Biodiesel Production

4 Biodiesel Production

■

■

4.1

■

■

Basics of the Transesterification Reaction Jon Van Gerpen1 and Gerhard Knothe2 1

Department of Biological and Agricultural Engineering, Unversity of Idaho; and 2USDA, ARS, NCAUR, Peoria, IL

Introduction Four methods to reduce the high viscosity of vegetable oils to enable their use in common diesel engines without operational problems such as engine deposits have been investigated: blending with petrodiesel, pyrolysis, microemulsification (cosolvent blending), and transesterification Schwab et al., 1987. Transesterification is by far the most common method and will be dealt with in this chapter. Only the transesterification reaction leads to the products commonly known as biodiesel, i.e., alkyl esters of oils and fats. The other three methods are discussed in Chapter 10. The most commonly prepared esters are methyl esters, largely because methanol is the least expensive alcohol, although there are exceptions in some countries. In Brazil, for example, where ethanol is less expensive, ethyl esters are used as fuel. In addition to methanol and ethanol, esters of vegetable oils and animal fats with other low molecular weight alcohols were investigated for potential production and their biodiesel properties. Properties of various esters are listed in the tables in Appendix A. Table 4.1.A of this chapter contains a list of C –C alcohols and their relevant properties. Information 1 4 on vegetable oils and animal fats used as starting materials in the transesterification reaction as well as on resulting individual esters and esters of oils and fats appears in Appendix A. ■

31

■

32

Source: Weast et al., 1985–1986.

a

2-Methyl-2 propanol (tert-butanol)

Methanol Ethanol 1-Propanol 2-Propanol (iso-Propanol) 1-Butanol (n-Butanol) 2-Butanol 2-Methyl-1-propanol (iso-butanol)

CH3OH C2H5OH CH2OH-CH2-CH3 CH3-CHOH-CH3 CH3-CH2-CH2-CH2OH CH3-CHOH-CH2-CH3 CH2OH-CH-CH2-CH3 | CH3 CH3-CHOH-CH3 | CH3

Formula

Table 4.1.A. Properties of C1–C4 Alcoholsa.

74.123

32.042 46.069 60.096 60.096 74.123 74.123 74.123

Molecular weight

82.3

65 78.5 97.4 82.4 117.2 99.5 108

Boiling point (°C)

25.5

–93.9 –117.3 –126.5 –89.5 –89.5 – –

Melting point (°C)

0.788720/4

0.791420/4 0.789320/4 0.803520/4 0.785520/4 0.809820/4 0.808020/4 0.801820/4

Density (g.mL)

Basics of the Transesterification Reaction ■ 33

O CH2-O-C-R

CH 2-OH O

O Catalyst

CH-O-C-R

+

3 R ' OH

➞

3 R' -O-C-R

+

CH-OH

O CH2-O-C-R Triacylglycerol (Vegetable oil)

CH 2 -OH Alcohol

Alkyl ester (Biodiesel)

Glycerol

Fig. 4.1.1. The transesterification reaction. R is a mixture of various fatty acid chains. The alcohol used for producing biodiesel is usually methanol (R' = CH3).

In addition to vegetable oils and animal fats, other materials such as used frying oils can also be suitable for biodiesel production; however, changes in the reaction procedure frequently have to be made due to the presence of water or free fatty acids (FFA) in the materials. The present section discusses the transesterification reaction as it is most commonly applied to (refined) vegetable oils and related work. Alternative feedstocks and processes, briefly indicated here, will be discussed later. The general scheme of the transesterification reaction was presented in the introduction and is given here again in Fig. 4.1.1. Di- and monoacylglycerols are formed as intermediates in the transesterification reaction. Fig. 4.1.2 qualitatively depicts conversion vs. reaction time for a transesterification

Fig. 4.1.2. Qualitative plot of conversion in a progressing transesterification reaction indicating relative concentrations of vegetable oil (triacylglycerols), intermediary di- and monoacylglycerols, as well as methyl ester product. Actual details can vary from reaction to reaction as mentioned in the text.

34

■

J. Van Gerpen and G. Knothe

reaction taking into account the intermediary di- and monoacylglycerols. Actual details in this figure, such as the final order of concentration of the various glycerides at the end of the reaction and concentration maximums for di- and monoacylglycerols, may vary from reaction to reaction depending on conditions. The scale of the figure can also vary if concentration (in mol/L) is plotted vs. time instead of conversion. Several reviews dealing with the production of biodiesel by transesterification have been published (Bondioli, 2004; Hoydonckx et al., 2004; Demirbas, 2003; Shah et al., 2003; Haas et al., 2002; Fukuda et al., 2001; Ma & Hanna, 1999; Schuchardt et al., 1998; Gutsche, 1997). Accordingly, the production of biodiesel by transesterification has been the subject of numerous research papers. Generally, transesterification can proceed by base or acid catalysis (for other transesterification processes, see the next section). However, in homogeneous catalysis, alkali catalysis (sodium or potassium hydroxide; or the corresponding alkoxides) is a much more rapid process than acid catalysis (Freedman & Pryde, 1982; Freedman et al., 1984; Canakci & Van Gerpen, 1999). In addition to the type of catalyst (alkaline vs. acidic), reaction parameters of base-catalyzed transesterification that were studied include the molar ratio of alcohol to vegetable oil, temperature, reaction time, degree of refinement of the vegetable oil, and effect of the presence of moisture and FFA (Freedman et al., 1984). For the transesterification to give maximum yield, the alcohol should be free of moisture and the FFA content of the oil should be <0.5% (Freedman et al., 1984). The absence of moisture in the transesterification reaction is important because according to the equation (shown for methyl esters), R-COOCH + H O → R-COOH + CH OH (R = alkyl) 3

2

3

hydrolysis of the formed alkyl esters to FFA can occur. Similarly, because triacylglycerols are also esters, the reaction of the triacylglycerols with water can form FFA. At 32°C, transesterification was 99% complete in 4 h when using an alkaline catalyst (NaOH or NaOMe) (Freedman et al., 1984). At ≥60°C, using an alcohol:oil molar ratio of at least 6:1 and fully refined oils, the reaction was complete in 1 h, yielding methyl, ethyl, or butyl esters (Freedman et al., 1984). Although the crude oils could be transesterified, ester yields were reduced because of gums and extraneous material present in the crude oils. These parameters (60°C reaction temperature and 6:1 methanol:oil molar ratio) have become a standard for methanol-based transesterification. Similar molar ratios and temperatures were reported in earlier literature (Feuge & Gros, 1949; Gauglitz & Lehman, 1963; Lehman & Gauglitz, Jr., 1966; Kurz, 1937). Other alcohols (ethanol and butanol) require higher temperatures (75 and 114°C, respectively) for optimum conversion (Freedman et al., 1984). Alkoxides in solution with the corresponding alcohol [made either by reacting the metal directly with alcohol or by electrolysis of salts and subsequent reaction with alcohol (Markolwitz, 2004)] have the advantage over hydroxides that the water-forming reaction according to the equation: R′OH + XOH → R′OX + H O (R′ = alkyl; X = Na or K) 2

cannot occur in the reaction system, thus ensuring that the transesterification reaction system remains as water free as possible. This reaction, however, is the one forming the transesterification-causing alkoxide when using NaOH or KOH as catalysts. The catalysts

Basics of the Transesterification Reaction ■ 35

are hygroscopic; precautions, such as blanketing with nitro gen, must be taken to prevent contact with moisture. The use of alkoxides reportedly also results in glycerol of higher purity after the reaction. Effects similar to those discussed above were observed in studies on the transesterification of beef tallow (Ma et al., 1998; Ma et al., 1999. FFA and, even more importantly, water should be kept as low as possible (Ma et al., 1998). NaOH reportedly was more effective than the alkoxide (Ma et al., 1998); however, this may have been a result of the reaction conditions. Mixing was important due to the immiscibility of NaOH/MeOH with beef tallow, with smaller NaOH/MeOH droplets resulting in faster transesterification (Ma et al., 1999). Ethanol is more soluble in beef tallow which increased yield (Ma et al., 1998), an observation that should hold for other feedstocks as well. Other work reported the use of both NaOH and KOH in the transesterification of rapeseed oil (Mittelbach et al., 1983). Recent work on producing biodiesel from waste frying oils employed KOH. With the reaction conducted at ambient pressure and temperature, conversion rates of 80–90% were achieved within 5 min, even when stoichiometric amounts of methanol were employed (Ahn et al., 1995). In two transesterifications (with more MeOH/KOH steps added to the methyl esters after the first step), the ester yields were 99%. It was concluded that an FFA content up to 3% in the feedstock did not affect the process negatively, and phosphatides up to 300 ppm phosphorus were acceptable. The resulting methyl ester met the quality requirements for Austrian and European biodiesel without further treatment. In a study similar to previous work on the transesterification of soybean oil (Freedman & Pryde, 1982; Freedman et al., 1984, it was concluded that KOH is preferable to NaOH in the transesterification of safflower oil of Turkish origin (Isigigur et al., 1994). The optimal conditions were given as 1 wt% KOH at 69 ± 1°C with a 7:1 alcohol:vegetable oil molar ratio to give 97.7% methyl ester yield in 18 min. Depending on the vegetable oil and its component fatty acids influencing FFA content, adjustments to the alcohol:oil molar ratio and the amount of catalyst may be required as was reported for the alkaline transesterification of Brassica carinata oil (Dorado et al., 2004). In principle, transesterification is a reversible reaction, although in the production of vegetable oil alkyl esters, i.e., biodiesel, the back reaction does not occur or is negligible largely because the glycerol formed is not miscible with the product, leading to a two-phase system. The transesterification of soybean oil with methanol or 1-butanol was reported to proceed (Freedman et al., 1986) with pseudo-first-order or second-order kinetics, depending on the molar ratio of alcohol to soybean oil (30:1 pseudo first order, 6:1 second order; NaOBu catalyst), whereas the reverse reaction was second order (Freedman et al., 1986). However, the originally reported kinetics (Freedman et al., 1986) were reinvestigated (Mittelbach & Trathnigg, 1990; Noureddini & Zhu, 1997; Boocock et al., 1996; Boocock et al., 1998) and differences were found. The methanolysis of sunflower oil at a molar ratio of methanol:sunflower oil of 3:1 was reported to begin with second-order kinetics but then the rate decreased due to the formation of glycerol (Mittelbach & Trathnigg, 1990). A shunt reaction (a reaction in which all three positions of the triacylglycerol react virtually simultaneously to give three alkyl ester molecules and glycerol) originally proposed (Freedman et al., 1986) as part of the forward reaction was shown to be unlikely, that second-order kinetics are not followed, and that miscibility phenomena (Mittelbach

36

■

J. Van Gerpen and G. Knothe

& Trathnigg, 1990; Noureddini & Zhu, 1997; Boocock et al., 1996; Boocock et al., 1998) play a significant role. The reason is that the vegetable oil starting material and methanol are not well miscible. The miscibility phenomenon results in a lag time in the formation of methyl esters as indicated qualitatively in Fig. 4.1.2. The formation of glycerol from triacylglycerols proceeds stepwise via the di- and monoacylglycerols, with a fatty acid alkyl ester molecule being formed in each step. From the observation that diacylglycerols reach their maximum concentration before the monoacylglycerols, it was concluded that the last step, formation of glycerol from monoacylglycerols, proceeds more rapidly than the formation of monoacylglycerols from diacylglycerols (Komers et al., 2001). The addition of cosolvents such as tetrahydrofuran (THF) or methyl tert-butyl ether (MTBE) to the methanolysis reaction was reported to significantly accelerate the methanolysis of vegetable oils as a result of solubilizing methanol in the oil to a rate comparable to that of the faster butanolysis (Boocock et al., 1996; Boocock et al., 1998; Komers et al., 2001; Boocock et al., 1996; Zhou et al., 2003; Boocock, 2001). This is to overcome the limited miscibility of alcohol and oil at the early reaction stage, creating a single phase. The technique is applicable for use with other alcohols and for acid-catalyzed pretreatment of high FFA feedstocks. However, molar ratios of alcohol:oil and other parameters are affected by the addition of the cosolvents. There is also some additional complexity due to recovering and recycling the cosolvent, although this can be simplified by choosing a cosolvent with a boiling point near that of the alcohol being used. In addition, there may be some hazards associated with its most common cosolvents, THF and MTBE. Other possibilities for accelerating the transesterification are microwave (Breccia et al., 1999) or ultrasonic (Stavarache et al., 2003; Lifka & Ondruschka, 2004) irradiation. Factorial experiment design and surface response methodology were applied to different production systems (Vicente et al., 1998) and are also discussed in the next section. A continuous pilot plant-scale process for producing methyl esters with conversion rates >98% was reported (Noureddini et al., 1998; Peterson et al., 2002) as well as a discontinuous two-stage process with a total methanol:acyl (from triacylglycerols) ratio of 4:3 (Cvengros˘ & Povazanec, 1996). Other basic materials, such as alkylguanidines, which were anchored to or entrapped in various supporting materials such as polystyrene and zeolite (Sercheli et al., 1999), also catalyze transesterification. Such systems may provide for easier catalyst recovery and reuse.

Industrial Production The chemistry described above forms the basis of the industrial production of biodiesel. Also, biodiesel processing and quality are closely related. The processes used to refine the feedstock and convert it to biodiesel determine whether the fuel will meet the applicable specifications. This section briefly describes the processing and production of biodiesel and how these determine fuel quality. The emphasis is on processing as it is conducted in the United States, where most biodiesel is produced by reacting soybean oil or used cooking oils with methanol and the standard for fuel quality is ASTM D 6751. For alkali-catalyzed transesterification, Fig. 4.1.3 shows a schematic diagram of the processes involved in biodiesel production from feedstocks containing low levels of FFA.

Basics of the Transesterification Reaction ■ 37

Fig. 4.1.3. Process flow scheme for biodiesel production.

These include soybean oil, canola (rapeseed) oil, and the higher grades of waste restaurant oils. Alcohol, catalyst, and oil are combined in a reactor and agitated for ~1 h at 60°C. Smaller plants often use batch reactors (Stidham et al., 2000) but most larger plants (>4 million L/yr) use continuous flow processes involving continuous stirred-tank reactors (CSTR) or plug flow reactors (Assman et al., 1996). The reaction is sometimes done in two steps in which ~80% of the alcohol and catalyst is added to the oil in a first-stage CSTR. Then, the product stream from this reactor goes through a glycerol removal step before entering a second CSTR. The remaining 20% of the alcohol and catalyst is added in this second reactor. This system provides a very complete reaction with the potential of using less alcohol than single-step systems. After the reaction, glycerol is removed from the methyl esters. Due to the low solubility of glycerol in the esters, this separation generally occurs quickly and can be accomplished with either a settling tank or a centrifuge. The excess methanol tends to act as a solubilizer and can slow the separation. However, this excess methanol is usually not removed from the reaction stream until after the glycerol and methyl esters are separated due to concern about reversing the transesterification reaction. Water may be added to the reaction mixture after the transesterification is complete to improve the separation of glycerol (Stidham et al., 2000; Wimmer, 1995). Some authors (Saka & Dadan, 1999; Saka & Kusdiana, 2001; Kusdiana & Saka, 2001; Dasari et al., 2003; Warabi et al., 2004; Diasakou et al., 1998) state that it is possible to react the oil and methanol without a catalyst, which eliminates the need for the water washing step. However, high temperatures and large excesses of methanol are required. The difficulty of reproducing the reaction kinetics results of other researchers was noted (Dasari et al., 2003) and was attributed to catalytic effects at the surfaces of the reaction vessels; it was also noted that these effects would be exacerbated at higher temperatures. Not including the effect of surface reactions could cause difficulties when scaling up reactors due to the decrease in the ratio of reactor surface area to volume. Kreutzer (Kreutzer, 1984) described

38

■

J. Van Gerpen and G. Knothe

how higher pressures and temperatures (90 bar, 240°C) can transesterify the fats without prior removal or conversion of the FFA. However, most biodiesel plants use lower temperatures, near atmospheric pressure, and longer reaction times to reduce equipment costs. Returning to Fig. 4.1.3, after separation from the glycerol, the methyl esters enter a neutralization step and then pass through a methanol stripper, usually a vacuum flash process or a falling film evaporator, before water washing. Acid is added to the biodiesel product to neutralize any residual catalyst and to split any soap that may have formed during the reaction. Soaps will react with the acid to form water-soluble salts and FFA according to the following equation: R-COONa + Sodium soap

HAc Acid

→

R-COOH Fatty acid

+

NaAc Salt

The salts will be removed during the water washing step and the FFA will stay in the biodiesel. The water washing step is intended to remove any remaining catalyst, soap, salts, methanol, or free glycerol from the biodiesel. Neutralization before washing reduces the amount of water required and minimizes the potential for emulsions to form when the wash water is added to the biodiesel. After the wash process, any remaining water is removed from the biodiesel by a vacuum flash process. The glycerol stream leaving the separator is only ~50% glycerol. It contains some of the excess methanol and most of the catalyst and soap. In this form, the glycerol has little value and disposal may be difficult. The methanol content requires the glycerol to be treated as hazardous waste. The first step in refining the glycerol is usually to add acid to split the soaps into FFA and salts. The FFA are not soluble in the glycerol and will rise to the top where they can be removed and recycled. Mittelbach and Koncar (1998) described a process for esterifying these FFA and then returning them to the transesterification reaction stream. The salts remain with the glycerol, although depending on the chemical compounds present, some may precipitate out. One frequently touted option is to use potassium hydroxide as the reaction catalyst and phosphoric acid for neutralization so that the salt formed is potassium phosphate, which can be used for fertilizer. After acidulation and separation of the FFA, the methanol in the glycerol is removed by a vacuum flash process, or another type of evaporator. At this point, the glycerol should have a purity of ~85% and is typically sold to a glycerol refiner. The glycerol refining process takes the purity up to 99.5–99.7% using vacuum distillation or ion exchange processes. Methanol that is removed from the methyl ester and glycerol streams will tend to collect any water that may have entered the process. This water should be removed in a distillation column before the methanol is returned to the process. This step is more difficult if an alcohol such as ethanol or isopropanol is used that forms an azeotrope with water. Then, a molecular sieve is used to remove the water.

Acid-Catalyzed Pretreatment Special processes are required if the oil or fat contains significant amounts of FFA. Used cooking oils typically contain 2–7% FFA, and animal fats contain 5–30% FFA. Some very low-quality feedstocks, such as trap grease, can approach 100% FFA. When an alkali

Basics of the Transesterification Reaction ■ 39

catalyst is added to these feedstocks, the FFA react with the catalyst to form soap and water as shown in the reaction below: R-COOH

+ KOH → R-COOK + Fatty acid Potassium hydroxide Potassium soap Water

H2O

Up to ~5% FFA, the reaction can still be catalyzed with an alkali catalyst, but additional catalyst must be added to compensate for that lost to soap. The soap created during the reaction is either removed with the glycerol or washed out during the water wash. When the FFA level is >5%, the soap inhibits separation of the glycerol from the methyl esters and contributes to emulsion formation during the water wash. For these cases, an acid catalyst such as sulfuric acid can be used to esterify the FFA to methyl esters as shown in the following reaction: R-COOH

+

CH3OH → R-COOCH3 Fatty acid Methanol Methyl ester Water

+

H2O

This process can be used as a pretreatment to convert the FFA to methyl esters, thereby reducing the FFA level (Fig. 4.1.4). Then, the low-FFA pretreated oil can be transesterified with an alkali catalyst to convert the triglycerides to methyl esters (Keim, 1945). As shown in the reaction, water is formed and, if it accumulates, it can stop the reaction well before completion. It was proposed (Kawahara & Ono, 1979) to allow the alcohol to separate from the pretreated oil or fat after the reaction. Removal of this alcohol also removes the water formed by the esterification reaction and allows for a second step of esterification; alternatively, one may proceed directly to alkali-catalyzed transesterification. Note that the methanol-water mixture will also contain some dissolved oil and FFA that should be

Fig. 4.1.4. Pretreatment process for feedstocks high in free fatty acids (FFA).

40

■

J. Van Gerpen and G. Knothe

recovered and reprocessed. Pretreatment with an acidic ion-exchange resin has also been described (Jeromin et al., 1987). It was shown (Haas et al., 2003; Haas et al., 2002) that acid-catalyzed esterification can be used to produce biodiesel from low-grade by-products of the oil refining industry such as soapstock. Soapstock, a mixture of water, soaps, and oil, is dried, saponified, and then esterified with methanol or some other simple alcohol using an inorganic acid as a catalyst. The procedure relies on a large excess of alcohol, and the cost of recovering this alcohol determines the feasibility of the process. More information is given in the next section.

Fuel Quality The primary criterion for biodiesel quality is adherence to the appropriate standard. Standards are listed in Appendix B. Generally, the fuel quality of biodiesel can be influenced by several factors, including the quality of the feedstock, the fatty acid composition of the parent vegetable oil or animal fat, the production process, the other materials used in this process, and postproduction parameters. When specifications are met, the biodiesel can be used in most modern engines without modifications while maintaining the engine’s durability and reliability. Even when used in low-level blends with petrodiesel fuel, biodiesel is expected to meet the standard before being blended. Although some properties in the standards, such as cetane number and density, reflect the properties of the chemical compounds that make up biodiesel, other properties provide indications of the quality of the production process. Generally, the parameters given in ASTM D6751 are defined by other ASTM standards and those in EN 14214 by other European or international (ISO) standards. However, other test methods, such as those developed by professional oleochemical organizations, such as the American Oil Chemists’ Society (AOCS), may also be suitable (or even more appropriate because they were developed for fats and oils and not for petroleum-derived materials addressed in the ASTM standards). This discussion will focus on the most important issues for ensuring product quality for biodiesel as it relates to production as well as some postproduction parameters.

