Polaina J., MacCabe A.P. (ed.). Industrial Enzymes. Structure, Function and Applications

Подождите немного. Документ загружается.

USE OF LIPASES FOR THE PRODUCTION OF BIODIESEL

319

the presence of a high content of free fatty acids an acid-catalysed process is used

(Fig. 1d). This employs strong acids such as H

, as well as high temperatures

and pressures, and requires longer reaction times than the alkaline process. Industrial

plants where both alkaline and acid processes are performed require that reactors

and accessories are resistant to these aggressive agents; moreover, high safety

standards are needed. Due to these drawbacks, alternative and more environmentally

sustainable routes for biodiesel production are being investigated.

2. BIODIESEL PRODUCTION USING LIPASES

Biodiesel production using lipases was first described by Mittlebach (1990) who

showed that the lipase from Pseudomonas fluorescens was superior to those from

Candida sp. and Mucor miehei for sunflower oil alcoholysis. The alcoholysis was

carried out both in the presence of solvent (petroleum ether) and in solvent free

conditions, and using five homologous alcohols with or without the addition of

water. Since then, subsequent studies have focused on different lipases, different

triglyceride feedstocks, different alcohols and different experimental conditions

(temperature, water content, stoichiometric ratio between reagents, enzyme concen-

tration, solvent use etc).

2.1. Lipases in Non-aqueous Media

It is now well established that enzymes can work with high activities in water-poor

environments usually called non-conventional media (Vermue and Tramper, 1995;

Salis et al., 2005a). The interest in using enzymes in non-aqueous media (organic

solvents, supercritical fluids, solvent-free systems, gaseous media and ionic liquids)

arises from the possibility to perform unusual reactions. Like other hydrolytic

enzymes, lipases can function differently in such media, and instead of triglyceride

hydrolysis they can catalyse transesterification reactions such as the alcoholysis

involved in biodiesel production when suitable reagents and only limited amounts

of water are present.

2.2. Sources of Lipases

Lipases used in biotechnology are normally of microbial origin (Jaeger and Eggert,

2002) and are produced by fermentative processes. A number of commercial lipases

are available for applied biocatalysis (Pandey et al., 1999). Table 1 lists those

most often utilised for biodiesel production. Whilst some are employed as free

powders the majority are used as immobilized preparations. Some of the latter

are commercially available, and in a number of cases the enzymes have been

immobilized on different supports. References given in Table 1 cite first use of a

lipase or its best performance.

320 SALIS ET AL.

Table 1. Source of free and immobilised lipases used for biodiesel production

Lipase source Commercial name Supplier Support Reference

Candida antarctica SP435 Novo Acrylic resin

a

Nelson et al., 1996

Novozym 435 Novo Acrylic resin

a

Shimada et al., 1999

Chirazyme L-2 Roche None (Lee et al., 2002)

Candida cylindracea OF Meito Sangyo None Lara and Park 2004

Candida rugosa - Meito Sangyo None Kaieda et al., 2001

Chromobacterium viscoum - Asahi Celite-545

b

Shah et al., 2004

Cryptococcus spp. S-2 Lipase produced in the researchers’ laboratory None Kamini and Iefuji 2001

Porcine pancreatic - Sigma Anion exchange resin

a

Yesiloglu, 2004

Pseudomonas cepacia PS Amano Sol-gel matrix

b

Noureddini et al., 2005

PS Amano None Kaieda et al., 2001

PS-30 Amano None Abigor et al., 2000

PS-30 Amano Pyllosilicate sol-gel matrix

b

Hsu et al., 2002

PS-D Amano Diatomaceous earth

a

Salis et al., 2005b

Pseudomonas fluorescens - Rhöm GmbH None Mittlebach, 1990

AK Amano None Kaieda et al., 2001

AK Amano Porous kaolinite

b

Iso et al., 2001

AK Amano Polypropylene EP100

b

Soumanou and Bornscheuer, 2003b

Mucor Miehei Lipozyme IM60 Novo Anion exchange resin

a

Nelson et al., 1996

Rhizopus oryzae F-AP15 Amano None Kaieda et al., 1999

Thermomyces lanuginosa Lipozyme TL IM Novo Acrylic resin

a

Du et al., 2003

- Novo Pyllosilicate sol-gel matrix

b

Hsu et al., 2004b

a

: Commercially available immobilised lipases.

b

: Lipases immobilised by researchers in their own laboratories.

