Fluidized bed reactors are theoretically an excellent reactor type choice for highly exothermic reactions such as the FT reaction. Fluidized bed reactors offer a much higher efficiency in heat exchange, compared to fixed beds, and better temperature control due to the turbulent gas flow and rapid circulation. At the same time, the high gas velocities do not cause any pressure drop issues and smaller catalyst particles can be employed. This translates to high cost reduction due to smaller required heat exchange area, lower gas compression costs and easier construction. Moreover, fluidized beds permit on-line catalyst removal; thus, no down time for catalyst change is necessary as opposed to the fixed bed reactor (Dry, 1996). However, the fluidized bed reactor is only suitable for HTFT, as it can only operate with two phases, solid and gas. If not, liquid and heavy components deposit on the catalyst, leading to solid agglomeration and loss of the fluid phase (Davis, 2002). This means that fluidized bed reactors cannot be used for maximized production of products heavier than gasoline/naphtha (Steynberg et al., 2004). Moreover, according to Geerlings et al. (1999), fluidized bed reactors are more
suitable for coal conversion, as opposed to the fixed bed and slurry reactors that operate well in natural gas conversion processes.

Some disadvantages of the fluidized beds are the complexity in operability, difficult separation of the fine catalyst particles from the exhaust gas (imposing significant capital costs for cyclones and oil scrubbers) and erosion problems due to the high linear velocities (Dry, 1996). Moreover, H2S contamination of the synthesis gas feed means complete deactivation.

Currently, two types of fluidized bed reactors have been developed and used mainly by Sasol: the CFB and the fixed fluidized bed (FFB). In the CFB reactor, the fine catalyst particles are entrained by high velocity gas stream through a riser reactor. The catalyst is separated from the effluent by cyclones and is returned to the reactor inlet. Due to fluidization problems observed in the CFB reactor, Sasol developed the FFB version, which operates in the bubbling regime and is internally cooled by cooling tubes, as shown in Fig. 19.5b (Sie and Krishna, 1999). The main advantages of the FFB reactor versus the CFB type are the lower construction costs, increased capacity per reactor, less energy required for gas circulation, less catalyst attrition and easier operation and maintenance (Dry, 2002; Sie and Krishna, 1999).

—— ► Product gases

—— Cyclones

Fluidized —i — bed

*—— Boiler feed

water

—— Gas distributor

<—— Total feed

Slurry reactors

Slurry bubble reactors are a version of the fluidized bed reactors, however, in a three-phase system, that is the catalyst is suspended in a liquid through which the feed gas is bubbled as shown in Fig. 19.6. They are therefore employed for LTFT with high molecular mass liquid waxes as the main product, which naturally serves as the liquid phase of the reactor (Dry, 1996). Slurry reactors share many

of the advantages of the fluidized bed reactors, such as good isothermal operation due to excellent heat transfer both within the slurry and to the cooling system, no intra-particle diffusion limitations as the catalyst particles are small, lower pressure drop and thus lower compression costs and, of course, easier catalyst replacement (Dry, 1996; Geerlings et al., 1999; Tijmensen et al., 2002). The main disadvantage of slurry reactors is the difficult catalyst/wax separation. The removal of wax, but not catalyst, is a critical aspect of bubble column reactor operation. Sasol, which is the main company operating slurry bubble reactors, uses wax/slurry separation considered to be proprietary information, paying special attention to the production of the catalyst and its physical characteristics as well as to the separation processes (Davis, 2002).

The different reactor types discussed above for the FT synthesis reaction all seem to have limitations and advantages. Therefore, there is no universal optimum FT reactor; the choice rather depends on the target product and the process conditions. According to different modelling studies in literature (de Swart et al., 1997; Iglesia et al., 1991), slurry reactors are more suitable for the FT synthesis and result in up to 60% lower capital costs. Shell, on the other hand, operates a multitubular fixed bed reactor and claims that the superior performance of the Shell catalyst invalidates most of the slurry reactor advantages, rendering the fixed bed technology competitive with the current slurry technology (Geerlings et al., 1999). Therefore, FT reaction selection should be based on process conditions and products, aiming at achieving optimized reactor/catalyst combination, based on the physico-chemical characteristics and activity performance of each type of catalyst.

