Biodegradative pathways are strongly involved in vegetable oil quality. Lipolytic activities in the oil palm fruit mesocarp are important for the acidity of the final oil product. Oil palm fruit lipases are particularly important because they are responsible for the increase in oil acidity in certain conditions. While polyunsaturated fatty acids are not usually in a free state in plants, but mainly esterified to glycerol in the form of membrane or storage lipids, their prior release by the action of lipolytic acylhydrolases is necessary [11]. Hydrolysis of lipids by lipase is induced by fruit bruising and/or poor storage. Once the polyunsaturated fatty acids have been released, they can be oxidised by the action of lipoxygenase (LOX) [38]. The fact that olive LOX oxidises free unsaturated fatty acids at a much higher rate than esterified ones confirms that action of an acylhydrolase releasing C18:2 or C18:3 acids for the glycerolipid fraction is necessary prior to their oxygenation.

Among the catabolic pathways related to lipids, the so-called lipoxygenase pathway is also particularly important in view of sensory quality (relevant for oil fruits, e. g. olive). Virgin oil aroma consists of a complex mixture of more than 100 compounds, mainly volatile saturated and unsaturated aldehydes, alcohols and esters [39]. These compounds are produced from polyunsaturated fatty acids through a cascade of reactions (lipoxygenase pathway). The LOX pathway involves oxidative degradation of polyunsaturated linoleic and a-linolenic acids, which are split into volatile C6 or C9 carbonyl fragments, which can further be modified by isomerisation, reduction and esterification (Fig. 3.5). In the case of olive oil, the lipoxygenase pathway is principally triggered during crushing of olive fruits and malaxation of the resulting pulp that takes place in the extraction process. The aroma of a given oil is a function of the activity levels and characteristics of the enzymes involved in that cascade of reactions. There exists a relationship between the activity level and the characteristics of the LOX pathway enzymes in olive pulp with the quality of the sensory components of the resulting olive oil [11]. The same phenomenon does not occur in other oil fruits (palm fruit, avocado). The extraction process of oil palm includes sterilisation (by heating) to inactivate enzymes such as endogenous lipases, which also cause inactivation of LOX pathway enzymes [40]. Aroma is not a quality parameter of palm oil.

The economics of biodiesel production may be improved through the use of low-value lipids as (extremely) low-cost feedstocks (see Table 5.31). Moreover, as no one raw material is in adequate supply to be used solely for biodiesel production, fats from the rendering industry are expected to play an increasingly important role as a highly desirable feedstock [246a]. Recently, a growing number of studies have reported use of cheap waste oils and fats (animal tallow) as a raw material [108, 210, 247-272]. By-products of the vegetable oil industry which also show promise as raw materials for alkyl ester production are rice bran oil [229, 273, 274] (see Section 5.5), palm-fruit pulp oil [275] and waste activated bleaching earth [276]. Biodiesel can be made out of 80% FFA oil.

The environmental benefits of recycling a lower-value material into a high-value energy product are obvious. This biodiesel will leave the lowest carbon footprint, provide the highest energy balance and is food — and forest — friendly.

As shown in Chapter 9, low-cost feedstocks must usually undergo some form of pre-treatment (filtration, bleaching, deodorisation, steam distillation, gravity separation, drying, pre-esterification) before they can be transesterified

to biodiesel. Figure 5.3 shows the fractionation of crude animal fat. Some esterification methods applicable to vegetable oils are not indicated for animal fats or fried oils because of chemical property differences (high amounts of FFAs, water and oxidised components). Waste oils are particularly suitable for two-step catalytic esterification-transesterification or enzymatic catalysis. Lipase catalysis allows the simultaneous transesterification of triglycerides and esterification of FFAs in a one-step process, and many lipases work well in the presence of water. As conversion of low-value feedstocks with high levels of FFAs into biodiesel requires more heat and pressure, supercritical methods may also be considered. Accordingly, Babcock and Schulte [277] used such a one-step non-catalytic method to convert low-grade chicken fat (6% FFA) and crude tall oil at 598 K into biodiesel with 94% and 89% yields, respectively. Crude tall oil contains rosins and resin acids that do not combine with methanol to form biodiesel under the test conditions. In the case of rice bran oil, in-situ transesterification (a combined extraction and transesterification process, see Section 7.4) has been reported [273]. The same procedure is applicable to palm-fruit pulp [275].

An overview of biodiesel production for RSO (canola), SBO, WVO and animal fats is available [278]. Quality assurance of biodiesels based on low- cost feedstocks is obviously more problematic than in the case of refined starting materials.

In-situ transesterification of lipids in oilseeds and other biological materials may be carried out in the presence of an extraction solvent or in solvent-free mode. Whereas in the former procedure extraction of oleaginous seeds and transesterification of the extracted oil take place in an integrated process [119], the latter procedure eliminates the air-polluting chemical hexane from

the process by simply skipping the oil extraction step. Differences between conventional and in-situ transesterification are shown in Fig. 7.7. Stern et al. [15] have described in-situ transesterification by using an extraction solvent and a reactant (aqueous EtOH) in contact with cottonseeds. Acid-catalysed in-situ transesterification has been reported for rice bran oil, a highly acidic oil obtained from the waste material produced during rice-dehulling [120, 121]. The lipid contents of the hull contribute to the overall yield of esters from the seed and lipid losses due to hull-kernel separation are avoided. In the esterification with methanol, all free fatty acids dissolved in methanol were interesterified within 15 min. Similarly, acid-catalysed in-situ transesterification of soybean [122] and sunflower seeds [123] was reported. The same procedure is applicable to palm-fruit pulp [124]. In in-situ alcoholysis with a variety of alcohols, ethyl, n-propyl and n-butyl esters of soybean fatty acids could be obtained in high yields, at variance to methyl esters, as methanol is a poor solvent for soybean oil [122]. Reactive extraction of biodiesel from rapeseed using methanol, ethanol and methanol/ethanol mixtures was reported [125]. Also in-situ transesterification of Cynara cardunculus seeds was described, with a biodiesel yield of 36% and 75% FAME content [126]. Similarly, an integrated process has been reported for producing biodiesel from castor bean seeds, comprising alkaline-catalysed ethanolysis (using anhydrous EtOH) [18]. Whereas the resulting FAEEs are used as biodiesel, the solid fractions may be used as fertilisers, for feeding cattle and as a raw material for producing ethyl alcohol. The efficiency of production of FAME from

materials not used so far for this purpose, such as seed cakes, has been improved by the application of ultrasound [127]. In in-situ derivatisation assisted by ultrasounds (ultrasonically assisted extraction transesterification, USAET) for biodiesel production, the TGs contained in solid material are extracted and immediately transesterified in a methanolic solution of 1 M NaOH in an ultrasonic field, thereby greatly increasing the total yield in most instances: from 46% to 85.5% for CSO, from 67.2% to 93% for SNO, and from 43.2% to 83.5% for SEO [127]. Overall advantages of the proposed methodology include the elimination of saponification, low reaction time, milder reaction conditions, and higher FAME yields.