Production Process Factors The most important issue during biodiesel production is the completeness of the transesterification reaction. The basic chemical process that occurs during the reaction is indicated in Fig. 4.1.2 with the reaction proceeding stepwise from triacylglycerols to glycerol and alkyl esters with each step producing a fatty acid alkyl ester. Even after a fully “complete” transesterification reaction, small amounts of tri-, di-, and monoacylglycerols will remain in the biodiesel product. The glycerol portion of the acylglycerols is summarily referred to as bound glycerol. When the bound glycerol is added to the free glycerol remaining in the product, the sum is known as the total glycerol. Limits for bound and total glycerol are usually included in biodiesel standards. For example, ASTM D6751 requires <0.24% total glycerol in the final biodiesel product as measured

Basics of the Transesterification Reaction ■ 41

using a gas chromatographic (GC) method described in ASTM D 6584. Because the glycerol portion of the original oil is usually ~10.5%, this level of total glycerol corresponds to 97.7% reaction completion. Other methods can be used to measure total glycerol such as high-performance liquid chromatography (HPLC) (e.g., AOCS Recommended Practice Ca 14b-96: Quantification of Free Glycerine in Selected Glycerides and Fatty Acid Methyl Esters by HPLC with Laser Light-Scattering Detection) or a chemical procedure such as that described in AOCS Official Method Ca 14–56 (Total, Free and Combined Glycerol Iodometric Method). However, only the GC procedures are acceptable for demonstrating compliance with standards.

Free Glycerol Glycerol is essentially insoluble in biodiesel so that almost all glycerol is easily removed by settling or centrifugation. Free glycerol may remain either as suspended droplets or as the very small amount that does dissolve in the biodiesel. Alcohols can act as cosolvents to increase the solubility of glycerol in the biodiesel. Most glycerol should be removed from the biodiesel product during the water washing process. Water-washed fuel is generally very low in free glycerol, especially if hot water is used for washing. Distilled biodiesel tends to have a greater problem with free glycerol due to glycerol carry-over during distillation. Fuel with excessive free glycerol will usually have a problem with glycerol settling out in storage tanks, creating a very viscous mixture that can plug fuel filters and cause combustion problems in the engine.

Residual Alcohol and Residual Catalyst Because alcohols such as methanol and ethanol as well as the alkaline catalysts are more soluble in the polar glycerol phase, most will be removed when the glycerol is separated from the biodiesel. However, the biodiesel typically contains 2–4% methanol after the separation, which may constitute as much as 40% of the excess methanol from the reaction. Most processors will recover this methanol using a vacuum stripping process. Any methanol remaining after this stripping process should be removed by the water washing process. Therefore, the residual alcohol level in the biodiesel should be very low. A specific value for the allowable methanol level is specified in European biodiesel standards (0.2% in EN 14214), but is not included in the ASTM standard; however, the flash point specification in both standards limits the alcohol level. Tests showed that as little as 1% methanol in the biodiesel can lower its flashpoint from 170°C to <40ºC. Therefore, by including a flashpoint specification of 130ºC, the ASTM standard limits the amount of alcohol to a very low level (<0.1%). Residual alcohol left in biodiesel will generally be too small to have a negative effect on fuel performance. However, lowering the flashpoint presents a potential safety hazard because the fuel may have to be treated more like gasoline (which has a low flashpoint) than diesel fuel. Most of the residual catalyst is removed with the glycerol. Like the alcohol, remaining catalyst in the biodiesel should be removed during the water washing. Although a value for residual catalyst is not included in the ASTM standard, it will be limited by the

42

■

J. Van Gerpen and G. Knothe

specification on sulfated ash. Excessive ash in the fuel can lead to engine deposits and high abrasive wear levels. The European standard EN 14214 places limits on calcium and magnesium as well as the alkali metals sodium and potassium.

Postproduction Factors Water and Sediment These two items are largely housekeeping issues for biodiesel. Water can be present in two forms, either as dissolved water or as suspended water droplets. Although biodiesel is generally insoluble in water, it actually takes up considerably more water than petrodiesel fuel. Biodiesel can contain as much as 1500 ppm of dissolved water, whereas diesel fuel usually takes up only ~50 ppm (Van Gerpen et al., 1997). The standards for diesel fuel (ASTM D 975) and biodiesel (ASTM D 6751) both limit the amount of water to 500 ppm. For petrodiesel fuel, this actually allows a small amount of suspended water. However, biodiesel must be kept dry. This is a challenge because many diesel storage tanks have water on the bottom due to condensation. Suspended water is a problem in fuel injection equipment because it contributes to the corrosion of the closely fitting parts in the fuel injection system. Water can also contribute to microbial growth in the fuel. This problem can occur in both biodiesel and petrodiesel fuel and can result in acidic fuel and sludge that will plug fuel filters. Sediment may consist of suspended rust and dirt particles or it may originate from the fuel as insoluble compounds formed during fuel oxidation. Some biodiesel users have noted that switching from petrodiesel to biodiesel causes an increase in sediment originating from deposits on the walls of fuel tanks that had previously contained petrodiesel fuel. Because its solvent properties are different from those of petrodiesel fuel, biodiesel may loosen these sediments and cause fuel filter plugging during the transition period.

Storage Stability Storage stability refers to the ability of the fuel to resist chemical changes during long-term storage; it is a major issue with biodiesel and is discussed at length in Chapter 6.4, Oxidative Stability of Biodiesel. Contact with air (oxidative stability) and water (hydrolytic stability) are the major factors affecting storage stability. Oxidation is usually accompanied by an increase in the acid value and viscosity of the fuel. Often these changes are accompanied by a darkening of the biodiesel color from yellow to brown and the development of a “paint” smell. In the presence of water, the esters can hydrolyze to long-chain FFA, which also cause the acid value to increase. The methods generally applied to petrodiesel fuels for assessing this issue, such as ASTM D 2274, were shown to be incompatible with biodiesel, and this remains an issue for research. Chapter 6.4 discusses some methods for assessing the oxidative stability of biodiesel that were or are being evaluated. Antioxidant additives such as butylated hydroxytoluene and t-butylhydroquinone were found to enhance the storage stability of biodiesel. Biodiesel produced from

Basics of the Transesterification Reaction ■ 43

soybean oil naturally contains some antioxidants (tocopherols, e.g., vitamin E), providing some protection against oxidation (some tocopherol is lost during refining of the oil before biodiesel production). Any fuel that is going to be stored for an extended period of time, whether it is petrodiesel or biodiesel, should be treated with an antioxidant additive.

Quality Control All biodiesel production facilities should be equipped with a laboratory so that the quality of the final biodiesel product can be monitored. To monitor the completeness of the reaction according to the total glycerol level specified requires GC analysis as called for in biodiesel standards. Analytical methods, including GC and other procedures, are discussed in more detail in Chapter 5. It is also important to monitor the quality of the feedstocks, which can often be limited to acid value and water contents, tests that are not too expensive. Another strategy used by many producers is to draw a sample of the oil (or alcohol) from each delivery and use that sample to produce biodiesel in the laboratory. This test can be fairly rapid (1–2 h) and can indicate whether serious problems may occur in the plant.

References Ahn, E., M. Koncar, M. Mittelbach, R. Marr. A low-waste process for the production of biodiesel. Sep. Sci. Technol. 1995, 30, 2021–2033. Assman, G., G. Blasey, B. Gutsche, L. Jeromin, J. Rigal, R. Armengand, B. Cormary. Continuous Progress for the Production of Lower Alkyl Esters, U.S. Patent 5,514,820, 1996. Bondioli, P. The preparation of fatty acid esters by means of catalytic reactions. Topics Catalysis 2004, 27, 77–82. Boocock, D.G.B., S.K. Konar, H. Sidi. phase diagrams for oil/methanol/ether mixtures. J. Am. Oil Chem. Soc. 1996, 73, 247–1251. Boocock, D.G.B., S.K. Konar, V. Mao, H. Sidi. Fast one-phase oil-rich processes for the preparation of vegetable oil methyl esters. Biomass Bioenergy 1996, 11, 43–50. Boocock, D.G.B., S.K. Konar, V. Mao, C. Lee, S. Buligan, Fast formation of high-purity methyl esters from vegetable oils. J. Am. Oil Chem. Soc. 1996, 75, 1167–1172. Boocock, D.G.B., Single-Phase Process for Production of Fatty Acid Methyl Esters from Mixtures of Triglycerides and Fatty Acids, Canadian Patent 2,381,394, 2001. Breccia, A., B. Esposito, G. Breccia Fratadocchi, A. Fini. Reaction between methanol and commercial seed oils under microwave irradiation. J. Microwave Power Electromagn. Energy 1999, 34, 3–8. Canakci, M., J. Van Gerpen. Biodiesel production via acid catalysis, Trans. ASAE 1999, 42, 1203–1210. Cvengros˘, J., F. Povazanec. Production and treatment of rapeseed oil methyl esters as alternative fuels for diesel engines. Bioresour. Technol. 1996, 55, 145–152. Dasari, M.A., M.J. Goff, G.J. Suppes, Non-catalytic alcoholysis kinetics of soybean oil. J. Am. Oil Chem. Soc. 2003, 80, 189–192.

44

■

J. Van Gerpen and G. Knothe

Demirbas, A. Biodiesel fuels from vegetable oils via catalytic and non-catalytic supercritical alcohol transesterifications and other methods: A survey. Energy Convers. Manag. 2003, 44, 2093–2109. Diasakou, M., A. Louloudi, N. Papayannakos. Kinetics of the non-catalytic transesterification of soybean oil. Fuel 1998, 77, 1297–1302. Dorado, M.P., E. Ballisteros, F.J. Lopez, M. Mittelbach. Optimization of alkali-catalyzed transesterification of Brassica carinata oil for biodiesel production. Energy Fuels 2004, 18, 77–83. Feuge, R.O., A.T. Gros. Modification of vegetable oils. vii.Alkali catalyzed interesterification of peanut oil with ethanol. J. Am. Oil Chem. Soc. 1949, 26, 97–102. Freedman, B., E.H. Pryde. Fatty Esters from Vegetable Oils for Use as a Diesel Fuel, in Vegetable Oil Fuels, Proceedings of the International Conference on Plant and Vegetable Oils as Fuels, Fargo, ND, 1982, ASAE Publication 4-82, pp. 117–122. Freedman, B., E.H. Pryde, T.L. Mounts. Variables affecting the yields of fatty esters from transesterified vegetable oils. J. Am. Oil Chem. Soc. 1984, 61, 1638–1643. Freedman, B., R.O. Butterfield, E.H. Pryde. Transesterification kinetics of soybean oil, J. Am. Oil Chem. Soc. 1986, 63, 1375–1380. Fukuda, H., A. Kondo, H. Noda. Biodiesel fuel production by transesterification of oils. J. Biosci. Bioeng. 2001, 92, 405–416. Gauglitz, E.J., Jr., L.W. Lehman. The preparation of alkyl esters from highly unsaturated triglycerides. J. Am. Oil Chem. Soc. 1963, 40, 197–198. Gutsche, B. Technologie der Methylesterherstellung—Anwendung für die Biodieselproduktion (Technology of methyl ester production and its application to biofuels), Fett/Lipid 1997, 99, 418–427. Haas, M.J., G.J. Piazza, T.A. Foglia. Enzymatic approaches to the production of biodiesel fuels, in Lipid Biotechnology, edited by T.M. Kuo H.W. Gardner, Marcel Dekker, New York, 2002, pp. 587–598. Haas, M.J., P.J. Michalski, S. Runyon, A. Nunez, K.M. Scott. Production of FAME from acid oil, A by-product of vegetable oil refining, J. Am. Oil Chem. Soc. 2003, 80, 97–102. Haas, M.J., S. Bloomer, K. Scott. Process for the Production of Fatty Acid Alkyl Esters, U.S. Patent 6,399,800, 2002. Hoydonckx, H.E., D.E. De Vos, S.A. Chavan, P.A. Jacobs. Esterification and transesterification of renewable chemicals. Topics Catalysis 2004, 27, 83–96. Isigigur, A., F. Karaosmanoôlu, H.A. Aksoy. Methyl ester from safflower seed oil of turkish origin as a biofuel for diesel engines, Appl. Biochem. Biotechnol. 1994, 45–46, 103–122. Jeromin, L., E. Peukert, G. Wollman. Process for the Pre-Esterification of Free Fatty Acids in Fats and Oils, U.S. Patent 4,698,186, 1987. Kawahara, Y., T. Ono. Process for Producing Lower Alcohol Esters of Fatty Acids, U.S. Patent 4,164,506, 1979. Keim, G.I. Treating Fats and Fatty Oils, U.S. Patent 2,383,601,1945. Komers, K., R. Stloukal, J. Machek, F. Skopal. Biodiesel from rapeseed oil, methanol and koh 3. Analysis of composition of actual reaction mixture. Eur. J. Lipid. Sci. Technol. 2001, 103, 359–362. Kreutzer, U.R. Manufacture of fatty alcohols based on natural fats and oils. J. Am. Oil Chem. Soc. 1984, 61, 343–348. Kurz, H. Zur katalytischen Umesterung fetter Oele durch alkoholische Kalilauge (The catalytic alcoholysis of fatty oils with alcoholic potash). Fette Seifen 1937, 44, 144–145.

Basics of the Transesterification Reaction ■ 45

Kusdiana, D., S. Saka. Kinetics of transesterification in rapeseed oil to biodiesel fuel as treated in supercritical methanol. Fuel 2001, 80, 693–698. Lehman, L.W., E.J. Gauglitz, Jr. The preparation of alkyl esters from highly unsaturated triglycerides. II. J. Am. Oil Chem. Soc. 1966, 43, 383–384. Lifka, J., B. Ondruschka, Einfluss des Stofftransportes auf die Herstellung von Biodiesel (Influence of mass transfer on the production of biodiesel), Chem. Ing. Techn. 2004, 76, 168–171 (). Ma, F., M.A. Hanna, Biodiesel Production: A Review, Bioresour. Technol. 1998, 70, 1–15. Ma, F., L.D. Clements, M.A. Hanna, biodiesel fuel from animal fat. Ancillary studies on transesterification from beef tallow, Ind. Eng. Chem. Res. 1998, 37, 3768–3771. Ma, F., L.D. Clements, M.A. Hanna, The effect of mixing on transesterification of beef tallow, Bioresour. Technol. 1999, 69, 289–293. Ma, F., L.D. Clements, M.A. Hanna, The effects of catalyst, free fatty acids, and water on transesterification of beef tallow, Trans. ASAE, 1998, 41, 1261–1264. Markolwitz, M., Consider Europe’s most popular catalyst, Biodiesel Magazine 2004, 1, 20–22. Mittelbach, M., B. Trathnigg, kinetics of alkaline catalyzed methanolysis of sunflower oil,. J. Am. Oil Chem. Soc. 1990, 92, 145–148. Mittelbach, M., M. Koncar, Method for the Preparation of Fatty Acid Alkyl Esters, U.S. Patent 5,849,939, 1998. Mittelbach, M., M. Wörgetter, J. Pernkopf, H. Junek, diesel fuel derived from vegetable oils: preparation and use of rape oil methyl ester, Energy Agric. 1983, 2, 369–384. Noureddini, H., D. Zhu, Kinetics of transesterification of soybean oil, J. Am. Oil Chem. Soc. 1997, 74, 1457–1463. Noureddini, H., D. Harkey, V. Medikonduru. A continuous process for the conversion of vegetable oils into methyl esters of fatty acids, J. Am. Oil Chem. Soc. 1998, 75, 1775–1783. Peterson, C.L., J.I. Cook, J.C. Thompson, J.S. Taberski. Continuous flow biodiesel production. Appl. Eng. Agric. 2002, 18, 5–11. Saka, S., D. Kusdiana. Biodiesel fuel from rapeseed oil as prepared in supercritical methanol. Fuel 2001, 80, 225–231. Saka, S., K. Dadan. Transesterification of Rapeseed Oils in Supercritical Methanol to Biodiesel Fuels, in Proceedings of the 4th Biomass Conference of the Americas, edited by R.P. Overend, E. Chornet, Oakland, CA, 1999. Schuchardt, U., R. Sercheli, R.M. Vargas. Transesterification of vegetable oils: A review, J. Braz. Chem. Soc. 1998, 9, 199–210. Schwab, A.W., M.O. Bagby, B. Freedman, Preparation and properties of diesel fuels from vegetable oils, Fuel 1987, 66, 1372–1378. Sercheli, R., R.M. Vargas, U. Schuchardt. Alkylguanidine-catalyzed heterogeneous transesterification of soybean oil. J. Am. Oil Chem. Soc. 1999, 76, 1207–1210. Shah, S., S. Sharma, M.N. Gupta, Enzymatic transesterification for biodiesel production, Indian J. Biochem. Biophys. 2003, 40, 392–399. Stavarache, C., M. Vinatoru, R. Nishimura, Y. Maeda, Conversion of vegetable oil to biodiesel using ultrasonic irradiation, Chem. Lett. 2003, 32, 716–717. Stidham, W.D., D.W. Seaman, M.F. Danzer, Method for Preparing a Lower Alkyl Ester Product from Vegetable Oil, U.S. Patent 6,127,560 (2000). Van Gerpen, J.H., E.H. Hammond, L. Yu, A. Monyem, Determining the Influence of Contaminants on Biodiesel Properties, SAE Technical Paper Series 971685, SAE, Warrendale, PA, 1997.

46

■

J. Van Gerpen and G. Knothe

Vicente, G., A. Coteron, M. Martinez, J. Aracil, Application of the Factorial Design of Experiments and Response Methodology to Optimize Biodiesel Production, Ind. Crops Prod. 2004, 8, 29–35. Warabi, Y., D. Kusdiana, S. Saka, Reactivity of triglycerides and fatty acids of rapeseed oil in supercritical alcohols, Bioresour. Technol. 2004, 91, 283–287. Weast, R.C., M.J. Astle, W.H. Beyer, eds., Handbook of Chemistry and Physics, 66th edn., CRC Press, Boca Raton, FL, 1985–1986. Wimmer, T., Process for the Production of Fatty Acid Esters of Lower Alcohols, U.S. Patent 5,399,731, 1995. Zhou, W., S.K. Konar, D.G.B. Boocock, ethyl esters from the single-phase base-catalyzed ethanolysis of vegetable oils. J. Am. Oil Chem. Soc. 2003, 80, 367–371.

■

■

4.2

■

■

Alternate Feedstocks and Technologies for Biodiesel Production Michael J. Haas U.S. Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center1 Wyndmoor, PA

Introduction Modern society is heavily dependent on the movement of materials under the power of internal combustion engines. National consumptions of petroleum fuels for this and other purposes in the more industrialized countries is expressed in units of hundreds of millions of metric tons/year. To introduce a liquid transportation fuel into this infrastructure is no simple feat, and will have numerous and substantial repercussions. The five years since publication of the first edition of this chapter (Haas & Foglia, 2005) were dynamic ones in the biodiesel industry as production volumes grew exponentially and the fuel went increasingly from a research and demonstration item to a component of national fuel systems. Such growth put stresses on conventional feedstock supplies and also stimulated a search for superior catalysts and technologies for biodiesel production. This chapter will examine some of the major initiatives in these areas. It is not possible to be all-inclusive in coverage of the scientific or patent literature in this area. Additional information can be found in other reviews of biodiesel production (including but not limited to Ma & Hanna, 1999; Narasimharao et al., 2007; Vasudevan & Briggs, 2008; Pinto et al., 2005; Sharma et al., 2008) and via web-based literature searches.