Much research, development and demonstration (RD&D) worldwide focuses on ways to produce advanced biofuels via thermo-chemical conversion of biomass feedstock. Making H2 and CO (syngas) from biomass is a crucial step in the production of most thermo-chemically derived advanced biofuels, for example Fischer-Tropsch (FT) diesel and methanol (Drift et al., 2006). It is however not always necessary to convert all biomass into syngas; production of a gas containing H2 and CO, as well as several hydrocarbons (product gas), can also be sufficient (e. g. SNG).

As the approach towards advanced biofuels like FT-diesel and methanol is different from that towards SNG, both biomass value chains are discussed separately. This, however, does not mean that there are no synergies between them. In fact, combined production of both types of fuel can be an interesting alternative, in particular when SNG is one of the advanced fuels considered (Zwart and Boerrigter, 2005).

Biomass-to-liquids (FT-diesel, methanol, DME, mixed alcohols) from lignocellulosic biomass

For optimal synthesis of the advanced biofuels FT-diesel, methanol, dimethylether (DME), mixed alcohols and even pure H2, a cleaned and conditioned bio-syngas is required. There are two major approaches in gasification to convert biomass into such a bio-syngas (Drift et al., 2006): (1) Fluidised bed gasification with subsequent catalytic reforming, both at 900°C, and (2) Entrained flow gasification at approximately 1300°C with extensive pre-treatment. Both approaches require extensive syngas cleaning and conditioning, as shown in Fig. 21.3.

The fluidised bed gasification approach has the advantage that the gasification technology has been developed and already demonstrated with biomass for the production of heat and/or power. RD&D therefore mainly focuses on downstream catalytic reforming. The scale of implementation is, however, limited with the largest existing biomass gasification plants close to 100 MWth.

Entrained flow gasification processes have already been developed and demonstrated on much large-scale for coal, e. g. the 600 MWth Buggenum IGCC of NUON in The Netherlands. Biomass feedstock, however, needs to be pre-treated in

order to take full advantage of the coal-based technologies. RD&D therefore focuses on pre-treatment such as flash pyrolysis for the production of a high energy density slurry and torrefaction for the production of a ‘bio-coal’, and on advanced process integration in order to increase efficiency and reduce overall costs. This last option includes the combination of black liquor gasification, biofuel production and chemical recovery in a pulp and paper mill (Landalv, 2006).

The scale of the gasifier is an important issue for most biomass-to-liquids plants. The fuel synthesis in general needs to be as large as possible because of the dominant economy-of-scale effect in biofuel synthesis and upgrading. However, a potential increase of the scale of the gasifier will result in higher feedstock costs due to higher transportation costs, even when considering pre-treatment and densification of the biomass before transportation (Zwart et al., 2006a).

As raw bio-syngas resembles syngas produced from conventional fuels like coal and oil residues, the syngas cleaning will in most cases exist of relatively conventional systems based on filters, a Rectisol unit, and gas polishing by e. g. ZnO and active carbon filters. Also generally included will be a water gas shift reactor, providing the H2/CO ratio desired for the different end-products.

The example process chosen for illustration is transesterification of rapeseed oil to produce rapeseed oil methylester (RME) as the main product and glycerine as a co-product. Therefore, the environmental impacts have to be allocated between these two products, using an appropriate allocation basis. Two allocation approaches are considered here — mass and energy basis — to illustrate the difference in the approach as well as any difference in the LCA results. GHG emissions are used for illustration purposes. This example is based on that found in Fehrenbach et al. (2007).

The process produces 1 kg of RME and 0.092 kg of glycerine. The total GHG emissions from the process are 0.307 kg CO2 eq. The lower heating value (LHV) of RME is 37.2 MJ/kg and of glycerine is 17 MJ/kg.

