The method of production of biodiesel using lipase as catalyst has not yet been implemented in industrial scale due to certain constrains like high cost of enzyme,
exhaustion of enzyme activity and enzyme inhibition by methanol. Enzymes such as proteases and carbohydrases have been used industrially for a number of years and corner the largest share of the worldwide enzyme market. While lipases at present account for less than 5% of the market, this share has the potential to increase dramatically via a wide range of different applications. The higher production costs of industrial lipase as compared with proteases and carbohydrases seem to be the main obstacle that hampers its wider industrial application. In order to overcome this limitation, lipase has to be repeatedly used, which is achieved by using it in immobilized form. Details of incentives of lipase immobilization are explained in Section 6.8.1. However, when lipase is used in immobilized form, another problem arises. The deposit of the by-product glycerol coating the immobilized lipase is formed during the process due to the low solubility of glycerol in biodiesel, which competitively inhibits the enzyme and reduces its activity by blocking the active sites (Dossat et al., 1999; Du et al., 2004; Al-Zuhair et al., 2008).

Another hindrance of biodiesel production by lipase is the inhibition of the enzyme by methanol. The effect of alcohol, specifically methanol, on the enzymatic production of biodiesel has been thoroughly discussed in literature. While it is a reactant, it also inhibits the enzyme. It has been found that biodiesel production increases with increasing methanol concentration up to oil to methanol ratio of 3:1 and then decreases when methanol concentration is further increased (Shimada et al., 1999; Al-Zuhair et al, 2007; Al-Zuhair et al, 2008). This was also found by Noureddini et al (2005), although the ratio was higher (7.5:1). In general, it is widely accepted that methanol which is completely dissolved in the substrate mixture does not inactivate the lipases (Shimada et al., 1999; Shimada et al., 2002; Al-Zuhair et al., 2007). Lipases, however, are inactivated by contact with insoluble methanol that exists as drops in the oil; thereby the catalytic activity of the transesterification reaction is decreased. The deactivation of lipase with contact with insoluble methanol is due to the strong polarity of the latter, which tends to strip the active water from the active sites of the enzyme (Lara and Park, 2004). The inhibitory effect of methanol is large at the beginning of the reaction, but with increasing oil conversion it decreases because it is consumed in the reaction and hence its concentration decreases, in addition its solubility is higher in the product methyl ester than in the triglyceride (Shimada et al., 1999). On the other hand, the inhibition due to the blocking of the active sites of the catalyst by glycerol is absent at the beginning of the reaction and becomes larger at higher oil conversions.

Lipase is also sensitive towards the water contents. It has been reported that up to 500 ppm water in reaction mixture decreased the rate of methanolysis; however the equilibrium of the reaction was not affected (Shimada et al., 1999). The effect of water content on the production of biodiesel from soybean oil using lipases from R. Oryzae (Kaieda et al., 1999), C. rugosa and P. Fluorescens (Kaieda et al., 2001), Novozym 435 (Shimada et al., 1999) and Burkholderia cepacia (Noureddini et al., 2005) have all shown that enzyme activity was low in absence of water; with the addition of water a considerable increase in lipase activity was observed, which is explained by the unique property of interfacial activation of lipase (Verger et al., 1973; Brady et al., 1990). The activity of lipases is low in monomeric solutions of lipid substrates but a configuration change and activity enhances strongly at the water-lipid interface. Activation of the enzyme involves unmasking and restructuring of the active site through conformational changes of the lipase molecule, which requires the presence of oil-water interface. An experimental approach to determine the activation of the lipase at the interface, proposed by Rooney and Weartherley (2001), was used by Al-Zuhair et al. (2003) to determine that the activity of lipase from C. rugusa at the oil interface, and was found to be 15.7% higher than that in the bulk. With the increased addition of water, the amount of water available for oil to form oil-water droplets increases, thereby, increasing the available interfacial area. However, excess water stimulates the competing hydrolysis reaction, since lipases usually catalyze hydrolysis in aqueous media. The optimum water content is a compromise between minimizing hydrolysis and maximizing enzyme activity for the transesterification reaction. The range of water content at which the enzyme maintains its methanolysis activity varies significantly from one type of lipase to another. For example, the activity of Novozym 435 significantly drops at water contents higher than only

O. 1% (Shimada et al., 1999), whereas lipase from R. meihei maintains its methanolysis activity at water contents of up to 20% (Al-Zuhair et al., 2006; Tweddell et al., 1998).