USE OF LIPASES FOR THE PRODUCTION OF BIODIESEL

321

2.3. Use of Immobilized Lipases

The use of immobilized enzymes confers two important advantages: i) the ability

to recycle the catalyst and ii) the ability to perform continuous processes. Several

reviews on this topic have been published (Fukuda et al., 2001; Shimada et al.,

2002; Shah et al., 2003). A number of methods for the immobilisation of lipases

on solid supports have been reported (Adlercreutz et al., 1996; Pedersen and

Christensen, 2000). Among these, the best seem to be based on entrapment of

the enzyme in hydrophobic sol-gel matrices (Reetz, 1997) or its adsorption onto

hydrophobic supports such as polypropylene (Bosley and Peilow, 1997; Salis et al.,

2003a). Commercially available lipases are supplied both as lyophilised powders,

which contain other components in addition to the lipase (Salis et al., 2005c), and

immobilied preparations. The immobilized lipase most frequently used for biodiesel

production is lipase B from Candida antarctica (Nelson et al., 1996; Shimada

et al., 1999; Samukawa et al., 2000; Watanabe et al., 2000; Watanabe et al., 2001;

Bélafi-Bakó et al., 2002; Köse et al., 2002; Watanabe et al., 2002; Chen and

Wu, 2003; De Oliveira et al., 2004; Du et al., 2004b; Tuter et al., 2004;

Chang et al., 2005; Lai et al., 2005). This is supplied by Novozymes under

the commercial name Novozym 435 (previously called SP435) and is immobi-

lized on an acrylic resin. The Mucor miehei commercial lipase (Lipozyme

IM60 - Novozymes) immobilized on a macroporous anionic exchange resin has

also been extensively used for the same purpose (Mittlebach, 1990; Nelson

et al., 1996; Selmi and Thomas, 1998; Dossat et al., 1999; Shieh et al.,

2003; De Oliveira et al., 2004). Although commercially immobilized prepara-

tions may find immediate application, the development of new supports is of

considerable interest.

Pseudomonas fluorescens lipase immobilized on porous kaolinite (Toyonite

200-M) gave high conversion ratios for propyl oleate and butyl oleate compared

to those obtained with the lipases from Pseudomonas cepacia, Mucor javanicus,

Candida rugosa and Rhizopus niveus. The Pseudomonas cepacia lipase (PS-30)

immobilized on a phyllosilicate sol-gel matrix was found to be more active than the

lipases of Candida antarctica and Thermomyces lanuginosa immobilized on granu-

lated silica. It was suggested that the higher ester yields of lipase PS-30 may be due

to entrapment of the lipase within the clay sol-gel matrix and its protection from

methanol inactivation. Granulated lipase preparations do not protect the enzymes

from inactivation by polar substrates (i.e. methanol) since they are adsorbed onto

the support (Hsu et al., 2002). Thermomyces lanuginosa and Pseudomonas cepacia

lipases immobilized on a phyllosilicate sol-gel matrix were shown to catalyse ester

formation (80–90% yield) from greases containing a range of free fatty acids from

2.6 to 36% (Hsu et al., 2004b). Porcine pancreatic lipase immobilized by ionic

linkage to a macroporous anion exchange resin was used for the ethanolysis of

sunflower oil in a solvent-free system. High substrate conversion was obtained by

performing the reaction with an oil:alcohol molar ratio of 1:3, at a temperature of



C, 0% of added water and 10% wt of lipase based on the weight of the substrate

(Yesiloglu, 2004). The choice of support seems to influence the methanolysis of

322 SALIS ET AL.

triolein in n-hexane. Although not described in detail, it has been reported that

Pseudomonas fluorescens lipase was significantly more active when immobilized

on polypropylene EP100 compared to celite. A conversion of 72.4% was achieved

in the former case but only 1.5% in the latter (Soumanou and Bornscheuer, 2003b).