[2] Mass-based allocation:

Total mass of products in the process: 1 kg RME + 0.092 kg glycerine = 1.092 kg Allocation factor:

RME: 1 kg/1.092 kg x 100 = 91.6%

Glycerine: 0.092 kg/1.092 x 100 = 8.4%

Most LCA studies of biofuels focus on GHG emissions and energy balances. However, as already discussed, biofuels have wider environmental impacts, including resource depletion, biodiversity, acidification, eutrophication and toxicity. These have rarely been considered in LCA studies to date.

As an illustration, Fig. 3.6 compares the selected environmental impacts from bioethanol (from wheat) and petrol. Global warming potential is also shown.

3.6 Environmental impacts of petrol and bioethanol.

Note: Bioethanol is from wheat cultivated in Germany; comparison is made on the basis of the equivalent energy content in petrol and bioethanol; the unit of analysis is 1 litre of petrol and 1.6 litres of bioethanol (due to the lower energy content of bioethanol compared to petrol); the impacts are expressed per litre for petrol, per 1.6 litre for bioethanol; system boundary is from ‘cradle to grave’ (use stage not included except for GWP); DCB — dichlorobenzene.

These results illustrate that, while biofuels can provide GHG savings, their wider impacts can be higher than that of conventional fossil fuels. For the example considered here, bioethanol has three times higher acidification and 27 times higher eutrophication than petrol (note that the use stage of fuels is not considered). These are mainly due to the use of fertilisers and fuel in the agricultural machinery. Its terrestrial toxicity is 1.6 times higher, again mainly due to the assumed use of synthetic fertilisers.

From this and the earlier discussion, it is clear that evaluation of environmental sustainability of biofuels should involve consideration of all relevant environmental impacts along the whole life cycle to avoid shifting the burdens and making unsustainable choices.

We now turn our attention to the economic sustainability of biofuels.

R. VERHb and C. ECHIM, Ghent University, Belgium, W. DE GREYT, Desmet Ballestra Group, Belgium and C. STEVENS, Ghent University, Belgium

Abstract: The objective of this chapter is to provide an overview of the conversion of various lipid sources (edible and non-edible) into biodiesel using traditional and new technologies emphasizing the quality standards mainly dependent on the used feedstock and technology, processing and purification.

Key words: heterogeneous catalysis, homogeneous catalysis, influence of the feedstock and technology on biodiesel properties, purification of biodiesel, industrial production of biodiesel.

5.1 Introduction

Renewable resources and biomass are becoming major raw materials for the energy supply. Vegetable oils and animal fats are considered as sources for the green energy supply. Therefore, the demand of lipids for food (80%), feed (5%) and industrial applications (15%) such as detergents, surfactants, biolubrificants and biofuels is leading to a shortage and higher prices. During the last few years, the production of biodiesel from edible lipids has been blamed for raising the cost of food products.

Traditionally, fully refined edible oils and fats were used for biodiesel production, namely rapeseed oil in Europe, soybean oil in North and South America and palm oil in Southeast Asia. Since nearly 85% of the total production cost is originating from the feedstock cost, new starting materials, not competing with local supply, have to be explored.

The objective of this chapter is to provide an overview of the conversion of various lipid sources (edible and non-edible) into biodiesel using traditional and new technologies emphasizing the quality standards mainly dependent on the used feedstock and technology, processing and purification.

Excellent books and reviews on the production of biodiesel have been published mainly emphasizing the traditional processing (Knothe and Dunn, 2001; Verhe 2004; Knothe et al., 2005; Mittelbach and Remschmidt, 2004; Mittelbach 2009; Demirbas 2009).

The use of vegetable oils, animal fats and their derivatives as a diesel fuel is nearly 100 years old. The inventor of the diesel engine, Rudolf Diesel, used peanut oil for the engine and he stated: ‘vegetable oils make it certain that motor power can still be produced from the heat of the sun, which is always available for agricultural purposes, even when all our natural stores solid and liquid fuels are exhausted’ (Diesel, 1912). However, the use of heat in vegetable oils due to poor atomization upon injection was leading to deposits in the injection system and the cylinders, causing operational problems.