Solvent extraction of sunflower seeds leads to higher yields (44-45%) than pressure extraction (38%) [128]; whole sunflower seeds contain some 5.5% of water. In-situ transesterification of sunflower seeds with acidified methanol produces FAME in yields significantly higher than with the conventional procedure with pre-extracted oilseed [123, 129-131]. Moisture in the seeds reduces the methyl ester yields. By drying oilseed flakes before starting the biodiesel synthesis, the required methanol volume can be greatly reduced [132]. Acidified hexane extracts 11% more total lipids from soybean than hexane alone [133]. Acidic conditions of the in-situ reaction apparently have a similar effect. The lipid fraction of whole oleic sunflower seeds has been transformed in-situ into fatty alkyl esters used as lubricants in a thermo — mechanico-chemical twin-screw system [134].

High levels of transesterification of lipids resident in biological materials can be achieved under mild temperatures and ambient pressures with relatively short incubation times in alkaline alcohol solutions [135]. The approach, which eliminates the need for lipid isolation by solvent extraction or expelling, and the use of extracting solvents, reduces process steps, and expands potential biodiesel production by allowing the use of lipid-bearing materials not currently used as such (e. g. distillers dried grains, meat and bone meal). Nevertheless, the cost of biodiesel produced through in-situ transesterification is higher than that made using the traditional route (typically US$1.02/gal and US$0.38/gal, respectively for soy diesel) [132].

Finally, reactive extraction of triglycerides and FAME formation was reported using fungal (Aspergillus flavus) resting cells and oilseeds at moderate temperature, either with solvents or in a solvent-free system [136].

Abstract: Although direct use of high-viscosity straight vegetable oils in standard modern diesel engines has been reported, the practice of using such non-regulated fuels is generally not recommended as short — and long-term problems do result. Engine compatibility for vegetable oils and animal fats may be enhanced chemically (by derivatisation, pyrolysis/gasification, ozonation) or physically (by dilution, blending, microemulsification). Fuel properties of oils and fats are outlined.

Key words: Fuel properties, straight vegetable oils, low-viscosity formulations, co-solvent blending, microemulsions.

4.1 Introduction

The term biodiesel (now in use to denote monoalkyl esters of long-chain fatty acids derived from renewable feedstocks) was originally coined to describe unmodified vegetable oils that could substitute for diesel fuel (DF). Industrial use of biofuels started in the 1880s. Rudolf Diesel designed a prototype of the diesel engine, received a German patent (28 Feb. 1892), and demonstrated a workable engine in 1897 [1]. The first public demonstration by the French Otto Company of a small diesel engine operated on straight peanut (Arachis hypogaea) oil was seen at the World Fair in Paris of 1900 [2]. At that time the inventor, Diesel, held the view that the future of his engine (in contrast to those operating on steam) would be connected to fuel use derived from biomass, in particular plant oils (such as peanut and castor oil) and animal fats. Diesel’s compression-ignition engine used large injectors to prevent clogging by viscous, heavy fuels, such as unrefined vegetable oil. Because of the size of the engines, large warships were among the first users of Diesel’s technology. In the 1920s, technological changes made possible much smaller diesel engines, which required lower viscosity fuels. At the same time, the advent of relatively cheap medium-weight diesel fuels from fossil origin produced by the upcoming petroleum industry temporarily put a halt to the commercial viability of biofuels. Since about 1930 the diesel engine has been fine-tuned to run on the diesel fraction of crude oil, which consists mainly of saturated hydrocarbons. As a result, modern diesel engines do not run satisfactorily on a pure vegetable oil feedstock because of problems of high viscosity, deposit formation in the injection system and poor cold-start problems. Pure rape oil can only be used in indirect injection (IDI) compression engines and very few diesel engines are now of this type.

As petroleum has historically been in short supply, vegetable oils and their derivatives have been proposed as alternative diesel fuels during such emergency situations. The concept of using biomass-based fuels, specifically straight vegetable oils (SVOs) as diesel fuel alternatives, is particularly interesting and has been pursued on various occasions. Since 1930, the use of SVOs or their blends and derivatives, such as hydrocarbons obtained by thermal-catalytic cracking, and fatty acid esters (nowadays known as biodiesel) has been proposed. Edible refined soybean oil was used to power Japanese vessels in the Battle of Okinawa (1945). Coconut oil had also been used. During World War II various Indian vegetable fuel oils were also considered for diesel engines [3]. Pyrolysis of different triglycerides was also used for liquid fuel supply in different countries. For example, hydrocarbons were produced in China by a tung oil pyrolysis batch system and used as liquid fuels [4]; in another approach, vegetable oils were transesterified [5-7] and fatty acids were esterified [7]. Neat vegetable oils, such as babassu, coconut, castor seed and cottonseed were used in internal combustion engines in Brazil during the 1940s [8] or hydrocarbons produced by thermal-catalytic cracking [9]. Export of CSO, which was the main Brazilian vegetable oil at that time, was forbidden in order to force a drop in price and, thus, make possible its use as fuel in trains. Cottonseed oil was regarded as a strategic material as a substitute for imported diesel fuel [10]. Later, grapeseed oil was tested as a petroleum substitute. Quite recently, the Boungainville Revolutionary Army again used straight CNO.

Since Diesel’s first experiments various other neat vegetable oils (mainly RSO, PMO, CNO) have been tested and used as diesel fuel alternatives [11-15]. All these fuels are non-toxic, renewable sources of energy, which do not contribute to the net global CO2 build-up. Direct use of renewable feedstocks, such as SVOs, in diesels may seem attractive (see Table 4.2), but there are risks. In fact, early tests already showed that using unmodified vegetable oils in diesel engines caused serious problems [16].