It should be pointed out that these two supports have very different morphological

features in terms of surface area, pore size distribution and chemical nature (Bosley

and Peilow, 1997; Barros et al., 1998). These parameters strongly influence enzyme

performance but this interesting subject has not been further investigated. Shah

et al. (2004) immobilized Chromobacterium viscosum lipase on Celite 545 for the

ethanolysis of Jatropha oil and found that this procedure increased ester yield from

62 %, obtained with the free lipase, to 71%.

A procedure for the immobilisation of Pseudomonas cepacia lipase was recently

proposed by Noureddini et al. (2005). The lipase was gel-entrapped by polycon-

densation of hydrolysed tetramethoxysilane and iso-butyltrimethoxysilane. The

immobilized lipase catalysed full triglyceride conversion in a very short time

(30 min), it was very stable and lost little activity when subjected to repeated use.

3. SUBSTRATES USED FOR BIODIESEL PRODUCTION

3.1. Use of Different Oil/Fat Sources

The use of a triglyceride feedstock for biodiesel production depends on regional

availability and economics. Rapeseed oil is the most widely used feedstock in

Europe; soybean is mainly used in the United States, and palm oil is used in tropical

regions (i.e. Malaysia). The main difference between these oils is their fatty acid

composition, and this strongly affects some important features of the final biodiesel

mixture. Table 2 details the compositions of the most common oils suitable for

biodiesel production. The most abundant fatty acids are palmitic, stearic, oleic and

linoleic acids. The main physical and chemical properties of an oil/fat depend on

the chemical structures of its fatty acids (Fig. 2). In this regard, a frequent problem

with biodiesel is its stability to oxidation. In linseed, sunflower and soybean oils

the high contents of linoleic acid confers low stability to oxidation as a result of

the presence of two double bonds. Indeed, the oxidation of unsaturated compounds

Table 2. Main fatty acids constituent of the most common oil/fat sources for biodiesel production

Oil/fat Palmitic acid Stearic acid Oleic acid Linoleic acid Problems

source (C16:0) (C18:0) (C18:1) (C18:2)

Soy bean oil 8 4 28 53 Oxidation stability

Palm oil 42 5 41 10 Low temperature

Rape seed oil 4 1 60 20 –

Sun flower oil 6 4 28 61 Oxidation stability

Beef tallow 26 18 37 10 Low temperature

Jatropha oil 13 7 45 34 Low temperature

USE OF LIPASES FOR THE PRODUCTION OF BIODIESEL

323

C OH

Linoleic acid (C18:2)

C OH

Palmitic acid (C16:0)

Stearic acid (C18:0)

Oleic acid (C18:1)

Figure 2. Chemical structure of the most common fatty acids occurring in oils and fats

proceeds at different rates depending on the numbers and positions of the double

bonds. The CH

groups in allylic positions relative to the double bonds in the fatty

acid chains are those susceptible to oxidation (Knothe, 2005b). By comparison,

palm oil and animal fats contain high percentages of saturated fatty acids that are

responsible for the poor low-temperature properties (i.e. high cloud point and pour

point values) of biodiesel fuel. This constitutes a problem in cold regions during

winter. From these considerations and the data in Table 2, it can be concluded that

rapeseed oil is one of the most suitable sources for biodiesel production. Clearly,

triglyceride source affects the properties of the biodiesel blends. Indeed, it was

found that palm kernel oil ethyl esters have a viscosity of 933mm

/s, a cloud point

of 12



C and a pour point of 8



C, whereas coconut 1-butyl esters have a viscosity

of 734 mm

/s, a cloud point of 5



C and a pour point of −8



C. The properties of

these fuels are likely to be dependent on both the oil and the alcohol used in the

transesterification (Abigor et al., 2000).