In order to reduce the viscosity, Chavanne (1937, 1943) converted a mixture of fatty acids and glycerol esters from palm oil into esters in the presence of mainly ethyl alcohol and acid catalysts. These experiments were performed in colonial Africa in order to be independent from external sources of fuels. In fact, Chavanne produced a fuel which we should call now biodiesel. In Europe, the first experiments with ethyl ester of palm oil were carried out in buses plying between Leuven and Brussels in 1938.

The word ‘biodiesel’ was used for the first time in 1988 (Wang) and expanded from 1991 (Bailer and De Hueber). The first plant for biodiesel was installed in 1987 in Silverberg, Austria mainly for the purpose of research at the University of Graz, by Prof. Mittelbach.

From 1990 the production of biodiesel in Europe has increased dramatically with a capacity in 2009 of nearly nine million tons per year (Fig. 5.1).

A. MACARIO and G. GIORDANO, University of Calabria, Italy, F. M. BAUTISTA, D. LUNA, R. LUQUE and A. A. ROMERO, University of Cordoba, Spain

Abstract: In this chapter, the characteristics of novel type of biofuels integrating glycerol into their composition are described. The advantages of using biofuels integrating glycerol (Ecodiesel®, DMC-Biod®, Gliperol®) and the respective technologies to produce them are reported. In addition, the production of high-quality diesel fuel from vegetable oils by hydrotreating of triglycerides in conventional oil refineries is, also, reported.

Key words: Ecodiesel, DMC-Biod, Gliperol, hydrotreating of triglycerides.

7.1 Introduction

The soaring oil price has drastically increased the demand of fuels from renewable and biological sources. Consequently, the world research efforts are devoted to the study of new processes to efficiently produce these novel fuels. Current industrial production of biodiesel (‘mono alkyl esters of long chain fatty acids derived from renewable lipid feedstock, such as vegetable oils or animal fats, for use in compression diesel engines’ — ASTM definition1) is carried out by homogeneous alkali-catalyzed transesterification of vegetable oils with methanol, using sodium hydroxide, potassium hydroxide or potassium methoxide as catalysts.2 The homogeneous basic transesterification reaction shows a very fast kinetic rate, but unfortunately, there are, also, several environmental and economic problems associated with the process. A collateral saponification reaction takes place, reducing the biodiesel production efficiency. To prevent the biodiesel yield loss due to the saponification reaction, oil and alcohol must be dry and the oil should have a minimum amount of free fatty acids (FFAs) (less than 0.1% wt.). Biodiesel is finally recovered by repeated washing with water to remove glycerol, soap and excess of methanol.

In contrast, the acid transesterification allows to obtain a biodiesel production without formation of by-products. Drawbacks of an acid homogeneous transesterification include the use of corrosive catalysts (H2SO4, H3PO4, HCl) and slow reaction rates. These may be increased at high temperatures and pressures, involving larger costs.3 Methanol and oil are poorly soluble, so the reaction mixture contains two liquid phases. Other alcohols can be used, but they are generally more expensive. Moreover, an acid pre-treatment is often needed in the homogeneous alkaline transesterification for oils having more than a 5 wt.% of FFAs in order to improve the biodiesel efficiency production.2-4