Vegetable oil sold for culinary purposes requires no stringent viscosity specification. Vehicles can be adapted to run on this sort of oil in unused or in waste form, though in the latter case filtration is necessary. Adaptations to make a modern diesel vehicle run reliably on SVO are fairly major. Vehicular use of SVO purchased for kitchen use is illegal in many countries, including the UK. Spent cooking oil from fast food outlets has been used to power vehicles with compression ignition engines. Some potential for serious usage of plant-derived waste cooking oils can be envisaged as the product is carbon neutral. R&D into the use of spent cooking oil as an automotive fuel is taking place in several countries (e. g. Greece and Spain).

116 Biodiesel science and technology

India consumed 40.3 Mt diesel in 2000-01, or 43.2% of the total consumption of petroleum products (two-thirds imported). The country stands in sixth place in the world in energy need and overall demand for crude oil, which is expected to rise annually by 5.6% up to 2011.

Being a tropical country, India has rich and abundant forest resources with a wide range of plants and oilseeds of sufficient volume potential. The country has enormous potential of oilseeds of tree origin (TBOs) like cheura (Diploknema butyracea), jatropha (Jatropha curcas), jojoba (Simmondsia chinensis), karanja (Pongamia pinnata), kokum (Garcinia indica), mahua (Madhuca indica), neem (Azadirachta indica), simarouba (Simarouba glauca), tung (Aleurites spp.), wild apricot (Prunus armeniaca) and Chinese tallow (Triadica sebifera), which can be grown and established in wasteland and varied agroclimatic conditions. These have domestic and industrial utility in agriculture, cosmetics, pharmaceutics, cocoa-butter substitute and as diesel substitute. Most of these tree-borne oilseeds are scattered in forest and non­forest areas and scarcely 20% of the existing potential is being crushed and utilised. Indian oil and fat sources are still largely underutilised [66]. The estimated Indian annual production capacity of various non-edible tree — borne oilseeds exceeds 20 Mt [67]. The production of these oilseeds can be stepped up many fold if the government takes the decision to use them for producing diesel fuels. The Indian policy is largely to favour non-edible oils and fats for biodiesel purposes. The prohibitive cost of edible oils prevents their use in biodiesel application, at variance to non-edible oils. Simarouba (60-75% oil, 1000-1500 kg oil/ha; 36% SFA : C16 : 0 + C18 : 0) would need cracking in cold climates; oil and pulp are edible, at variance to many other species.

Various vegetable oils from India (peanut, karanja, punnal, polang, castor, kapok, mahua, cottonseed, rapeseed, coconut and sesame) were investigated more than 60 years ago as fuel oils for diesel engines [68]. While castor bean is nowadays grown on a large scale on marginal and wasteland in South Asia, it appears that castor oil methyl ester would cause problems for most practical diesel engines because of its high viscosity (13.75 mm2/s compared to 3.2 mm2/s for mineral diesel) [69].

India’s biofuels industry is still in its infancy [70]. Initiatives are mainly at R&D level and demonstration plants with few small commercial units. India operates a 100 kt/yr crude palm oil/PFAD biodiesel plant. The prospects of biodiesel production from vegetable oils in India were recently described by Barnwal and Sharma [71]. Pongamia [72-77], Jatropha [78-82], and mahua [83, 84] have already shown their potential for biodiesel use. A method and system to produce ethanol and biodiesel from a combination of corn (maize) and other agro feedstocks (simarouba, mahua, rice, pongamia, etc.) has been described [85].

Jatropha is considered to be one of the best alternative non-edible oilseed crops for biodiesel production in India (and elsewhere), though with low power produced per unit area, and is now increasingly being used in reforestation programmes in tropical countries. India has up to 65 Mha of non-arable land available to produce jatropha, and intends to replace 20% of diesel fuels with jatro biodiesel. Numerous jatropha plantation projects are under way (see Section 5.9.1). Genetic improvement of energy crops such as Jatropha has barely begun. The estimated production potential of mahua oil is 181 kt/yr. India also grows 80 000 acres of its native Moringa oleifera (Moringa, Zogale). The tree oil crop moringa has multiple food uses (leaves, meal, seed). Moringa yields 3 t seeds/ha, producing 38-40% oil that contains 65% oleic acid, not unlike olive oil (see also Section 5.5). India also produces 5 kt/yr of highly acidic (17% FFA, unrefined), non-edible rubber (Hevea brasiliensis) seed oil, which requires a two-step transesterification for biodiesel production [86]. The overall potential of sal (Shorea robusta) seeds, which contain 18.5% of hard fat, is about 5.5 Mt/yr.

Azam et al. [56] examined the fatty acid profiles of seed oils of 75 Indian plant species having 30% or more fixed oil in their seed/kernel. Fatty acid compositions, experimental or calculated values of SN, IV and CN were used to predict the quality of the corresponding FAME for use as biodiesel. Some 26 species, including Azadirachta indica (neem), Calophyllum inophyllum L. (undi), Jatropha curcas (physic nut), Pongamia pinnata (karanja) and Ziziphus mauritiana (Indian jujube or Chinese apple) (Table 6.7), did meet the major (inter)national biodiesel specifications; another 11 species conformed to the more permissive US standard only. The fatty acid composition of C. elatum (Guttiferae family) [87] is similar to that of C. inophyllum [56]. Cultivation of these plants on (part of) the Indian wastelands (totalling 94 Mha [88]) would satisfy the energy needs of India, currently on imports for 70% dependent (87.5 Mt imported petroleum/yr). Neem oil is being used as lamp oil in India. Neem and pongamia are also being used as a source of safe and renewable natural pesticides (not unlike mustard seed). However, as a result of W. R. Grace’s patent claims, the price of neem seeds has sky­rocketed.

The other qualifying non-traditional crops with seed oils for potential use as biodiesel deserve to be evaluated and could be developed and expanded for other regions of the world with different climatic and environmental conditions. For example, Mesua ferrea L. (nahor), also belonging to the Guttiferae plant family and rated positively by Azam et al. [56], contains a viscous crude oil which was already proposed as a diesel fuel raw material (fraction distilling between 473 K and 573 K) more than 20 years ago [89]. More recently, both karanja (CN 56.2) and nahor (CN 54.6) were evaluated positively as cheap raw materials for biodiesel fuel [90].