3.2. Vegetable Oils

The vegetable oils most studied for use in biodiesel production by biocatalysis

(see Table 3) originate from: soy bean (Nelson et al., 1996; Kaieda et al., 1999;

Kaieda et al., 2001; Shieh et al., 2003), sunflower (Mittlebach, 1990; Bélafi-Bakó

324 SALIS ET AL.

et al., 2002; Soumanou and Bornscheuer, 2003b) and rapeseed (Nelson et al., 1996).

However, some other oleaginous, non-edible species should be mentioned: Jatropha

(euphorbiaceae) is a plant that grows in harsh soils and its seed kernel is 40–60 %

(w/w) oil. This plant cannot be used for edible purposes since its oil contains some

toxic substances, i.e. phorbol esters, that render the oil unsuitable for use in cooking

(Shah et al., 2004). Other triglyceride sources have been explored including the

Nigerian lauric oils palm kernel oil and coconut oil (Abigor et al., 2000), rice

bran oil (Kamini and Iefuji, 2001; Lai et al., 2005), refined cotton seed oil (Köse

et al., 2002), peanut palmolein oil (Soumanou and Bornscheuer, 2003b) and castor

oil (De Oliveira et al., 2004). Regarding biocatalytic processes, almost all sources

of triglycerides can be considered equivalent as enzyme substrates. The different

conversion percentages obtained from the transesterification of palm kernel oil and

coconut oil with lipase PS30 are likely to be due to the different alcohols used

(ethanol and butanol respectively) (Abigor et al., 2000).

3.3. Low Value Triglyceride Feedstocks

The main hurdle in the commercialisation of biodiesel is the cost of the raw

material. Biodiesel – produced by base catalysis - cost more than 0.50 US$/dm

in 2001 as compared with 0.35 US$/dm

for petroleum-based diesel (Zhang et al.,

2003). It has been reported that 60–75% of the price of biodiesel derives from

the cost of the feedstock oil (Krawczyk, 1996). For this reason, low value triglyc-

eride feedstocks are interesting alternatives for biodiesel production. The principle

problem associated with their use is the necessity for preliminary treatments to

render the oil/fat suitable for the transesterification process. Some of these can be

performed by lipases. Attention has also been paid to the use of low-value triglyc-

erides such as those from restaurant grease (Hsu et al., 2002), waste edible oil

(Watanabe et al., 2001) and animal fats, i.e. tallow (Nelson et al., 1996).

Waste bleaching earths from crude vegetable oil refining processes contain

approximately 40% oil by weight. Efficient methanolysis of oils recovered by

organic solvent extraction - identified as originating from soybean, palm and

rapeseed – has been reported for Rhizopus oryzae lipase in the presence of a high

water content and a single injection of methanol (Lara and Park, 2003). In a follow-

up study the same authors found that Candida cylindracea lipase was the most

active enzyme in methanolysis of oil from waste activated bleaching earths when

n-hexane was used as the solvent (Lara and Park, 2004).

Sunflower acid oils mainly consist of 55.6% free fatty acids and 24.7% triacyl-

glycerols. They are the main by-product of the alkali refining process of crude

vegetable oils to produce edible oils, and are obtained by acidification of soapstocks.

This waste oil was transformed into FAME (65% yield) by means of immobilized

Candida antarctica lipase B (15% based on acid oil weight) at 40



C after 1.5 h and

using n-hexane as the solvent (Tuter et al., 2004).