The various biomass feedstocks that can be used for bioethanol production are divided into two major groups: first generation feedstocks and second (next) generation feedstocks. First generation feedstocks include sugar, sugar cane, sugar beet and starch crops like corn, wheat and barley. To the next generation feedstocks belong wood, grasses, forestry residues and other lignocellulosic materials as new, more sophisticated conversion technologies are developed to enable the production of bioethanol from cellulosic feedstocks. However, feedstock availability for ethanol production can be limited in some countries with low biomass resources, e. g. woody biomass resources in Finland and Sweden are huge however the woody biomass has been used in many ways such as fuel pellet production for combustion and electricity production and lignocellulosic materials have been used for many years in paper mills. In fact, the price of the feedstock is about one-third the cost of bioethanol production (Balat, Balat, and Oz, 2008) depending on feedstock. In addition, there are other aspects of feedstock production that should be considered such as national and international regulations and policies, environmental questions, protection of high-value habitats and competition between food production and biofuel feedstock. It is also important to be able to determine the chemical composition of a feedstock (i. e. sugar units, extractives, lignin, etc.) by fast and non-destructive methods such as near infrared spectroscopy (NIR; Sanderson, Agblevor, Collins, and Johns, 1996).

There are three different groups of feedstocks available for ethanol production: sugar feedstock such as sugarcane and sugar beet; starch feedstock such as cereal grains and potatoes; and cellulose feedstock such as forest products and agricultural residues. In general, the sucrose-containing materials such as sugarcane allow the production of ethanol for the lowest costs compared to the starchy materials and lignocellulosic feedstocks.

Anaerobic digestion is a complex bioprocess consisting of successive, often interactive steps carried out by groups of microorganisms with different growth rates and sensitivity to environmental conditions (pH, partial pressure of hydrogen, etc.). The process can be outlined as consisting of the following steps (Fig. 12.1):

• Disintegration: The complex particulate waste disintegrates to organic polymers such as carbohydrates, proteins and lipids. Disintegration lumps a number of steps such as lysis, non enzymatic decay, phase separation and physical breakdown (e. g. shearing; Batstone et al., 2002).

• Hydrolysis: The organic polymers (carbohydrates, proteins and fats) are hydrolysed (depolymerised) by extracellular enzymes to their respective monomers (sugars, amino acids, lipids), which can be taken up by the microorganisms for further degradation. In the case of particulate complex organic matter consisting of lignocellulosic material (mainly of plant origin),

pretreatment steps are necessary to enhance hydrolysis by rendering the substrate matrix more amenable to enzyme attack.

• Acidogenesis: A versatile group of microorganisms are able to convert the simple monomers to a mixture of volatile fatty acids, alcohols and other simpler organic compounds. This step is also often called fermentation. During acidogenesis, large amounts of carbon dioxide are produced as well as hydrogen. Especially in the case of sugars fermentation, the amount of hydrogen produced can be high and may be harvested for energy recovery. The growth rate of acidogens is quite high (doubling time of the order of one hour or even less) and low pH resistant (5-6) giving them the advantage of prevailing in the anaerobic consortium at adverse conditions. As a result of the rapid acid formation, there is a danger of acid accumulation (and concomitant pH drop) if the acids are not degraded in time in the steps that follow.

• Acetogenesis: The higher volatile fatty acids (propionate, butyrate, valerate, etc.) as well as the other organic molecules produced in the acidogenesis step are transformed to acetic acid, carbon dioxide and hydrogen by the acetogenic bacteria. This step is thermodynamically inhibited by hydrogen, meaning that, unless hydrogen is depleted by the hydrogen-consuming bacteria in other steps, there is an accumulation of mainly propionic and butyric acid. The acetogenic bacteria are slow growing microorganisms doubling time of the order of days.

• Methanogenesis: There are two distinct groups of microorganisms that produce methane and carbon dioxide: (1) the acetoclastic methanogens that grow on acetic acid and produce approximately 70% of methane in the biogas, and (2) the hydrogen utilising methanogens that consume hydrogen and carbon dioxide. The methane content of biogas depends on the oxidation state of the organic carbon in the initial substrate (ranging from -4 for methane to +4 for carbon dioxide); the more reduced the initial substrate is, the more methane will be produced, but on average the biogas contains 60% of methane. Acetoclastic methanogens are slow growing microorganisms (doubling time of the order of days) and are particularly sensitive to a number of factors such as pH, lack of nutrients, certain compounds, etc.

A. AZAPAGIC, The University of Manchester, UK and

H. STICHNOTHE, Johann Heinrich von Thunen Institut — Institute of Agricultural Technology and Biosystems Engineering, Germany

Abstract: Biofuels have a potential to reduce carbon dioxide emissions from transport because the biomass used in their production is considered carbon neutral. This is the main reason for a growing interest in biofuels. However, there are certain aspects, particularly of the first-generation biofuels, which may render them unsustainable, including the increased use of land and competition with food production. Therefore, sustainability of biofuels should be assessed carefully, considering all relevant environmental, economic and social aspects. To prevent shifting the impacts along the supply chains, sustainability should be assessed considering the whole life cycle of biofuels, including cultivation of the feedstock and biofuel production processes. This chapter reviews various sustainability aspects of biofuels and illustrates how environmental and economic sustainability can be assessed on a life cycle basis. The environmental impacts considered include water use, global warming, acidification, eutrophication and loss of biodiversity while economic aspects include feedstock costs, capital costs and biofuel prices. Future viability of biofuels is also discussed.