South-East Asian pili nut (Canarium ovatum) oil (Table 6.7) has poor nutritional value due to its low polyunsaturated fatty acid content (10.5%; compared with SFA and MUFA values of 44.5 and 45.0%, respectively) [91] and is a potential biodiesel raw material candidate. Kenaf (Hibiscus cannabinus L.) from the Malvaceae family is a valuable fibre plant native to India. Although kenaf is used mainly for its fibre, mass production of (edible) oil as a by-product could significantly increase the economic value of this crop [103]. A similar situation is encountered in Europe for cardoon. Because kenaf oil has lower linoleic and linolenic content (45.9% and 0.7%, respectively) and higher oleic acid (29.2%) than soybean oil (53.0%, 7.5% and 24.5%, respectively) it is expected to be more stable. Kenaf is related to cotton and okra; the kenaf oil composition is similar to cottonseed oil. Kenaf oil is characterised by a higher total phospholipid content (about 6.0%) than major oil seeds such as soybean (1.5 to 3.0%) and cottonseed (< 2.0%).

Salicornia brachiata is a halophyte (salt-tolerant plant), used as a vegetable by mangrove-dependent village people. Potential industrial use of the salt­water irrigated crop (Chenopodiaceae family) oil has been prospected [107]. Quite different fatty acid profiles for salicornia oil are found in the literature (see Table 6.7), with a reported high 10-undecenoic acid (C11 : 1) content

aOil content from kernel. bOil content from seed. cGenetically altered plant strain VS-320. dOil content from tubers.

(37.9%) being suspect. S. brachiata oil is highly unsaturated and very rich in linoleic acid (61.0%). Similary, S. europaea oil also contains an appreciable C18 : 2 content [112]. Saponification number is high, as in the case of S. bigelovii [113]. The high content of saponins in the seeds of Salicornia makes the oil unsuitable for edible purposes. Cassia marginata and C. corymbosa (Leguminoseae family) oilseeds (Table 3.6) are characterised by very low oil contents and unacceptably low CN values [114]. On the other hand, as Ficus benghalensis (banyan tree) oil contains only small amounts of various unusual (epoxy, cyclopropenoid) fatty acids (Table 3.6), its potential use as a biodiesel feedstock might be envisaged.

Today only a few percent of global transportation fuels are derived from biomass and are produced either by fermentation of food-derived carbohydrates (e. g. sugar or cornstarch) to ethanol or by processing of plant oils to produce biodiesel. Except for Brazil, where favourable agricultural conditions and a flexible processing infrastructure allow the majority of the country’s road transport to be powered economically with cane-derived ethanol, without mandates or subsidies biomass derived fuels are not economically competitive with fossil fuels even at crude oil prices of US$150/barrel and at correspondingly high vegetable oil prices.

Brazil needs some 534 M barrels/yr, or 85 Mt/yr. The Brazilian diesel oil consumption (2003) amounts to about 40 Mt. Estimated domestic biodiesel need in Brazil will be about 800 kt in 2008 (mandatory B2 blend) up to 2.5 Mt in 2013 (at B5 mandate). This requires 8-25 100 kt/yr plants. Installed capacity is increasing rapidly from 50 kt (2005) and 600 kt (2006) to 2 Mt (2007). Brazilian biodiesel production has just started, using mainly soybean oil, but many other options are being considered and implemented (SNO, CSO, PMO, jatropha, castor). Brazil’s initial involvement in biodiesel production (by ethanolysis of SBO) is clearly too restrictive. The regulation of biodiesel production in Brazil does not limit the use of vegetable oils [100]. Babassu palm (17 Mha) could produce at least 20 Mt oil/yr. The country could easily become a major renewable fuel supplier, both for bioethanol and biodiesel.

Palm diesel has been established as a diesel substitute since 1996. Malaysia is the second largest producer of palm oil and operates a 500 kt/yr biodiesel plant (2003). Malaysia intends to utilise crude palm oil and RBD palm oil (olein) at 5-10% blend level directly, or as palm diesel, as industrial and transport fuels, also in order to achieve price stabilisation of the palm oil market during periods of high stock levels. Malaysian palm oil methyl ester was already used in 1987 as biodiesel in buses. With a limitation in the CFPP of 284 K, this type of biodiesel faces a major obstacle in colder climatic conditions, where it may be used in multi-feedstock mode.

The use of coconut biodiesel is being tested by the Philippine Coconut Authority. In Thailand various mixes of CNO and PMO blended with diesel oil have been introduced without meeting the official standards for commercial use. P. R. China is one of the biggest diesel oil consumers worldwide (60-70 Mt) with a forecast growth of 6.4% from 1999 to 2020, thus increasing its share for world energy use for transportation to 9.1%. The country is highly import dependent and committed to non-food biodiesel feedstocks. Bioethanol development is much faster than that of biodiesel. Japan has interest in importation of coco-biodiesel. A recent overview of the Asian biodiesel industry can be found in Wahid [24]. See also Section 14.6.4.

Safflower or False saffron (Carthamus tinctorius L., Asteraceae family) is an annual drought-resistant niche crop, which grows in the temperate zone where wheat and barley do well. India is the largest producer of safflower, which is grown exclusively for its food oil, which is high in essential unsaturated fatty acids. Around half of US acreage is in California. The cost of growing safflower runs just a little higher per hectare than barley.

Traditional plant breeding and plant biotechnology play an important role in developing safflower, sunflower, soybean, rapeseed, linseed and peanut varieties with increased seed oil yields and fatty acid profiles that more closely meet the requirements for food and non-food use, including drought-tolerant varieties that have low fertiliser and pesticide requirements [94]. Genetic engineering, which enables development of crops producing seed oils with tailored fatty acid compositions [64, 66], may uncover new horizons in biodiesel production. Most alterations in the oil quality of cultivated plants have concentrated on rapeseed (B. napus). High-oleic (HO) canola, safflower, and sunflower oils, as well as low-linolenic (LoLn) canola and soybean oils, have been developed. HO safflower with 77.4% C18:1 (SFA 7.0%, MUFA 78.0%, PUFA 15.0%) is characterised by a low iodine value (80-100 g I2/ hg) and high CN number (52.2). For further details, see ref. [95].