As already mentioned, animal fat produces a biodiesel with poor low-temperature

properties. In order to improve cold temperature resistant biodiesel several strategies

USE OF LIPASES FOR THE PRODUCTION OF BIODIESEL 325

Table 3. Biodiesel production through different triglyceride feedstocks, alcohols, solvents, reactor types

Oil/fat source Alcohol Lipase source Solvent Type of reactor Conversion (c) or

yield (y)

(mol or wt%)

Reference

Sunflower oil Ethanol Pseudomonas

fluorescens

Petroleum

ether

Batch 82 (y) Mittlebach, 1990

Methanol Pseudomonas

fluorescens

Solvent free 3-step batch > 90 Soumanou and

Bornscheuer, 2003b

Methanol Candida antarctica Solvent free Membrane

reactor

97 (c) Bélafi-Bakó et al.,

2002

Ethanol Mucor miehei Solvent free Batch

(4 cycles)

83 (y) Selmi and Thomas,

1998

Ethanol Porcine pancreatic Solvent free Batch 81 (y) Yesiloglu, 2004

Methanol Rhizomucor miehei Solvent free 3-step batch

(8 cycles)

> 80 (c) Soumanou and

Bornscheuer, 2003a

High oleic sunflower

oil

Butanol Rhizomucor miehei n-Hexane Packed bed

reactor

> 80 (c) Dossat et al., 1999

Sunflower acid oil Methanol Candida antarctica n-Hexane Batch 63.6 (y) Tuter et al., 2004

Soybean oil Methanol Pseudomonas cepacia Solvent free Batch ∼ 60(y) Kaieda et al., 2001

Methanol Candida antarctica Solvent free 3-step batch

(20 cycles)

97 (y) Samukawa et al.,

2000

Methanol Thermomyces

lanuginosa

Solvent free Continuous batch 80–90 (y) Du et al., 2003

Methanol Thermomyces

lanuginosa

Solvent free 3-step batch

(15 cycles)

94 (y) Xu et al., 2004

Methanol –

ethanol

Pseudomonas cepacia Solvent free Batch (12 cycles) 67 (y) Noureddini et al.,

2005

(Continued)

326 SALIS ET AL.

Table 3. (Continued)

Oil/fat source Alcohol Lipase source Solvent Type of

reactor

Conversion (c) or

yield (y)

(mol or wt%)

Reference

Methanol Rhizopus oryzae Solvent free Batch 80 (y) Kaieda et al., 1999

Degummed soybean

oil

Methanol Candida antarctica Solvent free 3-step batch

(25 cycles)

93.8 (c) Watanabe et al.,

2002

Soybean and

rapeseed oil mixture

Methanol Candida antarctica Solvent free 3-step batch

50 cycles

98.4 (c) Shimada et al., 1999

Methanol Candida antarctica Solvent free 3-packed-bed

reactors (100

days)

93 (y) Watanabe et al.,

2000

Triolein – safflower

oil

1-propanol

Pseudomonas

fluorescens

1,4-dioxane

Solvent free

Batch

Batch (10 cycles)

Iso et al., 2001

Triolein Fusel oil-like

mixture

Pseudomonas

cepacia

Solvent free Batch 100 (c) Salis et al., 2005b

Nigerian lauric

oils (palm kernel

and coconut)

Ethanol

1-butanol

Pseudomonas

cepacia

Solvent free Batch 72 (c) Abigor et al., 2000

Pseudomonas

cepacia

Solvent free Batch 40 (c)

Castor oil Ethanol Rhizomucor miehei n-Hexane Batch 98 (c) De Oliveira et al.,

2004

Cotton seed-oil Primary and

secondary

alcohols

Candida antarctica Solvent free Batch 91.5 (methanol) Köse et al., 2002

Methanol Rhizomucor miehei Solvent free Batch

(8 cycles)

> 90 (c) Soumanou and

Bornscheuer, 2003b

USE OF LIPASES FOR THE PRODUCTION OF BIODIESEL

327

Rice bran oil Methanol Cryptococcus spp.