Key words: biofuels, environmental impacts, economic costs, life cycle assessment, sustainability assessment.

3.1 Introduction

Biofuels can be produced from a range of biomass sources using different production routes, as discussed throughout this book. Depending on the type of the bio-feedstock used, they are referred to as first-, second — or third-generation biofuels (OECD and IEA, 2008). First-generation biofuels are produced commercially from conventional food crops, including wheat, maize, corn, sugar cane, rapeseed, sunflower seeds and palm oil. The most common first-generation biofuels are bioethanol, biodiesel, vegetable oil and biogas.

Second-generation biofuels are produced from non-food sources and include dedicated energy crops (e. g. perennial grasses, short-rotation coppice willow and other lignocellulosic plants) and waste biomass (e. g. agricultural, forestry and municipal solid waste). Two main processing routes used to produce these fuels are: thermo-chemical and bio-chemical. The former is used mainly for the production of biodiesel and the latter for bioethanol. Other second-generation fuels under development include: biohydrogen, biomethanol, dimethylfuran (DMF), bio-dimethylether (bio-DME), Fischer-Tropsch diesel, biohydrogen diesel and mixed alcohols (Brigenzu et al., 2009).

Third-generation biofuels are still under development and the main bio­feedstock being considered are algae for the production of biodiesel via the thermo-chemical route. Other sources of third-generation biofuels could include alcohols such as bio-propanol or bio-butanol; however, they are not expected to enter the market before 2050 (OECD and IEA, 2008).

Currently, the majority of the global biofuel production is from food crops with bioethanol representing over 80% of liquid biofuels by energy content (Brigenzu et al, 2009); however, the importance of the second — and third-generation fuels is growing.

Biofuels have a potential to reduce the carbon dioxide (CO2) emissions because the biomass used in their production is considered carbon neutral. This is based on the assumption that the amount of carbon released during combustion of biofuels in the use phase is equivalent to the amount of carbon sequestered during the growth of biomass from which the fuels were derived. Further attractive features of biofuels over fossil fuels are that they provide security of supply as they can be produced domestically by many countries. Furthermore, they require only minimal changes in the distribution system and production technologies. Biofuels also have a potential to stimulate rural development (Rajagopal and Zilberman,

2007) . Thus, the expectations from biofuels as a source of ‘sustainable’ energy are high.

However, there are certain aspects, particularly of the first-generation biofuels, which may render them less sustainable. For example, while the intensification of agriculture to increase crop production per land unit may lead to lower greenhouse gas (GHG) emissions per unit of product, the increased use of land, energy, fertilisers and pesticides will reduce the net GHG benefits and cause further environmental damage, including release of soil carbon, leaching of nutrients and loss of biodiversity. Other risks associated with large-scale production of the first — generation biofuels include competition with food production, leading to increased costs of food and in some cases, food poverty (Bird et al., 2008; Escobar et al., 2009; Fargione et al, 2008; Searchinger et al, 2008).

Therefore, sustainability of biofuels should be assessed carefully, considering all relevant environmental, economic and social aspects (The Royal Society,

2008) . Furthermore, to prevent shifting the burdens along the supply chains, sustainability should be assessed taking a systems approach and considering the whole life cycle of biofuels, including cultivation of the feedstock and biofuel production processes (Azapagic, 2006; Fehrenbach et al, 2007; Stichnothe and Azapagic, 2009; The Royal Society, 2008; US EPA, 2009). The life cycle approach is also required by various legislative acts related to biofuels, including the European Union (EU) Renewable Energy Directive (EC, 2009), the German Sustainability Biofuel Ordinance (GFG, 2007), the Swiss Directive on Mineral Oil Tax Redemption for Biofuels (SFG, 2007), the UK Renewable Transport Fuel Obligation (DfT, 2008) and the US Energy Independency and Security Act (USFG, 2007).

This chapter discusses how the main sustainability issues associated with biofuels can be assessed on a life cycle basis, considering different bio-feedstocks and production routes.