Turkish safflower seed oil conversion to methyl esters as a biodiesel has been described [96]. Safflower oil does not meet European EN 14214 biodiesel specifications due to high IV (126-150 g I2/hg) and low CN (30).

The catalyst employed has a direct impact on the purity of the feedstock required, the kinetics of the reaction, and the extent of post-reaction processing necessary. Use of a solid catalyst simplifies downstream purification of biodiesel. The search for improved catalysts is being pursued intensely.

Sodium alkoxides are among the most efficient catalysts for transesterification, although KOH and NaOH can also be used. Transesterifications occur considerably faster in the presence of equivalent amounts of alkaline rather than (even more corrosive) homogeneous acidic catalysts. Suitable alkaline catalyst concentrations achieving high yields (>95%) are in the range of 0.5 to 1.0 wt%. Increase in catalyst concentration adds extra costs to the removal from the reaction medium at the end. Sodium methylate, used at approximately 0.5 wt% concentration for most transesterification processes, is estimated to represent 70% of the global biodiesel catalyst market. Sodium methylate is usually supplied as 25-30% solution in anhydrous methanol. Humidity can decrease the effectiveness of the catalyst.

Although higher conversions are achieved at higher catalyst concentrations under otherwise identical flow and stoichiometric conditions, there are drawbacks to the excessive use of alkaline catalyst in the transesterification reaction. Higher catalyst concentrations increase the solubility of methyl esters in the glycerol phase of the final product. As a result, a significant amount of methyl esters remains in the glycerol phase after the phase separation [47]. (See also Chapter 8 for biodiesel catalysis).

The difficulty of reproducing reaction kinetics results for transesterification of vegetable oils has been partly attributed to catalytic effects at the surfaces of the reaction vessels [49]. The effects are greatest at higher temperatures and may cause scaling-up problems in plant design.

It has also been shown that lipases are able to catalyse the alcoholysis of triglycerides in both aqueous and non-aqueous systems (see Chapter 10). The possibility of applying this procedure to the synthesis of biodiesel from low-value materials such as soapstock has been suggested [50].

As indicated in Table 3.3, lipid biosynthesis is also affected by environmental factors (light, temperature, water stress, soil and atmospheric constituents) and biotic factors such as physical damage and pest attack, which play an important quantitative and qualitative role in terms of oil yield and fatty acid profile. Ecological factors (and consequently some geographical areas) are

Po

Table 3.3 Some factors affecting lipid biosynthesis

• Genetic variations

• Enzyme activity (synthetases, desaturases, acyl-tranferases, thioesterases, etc.)

• Ecological and environmental factors (climate conditions: light temperature; geographical area: soil and atmospheric constituents)

• Water management (water regime and water stress)

• Souring date

• Seasonal variations

• Seed development stage (maturity, post-maturity)

• Bioetic factors (physical damage, pest attack)

more favourable than others for the production of a given botanical species as an industrial oilseed crop. Oil content varies by region due to climatic and other environmental factors. For example, spring canola in Oregon has significantly lower oil content (32%) than that of North Dakota and Canada (43%). Also the fatty acid profile of sunflower oil is greatly affected by climatic conditions (N. Dakota: C18 : 1, 15%; C18 : 2, 75%; Texas: C18 : 1, 50%; C18 : 2, 35%).

High irradiation may improve the oil content of a crop. Differences of oil content among olives located in different parts of the tree have been noticed [41]. Seasonal variations in vegetable oil yield and quality are observed. Temperature affects lipid metabolism. The effects of air temperature variations
on oil content and fatty acid composition of plants are well documented [9, 42-47], as shown for RSO, SNO, SFO, LSO, CAS and others, and generally show an increased synthesis of total cellular lipids and unsaturated fatty acids at lower temperature. The activity of diacylglycerol acyltransferase (DAGAT) is limited within the TAG production process in olives at T > 303 K [11], which agrees with the reported instability of this enzyme in Brassica napus at T > 313 K [48]. Temperature also has an effect on the catabolic pathways related to lipid metabolism. For example, lipolytic activity increases following exposure to chilling temperatures in oil palm and leads to increased free fatty acid levels [49]. HPL is rather unstable above 293 K.

As most of the palm oil is synthesised in the last two weeks of fruit development, a correct judgement of ripeness is essential to ensure good yield. Over-ripeness leads to biodeterioration and poor quality oil and lower mill efficiency. During seed development, the fatty acid composition changes considerably with maturity or with air temperature, as observed for Lesquerella fendleri in relation to harvest date [50]. Triglyceride and fatty acid compositions in the mesocarp of avocado (Persea americana) during fruit development were reported [51]. In the period between harvesting and full maturation (post-harvest period), the oil content increases and change occurs in the oil composition [52]. Water management is another important factor in seed production. Water stress during peak flowering and seed development reduces yields. Changes in seed yield and fatty acid composition of oils of high oleic sunflower (Helianthus annuus L.) hybrids in relation to the sowing date and water regime were reported [53]. Water stress influences the C18 : 1/C18 : 2 ratio. The attack of pests causes an appreciable loss of yield in most oil fruits and causes a loss of quality of the final product due to the increase of acidity and appearance of off-flavours.

The involvement of the LOX pathway in plant wound response mechanisms is well known [38]. In addition, the production of volatile aldehydes through the LOX pathway is also related to plant defence mechanisms.

Rendering is the (dry or wet) extraction of fat or oil from animal tissues using heat. The rendering industry classifies used cooking oils (UCOs) and fats into different categories, but not in a uniform way worldwide. Grades of fat produced by the rendering industry are ranked according to different specifications including total fatty acids, free fatty acids, moisture, unsaponifiable and insoluble matter. Trade practice is to designate animal fats with titres (solidification point of fatty acids) of at least 313 K as tallow and below 313 K as grease. Greases are also referred to as recovered vegetable oils. Used cooking oils typically contain 2-7 wt% FFA and crude animal fats generally some 10 wt%, but occasionally much more. Tallow (TLW), a volatile commodity, is beef fat originating from slaughterhouses. Feedstocks with FFA levels in excess of ~4 wt% are American lower-quality animal fats (prime tallow, special tallow, ‘A’ tallow and poultry fat) [279] and British tallows no. 3-6 [280]. Lard is an edible grease. Choice white grease is a specific grade of mostly pork fat.