In any case (either in acid or basic catalysis), the process is far from being environmentally friendly, since the final mixture needs to be separated, neutralized and thoroughly washed, generating a great amount of waste in terms of salt residues. Moreover, the catalyst also cannot be recycled. These several additional steps inevitably put the total overall biodiesel production costs up, reducing at the same time the quality of the glycerol obtained as a by-product.5 Several reports can be recently found on the production of biodiesel involving other chemical6,7 or enzymatic catalytic protocols as greener alternatives.8,9 The increasing environmental concerns have led to a growing interest in the use of enzyme catalysis, as these biocatalysts normally produce a cleaner biodiesel under milder conditions.10 It also generates less waste than the conventional chemical process. Recent work demonstrates that heterogeneous enzymatic catalysts represent a potential solution to produce biodiesel from very low-quality triglycerides (TGs) feedstocks,11-14 but in these cases, the cost of the enzymes has to be considered. The true limitation of the enzymatic method compared to the conventional base-catalyzed process deals with the alcoholysis of the 2-fatty acid esters of glycerol. Lipases have a peculiar 1,3-regioselectivity, which means that they selectively hydrolyze the more reactive 1 and 3 positions in the triglyceride.15 In this regard, the production of biodiesel using lipases needs to take into account such regiospecific character.16,17 In general, the challenging full alcoholysis of TGs involves long reaction times and gives conversions lower than 70 wt.% to fatty acid methyl or ethyl esters.18,19

A series of improvements in conversion levels and/or the use of methanol as alcohol to mimic the results of the base-catalyzed transesterification reaction are currently ongoing as a consequence of the present legal regulations for biodiesel (EN 14214). The current standard biodiesel production (under alkaline chemical conditions) is considered to be the most technically simple way to reduce the viscosity of vegetable oils from a range of 11-17 times20-22 to just about twice of that of petroleum diesel. Various fuel properties of pure soybean oil, three B100 biodiesel types (soybean methyl esters, rapeseed methyl esters and rapeseed ethyl esters) and high-grade petro-diesel are summarized in Table 7.1.

The viscosity is the only significant parameter that may affect the performance of the diesel engine, as the other parameters are very similar. Interestingly, diglycerides (DGs) and TGs are mainly responsible for the increase in viscosity of pure vegetable oils. A novel biofuel containing fatty acid methyl esters/ monoglyceride (FAMEs/MG) or fatty acid ethyl esters/monoglyceride (FAEEs/ MG) blend (in which we exclude the presence of significant quantities of DGs and TGs) can be expected to have similar physical properties to those of conventional biodiesel, eliminating the production of glycerol as a by-product. The achievement of glycerol-free biofuels could be most convenient and advantageous in a market flooded by the production of glycerol as a by-product23-27

in the preparation of biodiesel. The aim of this chapter is to describe the characteristics and preparation of these novel types of biofuels integrating glycerol into their composition and the advantages of their use.

10.3.1 ABE fermentation history

Industrial ABE fermentation started almost 100 years ago in 1913. At that time, the rapidly growing automobile industry required high amounts of rubber for tires. Since butanol could be used as a precursor of butadiene (the starting material for the synthetic rubber production), the British company Strange and Graham, Ltd. launched a project to investigate microbial butanol formation. They employed William Perkins and Charles Weizmann from Manchester University and Auguste Fernbach and Moi’se Schoen from Institute Pasteur in Paris. Soon after, Fernbach isolated a respective strain and Strange and Graham, Ltd. started ABE fermentation at a plant in Rainham, UK, and later on in King’s Lynn, UK (Fernbach and Strange, 1911a, 1911b, 1912).

However, due to increased production from Asian plantations, rubber prices dropped again, but the outbreak of World War I led to a sudden demand for acetone (as a solvent for the production of the smokeless propellant cordite). Therefore, the British government contracted Strange and Graham, Ltd., which produced an average of 970 pounds acetone per week (Gabriel, 1928). Weizmann, who meanwhile had left the group, succeeded with the isolation of C. acetobutylicum (see Section 10.2). Originally, he had planned to publish his results as scientific contribution, but when he realized that his discoveries might be helpful to the empire, he changed his mind and applied for a patent (Weizmann, 1915). Since Weizmann’s results were so promising, the British government decided to adapt six distilleries for the Weizmann process and also requested Strange and Graham, Ltd. to switch to the Weizmann process. Consequently, their production increased to over 2200 pounds acetone per week (Gabriel, 1928). However, due to the German submarine offensive in the Atlantic, maize and grain, which were needed as feedstock for the Weizmann process, could not be imported in required quantities anymore. Therefore, horse chestnuts collected by school children were used as alternative feedstock (Hastings, 1978; Imperial War Museum Collections, 2009), and the British government decided to build new plants in Canada and India, where sufficient raw materials were available. While the plant in India was never completed, a plant in Toronto began operation in 1916 and another plant was constructed in Terre Haute, Indiana, after the United States entered the war (Gabriel, 1928). This secured the constant supply of acetone and was a decisive factor for the allied victory. To express their gratefulness, the British government wanted to honor Weizmann. However, he rejected any acknowledgments, but as a Zionist addressed the issue of a Jewish homeland in Palestine. When the state of Israel was finally founded in 1948, Weizmann became its first president.