Miscanthus giganteus is a tall (up to 3 m) perennial sterile grass originating from Japan (Hodkinson et al., 2002). It can be harvested yearly with a sugar cane

harvester and can be grown under cool climates like that of northern Europe and USA. Like other bioenergy crops, stems may be burned for heat and electric power production or fermented to ethanol. It combines many of the desirable properties of a biofuel crop. As a perennial C4 plant, it produces consistently high biomass yields (8-15 tons/ha dry weight) over many years with little or no nitrogen application, shows good energy balance and low mineral content, which improves fuel quality. However, the yield potential might not be fully used when this variety is cultivated under varying climatic conditions. Interspecific crosses between sorghum and Miscanthus could complement Miscanthus in adaptation to stress conditions in arid climates. Similarly, Miscane, a hybrid between sugar cane and Miscanthus, could potentially combine the high productivity of both species with the perenniality and adaptation of Miscanthus to colder climates (Jakob et al, 2009).

Although the application of lipase in the production of biodiesel from vegetable oils has been thoroughly addressed in the literature, most of the studies were purely parametric. On the other hand, significant number of kinetic studies is found in the literature on the esterification of free fatty acids rather than the transesterification of vegetable oil. The industrial interest, however, is on the production of biodiesel from the triacylglyceride (oil), not the free fatty acids. The main difference between esterification of free fatty acids and transesterification of triglycerides (oils) is that in the first O-H bonds are broken, whereas in the second ester bonds are the ones that are broken. In addition, the by-product of esterification is water, whereas it is glycerol in transesterification. An attempt to model vegetable oil transesterification was done (Al-Zuhair, 2005), assuming that the reaction took place in two consecutive steps. In the first step, triglycerides are hydrolyzed to produce free fatty acids and in the second step, the free fatty acids produced in the first step are esterified to produce fatty acids methyl esters. This study combined the enzymatic kinetics models of hydrolysis of oils (Al-Zuhair et al., 2003) and esterification of FFA (Janssen et al, 1999; Krishna and Karanth, 2001). However, it was later shown that it was more accurate to assume that transesterification takes place by direct alcoholysis of the triglycerides (Al-Zuhair et al., 2007). In order to understand the reaction behavior and to propose suitable mechanismic steps, experimental determination of the separate effects of oil and methanol concentrations on the rate of enzymatic transesterification were determined. The proposed mechanism of alcoholysis of oils was based on the enzymatic hydrolysis mechanism (Bailey and Ollis, 1986) and presented by a Ping-Pong Bi Bi mechanism shown in Fig. 6.3. To account for the inhibition by alcohol, competitive inhibition was assumed when an alcohol molecule reacts with the enzyme directly to produce a dead-end enzyme-alcohol complex (E. A). And to account for the inhibition by the substrate, competitive inhibition was also assumed when a substrate molecule reacts with the acylated enzyme to produce another dead-end complex, namely, acylated enzyme-substrate complex (E-Ac. S). Based on this mechanism and assumptions, the reaction rate presented in Eq. [6.4] was derived:

where и is the initial reaction rate, V „ is the maximum reaction rate, K„ and K.

max s

are the dissociation constants for the substrate (S) and the alcohol (A), respectively, and KIS and KIA are the inhibition constants for the substrate and the alcohol, respectively. Numerical values of the parameters found in Eq. [6.4] are shown in Table 6.2 for lipases from different sources.

Equation [6.4] describes the initial reaction rate in the absence of any product inhibition, which is similar to the one proposed by Krishna and Karanth (2001) for the esterification of short-chain fatty acids with alcohol. On the other hand, Janssen et al. (1999) derived an equation to be used when the water, taken as one of the products, was assumed to inhibit the reaction. This modification was applicable when free fatty acids were considered as the substrate. However, when the substrate was the triglyceride, the product water is replaced with monoglyceride, diglyceride or glycerol. And unlike water which is usually present in the reaction medium at time zero, these products are not. Therefore, the product inhibition was neglected, especially when considering the initial rate of reaction.

6.6 Future trends

For many years the research on bioethanol production was based on laboratory scale experiments but since the demand for commercial bioethanol production from lignocellulosic feedstock it has become necessary to use and test the present knowledge of the process of ethanol production on a larger scale; a pilot scale (100 times larger scale than laboratory scale) before it can be applied on a large industrial scale (100 times pilot scale) production. There are several practical and technical challenges that need to be solved before industrial scale bioethanol production can be established. A continuously operated pilot plant for ethanol production from lignocellulosic feedstock was inaugurated in Sweden in May 2004. It is a complete pilot plant from a raw-material intake of wood chips in truck carried containers to a distillation column producing ethanol with a concentration up to 94%. In between, there are stages for rinsing the wood chips, steaming, impregnation, digesting, filtration and fermentation. The pilot plant is operated 24 h/day, 7 days a week and processes up to 2000 kg of dry raw material producing up to 400 litre of ethanol/day. The reactor system is continuously operated and at high pressures. This means that the wood chips have to be transported from atmospheric pressure in the raw material intake to the pressurised reactors without interrupting the material flow.