Recycled grease products are referred to as waste grease. Greases are generally classified into two categories: yellow and brown grease. Yellow grease (typically containing some 7-20% FFA), sometimes referred to simply as waste oil, may be of vegetable or animal origin and is usually obtained from waste cooking oils, primarily from fast food restaurants and school cafeterias. Another source is from rendered animal (feed) fats (poorer pork and beef sources) that have been heated and used for cooking a wide variety of meat, fish or vegetable products. Various chemical and physical properties of oils and fats are affected by frying. FFA contents increase (due to hydrolysis and oxidation) and typically change from 0.04% in fresh SBO to 1.51% after 70 h of frying at 463 K [281]. Viscosity also increases (due to polymerisation), as well as AV, SV and specific gravity, but IV decreases.

#2 Yellow grease cannot exceed 10 wt% FFA content, and #3 yellow grease can reach up to 20 wt%. The difference between yellow grease and raw used cooking oils is the amount of processing involved. Renderers remove solids and moisture until it meets industry specifications for yellow grease. To use waste cooking oils for biodiesel production, all that is required is filtering and dewatering. The spot price for yellow grease (max. 10 wt% FFA, tanks) on 1 March 2004 from Chemical Market Reporter was US$0.13/ lb. Meanwhile, fryer grease has become much more valuable: its value has increased to historic highs in recent months (US$0.40/lb by mid July 2008); see also Table 14.21. Not surprisingly, cooking oil theft has recently become a new phenomenon.

In the US, yellow grease is required to have an FFA content below 15 wt%, otherwise it is classified as brown (or trap) grease [279]; the British classification system identifies one category of grease with FFA < 20 wt% [280]. Trap grease refers to grease collected from traps installed in sewage facilities which separate water from oil and ‘black grease’ (sludge) in wastewater. Grease traps (interceptors) are installed in drain lines of restaurants and similar facilities. Trap grease is abundant in urban centres. Brown or trap grease is not suitable for animal feed due to its highly contaminated state. Brown grease generally has an FFA content of over 20 wt% and can reach levels as high as 50-100 wt%. Lipid content is minimal; brown grease is gelatinous at room temperature. Grease trap waste encompasses a highly variable mixture of organic fats, greases and vegetable oils, comprising a high free fatty acid and mono-, di — and triglyceride content, contaminated by detergents, water, soaps, inorganic and particulate matter. Typically, the grease trap waste is between 80 and 100 wt% free fatty acids and glycerides (MG, DG, TG). Major components of grease trap waste are octadecanoic acid and n-hexadecanoic acid [267].

The production of large quantities of waste vegetable oils (WVO) is common to almost all world cuisines. Though cheap, used frying oils are limited in their availability (mainly in urban areas) and require a recycling infrastructure involving citizens, companies and local administrations. Cooking oils may be blends (e. g. composed of SBO, RSO and CRO). Waste cooking oil offers some potential as an alternative low-cost biodiesel feedstock whose availability is not affected by land use policies. It is estimated that 0.7-1

Mt/yr could be collected within the EU. The first plant for the production of methyl esters from used frying oil was put into operation in Mureck (Austria) in 1994 [282]. Austria recycles over 1300 t/yr of used cooking oils into biodiesel. The UK transforms about 75 kt of waste vegetable oil each year (total potential: 100 kt/yr).

As much of Asian food is fried, there is locally plenty of yellow grease available. Until 1999, in Japan 90% of the considerable amounts of used edible oils (700 kt SBO, 80 kt CNO, 50 kt CSO, and 700 kt RSO yearly) were dumped as refuse without collection; only 10% of the waste edible oil was recycled to be employed as a raw material for oleochemical purposes (soaps or the like). This has brought calls for promotional action for the effective utilisation of food resources and environmental protection [283]. In recent years, biodiesel has been produced from waste edible oils on a pilot scale in Kyoto City. Leung and Chen [251] have investigated biodiesel production technologies using waste cooking oils from Hong Kong restaurants.

The US produces large amounts of used cooking oils and animal fats (about 13 billion lbs) [284], reflecting the dominance of fried restaurant foods in the country. The 13 700 McDonald’s restaurants in the US use more than 35 kt/yr of frying oil. A large fraction of low value (< US$1/gal) fats, oils and grease (FOGs) is used for soap, cattle feed and other products, whereas only a minor proportion of these available raw materials is used in biodiesel production. Biodiesel legislation under the 2004 America Jobs Creation Act offered fuel blenders up to US$0.50 of federal support for each gallon of biodiesel made from used cooking oil (‘from McDonald’s to McDiesel’); this has recently been extended to a full tax credit (US$1/gal).

Y ellow grease is readily available at low cost. Production costs, including collection and (pre)processing, are about US$2/gal. Yellow grease lends itself best to transesterification in small biodiesel facilities (up to 5 MMgy), located in urban centres [285]. Typically, a medium-sized US city (e. g. Philadelphia) generates about 1 Mgy of yellow grease. Tellurian and Golden State Foods have formed a partnership for use of UCO. In Nov. 2007 San Francisco launched the SFGreasecycle program to turn FOGs into 1.5 Mgy of fuel for its municipal fleets. Used cooking oils are collected free of charge from food-service establishments. The city also intends producing biodiesel from brown grease; a demo plant is under development (Tellurian Biodiesel Inc., Los Angeles, CA). More recently, a Renew Energy Resources (Tampa, FL) subsidiary (Legacy Oil LCC), partnering with the Doe Fund (New York), has announced production of 5 MMgy biodiesel from New York City waste cooking oil and trap grease. Used US cooking oil has also been shipped to The Netherlands for biodiesel production.

The biodiesel industry classifies feedstocks by their FFA content:

• refined oils, such as canola or soybean oil (FFA < 1.5 wt%),

• low free fatty acid yellow greases and animal fats (FFA < 5 wt%); and

• high free fatty acid greases and animal fats (FFA > 20 wt%).