After peace was established, no more acetone was needed and all plants were shut down. However, some effort was made to salvage the ‘useless’ butanol that had accumulated during the war and had simply been stored in large tanks. While some butanol was converted to methyl ethyl ketone (MEK), it was found that butanol and its ester butyl acetate can also be used directly as a replacement of fusel oil and amyl acetate, respectively (Gabriel, 1928; Killeffer, 1927). The latter was needed in large quantities by the booming automobile industry (as solvent for lacquers) and had so far been produced from amyl alcohol. However, since amyl alcohol is obtained as a side product of the ethanol fermentation, it became unavailable when the prohibition was introduced in the United States in 1919. Butanol fermentation became lucrative again, and the newly founded Commercial Solvents Corporation of Maryland (CSC) acquired the rights to the Weizmann process as well as the Terre Haute plant from the Allied War Board in the very same year. However, the general business slump of 1920 forced a nine-month shutdown and a bacteriophage infection (see Section 10.5) in 1923 cut yields in half for almost a year (Gabriel, 1928; Jones et al, 1986). Nevertheless, the capacity of the Terre Haute plant was increased from 40 to 52 fermenters and a new plant was built in Peoria, Illinois, consisting of thirty-two 50 000-gallon fermenters and eventually enlarged to 96 fermenters (Gabriel, 1928). Strange and Graham also tried to get back into business but went into liquidation after a lost lawsuit against CSC (Ross, 1961). When the Weizmann patent finally expired in 1936, new plants were opened in Philadelphia, Pennsylvania, and Baltimore, Maryland, the UK, the former Soviet Union, Brazil, Australia, South Africa, Puerto Rico, Egypt, Japan, and the former Japanese colony Formosa (today’s Taiwan) (Jones and Woods, 1986; McCutchan and Hickey, 1954). At that time, butanol was predominately produced biologically by ABE fermentation (Durre, 2005a). During World War II, the focus shifted to acetone production again, and semi-continuous fermentations and continuous distillation methods were successfully used for the first time (Hastings, 1971, 1978).

However, a few years after the end of the war, the petrochemical industry overcame the ABE fermentation due to rising substrate prices (Durre, 2005a). Thus, the process was only continued in politically isolated countries. In apartheid South Africa, the National Chemical Products (NCP) company was operating an ABE fermentation plant in Germiston with twelve 90 000-l fermenters until 1983 (Jones, 2001; Jones and Woods, 1986). Their process relied on batch fermentations with molasses as substrate and showed remarkable efficiency and reliability. Each run had a length of around 30 hours and yielded a solvent concentration of 17-19 g/l (with an acetone ratio of 32-36%) from 5.5-7% sugar (Jones, 2001; Jones and Woods, 1986). Later on, it was shown that the NCP production strains belonged to two species, C. beijerinckii and C. saccharobutylicum (Jones, 2001; Keis et al, 2001). In the former Soviet Union, ABE fermentation was also carried out until the late 1980s in at least eight plants (the biggest in Dokshukino). The initial process was similar to the one from Weizmann, utilizing C. acetobutylicum in starch-based batch fermentations (Jones, 2001; Zverlov et al, 2006). However, over the years, the process was developed to continuous mode with molasses and wheat or rye flour, but also agricultural waste hydrolysates as substrate. An average fermentation run yielded 10 g/l of butanol, 6.4 g/l of acetone and 1.5 g/l of ethanol from 4.7% sugar equivalents (mixtures of starch, maltodextrines, sucrose and pentoses). For the Evremovo plant, a yearly production of 15 000-ton solvents (8550 tons butanol, 4140 tons acetone and 2310 tons ethanol) was calculated (Zverlov et al., 2006). While most other countries shut down their ABE fermentation plants, the People’s Republic of China opened its first plant in 1950 at Shanghai (Chiao and Sun, 2007). In the following years, the Chinese ABE fermentation industry expanded rapidly. At its peak during the 1980s, more than 30 plants were operating, producing around 170 000 tons of solvents per year. Most plants were using starchy materials such as corn, cassava, potato or sweet potato as feedstock, whereas some plants also relied on molasses as substrate, utilizing different isolates of C. acetobutylicum. The process started as batch fermentation but was later developed into a continuous process up to ca. 200 hours (Chiao and Sun, 2007). However, during the 1990s, the number of ABE fermentation plants decreased, and in 2004, the last plant was closed.