The feedstock in this plant is spruce wood chips at the moment but other feedstocks such as bagasse and other agro-based feedstock will be tested in the future. The pilot plant has two thermo-chemical reactors giving the possibility to either perform a two-stage dilute acid hydrolysis or a pre-treatment for enzymatic hydrolysis. In the plant, possibilities for both SHF and SFF are available. Also, recycling of liquids is possible in order to reduce the amount of fresh water usage. Fermentation and enzymatic hydrolysis is performed in five bioreactors with the size of 10 m3. A flow chart of possible steps in bio-ethanol production from lignocellulosic feedstock shown in Fig. 9.7.

There are biogas plants worldwide with different degree of technical development. The overall world market was approximately two billion euro in 2006 and is expected to increase to more than 25 billion by 2020 total (http://www. hkc22. com/biogas. html, updated in July 2008). Measures are taken worldwide to promote biogas and its market development (Sakulin, 2009). Among the initiatives taken, the feed-in tariff is a motivation to promote the adoption of renewable energy policy through legislation. According to this, the regional or national

electricity utilities are obliged to buy electricity generated from renewable sources (solar power, wind power, hydropower, geothermal as well as biomass). Table 12.4 lists the feed-in tariff for several European countries.

Europe has high-tech biogas plants in operation, with Germany being the leader. In 2006, 900 plants were built, reaching 3600 in total (Helmut Kaiser) in Germany. A market size of 7.5 billion euro, 30% export and 85 000 jobs are expected by 2020 in Germany. As in Germany, Denmark also has a variety of biogas plants of different capacity. The digestion of manure and organic waste is a well-established practice in Denmark with 20 centralised plants and over 35 on-farm plants (Raven and Gregersen, 2007), although there is a decline in the construction of new plants. In Austria and Switzerland, there are mostly small farm scale plants due to the national agricultural structure. In Sweden, there are also quite a few large scale plants (Fischer and Krieg, 2001). Table 12.4 reports the status of biogas plants and capacity in Europe.

Unlike Europe, experience with anaerobic digestion in North America is quite limited. Farm-based anaerobic digestion in North America only began out of necessity for odour control due to urban encroachment (Lusk, 1998). In the USA, which rejected the Kyoto protocol, most of the methane from wastes is allowed to escape into the atmosphere where it contributes to global warming. However, there is a strong movement towards the use of renewable energy from biogas. The development of anaerobic digesters for livestock manure stabilisation and energy production has accelerated at a very fast pace over the past few years. According to the EPA, about 111 digesters operate at livestock facilities in USA up to 2007 (U. S. EPA, 2007). The energy production was 215 million KWh (electrical energy: 170 million KWh). It was estimated that besides electricity generation, the biogas was used in boilers and fed in the natural gas gridding (after upgrade) or flared for odour control. In Canada, approximately 16 farm-scale anaerobic digesters operate or are being built (Wohlgemut, 2006). In 2006, the Ontario government implemented a Renewable Energy Standard Offer Program, which guaranteed farmers a higher rate for biogas-produced electricity, along with a financial assistance programme designed to reduce the capital costs of digester construction (Hilborn et al., 2007). On the other hand, in the Manitoba province, anaerobic digestion is less promoted due to the well-established and cost-effective hydroelectricity industry (Wohlgemut, 2006).

In Australia, the installed capacity for biogas was 458 MW in 2001. Electricity generation from biogas has increased considerably from 23 GWh in 1995 to 729 GWh in 2001, an average growth rate of 78% per year. Wastes from food processing plants, livestock manure and human sewage are the primary feedstocks for biogas production. Most of the installed capacity is at sewage treatment plants, which are considered highly cost effective.

In developing or less developed areas of the world, anaerobic digestion is spreading fast. In Asia (mostly China and India, but also Vietnam, Thailand, etc.) there are millions of low-tech, hand-made, plants consisting of underground, non­insulated digesters in operation for decades (Fischer and Krieg, 2001). Manure and food residues are the main feedstocks used and the biogas energy generated is used for cooking and lighting. According to the ministry of agriculture in China, 15 million households in China were using biogas in 2004, with the aim to increasing this number to 27 million by 2010, which will account for over 10% of all rural households. By the end of 2005 there were 2492 medium and large-scale biogas digesters in livestock and poultry farms, while 137 000 biogas bioreactors had been constructed for the household wastewater treatment (van Nes, 2006). In order to support the development of renewable energy sources, China enforces suitable legislation and takes steps to promote industrialisation of the construction of biogas plants. In India, 3.67 million biogas units were installed in 2004. The ministry of non-conventional energy resources implements a programme (national biogas and manure management programme) for providing financial, training and technical support for the construction and maintenance of biogas plants. Similar initiatives have been taken in Nepal and Vietnam (van Nes, 2006). European companies and organisations, mainly from Germany, Denmark and Austria, have already entered the Japanese and Korean market and transferred high-tech anaerobic technology, while they promote anaerobic digestion to the developing countries such as China, India, etc.