S-2

Solvent free Batch 80.2 (y) Kamini and Iefuji,

2001

Methanol Candida antarctica Solvent free Batch 98 (c) Lai et al., 2005

Jatropha oil Ethanol Chromobacterium

viscoum

Solvent free Batch 92 (y) Shah et al., 2004

Waste activated

bleaching earths

(ABE) oil

Methanol Candida cylindracea Diesel fuel Batch ∼ 100 (y) Kojima et al., 2004

Methanol Rhizopus oryzae Water Batch 55 (y) Lara Pizarro and

Park, 2003

Waste edible-oil Methanol Candida antarctica Solvent free Packed-bed

reactor (100

days)

90 (y) Watanabe et al.,

2001

Tallow (other oils) Primary alcohols Mucor Miehei n-Hexane Batch > 90 (c) Nelson et al., 1996

Restaurant grease Methanol Pseudomonas

cepacia

Solvent free Batch 98 (y) Hsu et al., 2002

Methanol Candida antarctica Solvent free Batch 96 (c) Lee et al., 2002

Ethanol Burkholderia

cepacia

Solvent free Packed bed

reactor

> 96 (y) Hsu et al., 2004a

Fractionated lard Methanol Candida antarctica Solvent free Batch 58 (c) Lee et al., 2002

328 SALIS ET AL.

can be followed. Lee et al. (2002) decreased the content of saturated fatty acids

present in lard and restaurant grease by performing an acetone fractionation step

followed by methanolysis catalysed by Chirazyme L-2 (Candida antarctica lipase).

Methanolysis of rice bran oil having a free fatty acid content greater than

18% gave conversions < 68%. A two-step lipase-catalysed (Candida antarctica)

methanolysis of rice bran oil was developed for the efficient conversion of both free

fatty acids and acylglycerides to FAME. More than 98% conversion can be obtained

in 4–6 h, depending on the relative proportion of free fatty acids and acylglycerides

present (Lai et al., 2005).

3.4. Alcohols

As already discussed, for cost reasons methanol is the reagent most frequently used

for triglyceride transesterification. Nevertheless, other alcohols are also used. In

Brazil biodiesel is produced by ethanolysis of triglycerides since ethanol is obtained

cheaply by the fermentation of sucrose from sugarcane. The use of different alcohols

gives different results. Alcoholysis of Nigerian lauric oils catalysed by lipase PS-30

gave different oil conversions with methanol, ethanol, 1-propanol, iso-propanol,

1-butanol and iso-butanol (Abigor et al., 2000). However, this was not only related

to the alcohol since the conversion trend was different for palm kernel compared

to coconut oil. It is worth noting that methanol gave the lowest conversions in both

these cases.

Nelson et al. (1996) used linear and branched alcohols for the biocatalytic trans-

esterification of tallow using hexane as solvent. They found that Candida antarctica

lipase was the most efficient in the transesterification with secondary alcohols,

whereas lipase from Mucor miehei was the most efficient with primary alcohols.

The use of C

-C

linear and branched alcohols from fusel oil, a low-value residue

from ethanol distillation, might constitute an interesting and cheap alternative to

methanol. Salis et al. (2005b) carried out the biocatalytic alcoholysis of triolein

with a fusel-oil like mixture. On a molar basis, fusel oil mainly comprises: isoamyl

alcohol (64.4%), 2-butanol (27.6%), 2-methyl-1-propanol (12.3%), 1-propanol

(5.6%) and 1-butanol (1.3%). These alcohols are not enzyme denaturing and

their esters, mainly the branched ones, improve the low-temperature properties of

biodiesel blends (Dunn, 2005). It should be remarked that the absence of methanol

makes the whole process more environmentally friendly. A different result was

obtained by Kaieda et al. (1999). In their case the ester content decreased in

the series methanol > iso-butanol > ethanol > butanol > propanol in catalytic

alcoholysis of rice bran oil using crude Cryptococcus spp. S-2 lipase. A methyl

ester content of 80.2% was obtained in the presence of a high content of water

(80% of substrate weight).

Conversion of cottonseed oil in a 24 h reaction at 40



C in a solvent free

system has been performed with various alcohols. Low conversions (10%) were

observed with short chain alcohols especially with Thermomyces lanuginosus lipase.

Higher conversions were obtained with Rhizomucor miehei lipase (about 30%)