Waste oils and fats generally constitute a very heterogeneous mixture, including particulate matter, organic impurities, oxidative degradation and polymerisation products, free fatty acids and water. As levels of FFAs in feedstocks from a rendering plant may vary typically from 0.7% to 42%, and moisture from 0.01% to 55%, consequently a very robust biodiesel process is needed for conversion of waste grease to biodiesel, which tolerates a wide range of feedstock properties. Because of the large amounts of water, FFAs and debris (sediments) in yellow grease, most biodiesel producers are unable to use this inconsistent raw material, which requires pre-processing. It is not possible to convert such feedstocks to biodiesel using a single-step process; a two-step process is needed [286] (see Chapter 9). After reduction of the acid value to less than 2 mg KOH/g with an acid-catalysed pre-treatment process, the reaction may then be completed with alkali-catalysed transesterification. Water formation is the primary mechanism limiting the completion of acid — catalysed esterification with FFAs. Vegetable oils as diesel fuel have been investigated more extensively than animal fats. Waste vegetable oil contains more FFAs than virgin oils. Free fatty acids raise the cloud point of the fuel, so biodiesel made from used cooking oil or animal fat will cloud at higher temperatures than biodiesel made from virgin vegetable oil feedstock; yellow grease clouds at intermediate values.

Similar to waste cooking oils, animal fats are a cheap alternative feedstock source of limited availability only. Up to the year 2000, UCO and TLW went into animal feeds but this is no longer allowed according to EU rulings (Animal By-Products Regulation, EC 1774/2002), as a reaction to mad cow disease (bovine spongiform encephalopathy, BSE) in 2000-01. Rendered fats are now classified into three different risk categories: categories I and II can no longer be fed to animals, whereas category III is no-risk material. Import and export of rendered fats from category I and II material is nowadays limited by restrictive regulations and markets are country-specific. EU laws restrict import of tallow unless it is from Category III raw material (material that is edible); the import of used cooking oil for animal feedstocks is banned. EC Regulation 92/2005 has enabled use of Category I tallow in biodiesel production as long as the tallow is processed at 406 K and 0.3 MPa pressure for 20 min, a system referred to as ‘pressure cooking’. If the raw material is Category III, then pressure cooking is not required, but such fat can usually gain a premium in the marketplace, making it potentially uneconomic for biodiesel production. Prices vary considerably, from 740-150/t for risk material to 7230-270/t for no-risk material (Austria, 2005); cf. contemporary RSO price of 7550-650/t (UFOP, 2005). US tallow, which could be used for biodiesel production, often contains dead stock in the raw material blend, which defines the tallow as Category I under European legislation, thus requiring ‘pressure cooking’, a system not used in the US.

Animal fat products, reviewed in ref. [287], are excellent sources for biodiesel due to their high cetane number (typically 56-62) and good stability. As a result of the BSE problem, which has created alarm for all products of animal origin, a significant amount of animal fat from so-called high-risk material (tallow, yellow grease, etc.), possibly contaminated with infectious prions, has now become available at an attractive price for industrial purposes such as biodiesel production. Exposure to the infectious prions responsible for BSE disease in cattle, through use of animal fats in biodiesel production, has been of potential concern. Analysis by EC [288], the FDA [289] and the WHO [290] has indicated that rendered animal fat is not an agent of transmission of BSE. A risk assessment for the biodiesel production from possibly BSE-contaminated fat has been published [291]. Animal fats have also been a main raw material for glycerol. In view of BSE, glycerol of vegetable origin is now being favoured for human contact. The BSE problem has also led to an increased EU biodiesel production based on used cooking oils.

The supply of cooking oil and tallow in the EU is not able to provide all the necessary biodiesel feedstock required. While the quantities of waste feedstocks are not sufficient to supply a large market, they can be used as blending agents to lower the overall costs. Germany and Austria have the most practical experience with collection and processing of recycled vegetable oils and fats. The German rendering company Saria started processing rendered fats into biodiesel, which — although not fulfilling the European standard EN 14214 in all parameters — fueled Saria’s own truck fleet. Austria produces some 15 kt of rendered fat category I and 10 kt category III (2004). Biodiesel from Austrian animal fats (category I) would substitute only about 0.4% of the annual Austrian diesel consumption.

The annual production of animal fats (tallow, lard and butter) is about 20-22 Mt/yr [292]. Animal fats used for alkyl ester production are mainly beef tallow [253, 259, 293-295] and lard [263], but also poultry fat [296] has been proposed. The very high content of saturated fatty acids of such feedstocks yield methyl esters with poor cold temperature properties, although the impact on B5 blends is only very slight. The cloud point of biodiesel from inedible tallow is 289 K, compared with 276 K for soy biodiesel and 270 K for canola. During the winter and transition period (1 Oct.-15 April), RME has to be added to keep yellow-grease-based European biodiesel fully fluid. Both CFPP and CN can be improved by alcoholysis of oils and fats with low unsaturation (< 20%) with branched alcohols (e. g. isopropanol) [297].

In the early years of the biodiesel industry, animal fat-based biodiesel has gained a poor reputation because quality standards were not always adhered to. Today, there are many high-quality (ASTM spec) animal fat — based biodiesel producers.

Brown grease can be processed into biodiesel but at higher cost and lower biodiesel yield than yellow grease or tallow. Production of biodiesel from this low-cost urban lipid source is being considered. It is possible to produce biodiesel meeting all ASTM D 6751 specifications using a variety of different trap grease feedstocks [271]. Major chemical challenges are numerous: odour, heavy emulsification, solid contamination, cold flow, high water content, heterogeneity, etc. Processing brown grease into biodiesel requires several pre-treatment steps (filtering, deodorising, bleaching, vacuum distilling, etc.). FFAs and water must be removed from the grease (e. g. by solvent extraction) since these components serve as an inhibitor for biodiesel fuel production [298]. The technology requires further development (as at the aforementioned Tellurian’s demo plant in San Francisco). Brown grease biodiesel facilities should also be located in urban areas. Fry-o-Diesel’s pilot facility (now Black-Gold Biofuels) in Philadelphia, PA, has been producing biodiesel from brown grease at low volumes. Similarly, BioFuel BoxCorp. (San Jose, CA) is a 100% brown grease-based biodiesel producer.