In Africa, there are attempts by international organisations and foreign aid agencies to promote biogas technology. Some digesters have been installed in some sub-Saharan countries, making use of feedstocks such as slaughterhouse wastes, municipal wastes, industrial waste, animal dung and human excreta. Small-scale biogas plants have been established all over the continent (Table 12.5) but only few of them are operational (Parawira, 2009). Insufficient know-how concerning anaerobic technology is claimed to be the main reason for

inadequate operational potential of the installed plants. In some cases, the installation of the plant is of poor quality and the appropriate maintenance lacks.

In Latin America, many biogas plants operate in the agricultural, industrial and municipal sectors. Biogas is mainly used for cooking, lighting, as town gas or as vehicle fuel. The quantity of biogas produced in Latin America was estimated at 217 million m3 per year in 1993 (Ni et al., 1993).

The future of biogas as a competitive biofuel relies on the economic feasibility of the anaerobic technology. The income sources of a biogas plant are the energy and fertiliser sales as well as the tipping fees for receiving off-farm waste. Remuneration or subsidies from the government is an extra income. If the cost of energy production is too high, the biogas can be flared to eliminate odours and greenhouse gas emissions, but this is not a viable option. The costs of biogas production are distinguished into the capital (or investment) and operational costs for the installation of the plant and its maintenance, respectively. Capital costs are determined mainly by the size of the plant and the technology selected. The price of components (feeders, stirrers, CHP, etc.) and construction materials (concrete, steel) also affect the investment cost. The operational costs include maintenance of the biogas plant, labour costs, insurance and other utilities. Laaber et al. (2007) estimated that the capital costs vary between 3000 and 5000 €/kWelectricity for the anaerobic digestion of energy crops, while the operational costs range between 2

and 4.5 ct/kWh

Life cycle assessment (LCA) is an environmental management tool used for estimating the environmental burdens and impacts from a system — product, process or a service — over its whole life cycle. The life cycle stages normally included in LCA are extraction and refining of raw materials; product manufacture, distribution and use; disposal of wastes; and all transportation steps in between. The LCA methodology, as defined by the ISO 14040 and 14044 standards (ISO, 2006a, 2006b), is outlined in Fig 3.1A. It consists of four phases:

• Goal and scope definition;

• Life cycle inventory (LCI) analysis;

• Impact assessment (IA); and

• Interpretation.

Goal and scope definition is the first and most important phase of LCA. Here, the reasons for carrying out the LCA study as well as the intended audience are stated. The system boundary and the functional unit (unit of analysis) are defined as well as the impact assessment method to be used in the Impact assessment phase. Assumptions, limitations, cut-off rules, etc., are also described in this phase.

Life cycle inventory (LCI) analysis quantifies the environmental burdens in the system, i. e. materials and energy used and emissions discharged into the

environment. Allocation of environmental impacts is also carried out within LCI. Allocation is the process of assigning to each function of a multiple-function system only those environmental burdens that each functional output is responsible for. For example, if there are two or more co-products from a system, the environmental burdens should be allocated between them so as to reflect their contribution to those burdens. ISO 14044 (ISO, 2006b) recommends three methods for dealing with allocation:

• if possible, allocation should be avoided by disaggregating the given process into different sub-processes or by system expansion;

• if it is not possible to avoid allocation, then the allocation problem must be solved by using system modelling which reflects the underlying physical relationships among the functional units (e. g. mass or energy basis);

• where physical relationships cannot be established, other relationships, including economic value of the functional outputs, can be used.

The allocation method used will usually influence the results of the LCA study so that selection of an appropriate allocation method is crucial. Sensitivity analysis should be carried out in cases where the use of different allocation methods is possible to determine the influence of the allocation method on the results.

Impact assessment (IA) consists of several steps. First, categorisation of environmental impacts is carried out to determine which impacts will be considered.

Impact categories commonly considered in LCA are resource depletion, land use, global warming, acidification, ozone layer depletion, summer smog, eutrophication, human and eco-toxicity. This is followed by the characterisation step, to calculate the contribution of different burdens to the selected impact categories. The impacts are calculated by multiplying the ‘potency factors’ of each burden with its total life cycle emission. The potency factors indicate the potential of a burden to cause a particular impact and are expressed relative to a reference substance. For example, the potency factor for methane with respect to global warming is 25 kg CO2 eq./kg CH4, indicating that methane is 25 times more potent global warming agent than CO2, whose potency factor is defined as unity.