Only three years later, China resumed production at 11 plants with a total capacity of 410 000-ton solvents per year, which is expected to be extended to over 1 000 000 tons soon (Ni and Sun, 2009). Recently, a new plant was opened in Brazil too (Jones, 2008). Furthermore, several companies such as Butalco, Butamax™ (a joint venture of BP and DuPont), ButylFuel, Cathay Industrial Biotech, Cobalt Biofuels, Gevo, Green Biologics, METabolic EXplorer or Tetravitae Bioscience compete to commercialize ABE fermentation on a global basis again. While Cathay Industrial Biotech is already running a 30 000-ton solvents per year plant in Jilin, China (Ni and Sun, 2009), Gevo operates a one million gallon butanol per year demonstration plant in St. Joseph, Missouri (Gevo, 2009) and Butamax™ is constructing a pilot plant at Wissington, UK, to produce an annual 30 000-ton butanol from 2010 onwards (BP and DuPont, 2006; Butamax™, 2009).

Biological water gas shift reaction is a new concept for hydrogen production. Certain photo-heterotrophic bacteria such as Rhodospirillum rubrum and Rubrivivax gelatinosus can perform this reaction at ambient temperature and pressure. These bacteria can survive in the dark using CO as the sole carbon source, oxidizing it to CO2 and reducing H+ to H2, according to the reaction:

CO + H2O « CO2 + H2, AG0 = -20,1 kJ/mol [13.5]

The thermodynamics of this reaction are very favorable to CO-oxidation and H2 synthesis, since the equilibrium of this reaction lies strongly to the right. The purple non-sulfur bacteria perform CO-water gas shift reaction in darkness, converting 100% of CO into a near-stoichiometric amount of hydrogen and also assimilate CO into new cell mass in the light, when CO is the sole source of carbon (Maness and Weaver, 1997). They are also able to utilize CO in the presence of other organic substrates.

The need to reduce residual CO to very low levels makes the mass transport of gaseous CO into an aqueous bacterial suspension the rate limiting step in the process and is the main challenge for bioreactor design. This suggests the need for counter-current gas-liquid contacting systems, as in trickling filters used in waste treatment or plug-flow systems typical in commercial gas biofiltration processes (Andrews and Noah, 1995).

However, hydrogen production through biological water-gas shift reaction is still at laboratory scale and thus, intensive research, including scale-up, should be done in order to become an applicable technology. Genetic strain improvements and identification of suitable microorganisms that have high CO uptake ability are strategies that should further increase the obtained hydrogen rates and yields. Process economics are presently uncertain since they depend on the required size of the bioreactors and the losses inherent in such ambient pressure-temperature conversion process. Wolfrum et al. (2003) have conducted a detailed study to compare the biological water-gas shift reaction with the conventional water-gas shift process. Their analysis showed that the cost of biological water-gas shift process is lower due to the elimination of the need for a reformer and the expensive equipment required for the thermochemical process. Thus, the microbial water — gas shift reaction may be a good candidate for near-term biohydrogen process development. Indeed, some authors consider this process as the most promising for commercial application.