Biodiesel Production: The Lipid Reactant Refined Triglycerides: The ‘Traditional’ Feedstocks The predominant and established feedstocks for biodiesel production, worldwide, are refined vegetable oils. It is likely that this will remain so for some time. Within this group, the oil of choice varies geographically according to availability, the most abundant lipid generally being the most common feedstock. The reasons for this are not only the desire to have an 1

Mention of brand or firm names does not constitute an endorsement by the U.S. Department of Agriculture over others of a similar nature not mentioned. ■

47

■

48

■

M. J. Haas

ample supply of feedstock, but also the inverse relationship between supply and cost. The chemistry of biodiesel synthesis from refined lipids is relatively simple and inexpensive, and thus feedstock costs can constitute up to 80% or more of the total fuel production cost (Haas et al., 2006). In addition, because of their low levels of free fatty acids (FFA) and other contaminants, refined fats and oils can be converted to fatty acid alkyl esters (FAE) by simple alkaline transesterification, the lowest cost technology. These are strong drivers promoting the use of the cheapest available refined lipids as feedstocks. Thus, rapeseed and sunflower oils are used in the European Union, palm oil predominates in biodiesel production in tropical countries, and soybean oil and animal fats are major feedstocks in the United States. Refined vegetable oils (Freedman et al., 1984; Ali & Hanna, 1994a) and high quality animal fats (Ali & Hanna, 1994b) can be directly converted to FAE by alkali-catalyzed transesterification, the conventional biodiesel synthetic technology, in high efficiency and with good yields. However, the processing technology necessary to refine lipids to these degrees, as well as the demand for use as foods, confers prices above those of other lower quality lipids, discussed below. Of the two, animal fats are typically the less expensive because they are a by-product of animal agriculture, rather than a primary product, and because demand is lower. Animal fats contain a higher content of saturated fatty acids than do vegetable oils. These have relatively high melting points, a trait that may lead to precipitation and poor engine performance in cold weather (Dunn et al., 1996). Nonetheless, the successful use of animal fat based biodiesels in blends with petroleum diesel has been described (Technology Early Action Measures Program, 2003). The higher saturated fatty ester content of animal fat-derived biodiesels gives them higher cetane values than vegetable oil derived biodiesel, a desirable trait. There are numerous grades of tallow (BSI British Standards, 1987, American Fats & Oils Association, 2004), differentiated solely or largely on the basis of FFA content. Only the grades with lowest FFA levels are suitable for successful direct alkali-catalyzed transesterification and unique considerations necessary to obtain high degrees of reaction have been identified (Ma et al., 1998, 1999). Of potential concern with regard to the use of animal fats, especially bovine lipids is the possibility of exposure to prions, the infectious proteins responsible for mad cow disease in cattle and variant Creutzfeldt-Jacob (vCJD) disease in humans (Erdtmann & Sivitz, 2004). The Scientific Steering Committee of the European Commission has examined normal industrial tallow production processes and concluded that the resulting product is free of detectable BSE infectivity, even if the source material was highly infective (European Commission, 2000). The United States Food and Drug Administration has ruled that tallow and other rendered fats are safe, and specifically omitted them from regulations prohibiting rendered products in feeds for cattle and other ruminants (U.S. Food and Drug Administration, 1997). The United Nations World Health Organization has examined the issue, and concluded that since prions are proteinaceous they would partition with the cellular residues of meat and bone, rather than the non-polar lipid fraction during processing. The tallow fraction was therefore judged not a risk to human or animal health (World Health Organization, 2001). Similarly, when animal fat was spiked with scrapie material and converted to biodiesel, Western blotting demonstrated the progressive reductions of scrapie levels through the process, supporting the conclusion that

Alternate Feedstocks and Technologies ■ 49

the finished product did not pose a health risk (Seidel et al., 2006). An assessment of the danger of a human contracting vCJD due to the use of tallow as a fuel in diesel engines indicated that the risk was several orders of magnitude less than the rate of spontaneous appearance of CJD (Cummins et al., 2002). Thus, scientific analysis indicates that processed (i.e. rendered) animal fat is not an agent of transmission of BSE to biodiesel. Unrefined vegetable oils, that are those from which phospholipids (‘lecithin’, ‘gum’) have not been removed, can also be converted to FAE and can be 10 to 15% cheaper than highly refined oils. It is possible that there is some use of unrefined oils as biodiesel feedstocks in the industry, but it is not a widespread phenomenon. Gums readily precipitate from crude oils and thus they complicate feedstock handling and storage. Also, since they are good emulsifiers they can complicate the washing of crude biodiesel. If carried over into product they can cause a biodiesel to exceed allowed phosphorus levels. Thus, the degumming of vegetable oil feedstocks is generally practiced, though bleaching and deodorization, other common steps in the purification of crude edible oils, need not be conducted in order to produce an acceptable biodiesel feedstock (Kramer, 1995). Animal fats do not typically contain sufficient amounts of phospholipids to require degumming. Although this chapter will consider new sources of lipids for biodiesel production, the traditional sources will certainly remain in use into the foreseeable future. Lipid production from these will increase, as it has regularly for over a half century, due to yield improvements based on advances in horticultural practices and the selection and engineering of high yield lines.

Drivers for Expansion and Change in Feedstock Type The following factors stimulate the search for new biodiesel feedstocks: 1. Feedstock cost. Although the conventional feedstocks are among the lowest priced refined oils, they contribute significantly to overall fuel cost. Fuel buyers may be seriously concerned regarding global warming and the use of biobased products, but these sentiments are rarely strong enough to support a fuel industry that is not competitive with fossil fuel prices. It is notable that the less expensive feedstocks are often of lower quality, and may require more costly chemical technologies to convert them to biodiesel. 2. Governmental policy: Use mandates. Local or Federal governmental policy can commit a country to biofuels usage targets. Thus, in the United States the Energy Independence and Security Act of 2007 established a Renewable Fuels Standard pledging that country to the production of 120 million metric tons of biofuels by 2022 (Anonymous, 2007). Not all of this is to be biodiesel, but this action has established biodiesel and other biofuels as real components of the future fuel mix, thereby stimulating the development of new fuel feedstocks. 3. Governmental policy: Support programs. Government programs favoring one or the other feedstock could seriously impact feedstock choices. Thus, early support programs in the United States favored the use of first-use refined soybean oil as a

50

■

M. J. Haas

feedstock, though the program was subsequently modified to also support animal fats and ‘second use’ (previously used) fats and oils. Conversely, although Brazil is the world’s second largest producer of soybeans, the government has supported a castor oil-based feedstock initiative on the rationale that adequate markets for soy oil exist while the sale of castor oil into the biodiesel market would provide income to impoverished regions of the country where soy cannot be grown. 4. Food-fuel debate. Recent transient increases in edible oil prices, and the perception that their use in biofuels production was responsible for the increases, spawned accusations that the diversion of edible lipids into fuel had reduced their availability to humans at the lowest economic income levels. For example, Indian law forbids the production of biodiesel from materials that are human foods. This has led to government initiatives to develop a biodiesel industry around such inedible species as Jatropha curcas. Overall, this ‘food-vs.-fuel’ issue has caused many in the industry to turn their attention to the development of non-food lipids, be they inedible refined fats and oils, or lower quality lipids such as waste greases.

New Sources of High Quality Lipid Feedstocks In addition to the common lipids previously named, FAE production for use as biodiesel has been demonstrated from a variety of refined lipids. These include but are not restricted to the oils of coconut (Solly, 1980), rice bran (Kamini & Iefuji; 2003; Özgül-Yücel & Türkay, 2003), safflower (Isigigür et al., 1994), palm kernel (Choo et al., 1991; Cardone et al., 2003), Jatropha curcas (Foidl et al., 1996; Achten et al., 2008), mustard (Thompson & He, 2006), rapeseed (Thompson & He, 2006), canola (Thompson & He, 2006), crambe (Thompson & He, 2006), camelina (Fröhlich & Rice, 2005) and the oil of the halophytic plant salicornia (Desai et al., 2006). A thorough bibliography of reports of biodiesel production from these and other raw materials was presented by Mittelbach and Remschmidt (2004). In principle any animal or plant lipid should be a ready substrate for the production of FAE, and it is becoming difficult to identify a domesticated plant or animal species with any substantial amount of lipid for which the production of esters for alleged use as biodiesel has not been described. In reality, not only the ability to be converted to FAE, but also such factors as cost, supply, ease of collection, feedstock and fuel storage stability, fuel quality, engine performance, and outlets for the non-lipid portions of the feedstock determine whether a lipid is adopted for commercial fuel production. Two of the most promising and/or actively investigated potential new sources of relatively pure triglycerides are:

1. Jatropha curcas (Achten et al., 2008; Kumar, 2008) This species, a member of the genus Euphorbiaceae, is a drought resistant perennial bush or small tree that is able to survive on poor soils in arid to moist climates. A native of Central America, it has been successfully transplanted into tropical and temperate regions worldwide, doing best in the drier regions of the tropics. The seed is approximately 35% lipid. With slightly over 20% saturated fatty acids, and approximately 40% mono- and 35%

Alternate Feedstocks and Technologies ■ 51

diunsaturated fatty acids, this oil has acceptable stability and low temperature fluidity for use as a biodiesel feedstock. Seed yields range from 1.5 - 7.8 t /ha yr, depending on site quality, suggesting a potential biodiesel production of 159 - 825 l./ha yr. This compares favorably with approximately 500 l biodiesel / ha yr for soybean in a good environment. The meal and oil from most Jatropha cultivars are inedible due to the presence of toxins, and thus food use does not contribute to oil demand or price. Since it can be cultivated in arid environments with poor soil and low agricultural potential, fuel feedstock production with this species should not interfere with the production of edible oils for human consumption. Indonesia and the Philippines are vigorously exploring the production of Jatropha as an oil producer. India has undertaken a vigorous program targeting the installation of tens of thousands of acres of the species, and a large research program to cover all areas related to Jatropha biodiesel. Jatropha oil is commercially available, its conversion to biodiesel has been investigated, and the combustion of the fuel has been characterized. Research challenges remain nonetheless. Toxicity issues are being addressed, with a recent report indicating that the phorbol esters can be degraded by steam stripping/deodorization of the oil (Makkar et al., 2009). Another approach could be the use of a reported nontoxic line of Mexican origin (Martínez-Herrera et al., 2006). However, demand for food use would elevate the cost of the oil from such a cultivar, and it is possible that increased consumption of seeds by wild or domestic animals could reduce yields. Work is also needed on agronomic improvement. Production is labor intensive since the fruits do not ripen all at once (thus multiple harvests are required) and since appropriate mechanical harvesting equipment has yet to be developed. Although some plantings have been made in the southern United States, there is at this time no commercial Jatropha production in that country.

2. Microalgae The microalgae constitute a multi-species group of unicellular water-borne photosynthetic organisms, some of which accumulate intracellular lipids to as much as 70 wt% of their mass. Considerable interest, investment, research and technology development has recently arisen regarding the potential of oil-producing photosynthetic microalgae as a source of lipids for biofuel production (Chisti, 2007; Huntley & Redalje, 2007; Hu et al., 2008; Li et al., 2008; Vasudevan & Briggs, 2008). More than 200 firms are estimated to be working in this area, which is expected to first produce fuel in commercial amounts in 2009 (Kram, 2008). Attractive features of algae as an oil source are the reported high lipid contents, proposals to sequester CO2 by using it as the carbon source for algal growth, lack of the need to use fertile agricultural land for production, in some instances the possibility of using low purity water as the growth medium, and an absence of human food use. The potential of these organisms seems exceptional: lipid production capabilities between 100 and 300 times greater than that of soybean have been calculated (Chisti, 2007) with some production estimates reaching 50 metric tons /acre. This leads to estimates that total U.S. liquid transportation fuel needs could be met by an algal farming system occupying

52

■

M. J. Haas

29 million acres, equivalent to only approximately 6% of the land mass presently used for crop production in the United States (Briggs, 2004). However, a number of challenges and unknowns remain in this area, and individuals familiar with the field suggest that initial commercial scale production will more likely achieve production levels on the order of 3 metric tons/acre. Between 1978 and 1996 the U.S. Department of Energy funded work in this area and its results are among the few concise reports published (Sheehan et al., 1998). The program was terminated due largely to the high cost of oil production. Recent increases in petroleum prices have sparked vigorous new interest in this area. Several geometries are being investigated for growth of the organisms: open and covered outdoor ponds employing solar illumination as energy source, and closed systems, some of which use solar radiation as the energy source while other approaches rely on simple carbohydrates for energy. It is not clear yet which of these technology options are superior, and it is certain that even the best will not consistently give production scale oil yields near the maxima achieved in laboratory tests. Whether the resulting ‘algal oil’ will be an economically or technologically viable biodiesel feedstock is also uncertain at present (Ratledge & Cohen, 2008; Reijnders, 2008; Chisti, 2008). Some calculations suggest that using present approaches the oil may cost 5 times more than soy oil (Chisti, 2007). Among other substantial questions remaining to be answered are whether high lipid levels can be stably achieved and maintained under real world conditions, the feasibility and optimization of long term pond culture systems, cell harvest and oil recovery technologies, uses of the lipid-depleted meal, and overall economics. For these current and developmental triacylglycerol biodiesel feedstocks, homogenous liquid phase alkali-catalyzed transesterification is the predominant and preferred means of conversion to simple alkyl esters for use as biodiesel. The method is simple, economical, and effective, and likely to stay the preferred technology for biodiesel production from high quality feedstocks for at least the near term.

New Sources of Lower Quality Lipid Feedstocks Traditional ‘Lower Quality’ Lipid Feedstocks Used here, the term ‘lower quality’ refers to the degree of purity of an acylglyceride, especially the level of contaminating free fatty acids (FFA). Because the latter are not converted to esters by alkaline-catalyzed transesterification, feedstocks containing significant levels of FFA require different processing technologies than do those lacking them (Kulkarni & Dalai, 2006). It is vital that the FFA be removed or converted to FAE since they can be detrimental to fuel storage tanks and fuel injection systems and pumps. All current biodiesel quality specifications impose strict limits on the ‘acid number’ of the fuel, which is a reflection of the content of FFA and other acidic species. Feedstocks with FFA levels in excess of 4%, (and lower in some situations) fall into this category of low quality lipids. Among these are lower quality animal fats, such as prime tallow, special tallow, ‘A’ tallow, and poultry fat according to U.S. standards (American Fats & Oils Association, 2004) and tallows No. 3 through 6 of the British classification

Alternate Feedstocks and Technologies ■ 53

scheme (BSI British Standards, 1987). Greases also fall into this feedstock category. British Standards identify one category, ‘grease’, with a maximum FFA content of 20% (BSI British Standards, 1987). In the United States, yellow (FFA to 15%) and brown grease (FFA >15%), are available (American Fats & Oils Association, 2004). It has been estimated that the supply of yellow grease in U.S. urban areas is approximately 9 lbs (4.1 kg)/ person per year (Wiltsee, 1998). This material is already used in some industrial scale biodiesel production. The greases are sometimes referred to as ‘recovered vegetable oil’ and typically consist of partially hydrogenated vegetable oils disposed after use in deep fat frying. Their costs are indexed relative to those of the refined vegetable oils, and are generally one half to one third that of the lowest priced refined oil.

By-products of Ethanol Production Corn contains approximately 3.6% oil, essentially all of it in the germ. In the production of ethanol from corn by yeast fermentation this lipid is not converted to alcohol. With U.S. production of corn-derived ethanol now at approximately 25 million metric tons annually the corn oil side stream is approximately 2.82 million metric tons annually. In response to the 2007 Renewable Fuels Act, U.S. corn ethanol production is expected to grow to 50 million metric tons by 2015, generating as a by-product sufficient corn oil to produce greater than 3 million metric tons of biodiesel. This is a substantial volume, given the estimated 2007 U.S. biodiesel production of 1.81 million metric tons. Research challenges in this area relate to enhancing the degree of oil recovery and in dealing with the fact that oil exiting the fermentation process is high in free fatty acids. It was not recovered for further use until recently, when at least one firm developed and deployed appropriate technologies. These use centrifugation to recover the corn oil present in the liquid phase following fermentation. Currently it is claimed that approximately two thirds of the oil can be recovered in this manner. The quality of the oil is inadequate for edible use, but acceptable for use in biodiesel production. The technology has been installed in a limited number of ethanol plants in the United States. Other researchers have described the recovery of oil from the fermentation residue, after removal of its ethanol by distillation, and conversion of the oil to fatty acid methyl esters (Noureddini et al., 2009; Balan et al., 2009). Direct transesterification has also been shown to be an effective means of converting the oil in ethanol fermentation by-products to biodiesel, although it has not been commercialized to date (see ‘in situ transesterification’ in the following paragraphs.).

Other New Low Quality Feedstocks Among the high-FFA lipids that are potentially suitable for biodiesel production is soapstock (SS), a by-product of vegetable oil refining. Annual U.S. production exceeds 100 million pounds. Soapstock is a rich source of fatty acids, consisting of approximately 12% acylglycerols, 10% FFA, and 8% phospholipids. It also contains nearly 50% water and is quite alkaline (pH typically > 9). Due to the high pH and substantial content of

54

■

M. J. Haas

polar lipids, the lipids and water in SS are thoroughly emulsified, forming a dense, stable, viscous mass that is solid at room temperatures. The largest use of this material at present is in animal nutrition, and there is interest in finding higher value applications. Its conversion to biodiesel via both homogenous (below) and heterogeneous (Shultz et al., 2009) catalysis has been described, although to date it does not appear that soapstock has become a feedstock for commercial biodiesel production. A less expensive and correspondingly more hydrolyzed and heterogenous potential feedstock for biodiesel production is ‘trap grease’, the lipid captured by in-line traps in the effluent streams exiting restaurant kitchens and other operations that use substantial amounts of fats and oils. Annual trap grease production in U.S. cites has been estimated at 13 lb/person (Wiltsee, 1998), suggesting a considerable stream of potential biodiesel feedstock. The material can be very low in cost, in some cases in the range of US$ 0.05 - 0.10/lb lipid. However, there have been only limited efforts to develop it as a biodiesel feedstock to date. Excessive levels of water, free fatty acids, and contaminants challenge material handling, biodiesel synthetic technologies, and product cleanup, respectively. Claims of successful biodiesel production from trap grease have been made, and it appears that this technology will soon be implemented at small production scale. There is interest in the use of even lower value lipid sources, such as those from sewage treatment facilities, but little reported success in this area. There is growing awareness that any lipid source is a potential biodiesel feedstock. Thus, for example, the extraction of oil from spent coffee grounds, its conversion to FAME, and demonstration that this product met the ASTM specifications for biodiesel has been recently reported (Kondamudi et al., 2008). Matters of logistics, volume, cost and quality will determine whether each such approach is adopted by industry.

Alcohol Reactant Being generally the least expensive alcohol, methanol is the alcohol of choice for the production of biodiesel in most parts of the world. Indeed, EN 14214, the European Standard for Biodiesel (European Committee for Standardization, 2003), specifies it to be composed of fatty acid methyl esters (FAME). In some regions of the world the climate favors the growth of sugar cane, an excellent source for the production of ethanol. Brazil is the world leader in ethanol production by fermentation. In such a region, it may be economical to produce fatty acid ethyl esters (FAEE) for use as biodiesel. Ethanol also has been used in biodiesel production in test situations in the United States under circumstances where it was available from the fermentation of starch-rich food processing waste streams (Lowe et al., 1998). Ethanol offers the opportunity to produce a truly biobased fatty acid ester, whereas methanol is typically produced from nonrenewable natural gas. The chemical technologies used for the synthesis of fatty acid methyl esters can generally be used in FAEE production. However, ethanol presents additional technical challenges, among which are the need to use larger volumes of alcohol (due to the higher molecular weight of ethanol), longer reaction times, higher temperatures, a greater tendency for the formation of difficult emulsions, and the need to recover

Alternate Feedstocks and Technologies ■ 55

un-reacted ethanol from its azeotrope with water. In addition, some of the accepted quality specifications for biodiesel rely on gas chromatographic methods to determine fatty acid ester content, and specify FAME-characteristic retention times to measure this parameter. However, FAEE exhibit mobilities in these systems that are outside of this specified region. These features, and the relatively high cost of ethanol, have restricted the adoption of ethyl ester biodiesel fuels in most regions of the world. This could change if methanol prices rise substantially due to fossil fuel scarcity or high cost, and ethanol prices fall due to increased production. The use of longer chain alcohols, either straight- or branched-chain, in biodiesel production has been described. The fatty acid esters of these alcohols generally exhibit reductions of 1 to 10 oC in their low temperature fluidities relative to the methyl esters (Lee et al., 1995; Foglia, et al., 1997). This may facilitate the use of tallow-based fuels at lower temperatures without the danger of fuel solidification and engine failure. The matter of fuel solidification, however, may be more economically addressed with available commercial fuel additives (Dunn et al., 1996). In addition, the higher prices of the longer chain alcohols render biodiesel made from them impractical as commercial fuels. Another challenge is low solubilities of the conventional alkaline transesterification catalysts in long chain alcohols, which hinders rapid, high efficiency reaction. Although ethanol production by fermentation is a well developed technology in some countries, and recent initiatives to expand industrial-scale butanol production (Ezeji et al., 2007) could lower the price of this alcohol, the near-term displacement of methanol as the preferred alcohol for biodiesel production seems unlikely.