The remaining two steps in IA — normalisation and valuation — are optional. The former normalises each impact to the total impact in a region or the world over a certain period of time, normally one year. In the valuation step, the impacts are aggregated into a single environmental impact index by assigning the weights of importance to the different impacts. This is the most subjective step in LCA and requires elicitation of preferences by stakeholders or decision makers.

Interpretation is the final LCA phase, whereby the results are interpreted depending on the goal of the study. This may include identification of the most significant impacts, ‘hot spots’ in the system and opportunities for improvements. Sensitivity analysis is also carried out within this phase.

Generally, there are two types of LCA studies: attributional and consequential (Curran et al, 2001; Ekvall and Weidema, 2004). In attributional studies, the impacts are attributed to the system of interest (e. g. product) based on the flows in and out of the system as they are. For example in attributional LCA, impacts from the production of biofuel from wheat in the UK are estimated (attributed) based on the inputs and outputs from this system, not taking into account what happens with the other related activities in the economy, for example if the supply of wheat is constrained, e. g. due to its use for bread production. In consequential LCA studies, the aim is to estimate how the flows to and from the system would change as a result of different potential decisions. For example in the case of biofuels, a consequential LCA study would attempt to quantify the impacts of diverting wheat in the UK into biofuel production and having to supply food (e. g. bread) from alternative sources or from elsewhere in the world.

The attributional approach is used for labelling purposes (e. g. PAS2050 [BSI, 2008]) and certification systems (e. g. EU RED [EU, 2009]; Renewable Transport Fuel Obligation, RTFO [DfT, 2008]). Most biofuel LCA studies are also based on the attributional approach.

3.4 Sources of further information

CO2 tool for estimating GHG emissions from the production of transport fuels, electricity and heat from biomass

http://www. senternovem. nl/gave_english/co2_tool/index. asp

IEA Bioenergy

http://www. ieabioenergy-task38.org

Well-to-wheels evaluation of biofuels http://ies. jrc. ec. europa. eu/WTW

Biofuels sustainability scorecard

http://idbdocs. iadb. org/wsdocs/getdocument. aspx? docnum=2152669

Biofuels and sustainability in Europe: http://www. biofuelstp. eu/sustainability. html

Many heterogeneous alkaline catalysts are available but the most frequently used are alkali metal, alkaline earth and metal salts. An overview is given in Table 5.3 (Bacovsky et al., 2007).

Heterogeneous basic catalysts can be classified as Brpnsted or Lewis catalyst. As in the case of homogeneous Brpnsted basic catalyst such as basic zeolites, the formed catalytically active compound is a homogeneous alkoxide (Fig. 5.11) (Lotero et al., 2006).

In the case of heterogeneous Brpnsted basic catalysts (e. g. resins with quaternary ammonium functions, QN+OH), the positive organic ammonium groups being bonded to the support surface electronically retain the catalytic anion on the solid surface.

Table 5.3 Overview on heterogeneous catalysts (Bacovsky et al., 2007)

Catalyst type examples

0“Na + CH3OH——————— ► O-H + CH30“Na+

5.11 Reaction mechanism for heterogeneous catalysis (I).

The reaction occurs between the methanol adsorbed on the cation and the ester from the liquid (Fig. 5.12).

CTN+OH” + CH3OH —————— ► 0-N+0CH3- + H20

5.12 Reaction mechanism for heterogeneous catalysis (II).

The formation of alkoxide functions is the fundamental step for heterogeneous Lewis basic catalyzed reactions. In the case of MgO, the reaction occurs between the methanol molecules adsorbed on magnesium oxide free basic sites and the esters functions in the triglycerides in the liquid phase (Fig. 5.13).

In the review by Di Serio et al. (2008) many applications of heterogeneous alkaline catalyst have been described.

0“R H+

> —— Mg—— 6——-

5.13 Reaction mechanism for heterogeneous catalysis (III).

In comparison with homogeneous catalysts, in order to obtain similar conversion rates, more severe reaction conditions have to be used:

• Temperatures up to 300°C under supercritical conditions are a pre-requisite to achieve conversions higher than 90%.

• High molar rates methanol:oil have to be utilized: 15-40:1.

• Longer reaction times (except in methanol supercritical conditions or microwave irradiation).

• Higher amounts of catalyst.

• Leaching of the catalyst into the biodiesel.

Good results using alkaline-earth metal hydrides, oxides and alkoxides have been reported by Gryglewicz (1999) and Demirbas (2007). The order of reactivity Ca(OH)2 < CaO < Ca(OCH3)2 is in agreement with the Lewis theory stating that methoxides of alkaline-earth metals are stronger bases than the corresponding oxides and hydroxides. Good biodiesel yields were also obtained in the transesterification of soybean oil using ZnO, loaded Sr(NO3)2 followed by
calcination. The active catalyst is SrO. When the reaction is carried out at reflux, five per cent catalyst and 12:1 mol ratio of methanol:oil, a conversion of 95% can be reached (Lopez Granados et al, 2007).