A resource assessment of edible and inedible beef tallow generations in the US has indicated more than 1.8 Mt/yr (for the period 1997-2001) in the 11 largest commercial cattle slaughtering states, which would at most equate to some 550 Mgy of biodiesel [299]. The potential of restaurant waste lipids as biodiesel feedstocks in the US was recently also evaluated by Canakci [272]. As shown in Table 5.32, there are large amounts of low-cost feedstocks, such as greases and rendered animal fat (11.6 billion lbs), which can be used in biodiesel production. Some estimates state that 2.3 billion lbs/yr of poultry fat may account for 300 MMgy of biodiesel, while 2.75 billion lbs/yr of yellow grease account for similar quantities. The inexpensive feedstocks, waste cooking oils, restaurant grease and animal fats, represent about one-third of the US total fats and oil production (about 35 billion lbs,

with SBO making up some 50%). If all of the US greases and animal fats were converted to biodiesel it would replace about 1.5 Bgy of diesel fuel. However, these materials are currently devoted to other industrial markets and animal feed. In practice, out of the 0.35 Bgy of yellow grease and 1.2 Bgy of other animal fats collected in the US (see Table 5.32) only 3-8% or 45-125 MMgy will eventually be available for biodiesel [301]. The US Census Bureau has recently begun tracking the use of animal fats and greases in biodiesel production. In 2007, 173.1 million lbs of animal fats and greases were used for US biodiesel production (or about 4% of total feedstock). The US biodiesel industry consumed approximately 2% of total US production of fats and greases in 2007. In June 2008 only 25 million lbs of yellow grease was used for biodiesel production. There is even more trap grease than yellow grease in the US. Compared with yellow grease, the brown grease industry is even more highly fragmented. Collection is problematic (with up to 95% consisting of water), as well as the transformation into biodiesel.

With the new US blenders tax credit extension applicable to recycled vegetable oils, more waste vegetable oils will be converted. In a 300 kt/yr Tyson Foods, Inc./Syntroleum (USA) joint venture fat-to-liquids project, tallow will first be gasified to CO and H2 and then catalytically converted to alkanes (Fischer-Tropsch). Tyson Foods (USA) produces 1 Mt/yr animal fat.

In Europe, Argent Energy operates one of the world’s largest biodiesel plants in Motherwell (Scotland), where it converts tallow from the meat industry and used cooking oil into biodiesel, which supplies 5% of Scotland’s diesel requirements. Ireland’s first commercial-scale 30 kt/yr biodiesel plant operated by Green Biofuels Ireland Ltd (Wexford County) uses inedible and waste oils.

In Abu Dhabi, the 3 MMgy facility of Emirates Biodiesel LLC (Al Ain) will use waste cooking oil and other inedible oils as feedstocks.

Although various oilseeds are grown in Australia and a multi-feedstock blend of oils (RSO, SNO, SBO) may be used for biodiesel, Australian production currently (2005) originates mainly from recycled frying oil and animal fats. A Transpacific Industries Group (TPI)/Australian Renewable Fuels (ARF) joint venture (50/50) intends producing 45 ML biodiesel in Brisbane using up to 50% of the feedstock through TPI’s used cooking oil collection business (2H 2007). BP will produce 100 ML of biofuels at its Bulwer Island refinery in Queensland; at least 50% will be derived from animal fat using new technology that uses hydrogen to convert tallow into biodiesel (see Section 15.4).

Brazil’s Bertin Group, a leading cattle farmer in the beef industry, produces 100 kt/yr of bovine tallow diesel in the state of Sao Paulo (as from June 2006), or almost 14% of the biodiesel required to comply with the country’s B2 target effective January 2006. Brazil has a cattle herd of 190 million.

5.10.1 Highly acidic feedstocks

Soapstock (SS) is a co-product of the extraction of oilseeds to produce edible oils. Annual US soybean soapstock production exceeds 450 kt. Soapstock, consisting of an emulsion of water, acylglycerols, phosphoglycerols and free fatty acids, cannot be converted into FAME by conventional alkaline catalysis, but requires acid-catalysed esterification and transesterification. As the resulting ester product contains high amounts of unsaponifiable matter [302], the need for purification in order to meet the required industry standards easily leads to uneconomic operation.

Also acid oils, by-products of alkali or physical refining of crude oils and fats, may be used as feedstock. Acid oil obtained by acidulation of soapstock contains free fatty acids, acylglycerols and other lipophilic compounds. Acid oils, which are cheaper than raw and refined oils, have been considered as an alternative source of biodiesel fuel [303, 304]. Transformation of soapstocks and acid oils into FAAEs is described in Sections 9.3.1 and 9.6.

Among the industrial waste oils suitable for production of biofuel, mention should be made of tall oil colophony (a mixture of free fatty acids and rosin acids with varying degrees of purity). Tall oil finds a number of uses in industries for the production of lacquers, paints, hydrophobic coatings, inks, etc. [305]. Some 25% of all plant-derived fatty acids used in the coatings industry comes from tall oil. Tall oil (from Swedish ‘tallolja’ or pine oil), also called liquid rosin, is a viscous yellow-black odorous liquid obtained as a by-product of the Kraft (or sulphate) process of wood pulp manufacture. Manufacturing of paper using the Kraft Mill process generates liquids called black liquor and black liquor soap. Black liquor is the resulting caustic from that process and consists of an aqueous solution of lignin residues, hemicellulose and inorganic chemicals used in pulping. Canada has a large black liquor resource (24 Mt of dry matter). Black liquor soaps are mainly composed of fatty and rosin acid soaps and unsaponifiable matter and minor amounts of partially soluble inorganic salts, lignin, mercaptans and polysulphides. Through acidulation black liquor soaps are transformed into crude tall oil (CTO), consisting of 20-60% fatty acids (TOFA) and 20-65% resin acids (TORA), together with some fatty alcohols, sterols and other alkyl hydrocarbon derivatives. In turn, CTO (AV 100-175 mg KOH/g; SN 120-180 mg KOH/g; IV 140-170 g I2/hg) can be processed, e. g. using distillation, to produce different fractions of distilled tall oil [306]. Tall oil fatty acid (TOFA) is cheap, consists mostly of oleic acid, and is a source of volatile fatty acids. TOFA is a low-cost alternative to tallow fatty acids for production of soaps and lubricants but without much of a price advantage over other such biodiesel feedstocks. World production of tall oil is 1.2 Mt/ yr (60% from the US). Tall oil fatty acids are easily converted into their methyl esters by reaction with methanol, whereas the resin acids are virtually

unesterified [277, 307]; see Section 9.7. Feng [308] described a tall oil-based

cetane enhancer for diesel fuels.