Catalysts for Biodiesel Production Traditional Approaches An attractive feature of the use of refined triacylglycerols as feedstocks, and another factor driving their choice as the predominant biodiesel feedstock, is the relative ease with which they are converted to simple alkyl esters by alkali-catalyzed chemical transesterification. This topic is amply treated elsewhere in volume (see Chapter 4.1, Basics of the Transesterification Reaction) The conventional transesterification technology is acceptable for feedstocks with low FFA levels (≤0.5%). At these levels the salts (‘soaps’) formed by the reaction of FFA with the cations of the catalyst are readily removed during purification of the crude biodiesel and represent only a minor loss of potential product. In the case of the use of metal hydroxides as catalysts, protons released from the FFA upon soap formation will combine with hydroxide ions to form water, a transesterification inhibitor. However, because the FFA levels are low the resulting reduction in available hydroxyl catalyst and the accumulation of inhibitory water are small and do not negatively impact reaction efficiency. For lipids with FFA levels between about 0.5-4 % the loss of catalyst accompanying soap formation during alkali-catalyzed transesterification is sufficient to reduce transesterification efficiencies if not compensated by the addition of make-up alkali catalyst. The approach in such cases is to conduct a pre-treatment with alkali to precipitate

56

■

M. J. Haas

the FFA as their soaps before beginning transesterification. This increases overall alkali costs, but converts the FFA to a form that can be removed and sold into other commercial outlets. The FFA-depleted lipid is then subjected to alkaline transesterification as for low-FFA feedstocks. Feedstock lipids amenable to this approach are those that are off-specification due to elevated FFA levels, intermediate grades of animal fats (top white, all beef packer, extra fancy, fancy, bleachable according to the U.S. classification scheme [American Fats & Oils Association, 2004], no. 1 tallow by the British scheme [BSI British Standards, 1987]), and lightly-used deep fat fryer greases (e.g., choice white grease in the U.S.) For feedstocks with FFA levels in excess of 4% the strategy of converting these to soaps and removing them is untenable due to the loss of potential product, the tendency of soaps to emulsify the FAME and glycerol layers, and the cost of the alkali required to compensate for losses to soap. Therefore the strategy is to convert both the FFA and the acylglycerol fractions to biodiesel. Multi-catalyst protocols involving sequential acidcatalyzed FFA esterification followed by alkali-catalyzed acylglycerol transesterification have been described (Canacki and Van Gerpen, 2001, 2003; Lepper & Friesenhagen, 1986). Even though this increases process complexity and cost, when greases are used as the feedstocks it has been estimated that the savings in feedstock cost can result in an overall cost reduction of 25 to 40% relative to the use of virgin soy oil (Canacki & Van Gerpen, 1999). An alternate approach employing only acid-catalyzed ester synthesis is alleged to be more economical (Zhang et al., 2003). The degree to which this method has been adopted by industry is difficult to assess, but is probably low, due to the corrosiveness of acid catalysts, the longer reaction times required, higher reaction temperatures, and lower reactivity than base catalysts (Boocock et al. 1996; Reid 1911; Srivastava & Prasad, 2000). Due to its low cost, sulfuric acid is the typical catalyst used in the FFA esterification step of the two-step process (Canacki & Van Gerpen, 1999; Goff et al., 2004). Water, a by-product of esterification, prevents quantitative ester synthesis. By conducting two or more sequential acid-catalyzed esterifications, with the removal of produced water after each, acceptable degrees of fatty acid esterification can be achieved. Final FFA levels below 0.5-1.0% are desired. The triacylglycerols of the oil or fat substrate also undergo some transesterification during acid catalyzed FFA esterification, yielding partial glycerides and FAME.

Alternate Technologies for Fatty Acid Ester Synthesis The acid- and alkali-catalyzed esterification and transesterification reactions discussed above are currently the predominant technologies for industrial-scale biodiesel production. However, the desire to reduce catalyst costs and waste output, and to eliminate extensive product purification has stimulated the investigation of alternate methods of fatty acid ester synthesis. This chapter will discuss only homogeneous catalysts, i.e. those that are soluble in the reaction system. A discussion of heterogeneous (i.e. insoluble) catalysts, including enzymes, is found elsewhere in this text (Schultz et al., 2009).

Alternate Feedstocks and Technologies ■ 57

Homogenous Catalysis Modifications of conventional alkali-catalyzed transesterification Monophasic transesterification. The insolubility of lipids in short chain alcohols reduces transesterification rates. Reaction occurs at the interface between the two phases, and rate is thus greatly impacted by phase boundary properties. Strategies have been developed to facilitate or overcome the resulting rate limitations. In what has been termed ‘solvent assisted methanolysis’, an organic solvent is added to the reaction in order to render it monophasic (Boocock et al., 1998; Doell et al., 2008). As a result, less catalyst is needed and reactions are completed in minutes rather than hours. Favored solvents are tetrahydrofuran, hexane, or methyl tert-butyl ether. In addition to the use of solvent to promote the miscibility of methanol and oil, a high methanol/ oil molar ratio (27:1) is also used. Advantages of this approach are the use of a one-step transesterification process; reduced catalyst requirements, methyl ester yields in excess of 98%, reaction times of <10 min, and lower reaction temperatures. Disadvantages are the necessity of recovering the tetrahydrofuran and the large molar excess of un-reacted methanol, and the inherent hazards associated with flammable solvents. Nonetheless, there are reports of the adoption of this technology in commercial biodiesel production (Caparella, 2002). Microwave enhanced transesterification. Microwave or radio frequency radiation has been applied to homogeneous alkali-catalyzed ester synthesis reactions (Hernando et al., 2007; Azcan & Danisman, 2007, 2008) and for both homogeneous and heterogeneous reactions catalyzed by either alkali or acid (Portnoff et al., 2005). Microwave irradiation results in a substantial increase in reaction rate, such that degrees of transesterification in excess of 90% are achieved in as little as 3 to 8 min. This acceleration is postulated to occur because the catalyst specifically absorbs the applied energy, increasing the local reaction rate (Portnoff et al., 2005). The capital and operating costs of this technology have not been made public. Acid-catalyzed FAE synthesis As opposed to the use of acid catalysis for esterification of the FFA in low quality feedstocks followed by alkali catalyzed transesterification of the acylglycerols, acid-catalysis can also be used alone for the combined alcoholysis of triglycerides and esterification of FFAs (Lotero et al., 2005). Compared with alkali catalyzed transesterification, higher reaction temperatures, longer reaction times, approximately 5 times more alcohol, and costlier materials of construction are required to achieve satisfactory yields (Freedman et al., 1984; Schwab et al., 1987; Canakci & Van Gerpen, 1999). Also, water, a by-product of FFA esterification, will accumulate during the reaction of low quality feedstocks and inhibit reaction if not removed. Nonetheless, the ability of acid catalysts to produce FAME from both FFA and acylglycerols offers sufficient advantages that this approach was shown to be superior to alkali- and to joint acid-/alkaline catalysis in an economic analysis of biodiesel production from grease (Zhang et al., 2003). Sulfuric acid is the

58

■

M. J. Haas

favored acid for this reaction: not only is it among the most economical of acids, but investigation of a series of Bronsted acids has shown only this acid to be effective at achieving high conversions in the transesterification of soy oil with methanol (Canacki & Van Gerpen, 1999; Goff et al., 2004). Though effective on a laboratory scale, the slow rate of the reaction under ambient conditions and the higher costs when implemented under higher pressure has hindered its adoption for industrial-scale biodiesel production. Another strategy employing acid catalysis in biodiesel production involves sequential acylglycerol hydrolysis followed by FFA esterification. Thus, using soapstock, an alkaline water-rich emulsion containing approximately equal amounts of acylglycerols, phosphoglycerols and FFA, an approach was described (Haas et al., 2000) wherein the material was first made more alkaline and heated to hydrolyze all acyl- and phosphoacyl- glycerol bonds. Following water removal, all FFA were then converted to FAME via sulfuric acid catalysis. The resulting ester product met the ASTM Provisional Specifications for biodiesel in effect at that time and gave emissions and performance in a heavy duty diesel engine that were comparable to biodiesel produced from soybean oil (Haas et al., 2001). A subsequent refinement eliminated the large amount of solid waste that was a feature of the first approach (Haas et al., 2003). Recently the use of the heteropolyacid H3PW12O40 as an alternative to sulfuric acid for the esterification of FFA at ambient pressure and reflux temperature was described (Cardoso et al., 2008). Though strong acids, such compounds are less corrosive and thus easier to handle than sulfuric acid, because less degradation and contamination of the product, can be recovered in a water wash, and are amenable to immobilization on solid carriers. Although this study used only pure substrates and reported degrees of esterification of only about 90%, with further development such catalysts could become useful for real-world biodiesel production from FFA. Direct (‘in situ’) transesterification of substrate-resident lipid. Rather than using isolated lipids as a feedstock, an alternate approach is to directly transesterify the oil while it resides in oil-bearing materials. A form of homogenous transesterification, this method eliminates the need to remove the lipid feedstock from the biological material in which it resides, and thus may be less costly than current technologies. Also, this direct approach may be capable of successfully transesterifying the lipids in materials from which these are not presently recovered economically by e.g. solvent extraction or pressing. In some cases pressing of lipid-bearing materials are used in place of solvent extraction to obtain oil for biodiesel production. These methods leave a substantial proportion of the lipid in the press cake. Direct transesterification has the potential to convert this material to FAE, thus further augmenting biodiesel production. Initial work in this area involved acid catalysis. Thus, using sulfuric acid the in-situ transesterification of homogenized whole sunflower seeds with methanol produced ester in yields up to 20% greater than that obtained with extracted oil (Harrington & d’ArcyEvans, 1985a,b). This was attributed to the transesterification of the seed hull lipids. The acid catalyzed methyl transesterification of homogenized sunflower seeds achieved up to 98% of theoretical maximum ester yield based on the oil content of the seeds

Alternate Feedstocks and Technologies ■ 59

(Siler-Marinkovic & Tomasevic, 1998). In the acid-catalyzed in situ transesterification of rice bran oil 90% of the oil was converted to ester, although the product contained high levels of free fatty acids (Özgül & Türkay, 1993; Özgül-Yücel & Türkay, 2002, 2003). Up to 40% of the oil in ground soybeans could be transesterified by this approach (Kildiran et al., 1996). The use of alkaline catalysis in this approach has also been explored, yielding generally better results than with acid catalysts. Methyl, ethyl and isopropyl esters of soybean lipids were successfully produced under mild conditions by this approach (Haas et al., 2004). Under optimized reaction conditions, greater than 95% of maximum theoretical transesterification was achieved in room temperature reactions. The methyl ester fraction contained minor amounts of fatty acids (<1%) and no acylglycerols. It was subsequently shown that drying the substrate reduced the methanol requirement by 60%, greatly improving the economics of the process, and that the resulting product met the quality specifications for biodiesel (Haas & Scott, 2007). The method has been shown to be generally applicable to virtually any lipid-containing material, with detailed studies being reported for distillers dried grains with soluble (a lipid-containing by-product of fermentative ethanol production from corn) and meat and bone meal (a by-product of the edible meat industry (Haas et al., 2007). The application of ultrasound in conjunction with hexane during in situ transesterification has been reported, and may enhance the rate of the reaction (Siatis et al., 2006). In order for any biodiesel production process to be affordable there must be some use and corresponding economic value for the non-lipid portion of the agricultural feedstock. In the case of soybeans, the suitability of the lipid-free meal co-product of in situ transesterification as a dietary component for trout (Barrows et al., 2008) and chickens (M. J. Haas, unpublished work) has been shown.

‘Uncatalyzed’ Ester Production Transesterification can also be achieved in the absence of any added catalyst. This requires high pressures and temperatures (e.g., 200 bar, 350 °C), which convert the reaction to a ‘supercritical’ fluid state. In such a system substances are neither liquid nor vapor, and the chemical properties of the system are between those of liquids and gases. Densities can vary between those of the liquid or the gaseous phase of the system, depending on small changes in the pressure applied. Mass transport properties and diffusivity are closer to those of gases. Hence, an acceleration of transport rates is often seen. In the case of biodiesel production, the acylglycerol and alcohol reactants form a single phase under such conditions, greatly accelerating reaction rates. The transesterification of rapeseed oil under supercritical conditions in a 42-fold excess of methanol at 350 °C and 450 to 650 bar has been described (Saka & Kusdiana, 2001; Kusdiana & Saka, 2001), although it is possible that in fact the metal walls of the reactor serve as a catalyst in such situations (Dasari et al., 2003). At the lower temperatures of 220 - 235 °C substantially longer reaction times (8-10 h) were required to achieve even 85% ester production (Diasakou et al., 1998). Under supercritical conditions FFA are also esterified, eliminating the need for feedstock pretreatment (Stern

60

■

M. J. Haas

et al., 1999). In addition to high reaction rates, the absence of added catalyst simplifies purification of the biodiesel and glycerol produced. Such technologies have been adopted to a limited degree at the industrial scale, especially in Europe (see for example, www .canentec.com/files/ipu.pdf ), but are not widely implemented. Due to the high capital, alcohol, and operational costs the per-gallon total costs of the process have been estimated to be nearly twice those of more conventional technologies (Marchetti & Errazu, 2008).

Direct Fatty Acid Alkyl Ester Production via Pathway Engineering Contemporary molecular genetic technology is sufficiently powerful to allow the engineering of cellular metabolism to produce desired biomolecules. Recently the standard laboratory bacterium Escherichia coli have been engineered to conduct the biosynthesis of fatty acid ethyl esters, a product given the name ‘microdiesel’ (Kalscheuer et al., 2006). This approach is very new, and ester yields of only 1.3 g/l were achieved. However, the basic approach has thus been shown feasible, and yield improvement is inevitable. At least one private sector firm has been formed to produce fuels and other industrial materials via genetic engineering of the producer organisms to enable them to use non-lipid feedstocks such as carbohydrates (Keasling et al., 2007). It is unclear at this time whether this approach, which uses sophisticated fermentation technology to produce material that must sell at commodity fuel prices, will prove economically viable.

Conclusions In many ways the history of biodiesel production to date has involved the harvest of the ‘low hanging fruit’: high purity feedstocks have been processed to biodiesel using known, straightforward chemical technologies. Increased biodiesel usage and production targets, rising feedstock costs, and the negatives of food vs. fuel competition for lipids has focused attention on the development of new resources for lipids, often with emphasis on inedible feedstocks. At the same time, desires to reduce the waste streams leaving production facilities, and to reduce production costs if possible, have stimulated attention to new catalytic approaches, especially reusable solid-phase catalysts. Given recent efforts in these areas it is reasonable to expect profound advances at the industrial scale in the near future.

References Achten WMJ, Verchot L, Franken YJ, Mathijs E, Singh VP, Aerts R, Muys B. Jatropha bio-diesel production and use. Biomass Bioenerg.. 32:1063-1084, 2008. Ali Y, Hanna MA. Alternative diesel fuels from vegetable oils. Bioresour. Technol. 50:153-163, 1994a. Ali Y, Hanna MH. Physical properties of tallow ester and diesel fuel blends. Bioresour. Technol. 47:131-134, 1994b. American Fats & Oils Association. Specifications for commercial grades of tallows, animal fats and greases. Columbia, South Carolina, U.S.A. www.afoaonline.org. 2004. Anonymous. Energy Independence and Security Act of 2007. http://en.wikipedia.org/wiki/ Clean_Energy_Act_of_2007. 2007

Alternate Feedstocks and Technologies ■ 61

Azcan N, Danisman A. Alkali Catalyzed Transesterification of Cottonseed Oil by Microwave Irradiation. Fuel 86:2639-2644, 2007. Azcan N, Danisman A. Microwave Assisted Transesterification of Rapeseed Oil. Ibid. 87:17811788, 2008. Balan V, Rogers CA, Chundawat SPS, da Costa Sousa L, Slininger PJ, Gupta R. Dale BE. Conversion of extracted oil cake fibers into bioethanol including DDGS, canola, sunflower, sesame, soy and peanut for integrated biodiesel processing. J. Am. Oil Chem. Soc. 86:157-165, 2009. Barrows FT, Gaylord TG, Sealey WM, Haas MJ, Stroup RL. Processing soybean meal for biodiesel production; effect of a new processing method on growth performance of rainbow trout, Oncorhynchus mykiss. Aquaculture. 283: 41-147, 2008. Boocock DGB, Konar SK, Mao V, Sidi H. Fast one-phase oil-rich processes for the preparation of vegetable oil methyl esters. Biomass Bioenerg. 11:43-50, 1996. Boocock DGB, Konar SK, Mao L, Lee C, Buligan S. Fast formation of high-purity methylesters from vegetable oils. J. Am. Oil Chem. Soc. 75:1167-1172, 1998. Briggs, M. Widescale biodiesel production from algae. Physics Department, University of New Hampshire, Durham, NH, http://www.unh.edu/p2/biodiesel/article_alge.html, 2004. BSI British Standards. Specification for technical tallow and animal grease. British Standard 3919:1987. BSI British Standards, London, UK. www.bsi-global.com. 1987. Canakci M, Van Gerpen J. Biodiesel production via acid catalysis. Trans. ASAE 42:1203-1210, 1999. Canakci M, Van Gerpen J. Biodiesel production from oils and fats with high free fatty acids. Trans. ASAE 44:1429-1436, 2001. Canakci M, Van Gerpen, J. A pilot plant to produce biodiesel from high free fatty acid feedstocks. Trans. ASAE 45:945-954, 2003. Caparella T. Biodiesel plants open in Germany, Render Mag. 37:16 2002. Cardone M, Mazzoncini M, Menini S, Rocco V, Senatore A, Seggiani M, Vitolo S. Brassica carinata as an alternative crop for the production of biodiesel in Italy: agronomic evaluation, fuel production by transesterification and characterization. Biomass Bioenerg. 25:623-636, 2003. Cardoso AL, Augusti R, Da Silva MJ. Investigation on the esterification of fatty acids catalyzed by the H3PW12O40 heteropolyacid. J. Am Oil Chem. Soc. 85:555-560, 2008. Chisti Y. Biodiesel from microalgae. Biotechnol. Adv. 25:294–306, 2007. Chisti Y. Response to Reijnders: Do biofuels from microalgae beat biofuels from terrestrial plants? Trends Biotechnol. 26(7):351-352, 2008. Choo YM, Cheah KY, Ma AN, Halim A. Conversion of crude palm kernel oil into its methyl esters on a pilot plant scale. In Applewhite TH (ed.), Proceedings, World Conference on Oleochemicals in the 21st Century. Champaign, IL. AOCS Press, 1991, pp. 292-295. Cummins E J, Colgan SF, Grace PM, Fry DJ, McDonnell KP, Ward SM. Human risks from the combustion of SRM-derived tallow in Ireland. Hum. Ecol. Risk Asses. 8:1177-1192, 2002. Dasari M, Goff MJ, Suppes GJ. Noncatalytic alcoholysis of soybean oil. J. Am. Oil Chem. Soc., 80:189-192, 2003. Desai PD, Dave AM, Devi S. Alcoholysis of salicornia oil using free and covalently bound lipase onto chitosan beads. Food Chem., 95:193-199, 2006. Diasakou M, Louloudi A, Papayannakos N. Kinetics of the non-catalytic transesterification of soybean oil. Fuel 77:1297-1302, 1998. Doell R, Konar SK, Boocock DGB. Kinetic parameters of a homogeneous transmethylation of soybean oil. J. Am. Oil Chem. Soc. 85:271-276, 2008.

62

■

M. J. Haas

Dunn RO, Shockley MW, Bagby MO. Improving the low-temperature properties of alternative diesel fuels: Vegetable oil-derived methyl esters. J. Am. Oil Chem. Soc. 73:1719-1728, 1996. Erdtmann R, Sivitz LB. (eds). Advancing Prion Science: Guidance for the National Prion Research Program. Washington, D.C.: The National Academies Press, 2004. European Commission. Preliminary Report on Quantitative Risk Assessment on the Use of the Vertebral Column for the Production of Gelatine and Tallow. Submitted to the Scientific Steering Committee at its meeting of 13-14 April, 2000, Brussels, Belgium. 2000. European Committee for Standardization (CEN). Automotive fuels - fatty acid methyl esters (FAME) for diesel engines - requirement methods, EN 14214:2003, Brussels, Belgium, 2003. Ezeji TC, Qureshi N, Blaschek HP. Bioproduction of butanol from biomass: from genes to bioreactors. Curr. Op. Biotechnol. 18(3):220-227 (2007)) Foglia, T. A., L. L. Nelson, R.O. Dunn, and W. N. Marmer. Low-temperature properties of alkyl esters of tallow and grease. J. Am. Oil Chem. Soc. 74:951-955, 1997. Foidl N, Foidl G. Sanchez M, Mittelbach M, Hackel S. Jatropha curcas L. as a source for the production of biofuel in Nicaragua. Bioresour. Technol. 58:77-82, 1996. Freedman B, Pryde EH, and Mounts, TL. Variables affecting the yields of fatty esters from transesterified vegetable oils. J. Am. Oil Chem.. Soc. 61:1638-1643, 1984. Frohlich A, Rice B. Evaluation of Camelina sativa oil as a feedstock for biodiesel production. Ind. Crops Prod. 21:25-31, 2005. Goff M J, Bauer NS, Lopes S, Sutterlin, WR, Suppes GJ. Acid-catalyzed alcoholysis of soybean oil. J. Am. Oil Chem. Soc. 81:415-420, 2004. Gubitz GM, Mittelbach M, Trabi M. Exploitation of the tropical oil seed plant Jatropha curcas L. Bioresour. Technol. 67:73-82, 1999. Haas MJ, Bloomer S, Scott K. Simple, high-efficiency synthesis of fatty acid methyl esters from soapstock. J. Am. Oil Chem. Soc. 77:373-379, 2000. Haas MJ, Scott KM, Alleman TL, McCormick RL. Engine performance of biodiesel fuel prepared from soybean soapstock: a high quality renewable fuel produced from a waste feedstock. Energ. Fuel. 15:1207-1212, 2001. Haas MJ, Michalski PJ, Runyon S, Nunez A, Scott KM. Production of fatty acid methyl esters from acid oil, a by-product of vegetable oil refining. J. Am. Oil Chem. Soc. 80:97-102, 2003. Haas MJ, Foglia TA. Alternate feedstocks and technologies for biodiesel production, in Knothe G, Krahl J, and Van Gerpen J (eds.). The Biodiesel Handbook, Champaign, IL.: AOCS Press, 2005, pp. 42-61. Haas MJ, McAloon AJ, Yee WC, Foglia TA. A process model to estimate biodiesel production costs. Bioresour. Technol. 97:671-678, 2006. Haas MJ, Scott KM. Moisture removal substantially improves the efficiency of in situ biodiesel production from soybeans. J. Am. Oil Chem. Soc. 84(2):197-204, 2007. Haas MJ, Scott KM, Foglia TA, Marmer WN. The general applicability of in situ transesterification for the production of fatty acid esters from a variety of feedstocks. Ibid. 84(10):963-970, 2007. Haas M J, Scott KM, Marmer WN, Foglia TA. In situ alkaline transesterification: an effective method for the production of fatty acid esters from vegetable oils. Ibid. 81:83-89, 2004. Harrington KJ, D’Arcy-Evans C. Transesterification in situ of sunflower seed oil. Ind. Eng. Chem. Prod. Res. Dev. 24:314-318, 1985a. Harrington KJ, D’Arcy-Evans C. A comparison of conventional and in situ methods of transesterification of seed oil from a series of sunflower cultivars. J. Am. Oil Chem. Soc. 62:1009-1013, 1985b.