The preparation of new materials obtained from the co-precipitation of aluminum, tin and zinc oxides and their use as catalytic systems for the alcoholysis of vegetable oils have been reported by Macedo et al., 2006. These (Al2O3)X (SnO)Y(ZnO)Z type of metal-oxides were found to be active for the alcoholysis of soybean oil, using several alcohols, including branched ones. Best results were achieved using methanol, with conversion yields up to 80% in 4 hours. It was also possible to recycle the catalysts without apparent loss of activity.

Sodium silicates can be used at 60-120°C but the use of microwave energy greatly increases the conversion (Portnoff et al., 2006).

Good results have also been obtained by MgO/Al2O3 hydrotalcites and industrial applications could be possible if the reaction was carried out at 180°C and 12:1 methanol:oil molar ratio (92% yield) (Leclercq et al., 2001; Di Serio et al., 2006).

Waste oil was converted in good yields using Mg-Al layered double hydroxide catalysts in 80-160°C temperature range with up to 48:1 molar methanol:oil ratio and high amounts of catalyst (up to 12%) (Brito et al., 2009).

In addition calcined Li-Al layered double hydroxides (Shumaker 2007), sodium zeolites, titanium containing zeolites (Xie et al., 2007), anion resins (Shibasaki — Kitakawa et al., 2006) and polystyrene supported guanidine and biguanidines (Gelbard and Vielfaure-Joly, 2001) have shown promising results.

The use of strong alkaline ion exchange resins is limited due to the loss of stability at temperatures higher than 40°C and the neutralization of the catalyst by the FFA present in the feedstock.

In addition, the glycerol formed is absorbed in the polymeric matrix causing deactivation of the active sites.

Very recently, an efficient laboratory procedure has been developed using CaO after appropriate treatment which allows reaching high conversions of triacylglyceride (TAG) into FAME in a one-stage operation, meeting the requirements of the EN 14214 (Kouzu et al., 2008; Lengyel et al., 2009). The catalyst was activated by drying for 24 hours at 105°C. Using 6:1-12:1 molar ratio of methanol:oil, reflux temperature and eight per cent catalyst, conversion rates of 99% were obtained. However, organosols are formed due to the presence of calcium soaps, leading to a yield of 70%.

An industrial applied heterogeneous catalysis process (Fig. 5.14) has been selected by Diester Industrie using Axens biodiesel technology, Esterfip-H™ for a new plant in Sete (France) with a capacity of 160 000 tonnes per year, followed by a plant in Sweden in 2007. The catalyst consists of a mixed oxide of Zn and Al coated on y-alumina, which promotes the transesterification without catalyst loss. The reaction is performed at higher temperatures and pressures compared to those of homogeneously catalyzed processes, also using an excess of methanol. This

methanol excess is removed by vaporization and reused in the reaction together with the fresh methanol. The conversion is reached in two successive stages and separation of the glycerol in order to shift the equilibrium to methanolysis.

The catalyst section includes two fixed bed reactors. Excess of methanol is removed after each reactor by partial evaporation and the esters and glycerol are separated in as settler. The residual methanol in the glycerol is evaporated. Biodiesel purification consists of methanol vaporization under vacuum and adsorption of the soluble glycerol (Bournay 2005).

The advantage of the process is a very high biodiesel quality, salt-free glycerol, no soaps formation and no handling of hazardous chemicals. This process can be considered as green technology.

On a semi-industrial scale a new type of heterogeneous catalyst has been introduced by Catalin using nanoparticles. The catalyst preparation involves the utilization of organotrialkoxysilanes with various anionic, hydrophobic or hydrophilic functional groups that could provide different noncovalent interactions, for example electrostatic attractions, hydrophobic interactions etc., with cationic cetyltrimethylammonium bromide (CTAB) surfactant micelles in a base catalyzed condensation reaction of tetraethoxysilane.

Catalin T300 catalyst differs from the most solid catalyst that requires a fixed bed and high temperature and pressure to operate. The T300 catalyst can be used in existing plants with minimal modification as it reacts at common operational temperatures and pressure. The reactor consists of a reactive vessel within a plate with a mesh. The catalyst is stocked on top of the mesh and the oil flows through. The T300 catalyst is a ‘drop-in catalyst’ that can be used as a direct replacement for the commonly used sodium methoxide. Therefore there is no need for a fixed bed and the catalyst in the form of a granular powder can be directly mixed with oil.

A filter system is used to keep the catalyst in the reactor and there is no need for water washing. The Catalin process flow diagram is depicted in Fig. 5.15.