Human activity requires considerable amounts of energy distributed more or less evenly between three types of activity: industrial, residential and transportation. This is a typical 20th century development that resulted from the growth of road transportation. Until the end of the 19th century land transportation was based essentially on the use of horses and represented less than 10% of all energy consumption. Biomass in the form of hay for feeding horses has been the main ‘fuel’ sustaining transportation.

The development of internal combustion engines by Diesel and Otto, approximately 100 years ago was based on the use of biofuels, but with discovery of abundant petroleum reserves there was a dramatic shift in fuels. Transportation now accounts for more than 30% of all energy used in the world and consumes around 83 million barrels of oil per day.

The 20th century saw an explosion in the use of automobiles for personal use and trucks for the transportation of goods. There are already more than 600 million automobiles in the world and the number is increasing steadily since the use of the automobile is not only very convenient, but it is also intimately associated with our cultural values. In the United States there are almost 800 automobiles per 1000 people; in China and India this number is 10 times less but quantities are increasing rapidly.

Unfortunately this is a situation that cannot last for very long because the fuels used for present modes of transportation are almost exclusively from petroleum, of which remaining reserves are being depleted rapidly. In addition to that, such fuels are the main source of environmental problems ranging from bad air quality in large cities to regional pollution and the increase of the concentration of greenhouse gases in the atmosphere.

It is therefore urgent to find fuels that could replace petroleum products or develop other methods of propulsion if one wishes to preserve individual transportation.

Of the several technologies in development for that purpose only electrical motors and biofuels seem to be promising solutions. Electrical motors using batteries, where the energy is stored or produced by fuel cells, are making limited progress and in any case will require large additional amounts of electricity to be produced mainly from fossil fuels which are not renewable. One tends to forget

xxiv Foreword that an automobile usually requires 30 kilowatts of power which mean 18 billion kilowatts for an entire fleet. For comparison the total installed capacity for electricity generation in the world is around 4 billion kilowatts.

Biofuels are therefore the more promising option: they are renewable, contribute little to the production of greenhouse gases and do not have the impurities that petroleum derived fuels have. Biofuels already represent a small percentage of the transportation fuels in the world. The ‘automobile age’, which started with biofuels, seems to be returning to its origins.

The Handbook of biofuels production provides a comprehensive discussion of all the aspects of the problem ranging from the feedstocks and production chain to chemical and biochemical production as well as the thermal and thermo-chemical conversion process. Sustainability assessment and policies surrounding the issue are also discussed.

Professor Jose Goldemberg University of Sao Paulo Sao Paulo, Brazil

Soybean (Glycine max Merr.) oil is used as both edible oil and transportation fuel (Fig. 4.3). However, oxidative instability and cold flow in northern climates have

limited usefulness of a soybean oil-derived biodiesel. Implementing the tools of biotechnology to modify the fatty acid profile of soybean for locale performance enhancement may increase the attractiveness of biodiesel derived from this commodity crop (Kinney and Clemente, 2005).

More than 20 years ago, researchers demonstrated the feasibility of the production of biodiesel from soybean oil using methanol in the presence of homogeneous catalysts (Freedman et al., 1986). Since then, research has been conducted under supercritical and subcritical methanol (Wang and Yang, 2007) and with heterogeneous catalysts (Xie and Huang, 2006; Liang et al., 2009). Diesel engines have been run on soybean oil biodiesel, straight or 20% blended with diesel fuel, showing an increase in BSFC up to 18% and 2.5% respectively, compared to the use of diesel fuel. It has been found that the oil origin has no influence over BSFC (Canakci and Van Gerpen, 2001; Hess et al, 2007). Furthermore, in the 1950s, 20% waste soybean oil blended with 80% diesel fuel was successfully used to run the University bus at Ohio State University (Fishinger, 1980).