Alternate Feedstocks and Technologies ■ 63

Hernando J, Leton P, Matia MP, Novella JL, Alvaarez-Builla J. Biodiesel and FAME synthesis assisted by microwaves: Homogeneous batch and flow processes. Fuel 86:1641-1644, 2007. Hu Q, Sommerfeld M, Jarvis E, Ghirardi M, Posewitz M, Seibert M, Darzins A. Microalgal triacylglycerols as feedstocks for biofuels production: perspectives and advances. The Plant Journal 54:621-639, 2008. Huntley M, Redalje D. CO2 mitigation and renewable oil from photosynthetic microbes: a new appraisal. Mitigat. Adapt. Stateg. Glob. Change 12:573-608, 2007. Isigigur A, Karaosmanoglu F, Aksoy HA. Methyl ester from safflower seed oil of Turkish origin as a biofuel for Diesel engines. Appl. Biochem. Biotechnol. 45/45:103-112. 1994. Kalscheuer R, Stolting T, Steinbuchel A. Microdiesel: Escherichia coli engineered for fuel production. Microbiology 152:2529-2536. 2006. Kamini NR, Iefuji H. Lipase catalyzed methanolysis of vegetable oils in aqueous medium by Cryptcoccus spp. S-2. Process. Biochem. 37:405-410. 2001. Keasling JD, Hu Z, Somerville C, Church G, Berry D, Friedman L, Schirmer A, Brubaker S, del Cardayré SB. Production of Fatty Acids and Derivatives Thereof, WO/2007/136762, http:// www.wipo.int/pctdb/en/wo.jsp?WO=2007136762, 2007. Kildiran G, Ozgul-Yucel S, Turkay S. In-situ alcoholysis of soybean oil. J. Am. Oil Chem. Soc. 73:225-228,1996. Kondamudi N, Mohapatra SK, Misra M. Spent coffee grounds as a versatile source of green energy. J. Agric. Food Chem. 56:11757-11760. 2008. Kram JW. Algae interests align, Biodiesel Mag. 5(11):60-65, 2008. Kramer W. The potential of biodiesel production. Oils and Fats International, 11:33-34, 1995. Kulkarni, M.G., Dalai, A.K. Waste cooking oil - An economical source for biodiesel: A review. Ind. Eng. Chem. Res. 45:2901-2913, 2006. Kumar A, Sharma S. An evaluation of multipurpose oil seed crop for industrial uses (Jatropha curcas L.): A review. Ind. Crops Prod. 28:1-10, 2008. Kusdiana D, Saka S. Kinetics of transesterification in rapeseed oil to biodiesel fuel as treated in supercritical methanol. Fuel 80:693-698, 2001. Lee I, Johnson LA, Hammond EG. Use of branched-chain esters to reduce the crystallization temperature of biodiesel. J. Am. Oil Chem. Soc. 72:1155-1160, 1995. Lepper H, Friesenhagen L. Process for the production of fatty acid esters of short-chain aliphatic alcohols from fats and/or oils containing free fatty acids. U.S. Patent No. 4,608, 202, 1986. Li Q, Du W, Lieu D. Perspectives of microbial oils. Appl. Microbiol. Biotechnol. 80:749-756, 2008. Lotero E, Liu Y, Lopez DE, Suwannakarn K, Bruce DA, Goodwin Jr. J.G. Synthesis of biodiesel via acid catalysis. Ind. Eng. Chem. Res. 44:5353-5363, 2005. Lowe GA, Peterson CL, Thompson JC, Taberski JS, Mann PT, and Chase CL. Producing HySEE biodiesel from used French fry oil and ethanol for an over-the-road truck. ASAE Paper No. 98-6081, American Society of Agricultural Engineers, St. Joseph, MI, 1998. Ma F, Clements LD, Hanna MA. The effects of catalysts, free fatty acids, and water on transesterification of beef tallow. Trans. ASAE 41:1261-1264, 1998. Ma F, Clements LD, Hanna MA. The effect of mixing on transesterification of beef tallow. Bioresour. Technol. 69:289-293, 1999. Ma F, Hanna MA. Biodiesel production: a review. Bioresour. Technol. 70:1-15, 1990. Makkar H, Maes J, De Greyt W, Becker K. Removal and degradation of phorbol esters during pretreatment and transesterification of Jatropha curcas oil. J. Am. Oil Chem. Soc. 86:173-81, 2009. Marchetti JM, Errazu AF. Technoeconomic study of supercritical biodiesel production plant. Energy Convers. Manage. 49:2160-2164, 2008.

64

■

M. J. Haas

Martínez-Herrera J, Siddhuraju P, Francis G, D´ avila-Ortíz G, Becker K. Chemical composition, toxic/antimetabolic constituents, and effects of different treatments on their levels, in four provenances of Jatropha curcas L. from Mexico. Food Chem. 96:80–89. 2006. Mittelbach M, Remschmidt C. Biodiesel. The Comprehensive Handbook. Graz, Austria, Martin Mittelbach (Publisher). 2004. Narasimharao K, Lee A, Wilson K. Catalysts in production of biodiesel: A review. J. Biobased Mater. Bioenerg. 1:19-30, 2007. Noureddini H, Bandlamudi SRP, Guthrie EA. A novel method for the production of biodiesel from the whole stillage-extracted corn oil. J. Am. Oil Chem. Soc. 86:83-91, 2009. Ozgul S, Turkay S. In situ esterification of rice bran oil with methanol and ethanol. Ibid. 70:145-147, 1993. Ozgul-Yucel S, & Turkay S. Variables affecting the yields of methyl esters derived from in situ esterification of rice bran oil. Ibid. 79:611-613, 2002. Ozgul-Yucel S, Turkay S. FA monoalkylesters from rice bran oil by in situ esterification. Ibid. 80:81-84, 2003. Pinto AC, Guarieriro LLN, Rezende MJC, Ribeiro NM, Torres EA, Lopes WA, del.Pereira PA, J. B. de Andrade, JB. Biodiesel: An overview. J. Braz. Chem. Soc. 16(6B):1313-1330, 2005. Portnoff MA, Purta DA, Nasta MA, Zhang J, Pourarian F. Methods for Producing Biodiesel U.S. Patent Application, Publication No. US 2005/0274065 A1, Dec. 15, 2005. Ratledge, C., Cohen, Z. Microbial and algal oils: do they have a future for biodiesel or as commodity oils? Lipid Technol. 20(7):155-160, 2008. Reid EE. Studies in esterification. IV. The interdependence of limits as exemplified in the transformation of esters. Am Chem J 45:479-516, 1911. Reijnders, L. Do biofuels from microalgae beat biofuels from terrestrial plants? Trends Biotechnol. 26:349-350, 2008. Saka S, Kusdiana D. Biodiesel fuel from rapeseed oil as prepared in supercritical methanol. Fuel 80:225-231, 2001. Schultz AK, Haas MJ, Banavali R. Catalysis in Biodiesel Processing, in Knothe G, Krahl J, and Van Gerpen J (eds.). The Biodiesel Handbook, 2nd Edn., Champaign, IL., AOCS Press, 2009. Schwab AW, Bagby MO, Freedman B. Preparation and properties of diesel fuels from vegetable oils. Fuel 66:1372-1378, 1987. Seidel B, Alm M, Peters R, Kordel W, Schaffer A. Safety evaluation for a biodiesel process using prion-contaminated animal fat as a source. Environ. Sci. Pollut. Res. 13:25-130, 2006. Sharma YC, Singh B, Upadhyay SN. Advancements in development and characterization of biodiesel : A review. Fuel 87:2355-2373, 2008. Sheehan, J., Dunahay, T., Benemann, J., Roessler, P. A Look Back at the U.S. Department of Energy’s Aquatic Species Program: Biodiesel from Algae NREL/TP-580-24190, National Renewable Energy Laboratory, Golden, CO, http://www.nrel.gov/docs/legosti/fy98/24190.pdf, 1998. Siatis NG, Kimbaris AC, Pappas CS, Tarantilis PA, Polissiou MG. Improvement of biodiesel production based on the application of ultrasound: monitoring of the procedure by FTIR spectroscopy. J. Am. Oil Chem. Soc. 83:53-57, 2006. Siler-Marinkovic S, Tomasevic A. Transesterification of sunflower oil in situ.. Fuel 77:1389-1391, 1998. Solly RK. Coconut oil and coconut oil-ethanol derivatives as fuel for diesel engines. J. Fiji Agric. 42:1-6, 1980. Srivastava A, Prasad R. Triglycerides-based diesel fuels. Renew Sust Energ Rev 4:111-133, 2000.

Alternate Feedstocks and Technologies ■ 65

Stern R, Hillion G, Rouxel J-J, & Leporq S. Process for the production of esters from vegetable oils or animal oils alcohols. US Patent 5,908,946, 1999. Swedish Standards Institute. European Biodiesel Standard, EN 14214:2008 Automotive fuels - Fatty acid methyl esters (FAME) for diesel engines - Requirements and test methods. Stockholm, http://www.sis.se/DefaultMain.aspx, 2008. Technology Early Action Measures Program. Biobus Project: Biodiesel demonstration and impact assessment with the Societ’e de Transport de Montreal (STM). Government of Canada, Final Report available at http://www.stm.info/English/info/a-biobus-final.pdf, 2003. Thompson JC, He BB. Characterization of crude glycerol from biodiesel production from multiple feedstocks. Appl. Eng. Agric. 22(2): 261-265, 2006. U. S. Food and Drug Administration. Substances prohibited from use in animal food or feed. 21 CFR Part 589, U.S. Fed. Regist., 62(108):30935-30978, 1997. Vasudevan PT, Briggs M. Biodiesel production - current state of the art and challenges. J. Ind. Microbiol. Biotechnol. 35:421-430, 2008. Wiltsee, G. Urban Waste Grease Resource Assessment. Springfield, VA: Publication NREL/ SR-570-26141, National Technical Information Service, U.S. Department of Commerce, or DOE Information Bridge, http://www.doe.gov/bridge/home.html, 1998. World Heath Organization. Report of a WHO consultation on medicinal and other products in relation to human and animal transmissible spongiform encephalopathies. Report WHO/ CDS/VPH/95.145, as reported in Inform. 12:588, 2001. Zhang Y., M. A. Dube, D. D. McLean, and M. Kates. Biodiesel production from waste cooking oil: 2. Economic assessment and sensitivity analysis. Bioresour. Technol. 90:229-240, 2003.

■

■

4.3

■

■

Catalysis in Biodiesel Processing Alfred K. Schultz1, Michael J. Haas2,3, and Rajiv Banavali1 1

The Dow Chemical Company, Spring House, PA 2 USDA, ARS, ERRC, Wyndmoor, PA

Introduction Catalyst, a term coined by Berzelius (Thomson & Webb, 1968), refers to a material that assists in a chemical reaction without being consumed during the reaction. Catalysts generally improve reactions by accelerating reaction rates or by facilitating reactions at lower temperatures. Catalysts function by lowering the activation energy (∆G‡) required for a chemical transformation. For acid or base catalysis, reactions can be divided into two categories: general and specific catalysis. If the rate of an acid-catalyzed reaction, run in solvent S, is proportional to the concentration of SH+, then the reaction is said to be specific acid catalysis. On the other hand, if the reaction is proportional to either SH+ or another acid, then the reaction is said to be general acid-catalyzed. The relationship between acid strength and catalytic ability can be expressed by employing the Bronsted catalysis equation. The esterification of fatty acids is an example of a general acid-catalyzed reaction. The same general/specific catalysis concept holds for the basecatalyzed reaction of a triglyceride with alcohol to produce biodiesel and glycerol (March, 1992). The esterification of fatty acids with alcohols can be catalyzed by acids, organometallic reagents, and enzymes. The transesterification reaction of triglycerides with low molecular weight alcohols can be carried out by using a variety of catalysts, including acids, bases, enzymes, organometallic reagents, clays, and other natural silicates. The catalyst can be supplied for the reaction in either a homogeneous or heterogeneous phase. In homogeneous catalysis, the catalyst is soluble in the reaction medium, whereas, in a heterogeneous system, the catalyst is in a different phase from the reactants. Typically, homogeneous catalysis refers to a liquid (or soluble) catalyst used to convert liquid (or soluble) reagents, whereas heterogeneous catalysis refers to a solid catalyst used to convert liquid or gaseous reagents. In the following sections, we will review both homogeneous and heterogeneous catalysts for the esterification of fatty acids with methanol to produce biodiesel and also for the transesterification of triglycerides with methanol to produce biodiesel. 3

Note: Mention of brand or firm names does not constitute an endorsement by the U.S. Department of Agriculture over others of a similar nature not mentioned. ■

67

■

68

■

A.K. Schultz et al.

Esterification The raw material used in biodiesel manufacturing accounts for approximately 80–90% of the final cost of the fuel. For this reason, many producers constantly review their raw material supply in search of a less expensive alternative. The reader is directed to Chapter 4.2, Alternate Feedstocks and Technologies for Biodiesel Production, in this volume for a more thorough examination of some contemporary alternate feedstocks. Generally speaking, the less expensive the raw material, the less refining the oil has experienced. Hence, the less expensive raw materials typically contain a variety of impurities, along with the highest levels of free fatty acids (FFAs). These FFA impurities cause problems during the base-catalyzed transesterification reaction. The FFAs are converted into soaps, which consume catalysts, and thus excess catalyst must be used. These same soaps cause problems during the separation step of biodiesel manufacturing, creating an emulsion between the biodiesel and the glycerol. This emulsion must then be held in storage for extended periods of time to separate completely. The necessary added catalyst and yield loss, due to emulsion, often outweigh any cost benefits associated with the less expensive raw material. Therefore, in order to take advantage of the less expensive, crude, higher FFA containing feedstocks, a method for FFA removal is necessary. Fatty acids can be removed from feedstocks by washing them out with diluted base during the degumming process to create soap stock. This step of the degumming process creates waste and biodiesel yield loss. We, along with other researchers, prefer reacting these FFAs with methanol to convert them into the corresponding methyl esters. This reaction serves two purposes: removing the FFA, and the processing issues associated with FFA, and increasing the biodiesel yield. The reaction of a carboxylic acid with an alcohol was first discovered by Fischer and is now better known as the Fischer Esterification Reaction. The reaction is controlled by equilibrium, and special steps are necessary to drive the reaction to completion. Typically, additional methanol is introduced to shift the equilibrium toward products. Another method of driving the equilibrium toward products involves continuously removing water during the reaction of glycerol with FFAs at high temperatures (Luxem et al., 2004). The production costs and environmental impact of biodiesel can be reduced by applying modern catalyst technology, which will allow increased process flexibility to incorporate the use of low-cost, high-FFA feedstock and reduce water and energy requirements. The traditional catalyst used for esterification of acids is sulfuric acid; however, this homogeneous catalyst has many downsides (Canakci & Van Gerpen, 1999). Considerable expense is required for Hastalloy and/or other specialty metals of construction when using sulfuric acid as catalyst. Using homogeneous catalysis, contamination of the product by sulfuric acid can result in neutralization, and the acid must be removed to meet biodiesel specifications and to protect the downstream transesterification process. There is also a possibility that the use of sulfuric acid will produce organic sulfur compounds. These products could then cause the resultant biodiesel to fail to meet sulfur specifications.

Catalysis in Biodiesel Processing ■ 69

Solid catalysts—e.g., synthetic polymeric catalysts, zeolites, and superacids, such as sulfated zirconia and niobic acid—have the strong potential to replace liquid acids, eliminating separation, corrosion, and environmental problems. Lotero and coworkers recently published a review that elaborates on the importance of solid acids for biodiesel production (Lotero et al., 2005). Apart from a few reports (Kiss et al., 2006; Lopez et al., 2005) on solid acid–catalyzed esterification of model compounds, the use of solid catalysts for biodiesel production from low quality real feed stocks has been explored only recently (Kulkarni, 2005). 12-Tungstophosphoric acid (TPA) impregnated on hydrous zirconia was evaluated as a solid acid catalyst for biodiesel production from canola oil containing up to 20 wt% free fatty acids and gave an ester yield of 90% at 200°C (Mbaraka & Shanks, 2006). Propylsulfonic acid–functionalized mesoporous silica catalyst for esterification of FFAs in flotation beef tallow showed a superior initial catalytic activity (90 percent yield) relative to a Nafion NR50 catalyst at 120°C. However, the presence of polar impurities in the beef tallow negatively affected the performance of the recycled acidic mesoporous catalyst (Mbaraka & Shanks, 2006). Supported acid catalysts, such as diarylammonium salts, immobilized on a porous, highly cross-linked organic polymer have been reported to be efficient catalysts for the esterification of FFAs (12 to 40 wt%) in yellow and brown greases to FAME ( > 99% yield). The pretreated greases, with less than one weight percent residual FFA, could be converted to FAME with sodium methylate catalyst. Although this work presents a potential sustainable solution for biodiesel synthesis using readily available high fatty acid feedstock, advances in the recovery and reusability of solid acid catalysts will be required to make the approach industrially useful (Zafiropolous et al., 2007; Ngo et al., 2008). All of the above work is based on laboratory-made catalysts and not commercially available ones. Recently, Özbay and coworkers (2008) studied several commercially available acidic polymeric catalysts for the esterification of fatty acids. The highest FFA conversion (45.7%) was obtained over strongly acidic macroreticular polymer catalysts; AMBERLYST 15 catalyst at 60°C was compared with AMBERLYST 35 catalyst, AMBERLYST 16 catalyst, and DOWEX HCR-W2 catalyst (Özbay et al., 2008). Typical AMBERLYST catalysts consist of cross-linked polystyrene beads that have been functionalized with sulfuric acid (or its equivalent) to produce the active catalytic species. These materials are supplied as spherical beads in a particle-size range of 200–800 microns. All of the above reports have shown that solid catalysts for the esterification of FFAs have one or more problems, such as high cost, severe reaction conditions, slow kinetics, low or incomplete conversions, and limited lifetime. Recently, AMBERLYST BD20 polymeric catalyst with a highly specialized morphology providing excellent accessibility of the supported catalytic sites, even for sterically demanding molecules such as fatty acids, was developed; this catalyst overcomes the traditional drawbacks of limited catalyst lifetime, slow reaction rates, and low conversions (Banavali et al., 2009). Its workingstate porosity has been optimized to achieve high production per unit of reactor volume, and its mechanical properties allow for its use in fixed-bed reactors, as well as stirredbatch reactors, without danger of deforming and clogging the catalyst bed. By comparing

70

■

A.K. Schultz et al.

Fig.4.3.1. Reaction profile comparing heterogeneous catalysis to homogeneous catalysis; differing FFA content.

the AMBERLYST BD20 catalyst with sulfuric acid, virtually identical behavior when employing low FFA feedstocks is observed, but when employing oils with higher FFA content, sulfuric acid catalysis becomes sluggish, and lower overall yields are achieved. This result occurs because the water produced in the esterification reaction hydrates the sulfuric acid. Fig. 4.3.1 compares the reaction profiles of sulfuric acid– and AMBERLYST BD20–catalyzed esterification using two different FFA content oils

Transesterification By far, the most important transformation necessary to produce biodiesel is the transesterification reaction of triglycerides with methanol or ethanol to produce the corresponding methyl or ethyl esters. The transesterification reaction can be run without a catalyst, but it requires excessive temperatures and pressures and produces very low yields, along with numerous by-products (Diasakou et al., 1998). As mentioned above, catalysts help speed reactions at reduced temperatures and pressures, often producing less waste and consuming less energy. Traditionally, the transesterification reaction of a triglyceride with methanol is catalyzed by acids, bases, metals, or enzymes. Each class of catalyst has associated strengths and weaknesses. The following sections will outline the most current research on catalysts used in biodiesel processing. Special attention should be paid to the recent flurry of research implementing heterogeneous catalysis. Heterogeneous catalysts offer a variety of benefits, including an improved environmental impact profile, when compared with

Catalysis in Biodiesel Processing ■ 71

their homogeneous counterparts, since they are reusable, relatively easy to use, and reduce byproduct formation and/or impurity levels.

