the high gasifier operating temperatures (1300-1500°C) yield a tar-free syngas.

After the removal of the tars, other contaminants can be removed from the syngas by either the conventional ‘wet’ low temperature or the ‘dry’ high temperature cleaning. The wet gas cleaning technology is proven and has been well commercialized for large-scale coal gasification systems (Zhang et al., 2007). The general approach involves the quenching of the raw hot gas with water to cool the gas and remove solid particles (e. g. dust, soot, ash) and the volatile alkaline metals (Boerrigter et al., 2004). NH3 is then removed by a water washer along with halides and H2S is removed either by absorption or the Claus process to elementary sulphur. In the final step, the gas passes through a ZnO and active carbon filters, which remove H2S and remaining trace impurities and act as guard beds for the FT catalyst. Although proven, this technology has efficiency penalties and requires additional waste-water treatment. Many research efforts have been focused on the development of dry hot syngas cleaning processes, which appear to be potentially more efficient and cleaner than the proven conventional wet technology (Sharma et al., 2010). Hot gas cleaning consists of candle or ceramic filters for removing solid contaminants and sorbents for fluid contaminants, through which the high temperature of the syngas can be maintained, achieving efficiency benefits and lower operational costs. Dry gas cleaning can be especially advantageous when preceding a reformer or shift reactor, as these processes have high inlet temperatures. However, as aforementioned, the performance and reliability of the filters and sorbents has still to be proven at high temperatures, especially above 400°C, for a commercial implementation of the dry gas cleaning technology. Recent developments and critical review of the different syngas cleaning technologies have been published and can be found in Sharma et al. (2008) and Sharma et al. (2010).

After the gas cleaning train, the biomass-derived syngas has to be conditioned in order to adjust the H2/CO ratio to that required for the FT reactor. Typical conditioning includes steam reforming of methane and light hydrocarbons to CO and H2 over a nickel catalyst, followed by a water gas shift (WGS ) reactor. Finally, as the concentration of inert gases must be kept below 15 vol.% (Boerrigter et al., 2004), CO2 is removed with amine treating. The purified and conditioned synthesis gas is then compressed to the required pressure and is fed to the FT reactor.

In literature, various types of biorefineries are dealt with (IEA Bioenergy, 2010), viz.:

• Green Biorefineries (GB), using ‘nature-wet’ biomass such as green grass, alfalfa, clover or immature cereals.

• Whole Crop Biorefineries (WCB), using raw materials such as cereals or maize.

• Lignocellulosic Feedstock Biorefineries (LCFB), using ‘nature-dry’ raw materials such as lignocellulose containing biomass and residues, including the more technology and/or main intermediate based concepts:

— Thermo-chemical Biorefineries (TCB)/Syngas Platform (SG)

— Bio Chemical Biorefineries (BCB)/Sugar Platform (SG)

— Two Platform Concept Biorefineries (TPCB)

— Forest Based Biorefineries (FBB)

• Marine Biorefineries (MB), using micro — or macro-algae (seaweeds), including:

— Micro Algae Biorefineries (MAB)

— Seaweeds (macro algae) Biorefineries (SB).

IEA Bioenergy Task 42 developed a more general classification system, better describing raw materials used, main intermediates (platforms) produced (a measure for the complexity of the biorefinery concept dealt with), and final products delivered.

The background for the proposed biorefinery classification system is the current main driver in biorefinery development, i. e. efficient and cost-effective production of transportation biofuels, to increase the biofuel share in transportation sector, whereas for the co-produced bio-based products additional economic and environmental benefits are gained. The classification system is based on a schematic representation of full biomass to end products value chains, distinguishing: raw materials, primary conversion processes, main biomass constituents (carbohydrates, lignin, proteins, fats, etc.), secondary conversion processes, platform intermediates, conversion processes and end products (see Fig. 21.1).

The platforms (e. g. C5/C6 sugars, syngas, biogas, bio-oil) are intermediates which are able to connect different biorefinery systems and their processes. The number of involved platforms is an indication of the system complexity of the biorefinery facility/concept. The two biorefinery product groups are energy (e. g. bioethanol, biodiesel, synthetic biofuels, power and heat) and products (e. g. chemicals, materials, food and feed). The two main feedstock groups are ‘energy crops’ from agriculture (e. g. starch crops, short rotation forestry) and ‘biomass residues’ from agriculture, forestry, trade and industry (e. g. straw, bark, wood chips from forest residues, used cooking oils, waste streams from biomass processing).

In the classification system, four main conversion processes are differentiated, including: biochemical (e. g. fermentation, enzymatic conversion), thermo-chemical (e. g. gasification, pyrolysis), chemical (e. g. acid hydrolysis, synthesis, esterification) and mechanical processes (e. g. fractionation, pressing, size reduction).

The biorefinery processes/concepts can be classified as:

A <specific platforms concerned> platform biorefinery for the production of final products produced> from <name raw materials used>.

Some examples of classifications are:

• A C6 sugar platform biorefinery for the production of bioethanol and animal feed from starch crops.

• A syngas platform biorefinery for the production of FT-diesel and phenols from straw.

• A C6 and C5 sugars and syngas platform biorefinery for the production of bioethanol, FT-diesel and furfural from saw mill residues.