Base Catalysis There are many processes for commercial biodiesel production, and the design and construction of biodiesel facilities vary widely, but the majority of producers utilize base catalysts for the transesterification of triglycerides with methanol. The catalyst is typically sodium methoxide (sodium methylate), sodium hydroxide (caustic soda), or potassium hydroxide (potash). The reaction mix is stirred and heated just above the boiling point of the alcohol (around 70°C) to speed up the reaction, and typical reaction times vary from one to three hours. Excess alcohol is normally used to ensure total conversion of the fat or oil into its corresponding esters. Typically, one to three percent weight ratio of the catalyst to oil is used. Recently, Dube and coworkers determined the actual amount of catalyst necessary to carry out transesterification in a membrane reactor (Trembley et al., 2008). Using highly purified raw materials and a methanol:oil molar ratio of 24:1, they determined that 0.05% catalyst was necessary to convert triglyceride to its steady state with a one hour reaction time. For reaction times of 2 hours, 0.03% catalyst led to steady state conversion. Unfortunately, processes such as these are very difficult to reproduce routinely on an industrial scale; therefore, excess homogeneous catalyst is necessary. Furthermore, excess base catalyst is necessary to counteract poisons such as FFA and water. The base-catalyzed transesterification reaction includes three steps: (1) the creation of an alkoxide ion from the reaction of the catalyst with (m)ethanol; (2) the reaction of (m) ethoxide with glyceride; and (3) the transfer of a proton from methanol to the glyceride anion, which creates more (m)ethoxide for further reaction. This reaction scheme must occur three times to achieve the overall conversion of TG → DG + FAME; DG → MG + FAME; and MG → glycerol + FAME. The rate-limiting step for the complete conversion of triglycerides to methyl esters and glycerol is the first reaction of TG→ DG (Lotero et al., 2006). Arzamendi and coworkers studied the effect of different alkali metal cations on the catalytic activity of hydroxides and carbonates. Within the periodic group, they saw little difference in activitybetween cations ; for example, Li+, Na+, K+, Rb+, and Cs+ all showed similar catalytic activity (Arzamendi et al., 2008). Hence, there is no productivity gain in using the more expensive hydroxides. Since bases are the most effective and commonly employed homogeneous catalysts for transesterification, many researchers have attempted to develop a heterogeneous catalyst counterpart. To this end, many researchers have studied solid-base catalysis. Metal oxides function as base catalysts, and this field of study has received the most attention from . The M2+ oxides have shown great promise as catalysts, yet their lifetimes are often too short for commercial viability. These materials, although not soluble in alcohol at room temperature often suffer from solubility at elevated temperatures, causing leaching of the catalytically active species and often leading to a homogeneous reaction mechanism. It has been shown that metal oxides, not the corresponding hydroxides, are

72

■

A.K. Schultz et al.

the catalytically active materials of alkaline earth metals. For example, Gryglewicz (1999) showed that Ca(OH)2 was not an active catalyst, whereas CaO is an excellent choice for catalysis. In a 12 hour reaction, the total conversion of oil using this catalyst was 95% (Peterson & Scarrah, 1984). When using CaO as catalyst, researchers found that a great deal of soap is generated. Presumably, the solubility of the catalyst creates a homogeneous catalysis environment. This soluble catalyst is responsible for the high degree of catalytic activity of calcium oxide. To overcome some of the shortcomings of the single metal oxides, researchers have turned to mixtures of oxides as a means of refining the basicity of the catalyst. For example, Ngamcharussrivichai and coworkers (2008) evaluated the use of a Ca and Zn mixed oxide for the transesterification of palm kernel oil. The catalysts were prepared by conventional co-precipitation of the corresponding mixed metal nitrate solutions. The resultant mixed metal oxide was calcined, and under suitable transesterification conditions—60°C, 10% t by weight catalyst, 30:1 MeOH:oil, and one-hour reaction time—94% conversion was achieved. Unfortunately, the catalyst has a limited lifetime, and the authors describe a catalyst regeneration scheme (Ngamcharussrivichai et al., 2008). Other examples of mixed metal oxides involve the use of a mixed magnesiumlanthanum metal oxide (Babu et al., 2008); mixed oxides of calcium with titanium, manganese, iron, zirconium, and cerium (Kawashima et al., 2008); and barium-zinc oxides (Xie & Yang, 2007). Harvey and coworkers have evaluated another potential method of improving the catalytic properties of metal oxides: doping the calcium oxides with alkali metals (MacLeod et al., 2008; D’Cruz et al., 2007). LiNO3/CaO, NaNO3/CaO, KNO3/ CaO, and LiNO3/Mgo exhibited the highest catalytic activity, achieving greater than 90 percent conversion in a standard three hour test. Metal leaching from the catalyst was detected, however, which caused the authors to suspect homogeneous activity. To help with catalyst lifetime and prevent the leaching of catalytically active species, many workers have opted to support the metal oxides on alumina and/or silica. As an example, researchers studied the effectiveness of calcium oxide supported on mesoporous silica (SBA-15, MCM-41, and fumed silica). The sample containing 14% CaO on SBA-15 was the most active (conversion greater than 94%), and no lixiviation was reported (Albuquerque et al., 2008). The use of loaded gamma Al2O3 has attracted much recent attention (Cui et al., 2007; Bo et al., 2007; Ma et al., 2008; Alonso et al., 2007; Arzamendi et al., 2007; Xie & Li, 2006; Xie et al., 2006; Kim et al., 2004). This approach involves adding a more active catalyst to an alumina support matrix. Of the aluminates tested, the most active catalysts were the materials that showed the highest base strength. The major issue became the leachability of the active catalyst from the alumina. For example, Alonso et al. (2007) added potassium carbonate to the alumina, and a K-aluminate-like species was created, as shown by evolved gas analysis, mass spectrometry (EGA-MS), and infrared spectroscopy of the resultant catalyst. The first use of the catalyst showed excellent activity, with conversions reaching 100% after only 1 hour. In subsequent runs, the catalytic activity was greatly diminished. The authors concluded that a homogeneous catalysis mechanism (from the leached material) was responsible for the activity. This result will often occur when using heterogeneous catalysis and greatly diminishes catalyst lifetime. This

Catalysis in Biodiesel Processing ■ 73

technology has lead to the first commercial application of heterogeneous catalysis for transesterification by IFP. Another approach to the heterogeneous base catalysis of transesterification is the use of anion exchange resins.

Acid Catalysis Acidic materials can be used as catalysts for the transesterification reaction, but the reaction rates are frequently too slow for these reactions, and therefore base catalysts are preferred (Meneghetti et al., 2006). However, some examples of potentially useful catalysts have been reported. Lachter and coworkers evaluated the use of strongly acidic polymeric materials as catalysts in the transesterification of various Brazilian vegetable oils (dos Reis et al., 2005). After an eight hour reaction at 60°C with a 300:1 molar ratio of methanol:TG, sulfuric acid homogeneous catalysis showed 44% conversion, whereas the use of strongly acidic polymeric materials showed 64 to 74% conversion.

Natural Clays/Silicates Many natural clays, minerals, and silicas exist. Much work has been devoted to these materials as catalysts in a variety of reactions, including transesterification. These catalyst materials are relatively inexpensive, readily available, and should lend themselves well to heterogeneous catalysis. In Table 4.3.A, we compare clays from different sources as catalysts for transesterification from various published reports.

Metallic Catalysts Many metallic materials have been evaluated as potential catalysts for transesterification reactions. Table 4.3.A. Summary of Clays/Silicates. Catalyst Mg-Al Talcites Montmorillonite KSF Hydrotalcite HT2 Mg-Al Hydrotalcite Mg-Al Hydrotalcite Dolomite Mg-Al Hydrotalcite

Catalyst Quantity Oil 1.5%

Rapeseed

4.0%

Palm

1.0% 3.0% 4.0% 10.0%

7.5%

MeOH: Rxn. Conv Oil Temp. Time (%) Reference 6:1

65°C

4 hr 90.5

Zeng et al., 2008

10:1

155°C

2 hr 78.7

Kansedo et al., 2008

Cottonseed

6:1

200°C

3 hr >95

Barakos et al., 2008

Canola

6:1

60°C

9 hr 71.9

Ilgen et al., 2007

Poultry Palm Kernel

6:1 15:1

120°C 60°C

44 hr 91 3 hr 99.9

Soybean

15:1

65°C

9 hr 67

Liu et al., 2007 Ngamcharussrivichai et al., 2007 Xie et al., 2006

74

■

A.K. Schultz et al.

Basu and Norris (1996) describe the use of a non-alkaline catalyst. Yellow grease with a 9.4% FFA content was converted to methyl ester by a calcium acetate/barium acetate catalyst. The reaction took place at a temperature of 220°C and a pressure of 575 psig for 3 hours. The catalyst was precipitated fromthe ester phase by adding petroleum ether, and it was then filtered for reuse. The catalyst does not produce soaps, allowing the two phases to be separated when the feedstock contains a high FFA content. The use of zinc hydroxide nitrate as catalyst led to effective alcoholysis of palm oil at a temperature range of 100–140°C and afforded methyl palmitate at 95% conversion. The lifetime of the catalyst was not discussed. In a related system, researchers evaluated the oxides of transition metals as catalysts for the transesterification reaction. Porous zirconia, titania, and alumina microparticulate heterogeneous catalysts were shown to be capable of continuous, rapid transesterification reactions, but at pressures of nearly 2,500 psi and temperatures exceeding 300°C (McNeff et al., 2008). Tin and titanium catalysts are also known to catalyze transesterification reactions. Typically, dialkyl Sn oxides, such as dibutyl-tin-oxides are excellent catalysts for transesterification reactions (Abreu et al., 2004). These tin catalysts are removed by distilling the ester from the catalyst. Many researchers have tried to find a way to simplify this cumbersome process, and heterogeneous Sn–containing catalysts have been studied. Researchers at Dow have developed an organometallic tin containing catalyst for the transesterification reaction of acrylate esters (Jiang et al., 1995). Banavali and coworkers (Banavali & Benderly, 2008) showed that supporting tin oxide catalysts on cation exchange resins enhanced performance at reduced temperatures. It appears that the transesterification process proceeded in the presence of the spiked stearic acid, which was confirmed by the presence of various methyl esters. As expected, the esterification of stearic acid also progressed, as the presence of methyl ester of stearic acid was confirmed quantitatively by GCMS. The analysis indicated that about 80 to 90 % conversion of stearic acid/triglycerides took place in first three hours, and after 11 hours, the conversion was about 92 to 100%.

Enzyme Catalysts Lipases are enzymes, ubiquitous throughout the living world, that hydrolyze the ester bonds of water-insoluble substrates. In low water activity systems, they can reverse this reaction, synthesizing esters from FFAs and alcohols. Reports (Zaks & Klibanov, 1984, 1985; Klibanov, 1989) that enzymes are able to retain activity in organic solvents containing small amounts of water have been used to conduct reactions not possible in aqueous systems; these results led to the birth of the substantial new discipline of nonaqueous enzymology. The advantages of enzymatic catalysis over chemical methods in the production of biodiesel include the ability to esterify both acylglycerol-linked and FFAs in one step; the production of a glycerol sidestream with minimal water content and little or no contaminating catalyst; operation at lower temperatures, which reduces process cost; and the potential for catalyst reuse. Barriers to the use of enzymes include high cost relative to inorganic catalysts (in the absence of effective schemes for multiple enzyme use), relatively slow reaction rates, and enzyme inactivation by contaminants in the lipid feedstock and by the polar short-chain alcohol reactants.

Catalysis in Biodiesel Processing ■ 75

Most commercially applied lipases are of fungal or bacterial origin. It is conceivable that one could use a preparation of these organisms as the catalyst in oil-alcohol systems for FAME synthesis, and some work of this kind has been reported (Du et al., 2008; Ranganathan et al., 2008). However, effective contact between the enzyme and its substrates can be achieved by immobilizing the enzyme on a solid carrier. Most reports of enzyme-catalyzed biodiesel production have employed various microbial enzymes immobilized on solid supports to which researchers added lipid feedstocks and alcohol. A substantial collection of literature was reviewed recently (Akoh et al., 2007; Du et al., 2008; Nielsen et al., 2008; Ranganathan et al., 2008).

Enzymatic Catalysis for FAE Production from High Purity Feedstocks For the enzymatic transesterification of refined vegetable oils with methanol as substrate, published reaction conditions generally involve a six-fold molar excess of alcohol over triglyceride (i.e., a two-fold molar excess of alcohol over fatty acid); catalyst loading levels of 1 to 10 wt-%, relative to the oil substrate; the possible addition of one half to a few percent of water; temperatures of 35 to 70oC; and reaction times in excess of 12 hours. The earliest work in this area used sunflower oil as the feedstock, along with three commercially available lipase preparations and petroleum ether as solvent (Mittelbach, 1990). An immobilized Pseudomonas sp. lipase preparation gave the best ester yields. Maximum conversion (99%) was obtained with ethanol. In the absence of solvent, FAE yield was only 3% with methanol as alcohol, whereas with absolute ethanol, 96% ethanol and 1-butanol conversions ranged between 70 and 82%. With alcohols other than methanol, the addition of water increased the transesterification rate by two- to five-fold. As in the paper above, some approaches have involved the addition of an organic solvent to reduce mass transfer limitations through the formation of a monophasic mixture of oil and alcohol. However, this solvent introduces additional cost and technology for purchase, use, and recovery. Thus, most work has been conducted without added solvent. Recently, the use of tert-butanol in this role has been shown to be advantageous (Du et al., 2007; Royon et al., 2007). This hindered alcohol is not an effective substrate for transesterification and thus does not lead to the synthesis of fatty acid butyl esters. However, it has been found to greatly improve enzyme stability (200 reuses and continuous use for over 500 hours have been reported) and to reduce inhibition by methanol and the glycerol released during transesterification. This result probably occurs due to the dissolution of these enzymes in the bulk liquid phase. In another early study (Linko et al., 1998), the lipase-catalyzed alcoholysis of lowerucic acid rapeseed oil without organic solvent in a stirred batch reactor was reported. The best results were obtained with a Candida rugosa lipase, and under optimal conditions, a nearly complete conversion to ester was obtained. Other studies (Selmi & Thomas, 1998) reported the ethanolysis of sunflower oil by LipozymeTM (a commercial immobilized R. miehei lipase) in a medium containing only sunflower oil and ethanol. Despite variations in substrate molar ratio, reaction temperature and time, and enzyme load, ethyl ester

76

■

A.K. Schultz et al.

yields did not exceed 85%. The addition of water (10 wt% ) resulted in lower yields of ester. Yields were improved with the addition of silica, presumably due to its adsorption of the polar glycerol co-product, thereby reducing enzyme inhibition. The enzyme did not retain sufficient activity for reuse. Other approaches to reduce enzyme inhibition have been to use esters such as methyl acetate in place of alcohol as the co-reactant (Xu et al., 2003). This alternative has the advantages of avoiding substrate inhibition of the enzyme and product inhibition by free glycerol because the acetate esters of glycerol are formed . It is unclear whether the process economics will be favorable with these more expensive methyl donors. Other approaches to reducing inhibition by glycerol are to remove it by dialysis (Bako et al., 2002) or by washing with isopropanol. The recent demonstration that enzyme inhibitors can be removed from oils destined for enzymatic interesterification may facilitate industrial scale enzyme-catalyzed biodiesel production (Ibrahim et al., 2008). Using sunflower oil as substrate in a batch-mode reaction geometry, Candida (C.) antarctica lipase immobilized on a marcoporous acrylic resin displayed the best transesterification activity of six enzymes studied (Deng et al., 2005). Among the seven alcohols tested, this enzyme displayed the highest activity with methanol, although all others demonstrated the best activity with 96% ethanol. Maximum FAE yield was slightly greater than 90%. In a study of the enzymatic transesterification of a mixture of soybean and rapeseed oils by various immobilized lipases in the presence of methanol (Shimada et al., 1999), C. antarctica lipase was the most effective catalyst, as has often been observed. The inhibitory effect of high methanol concentrations was realized, and the strategy of its stepwise addition over time was first employed. This procedure resulted in prolonged enzyme activity and converted more than 97% of oil to ester. Stepwise methanol addition subsequently became a common feature in investigations of enzymatic biodiesel production. In such a system, ester yields greater than 90% were reported for the transesterification of soybean oil by Rhizopus oryzae lipase (Kaieda et al., 1999), with it being essential to add water for enzyme activity. A stepwise methanol addition approach was also employed to achieve complete transesterification of soybean oil in a continuous batch reactor employing immobilized Thermomyces lanuginosa lipase (Du et al., 2004). Repeated reuse of the lipase was made possible by removing the bound glycerol with an isopropanol wash. When crude soybean oil was used as substrate, a much lower yield of methyl esters was obtained (Du et al., 2003). The decrease in yield was directly related to the phospholipid content of the oil, which apparently deactivated the lipase. Efficient esterification activity could be attained by pre-immersion of the lipase in crude oil. Several commercially available lipases were screened for their abilities to transesterify tallow with short-chain alcohols (Nelson et al., 1996). Immobilized R. miehei lipase was generally the most effective, resulting in greater than 95% conversion. For methanol and ethanol, but not larger alcohols, enzyme activity was inhibited by even minor amounts of water. n-Propyl, n-butyl, and isobutyl esters also were prepared at high conversion efficiencies (94 to 100%). Esters were also formed using secondary alcohols as alkyl donor, with the lipase from C. antarctica exhibiting the highest activity. The enzymatic conversion of

Catalysis in Biodiesel Processing ■ 77

lard and poultry fat to methyl and ethyl esters, again using the approach of stepwise alcohol addition, has also been reported (Lee et al., 2002).

Enzymatic Catalysis for FAE Production from Low Purity Feedstocks Since they accept both FFAs and acylglycerols as substrates, lipases are also of interest as catalysts for biodiesel production from feedstocks containing both FFAs and acylglycerols, such as waste greases. However, greases and other lower value feedstocks can contain contaminants that inhibit or inactivate enzymes. Since greases and animal fats have higher melting ranges than vegetable oils, solvent-free reactions with them must be run above at least 50oC so that the lipid reactant is a liquid. Studies have been conducted using recycled restaurant grease, a Pseudomonas cepacia lipase preparation, and 95% ethanol in batch reactions (Wu et al., 1999). Optimal reaction conditions were 38oC, 2.5 hours, 14 wt-% lipase, and a grease to ethanol molar ratio of 1:6.6. Ester yield was 85%of maximum theoretical. This yield increased to greater than 96% when 5 wt-% of a C. antarctica lipase preparation was added during the reaction. Subsequent work (Lee et al., 2002) showed that methyl and ethyl esters of lard and restaurant grease could be obtained by C. antarctica lipase-catalyzed alcoholysis. Maximum transesterification of the former reached 74%, while 96% transesterification could be achieved if the grease’s FFAs were removed prior to the reaction. Among a number of lipases tested, P. cepacia immobilized within a phyllosilicate sol-gel matrix gave the best methanolysis (98% yield) and ethanolysis (90%) of restaurant grease, containing 7 wt-% FFA (Hsu et al., 2002). The highest yields were obtained by adding molecular sieves, presumably because they bound the water produced in the esterification of the FFAs in the substrate. Other work has described the use of cotton-immobilized Candida sp. lipase to produce FAME from vegetable and waste oils in a continuous mode (Nie et al., 2006). The enzyme was contained in a series of nine columns, and methanol was added at three locations, while the co-product glycerol was removed with a hydrocyclone. Transesterification yields of 90 and 92%, respectively, were obtained with salad and waste oils (47 wt-% FFA), although reaction times of 22 to 27 hours were required to do so. The catalyst could be reused for more than 450 hours before losing substantial activity. Petroleum ether was present at 50 volume percent excess of the amount of oil. The cost of its removal, however, could reduce the economic viability of the process. Other low-cost feedstocks for biodiesel production include the residual oil present in spent bleaching earths and in soapstock, a byproduct of vegetable oil refining. These feedstocks contain approximately 40% and 50% by weight of oil, respectively. Residual oils present in the waste bleaching earths from soybean, rapeseed, and palm oil refining were extracted with hexane, recovered, and subjected to methanolysis by R. oryzae lipase in the presence of water. Although FAME were produced in reactions that contained water added up to the weight of the oil substrate, the maximum ester production was only 55% (Pizarro & Park, 2003). High viscosity was cited as a possible cause for the low conversions, but inactivation of the lipase by phospholipids in the oil, as reported for unrefined oils (Du et al., 2003), may have also played a role in the case of the soybean and rapeseed oils.

78

■

A.K. Schultz et al.

Lipases have been investigated for the direct production of FAE from the FFAs and acyl lipids present in soapstocks (Haas & Scott, 1996). Only low degrees of esterification were achieved, probably because the reactions were nearly solid, resulting in significant mass transfer limitations. Two-step enzymatic approaches for the conversion of acid oils, produced from soapstock, to FAE have been developed (Ghosh & Bhattacharyya, 1995; Watanabe et al., 2007). First, the acyllipids in the acid oil were hydrolyzed completely using C. cylindracea (now termed C. rugosa) lipase. Subsequently, these acyllipids were esterified using a second lipase (both Rhizomucor miehei and C. antarctica lipases were employed). Conditions were identified to drive this second reaction to a high conversion (greater than 98%), including the addition of glycerol to scavenge water from the organic layer. The lipase used in the esterification reaction was found to retain very high activity for at least 40 repeated reactions. These publications represent only a sampling of those describing enzymatic FAE production. The net effect of this work has been to demonstrate that high-level production is possible. The work has also clearly shown the barriers preventing the use of enzymatic catalysis for biodiesel production on an industrial scale. Among these are relatively slow reaction rates, low conversions, and sensitivity to inhibitors. In addition, enzyme costs far exceed those of the inorganic bases and acids used in conventional biodiesel synthesis. Substantial progress has been made in addressing these restrictions (Nielsen et al., 2008). A key part of this progress has been the realization, noted above, of the value of modulated addition of the alcohol reactant (Shimada et al., 1999; Watanabe et al., 2000). However, there are few universal rules to ensure a successful reaction in all cases. In some cases, FFAs increase the resistance of the enzyme to methanol, resulting in an increase in the observed reaction rate (Du et al., 2007). Other work has shown increased activity by an immobilized enzyme preparation when it had been pretreated with the lipid substrate (Samukawa et al., 2000). This result was postulated to load the interior of the solid catalyst carrier with substrate. Methods for regenerating the enzyme following reaction, e.g., by washing the catalyst with a long-chain alcohol capable of removing inhibitory methanol, have also been described (Chen & Wu, 2003). Progress has also been made in reducing catalyst cost, especially with the recent development of more affordable immobilized lipase preparations (Nielsen et al., 2008). Thus, the day when enzymatic catalysis is adopted for industrial scale biodiesel production may not be far in the future, although it has not yet arrived.

Heterogeneous versus Homogeneous Catalysis Environmental concerns associated with the safe handling and disposal of corrosive mineral acids, such as sulfuric acid, and the contamination and waste issues associated with sodium methoxide have encouraged the research and development of solid catalysts, which could be safer, non-waste producing alternatives for the production of biodiesel. There is also a strong economic impetus to use solid catalysts instead of liquids. The advantages of a solid include reduced equipment corrosion, easier product separation, less potential contamination in waste streams, and its recovery and reuse. Using a solid may

Catalysis in Biodiesel Processing ■ 79

Fig.4.3.2. Simplified flow sheet of the EsterFip-H process.

also increase the number of processing options, such as slurry or a fixed-bed reactor. The selectivity may also be improved by using a solid catalyst.

Commercial Processes Many researchers are working to develop a commercially viable heterogeneous transesterification process. The EsterFip-H™ (Bloch et al., 2008) process, developed at the Institut Français du Pétrole, was first introduced commercially in 2006. The process diagram is given in Fig.4.3.2. The process involves the use of a catalyst, consisting of a mixed oxide of zinc and aluminum, which promotes the transesterification reaction without catalyst loss (Stern et al., 1999). The desired degree of transesterification required to produce FAME on specification is reached in two successive stages, removing the glycerol between stages. Fixed-bed plug flow reactors are utilized in this process to achieve better conversion than stirred-tank reactors, and use of this column configuration allows for continuous biodiesel production. The reaction is carried out using excess methanol at ~200 oC. (Delfort et al., 2004)

Future Outlook Given that biodiesel itself represents a “green” fuel—i.e., one whose technology enhances global sustainability—“green” catalytic processes for its production would further augment its desirability. Features of such green catalysis would be its relatively benign environmental impact and its high yields and selectivities in the production of both biodiesel and co-product, glycerol. The term also implies that biodiesel production will occur with low

80

■

A.K. Schultz et al.

waste and high energy efficiency. The design of these types of catalysts will be driven by both the ingenuity of catalyst scientists and the creative input of the engineering community. These factors will drive conversion of biodiesel production processes from today’s use of more hazardous, waste-generating homogenous technologies to heterogeneous processes with environmental, as well as economic, benefits. Still relatively unexplored is the design of catalysts that will work at much reduced temperatures and pressures and with substantially higher catalytic rates. This development will allow biodiesel production to become more energy efficient and more competitive, in terms of capital costs. The recent progress described above, and the continuing evolution of solid catalysts by careful manipulation at the molecular level, will lead to the further improvement of catalysts and production processes in the years to come.

References Abreu, F.R.; D.G. Lima; E.A. Hamú; C. Wolf; P.A.Z. Suarez. Utilization of Metal Complexes as Catalysts in the Transesterification of Brazilian Vegetable Oils with Different Alcohols. J. Mol. Cat. A-Chem. 2004, 209, 29–33. Akoh, C.; S-W. Chang; G-C. Lee; J-F. Shaw. Enzymatic Approach to Biodiesel Production. J. Agric. Food Chem. 2007, 55, 8995–9005. Albuquerque, M.C.G.; J. Jimenez-Urbistondo; J. Santamaria-Gonzalez; J.M. Merida-Robles; R. Moreno-Tost; E. Rodriguez-Castellon; A. Jimenez-Lopez; D.C.S. Azevedo; C.L. Calalcante; P. Maireles-Torres. CaO Supported on Mesoporous Silicas as Basic Catalysts for Transesterification Reactions. Appl. Catal., A 2008, 334, 35–43. Alonso, D.M.; R. Mariscal; R. Moreno-Tost; M.D.Z. Poves; M.L. Granados. Potassium Leaching during Triglyceride Transesterification Using K/Gamma Al2O3 Catalysts. Catal. Comm. 2007, 8, 2074–2080. Arzamendi, G.; E. Arguinarena; I. Campo; S. Zabala; L.M. Gandia. Alkaline and Alkaline–Earth Metal Compounds as Catalysts for the Methanolysis of Sunflower Oil. Catal. Today 2008, 133, 305–313. Arzamendi, G.; I. Campo; E. Arguinarena; M. Sanchez; M. Montes; L.M. Gandia. Synthesis of Biodiesel with Heterogeneous NaOH/Alumina Catalysts; Comparison with Homogeneous NaOH. Chem. Eng. J. 2007, 134, 123–130. Babu, N.S.; R. Sree; P.S.S. Prasad; N. Lingaiah. Room-Temperature Transesterification of Edible and Nonedible Oils Using a Heterogeneous Strong Basic Mg/La Catalyst. Energy Fuels 2008, 22, 1965–1971. Bako, K.B.; F.C.S. Kova; L. Gubicza; J.K. Hansco. Enzymatic Biodiesel Production from Sunflower Oil by Candida antarctica Lipase in a Solvent Free System. Biocatal. Biotransform. 2002, 20, 437–439. Banavali, R.M.; A. Benderly. U.S. Patent 20,080,015,375, 2008. Banavali, R.M.; K. Jerabek; A.K. Schultz. Heterogeneous Catalysis and Process for the Production of Biodiesel from High Free-Fatty Acid-Containing Feedstocks. Catalysis of Organic Reactions; M. Prunier, Ed.; CRC Press: New York, 2009; pp 279–289. Barakos, N.; S. Pasias; N. Papayannakos. Transesterification of Triglycerides in High and Low Quality Oil Feeds over an HT2 Hydrotalcite Catalyst. Bioresour. Technol. 2008, 99, 5037–5042.

Catalysis in Biodiesel Processing ■ 81

Basu, H.N.; M.E. Norris. U.S. Patent 5,525,126, 1996. Bloch, M.; L. Bournay; D. Casanave; J.A. Chodorge; V. Coupard; G. Hillion; D. Lorne. Fatty Acid Esters in Europe: Market Trends and Technological Perspectives. Oil & Gas Science and Technology 2008, 63, 405–407. Bo, X.; G.M. Xiao; L.F. Cui; R.P. Wei; L.J. Gao. Transesterification of Palm Oil with Methanol to Biodiesel Over a KF/Al2O3 Heterogeneous Base Catalyst. Energy Fuels 2007, 21, 3109–3112. Canakci, M.; J. Van Gerpen. Biodiesel Production via Acid Catalysis. Trans. ASAE. 1999, 42, 1203–1210. Chen, J-W.; W-T. Wu. Regeneration of Immobilized Candida antarctica Lipase for Transesterification. J. Biosci. Bioeng. 2003, 95, 466–469. Cui, L.F.; G.M. Xiao; B. Xu; G.Y. Teng. Transesterification of Cottonseed Oil to Biodiesel by Using Heterogeneous Solid Basic Catalysts. Energy Fuels 2007, 21, 3740–3743. D’Cruz, A.; M.G. Kulkarni; L.C. Meher; A.K. Dalai. Synthesis of Biodiesel from Canola Oil Using Heterogeneous Base Catalyst. J. Am. Oil. Chem. Soc. 2007, 84, 937–943. Delfort, B.; G. Hillion; G. Martino. Biodiesel through Transesterification of Vegetable Oils—A New Solid Catalyst Based Process. Trans. ASAE. 2004, 28-028, Deng, L.; X. Xu; G.G. Haraldsson; T. Tan; F. Wang. Enzymatic Production of Alkyl Esters through Alcoholysis: A Critical Evaluation of Lipases and Alcohols. J. Am. Oil Chem. Soc. 2005, 82, 341–347. Diasakou, M.; A. Louloudi; N. Papayannakos. Kinetics of the Non-catalytic Transesterification of Soybean Oil. Fuel 1998, 77, 1297–1302. Dos Reis, S.C.M.; E.R. Lachter; R.S.V. Nascimento; J.A. Rodrigues Jr; M.G. Reid. Transesterification of Brazilian Vegetable Oils with Methanol over Ion-Exchange Resins, J. Am. Oil Chem. Soc. 2005, 82, 661–665. Du, W.; W. Li; T. Sun; X. Chen; D. Liu. Perspectives for Biotechnological Production of Biodiesel and Impacts. Appl. Microbiol. Biotechnol. 2008, 79, 331–337. Du, W.; D. Liu; L. Li; L. Dai. Mechanism Exploration during Lipase-mediated Methanolysis of Renewable Oils for Biodiesel Production in a tert-Butanol System. Biotechnol. Prog. 2007, 23, 1087–1090. Du, W.; L. Want; L. Dehua. Improved Methanol Tolerance during Novozyme 435-mediated Methanolysis of SODD for Biodiesel Production. Green Chem. 2007, 9, 173–176. Du, W.; Y. Xu; D. Liu. Lipase-catalyzed Transesterification of Soya Bean Oil for Biodiesel Production during Continuous Batch Operation. Biotechnol. Appl. Biochem. 2003, 38, 103–106. Du, W.; Y. Xu; J. Zing; D. Liu. Novozyme 435-catalyzed Transesterification from Crude Soybean Oils for Biodiesel Production in a Solvent-free Medium. Biotechnol. Appl. Biochem. 2004, 40, 187–190. Ghosh, S.; D.K. Bhattacharyya. Utilization of Acid Oils in Making Valuable Fatty Products by Microbial Lipase Technology. J. Am. Oil Chem. Soc. 1995, 72, 1541–1544. Gryglewicz, S. Rapeseed Oil Methyl Esters Preparation using Heterogeneous Catalysts. Bioresour. Technol. 1999, 70, 249–253. Haas, M.J.; K.M. Scott. Combined Nonenzymatic–Enzymatic Method for the Synthesis of Simple Alkyl Fatty Acid Esters from Soapstock. J. Am. Oil Chem. Soc. 1996, 73, 1393–1401. Hsu, A-F.; K. Jones; T.A. Foglia; W.N. Marmer. Immobilized Lipase-catalyzed Production of Alkyl Esters of Restaurant Grease as Biodiesel. Biotechnol. Appl. Biochem. 2002, 36, 181–186. Ibrahim, N.A.; S.T. Nielsen; V. Wigneswaran; H. Zhang; X. Xu. Online Pre-purification for the Continuous Enzymatic Interesterificaton of Bulk Fats Containing Omega-3 Oil. J. Am. Oil Chem. Soc. 2008, 85, 95–98.

82

■

A.K. Schultz et al.

Ilgen, O.; I. Dincer; M. Yildiz; E. Alptekin; N. Boz; M. Canakci; A.N. Akin. Investigation of Biodiesel Production from Canola Oil using Mg-Al Hydrotalcite Catalysts. Turkish J. Chem. 2007, 31, 509–514. Jiang, Q.; C. McDade; A.W. Gross. U.S. Patent 5,436,357, 1995. Kaieda, M.; T. Samukawa; T. Matsuumoto; K. Ban; A. Kondo; Y. Shimada; H. Noda; F. Nomoto; K. Ohtsuka; E. Izumoto; et al. Biodiesel Fuel Production from Plant Oil Catalyzed by Rhizopus oryzae Lipase in a Water-containing System without an Organic Solvent. J. Biosci. Bioeng. 1999, 88, 627–631. Kansedo, J.; K.T. Lee; S. Bhatia. Feasibility of Palm Oil as the Feedstock for Biodiesel Production via Heterogeneous Transesterification. Chem. Eng. Technol. 2008, 31, 993–999. Kawashima, A.; K. Matsubara; K. Honda. Development of Heterogeneous Base Catalysts for Biodiesel Production. Bioresour. Technol. 2008, 99, 3439–3443. Kim, H.J.; B.S. Kang; M.J. Kim; Y.M. Park; D.K. Kim; J.S. Lee; K.Y. Lee. Transesterification of Vegetable Oil to Biodiesel Using Heterogeneous Base Catalysis. Catal. Today 2004, 93, 315–320. Kiss, A.A.; A.C. Dimian; G. Rothenberg. Solid Acid Catalysts for Biodiesel Production—Towards Sustainable Energy. Adv. Synth. Catal. 2006, 348, 75–81. Klibanov, A.M. Enzymatic Catalysis in Nonaqueous Organic Solvents. Trends Biochem. Sci. 1989, 14, 141–144. Kulkarni, M.G.; R. Gopinath; L.C. Meher; A.K. Dalai. Solid Acid Catalyzed Biodiesel Production by Simultaneous Esterification and Transesterification. Green Chem. 2006, 8, 1056–1062. Lee, K-T.; T.A. Foglia; K-S. Chang. Production of Alkyl Esters as Biodiesel Fuel from Fractionated Lard and Restaurant Grease. J. Am. Oil Chem. Soc. 2002, 79, 191–195. Linko, Y-Y.; M. Lamsa; X. Wu; E. Uosukainen; J. Seppala; P. Linko. Biodegradable Products by Lipase Biocatalysis. J. Biotechnol. 1998, 66, 41–50. Lopez, D.E.; J.G. Goodwin Jr; D.A. Bruce; E. Lotero. Transesterification of Triacetin with Methanol on Solid Acid and Base Catalysts. Appl. Catal., A 2005, 295, 97–105. Lopez, D.E.; K. Suwannakarn; D.A. Bruce; J.G. Goodwin. Esterification and Transesterification on Tungstated Zirconia: Effect of Calcination Temperature. J. Catal. 2007, 247, 43–50. Lotero, E.; J.G. Goodwin Jr; D.A. Bruce; S. Suwannakarn; Y. Liu; D.E. Lopez. The Catalysis of Biodiesel Synthesis. Catalysis 2006, 19, 41–83. Lotero, E.; Y. Liu; D.E. Lopez; K. Suwannakarn; D.A. Bruce; J.G. Goodwin. Transesterification of Triacetin with Methanol on Solid Acid and Base Catalysts. Ind. Eng. Chem. Res. 2005, 44, 5353. Luxem, F.; J.H. Galante; W.M. Troy; R.R. Bernhardt. U.S. Patent 6,822,105, 2004. Ma, H.B.; S.F. Li; B.Y. Wang; R.H. Wang; S.J. Tian. Transesterification of Rapeseed Oil for Synthesizing Biodiesel by K/KOH/gamma-Al2O3 as Heterogeneous Base Catalyst. J. Am. Oil Chem. Soc. 2008, 85, 263–270. MacLoed, C.S.; A.P. Harvey; A.F. Lee; K. Wilson. Evaluation of the Activity and Stability of Alkali-doped Metal Oxide Catalysts for Application to an Intensified Method of Biodiesel Production. Chem. Eng. J. 2008, 135, 63–70. March, J. Advanced Organic Chemistry, Fourth Edition; John Wiley and Sons: New York, 1992. Mbaraka, I.K.; K.J. McGuire; B.H. Shanks. Acidic Mesoporous Silica for the Catalytic Conversion of Fatty Acids in Beef Tallow. Ind. Eng. Chem. Res. 2006, 45, 3022–3028. Mbaraka, I.K.; B.H. Shanks. Design of Multifunctionalized Mesoporous Silicas for Esterification of Fatty Acid. J. Catal. 2005, 229, 365–373.

Catalysis in Biodiesel Processing ■ 83

McNeff, C.V.; L.C. McNeff; B. Yan; D.T. Nowlan; M. Rasmussen; A.E. Gyberg; B.J. Krohn; R.L. Fedie; T.R. Hoye. A Continuous Catalytic System for Biodiesel Production. Appl. Catal., A 2008, 343, 39–48. Meneghetti, S.M.P.; M.R. Meneghetti; C.R. Wolf; E.C. Silva; G.E.S. Lima; M.A. Coimbra; J.I. Soletti; S.H.V. Carvalho. Ethanolysis of Castor and Cottonseed Oil: A Systematic Study Using Classical Catalysts. J. Am. Oil Chem. Soc. 2006, 83, 819–822. Mittelbach, M. Lipase-catalyzed Alcoholysis of Sunflower Oil. J. Am. Oil Chem. Soc. 1990, 61, 168–170. Nelson, L.L.; T.A. Foglia; W.N. Marmer. Lipase-catalyzed Production of Biodiesel. J. Am. Oil Chem. Soc. 1996, 73, 1191–1195. Ngamcharussrivichai, C.; P. Totarat; K. Bunyakiat. Ca and Zn Mixed Oxide as a Heterogeneous Base Catalyst for Transesterification of Palm Kernel Oil. Appl. Catal., A 2008, 341, 77–85. Ngamcharussrivichai, C.; W. Wiwatnimit; S. Wangnoi. Modified Dolomites as Catalysts for Palm Kernel Oil Transesterification. J. Mol. Catal. A-Chem. 2007, 276, 24–33. Ngo, H.L.; N.A. Zafiropoulos; T.A. Foglia; E.T. Samulski; W. Lin. Efficient Two-Step Synthesis of Biodiesel from Greases. Energy Fuels 2008, 22, 626–634. Nie, K.; F. Xie; F. Wang; T. Tan. Lipase Catalyzed Methanolysis to Produce Biodiesel: Optimization of the Biodiesel Production. J. Mol. Catal. B: Enzym. 2006, 43, 142–147. Nielsen, P.M.; J. Brask; L. Fjerbaek. Enzymatic Biodiesel Production: Technical and Economical Considerations. Eur. J. Lipid Sci. Technol. 2008, 110, 692–700. Özbay, N.; N. Oktar; N. Tapan. Esterification of Free Fatty Acids in Waste Cooking Oils (WCO): Role of Ion-Exchange Resins. Fuel 2008, 87, 1789–1798. Peterson, G.R.; W.P. Sacarrah. Rapeseed Oil Transesterification by Heterogeneous Catalysis. J. Am. Oil Chem. Soc. 1984, 61, 1593–1596. Pizarro, A.V.L.; E.Y. Park. Lipase-catalyzed Production of Biodiesel Fuel from Vegetable Oils Contained in Waste Activated Bleaching Earth. Process Biochem. 2003, 38, 1077–1082. Ranganathan, S.V.; S.L. Narasimhan; K. Muthukumar. An Overview of Enzymatic Production of Biodiesel. Bioresour. Technol. 2008, 99, 3975–3981. Royon, D.; M. Daz; G. Ellenrieder; S. Locatelli. Enzymatic Production of Biodiesel from Cotton Seed Oil Using t-Butanol as a Solvent. Bioresour. Technol. 2007, 98, 648–653. Samukawa, T.; M. Kaieda; T. Matsumoto; K. Ban; A. Kondo; Y. Shimada; H. Noda; H. Fukuda. Pretreatment of Immobilized Candida antarctica Lipase for Biodiesel Fuel Production from Plant Oil. J Biosci Bioeng. 2000, 90, 180–183. Selmi, B.; D. Thomas. Immobilized Lipase-catalyzed Ethanolysis of Sunflower Oil in a Solvent-free Medium. J. Am. Oil Chem. Soc. 1998, 75, 691–695. Shimada, Y.; Y. Watanabe; T. Samukawa; A. Sugihara; H. Noda; H. Fukuda; Y. Tominaga. Conversion of Vegetable Oil to Biodiesel Using Immobilized Candida antarctica Lipase. J. Am. Oil Chem. Soc. 1999, 76, 789–793. Stern, R.; G. Hillion; J.J. Rouxel; S. Leporq. U.S. Patent 5,908,946, 1999. Thomson, S.J.; G. Webb. Heterogeneous Catalysis; John Wiley and Sons: New York, 1968. Trembley, A.Y.; P.G. Cao ; M.A. Dube. Biodiesel Production Using Ultralow Catalyst Concentrations. Energy Fuels 2008, 22, 2748–2755. Watanabe, T.; Y. Nagao; Y. Nishida; Takagi; Y. Shimada. Enzymatic Production of Fatty Acid Methyl Esters by Hydrolysis of Acid Oil Followed by Esterification. J. Am. Oil Chem. Soc. 2007, 84, 1015–1021.

84

■

A.K. Schultz et al.

Watanabe, Y.; Y. Shimada; A. Sugihara; H. Noda; H. Fukuda; Y. Tominaga. Continuous Production of Biodiesel Fuel from Vegetable Oil Using Immobilized Candida antarctica Lipase. J. Am. Oil Chem. Soc. 2000, 77, 355–360. Wu, W.H.; T.A. Foglia; W.M. Marmer; J.G. Phillips. Optimizing Production of Ethyl Esters of Grease Using 95% Ethanol by Response Surface Methodology. J. Am. Oil Chem. Soc. 1999, 76, 517–521. Xie, W.L.; H.T. Li. Alumina-supported Potassium Iodide as a Heterogeneous Catalyst for Biodiesel Production from Soybean Oil. J. Mol. Catal. A-Chem. 2006, 255, 1–9. Xie, W.L.; H. Peng; L.G. Chen. Calcined Mg-Al Hydrotalcite as Solid Base Catalysts for Methanolysis of Soybean Oil. J. Mol. Catal. A-Chem. 2006a, 246, 24–32. Xie, W.L.; H. Peng; L.G. Chen. Transesterification of Soybean Oil Catalyzed by Potassium Loaded on Alumina as a Solid-base Catalyst. Appl. Catal. A 2006b, 300, 67–74. Xie, W.L.; Z.Q. Yang. Ba-ZnO Catalysts for Soybean Oil Transesterification. Catalysis Letters 2007, 117, 159–165. Xu, Y.; W. Du; D. Liu; J. Zeng. A Novel Enzymatic Route for Biodiesel Production from Renewable Oils in a Solvent-free Medium. Biotechnol. Lett. 2003, 25, 1239–1241. Zafiropoulos, N.A.; H.L. Ngo; T.A. Foglia; E.T. Samulski; W. Lin. Catalytic Synthesis of Biodiesel from High Free Fatty Acid-Containing Feedstocks. Chem. Commun. 2007, 3670–3672. Zaks, A.; A.M. Klibanov. Enzyme Catalyzed Processes in Organic Solvents. Proc. Natl. Acad. Sci. 1985, 82, 3192–3196. Zaks, A; A.M. Klibanov. Enzymatic Catalysis in Organic Media at 100 oC. Science 1984, 224, 1249–1251. Zeng, H.Y.; Z. Feng; X. Deng; Y.Q. Li. Activation of Mg-Al Hydrotalcite Catalysts for Transesterification of Rapeseed Oil. Fuel 2008, 87, 13–14.

■

■

4.4

■

■

Ion Exchange Resins in Biodiesel Processing Rajiv Banavali, Alfred K. Schultz, Klaus-Dieter Topp, and Mark T. Vandersall Dow Chemical Company, 727 Norristown Rd., Spring House, PA

Introduction There are three well-known routes to biodiesel production from oils and fats: alkalicatalyzed transesterification, acid-catalyzed transesterification, and hydrolysis of the triglycerides to fatty acids followed by esterification of these fatty acids. Most biodiesel is produced today via the base-catalyzed reaction owing to its relatively low temperature and pressure. High conversions are achieved in short reaction times with minimal side reactions. Use of homogeneous base catalysis yields a product from which residual catalyst as well as other impurities must be removed. Many commercial biodiesel production processes exist with the design and construction of biodiesel facilities varying widely. However, the base-catalyzed production of biodiesel generally occurs by mixing the alcohol and catalyst with the oil. The catalyst is typically sodium methoxide (methylate), hydroxide (caustic soda), or potassium hydroxide (potash). The reaction mixture is heated just above (around 70°C) the boiling point of the alcohol to accelerate the reaction. Reaction time varies from 1 to 3 hours. Excess alcohol is normally used to ensure very high conversion of the fat or oil to the esters. The amount of water and free fatty acids in the feedstock must be monitored. If the level of either is too high, problems with soap formation and downstream separation of glycerol may result. Once the reaction is complete, two major products, crude glycerol and crude biodiesel, exist. Each contains a substantial amount of excess methanol used in the reaction. The excess methanol is removed by flash distillation. After methanol removal, two distinct layers result. The glycerol phase is much denser than the biodiesel phase. The two phases can be gravity separated with glycerol drawn off the bottom of the settling vessel. In some cases, a centrifuge is used to separate the two materials more quickly. Once the glycerol and biodiesel phases have been separated, the residual alcohol in each phase is removed by flash evaporation or distillation. The alcohol is recovered using distillation equipment and is reused. Biodiesel often contains residual alkali metals from the catalyst, alkaline earth metals from either the oil source or hard water used in processing, soap/ ■

85

■

86

■

R. Banavali et al

FFA, and glycerol (Van Gerpen, 2005). To meet standards, the crude biodiesel must be further processed to remove these impurities. Ion exchange resin technology, sometimes referred to as dry wash, is well suited for this purpose. The materials are designed to remove unwanted ions from a variety of materials, including water and a wide range of organic streams. The glycerol by-product contains unused catalyst and soaps that are typically neutralized with an acid and sent to storage as crude glycerol. Crude glycerol can be upgraded by distillation or chromatography.

Ion Exchange Resin Technology The discovery of the concept of ion exchange dates back to the middle of the 19th century (Thompson, 1850) when it was noticed that ammonium sulfate fertilizer was transformed into calcium sulfate after percolation through a soil-filled tube. Ion exchange can occur in solution between two electrolyte solutions, or more commonly, it refers to exchange of ions from an electrolyte solution to an ion-containing solid material. Extensive industrial research on ion exchange resins has led to their current widespread use (Kunin, 1958; Helfferich, 1962). Ion exchange resins are supplied as cation exchange resins, anion exchange resins, or adsorbents. As the term implies, cation exchange resins exchange cations from solution onto a resin material. For example, sodium ions in an aqueous solution will be exchanged for H+ in a standard water deionization process. All three classes of resins have been used widely in a plethora of industrial applications such as water purification for nuclear reactors, enzyme support materials, solid-phase catalysis, chemical purification and separation, as well as purification of biodiesel and glycerol from the transesterification reaction.

Impurities in Crude Biodiesel As discussed earlier, both the biodiesel and glycerol layers contain various impurities after separation. Typical impurities in biodiesel include residual alkali metals, such as sodium or potassium, from the catalyst as well as alkaline earth metals from either the oil source or hard water used in processing. Residual alkali and alkaline earth metals can form deposits in fuel injection system components and poison exhaust emissions control aftertreatment systems. Residual soap, FFA and glycerol are often an artifact of the separation process. Often, high soap/FFA leads to very poor separation. Each of these impurities may be removed by a selected resin material. Residual catalyst, typically sodium or potassium ions, can be removed by using a cation exchange resin. Since they are divalent and can occupy two active sites on the ion exchange resin, alkaline earth materials such as calcium or magnesium are more difficult to remove but can also be exchanged onto cation exchange resins. Soaps and FFAs can traditionally be adsorbed by anion exchange resins, and residual glycerol can be removed by adsorption. A complicated system involving the use of three or four different ion exchange media would not be acceptable to the biodiesel industry. Therefore, considerable research has been conducted on multifunctional media capable of removing more than one impurity. Consequently, materials combining both

Ion Exchange Resins ■ 87

ion exchange and adsorption capacity were developed. Crude glycerol resulting from biodiesel production is typically contaminated with salts of fatty acids and inorganic salts. It can also contain methanol and water. Ion exchange technology can upgrade crude glycerol to technical grade.

Biodiesel Purification Using Ion Exchange Resins The use of ion exchange resin technology for purifying biodiesel was introduced by Mittelbach et al (1987). Using ion exchange resin technology, cationic impurities can be removed from crude biodiesel via an ion exchange mechanism. Glycerol can be removed via adsorption. The two mechanisms are related. It is proposed that the sodium atom of the sodium sulfonate salt group is chelated with the hydroxyl groups of the glycerol molecule. Up to three molecules of glycerol can be held by each sulfonate group. Another benefit of adsorption is the potential removal of soap molecules as they accumulate in the hydrophilic, glycerol-saturated boundary layer of the ion exchange resin. (Albus, 2006) To design an industrial scale plant, the useful capacity as well as leakage data for the specified resin materials to be used for biodiesel purification must be known. To determine these values, column runs were performed where a given quantity of ion exchange resin was placed in a column with crude biodiesel passing at a given flowrate through the bed. The resin was highly effective at removing cations, as shown by ultra low leakage. The resin material was efficient at removing the cations as shown by its long lifetime. The experiment below was carried out by exposing 55 g of resin to a flow of

Fig. 4.4.1. Glycerol removal mechanism.

88

■

R. Banavali et al

commercial biodiesel spiked with potassium oleate to give a concentration of 210 ppm potassium ion as a surrogate for all cations, 1000 ppm glycerol, and 2.7% methanol at a flowrate of 153 mL/hr (2.42 BV/h or 2.78 L biodiesel per kg ion exchange resin per hour) at 35°C. The experiment was run to complete ionic exhaustion, i.e., the inlet and outlet concentration of K+ was approximately the same. Glycerol was also effectively removed. Based on these results, the useful capacities for K+ and glycerol were calculated. The concentration of potassium and sodium ions in the effluent was determined by EN 14109 and glycerol by EN 14105.

Capacity for Potassium Ions The breakthrough curve for K+ is shown in Fig.4.4.2. As the useful capacity of the resin depends on the desired endpoint, Table 4.4.A shows the capacity for two different endpoints. To which endpoint an industrial scale plant can be operated depends on the plant design, particularly on the number of columns (see next section for more details). If the plant consists of only one column, then 5 ppm K+ in the effluent must be the endpoint (compare to biodiesel standards in Appendix B). However, the general recommendation is to use at least two columns in series so that one can run to a higher endpoint and achieve a higher useful capacity (in this example, achieving a 30 % higher capacity when running to a 40 ppm endpoint compared to a 5 ppm endpoint).

Fig. 4.4.2. Breakthrough curve for K.

Ion Exchange Resins ■ 89

Table 4.4.A. Capacity for K+ as a Function of Endpoint.

Endpoint (ppm)

Bed Volumes treated

5 40

215 285

Capacity (g) per kg resin 110.0 142.4

Quantity of Biodiesel treated with 1 kg resin (L) 600 793

Capacity for Glycerol The breakthrough curve for glycerol is shown in Fig. 4.4.3. As expected, under these conditions the glycerol breakthrough was observed before the ionic breakthrough and the capacity has been calculated for a 200 ppm endpoint (Table 4.4.B). It could be considered running to a higher endpoint but because the glycerol breakthrough is very sharp, this is not recommended. It should be emphasized that the capacity is valid for the first cycle only. Amberlite™ BD10DRY™ can be regenerated with methanol and will again be able to remove glycerol even when ionically exhausted, as discussed above. The capacity for glycerol may decrease, however, with consecutive cycles. Depending on the level of impurities and the temperature, one may regenerate 5 to 10 times.

Fig. 4.4.3. Breakthrough curve for glycerol.

90

■

R. Banavali et al

Table 4.4.B. Capacity for Glycerol.

Endpoint (ppm) 200

Bed Volumes treated

Capacity (g) per kg BD10DRY

Quantity of Biodiesel treated with 1 kg BD10DRY (L)

170

417

470

Effect of Flowrate on Capacity and Leakage In further studies, flow rate and overall impurity concentrations were varied to study their effect on the effectiveness and efficiency of ion and glycerol removal using ion exchange resins. The flowrate was varied in the range of 0.5 BV/h to 2 BV/h (BV = “Bed Volume” = volume of resin initially loaded into the column). No effect on either leakage or capacity for K+ was observed when operating at these flowrates. At the higher flowrate (2 BV/h), however, the capacity for glycerol drops by 14%. The different behavior for K+ and glycerol is related to the different mechanisms of removal (ion exchange vs. adsorption), as discussed earlier. At even higher flowrates (3 to 10 BV/h) the leakage increases substantially such that even after a short number of BVs the K+ leakage is >5 ppm (out of specification range). It is therefore not recommended to operate at flowrates >2 BV/h.

Effect of Glycerol Concentration The effect of glycerol concentration has been tested using a biodiesel feed with a relatively low glycerol level of 300 ppm. As expected, there is no effect on leakage and capacity for K+ which confirms that the ion exchange mechanism is independent of the glycerol uptake by adsorption. Because glycerol is removed by adsorption, the capacity for glycerol is about 65% lower at an inlet concentration of 300 ppm compared to 1000 ppm.

Effect of Cation (Sodium vs. Potassium) To study the nature of the cation on the effectiveness and efficiency of cation removal, in several experiments sodium was used instead of potassium. Thus, 130 ppm of Na+ was added to the crude biodiesel as sodium oleate. Table 4.4.C lists the capacities for sodium at a flow rate of 2 BV/h.

Table 4.4.C. Capacity for Na as a Function of Endpoint. Endpoint (ppm) 5 40

Bed Volumes treated 215 285

Capacity (g) per kg BD10DRY 71.2 g 87.9 g

Capacity (expressed as g K) per kg BD10DRY 120.4 149.0

Ion Exchange Resins ■ 91

The capacity for Na+ seems to be slightly higher than K+ but this may be an artifact of the analytical measurement techniques. The capacity for glycerol was similar to the run carried out with K+.

Use of Ion Exchange Resins for Biodiesel Purification in Commercial Units Dry wash units have been used commercially for nearly 5 years now. The economics of a dry wash unit versus conventional water wash systems depend on the local situation of a given biodiesel plant so a careful comparison of costs is necessary to determine which is more cost effective. Dry wash systems have been shown to be less costly in several units around the world. For a continuous process, it is recommended to install at least two columns to eliminate downtime when replacing spent resin and to achieve higher useful capacities as outlined earlier. Introducing a second column does not impact the overall resin consumption because the resin in either column will be operated to complete exhaustion. The two columns are switched by valves to alternate between “lead” and “lag” treatment positions (Fig. 4.4.4). The design will depend on the level of impurities in the crude biodiesel. A column design sketch based on 10 MM gallons (37.8 MM liters) per year throughput is shown in Fig. 4.4.5. It should be emphasized that ion exchange purification technology is designed to be used in a water-free process. The polymer beads will swell to up to three times their

Fig. 4.4.4. Alternating flow sequence of a typical “lead-lag” installation.

92

■

R. Banavali et al

Fig. 4.4.5. Design sketch for a purification column for a 10 MM gallon per year biodiesel plant.

original volume as they absorb water. The beads will also swell over their life cycle as they remove impurities, including methanol and glycerol. It is therefore strongly recommended to leave sufficient void space in the columns to accommodate this potential resin volume swell. The ability of the polymer beads to remove glycerol and remove soap depends on the presence of some amount of methanol. If methanol is absent, the polymer beads are not swelled sufficiently and the capacity for glycerol adsorption and cation exchange is low. Also, if methanol is absent, there is a risk that soap in the feed to the column may precipitate at the top of the resin bed, forming a layer of solid material and resulting in high flow resistance and high pressure drop. On the other hand, the presence of too much methanol causes lower glycerol capacity due to desorption effects. In practice, it was found that 1–3% concentration of methanol in crude biodiesel gives good performance for both glycerol adsorption and soap removal.

Ion Exchange Resins ■ 93

Extra Glycerol Removing Capacity An even higher capacity for glycerol can be achieved by adding one or two extra columns of resin dedicated exclusively to glycerol removal (Fig. 4.4.6). These additional glycerol removal columns are inserted just after the phase separation step. The ion exchange resin used in these columns will not impact overall resin consumption because when initially installing the glycerol removal columns not only glycerol but also cations are removed. Therefore consumption of resin in the downstream purification columns will be saved. In the arrangement shown in Fig. 4.4.6., the first set of two columns contains resin exhausted for cation removal but still effective for glycerol adsorption. The second set of two columns contains newer resin and therefore still has capacity for cation removal. When the resin in the first of the latter two columns is exhausted for soap removal, the biodiesel flow can then be switched by valves so that this column becomes one of those used for glycerol removal. The oldest bed of resin would then be taken offline, the resin replaced, and the new bed returned to service for soap removal, typically as the final bed of the four-column sequence. When the resin in a glycerol removal column is saturated with glycerol, the column can be regenerated by rinsing with methanol. Methanol from the column can then be recycled back to the transesterification unit where the methanol is reused. The primary reason for two columns dedicated to glycerol removal is that the plant can continue to operate while the first column is being regenerated. These columns can thus be regenerated several times to re-establish their glycerol removing capacity. However their capacity to remove cations is finite; in steady state operation, these columns will only serve to remove glycerol. Under industrial conditions, the resin in the glycerol columns is usually replaced from time to

Fig. 4.4.6. Typical four-column purification system with separate glycerol removal and ionic purification.

94

■

R. Banavali et al

Fig. 4.4.7. Purification system of a 30 kt/y biodiesel plant using 4 columns in series.

time by the spent resin from the downstream purification columns. Fig. 4.4.7 shows an example of a purification system consisting of four columns operated in series.

Comparison to Other Technologies Washing The removal of water-soluble contaminants is traditionally accomplished by waterwashing the biodiesel. This process purifies biodiesel by removing impurities when the crude biodiesel is contacted with water in a water-wash tower (column). Sinking water droplets, being heavier than biodiesel, dissolve the remaining glycerol along with soaps and salts. The glycerol settles to the bottom of the tank with the water. Washing is currently the predominant method for cleaning biodiesel. The current washing processes vary from a counter-current water-wash method to an air/water bubble wash. Washing is generally done through one of two methods–mist washing or bubble washing (or both). Mist washing consists of gently misting water down onto the biodiesel, so that as it falls through the biodiesel. Soaps dissolve into the water and are thereby removed from the biodiesel phase. Bubble washing consists of using an aerator to bubble water and air up through the biodiesel, with the soaps dissolving into the water as it travels up with the air and then falls back down. Bubble washing is more effective, but causes more agitation which can result in emulsification in case of too much soap. Washing is generally done multiple times (each wash requiring several hours), until further wash steps do not pull out any more soap, evidenced by the wash water remaining clear.

Ion Exchange Resins ■ 95

Additionally, it is not uncommon for a methyl ester/water emulsion to form during production. The regulatory permission and disposal of such wastewater may be difficult or impossible, depending on the plant location. The water-wash method has several limitations, including decreased yields due to methyl ester loss in effluent, high soap levels causing emulsification, high effluent treatment and disposal costs, and the time and cost of drying methyl esters. High soap levels in particular may lead to poor separation, contribute to yield losses and require multiple washes to achieve specification. In some cases, 24 hours are required to effect a single separation. Finally, the water-wash method does not remove water-insoluble impurities. The traditionally employed water-wash method may also necessitate either the purchase of centrifuges to supplement the normal gravityseparation of water from biodiesel. Furthermore, this step significantly increases the need for heated and conditioned water. The need to treat and discharge the resulting effluent water also raises environmental and sustainability concerns. These issues can be overcome with a contained, water recycle system.

Magnesium Silicate Magnesium silicate is a waterless wash method used by some small scale producers. It is an “adsorbent filter aid” that ensures biodiesel quality by removing contaminants. Magnesium silicate has an affinity for polar compounds, thus adsorbing methanol, free glycerol, mono- and diglycerides, metal contaminants, free fatty acids and soap from biodiesel. These materials are then removed from the process through filtration. Synthetic magnesium silicate has high numbers of acidic and basic adsorptive sites. The product can increase the oxidative stability of biodiesel and is used in conjunction with, or in place of, water-wash treatment in the biodiesel production process. With magnesium silicate, the water-wash step can be eliminated, along with the liquid separation and drying steps. It can also replace other methods of removing chlorophyll, metals, and color from biodiesel. Magnesium silicate also has a high affinity for methanol and water, removing their traces from biodiesel. Magnesium silicate being a “dry purification” process can greatly reduce dependency on water and resultant wastewater disposal issues. After glycerol separation and methanol removal in a standard biodiesel production process, magnesium silicate—a fine white powder—is mixed with the unwashed biodiesel in a mixing tank for 5 to 10 minutes. The powder, which can be used in either batch or continuous processes, removes residual methanol, providing a cost savings in the stripping step. The slurry mixture of biodiesel with magnesium silicate powder is passed through a filter press which removes the solid silicate particles. The biodiesel plant is left with a solid “filter cake” to dispose as solid waste.

Diatomaceous Earth Diatomaceous earth is a talcum powder-like naturally occurring siliceous sedimentary mineral compound obtained from skeletal remains of unicellular plants called diatoms. It is used as a filter for removing certain polar impurities from biodiesel. This technique is used very rarely in biodiesel plants.

96

■

R. Banavali et al

References Albus, S. New energy sources – Biodiesel” Specialty Chemicals Magazine, 2006, 38–40. Helfferich, F. Ion Exchange. McGraw-Hill Book Company, New York, 1962. Kunin, R. Ion Exchange Resins. Robert E. Krieger Publishing Company, Malabar, FL, 1958. Mittelbach, M.; Andreae, F.; Junek, H. Verfahren zur Herstellung eines als Kraft- bzw. Brennstoff geeigneten Fettsaeureestergemisches DE3727981C2, 1987. Simpson, D. W. J. Phys. Chem. 1956, 60, 518–21. Thompson, H. S. On the adsorbent power of soils. J. R. Agric. Soc. Engl. 1850, 11, 68. Van Gerpen, J. Biodiesel processing and production, Fuel Processing Technology, 2005, 86, 1097–1107.