Lipids as renewable resources: current state of chemical and biotechnological conversion and diversification
- First Online:
- Cite this article as:
- Metzger, J.O. & Bornscheuer, U. Appl Microbiol Biotechnol (2006) 71: 13. doi:10.1007/s00253-006-0335-4
Oils and fats are the most important renewable raw materials of the chemical industry. They make available fatty acids in such purity that they may be used for chemical conversions and for the synthesis of chemically pure compounds. Oleic acid (1) from “new sunflower,” linoleic acid (2) from soybean, linolenic acid (3) from linseed, erucic acid (4) from rape seed, and ricinoleic acid (5) from castor oil are most important for chemical transformations offering in addition to the carboxy group one or more C-C-double bonds. New plant oils containing fatty acids with new and interesting functionalities such as petroselinic acid (6) from Coriandrum sativum, calendic acid (7) from Calendula officinalis, α-eleostearic acid (8) from tung oil, santalbic acid (9) from Santalum album (Linn.), and vernolic acid (10) from Vernonia galamensis are becoming industrially available. The basic oleochemicals are free fatty acids, methyl esters, fatty alcohols, and fatty amines as well as glycerol as a by-product. Their interesting new industrial applications are the usage as environmentally friendly industrial fluids and lubricants, insulating fluid for electric utilities such as transformers and additive to asphalt. Modern methods of synthetic organic chemistry including enzymatic and microbial transformations were applied extensively to fatty compounds for the selective functionalization of the alkyl chain. Syntheses of long-chain diacids, ω-hydroxy fatty acids, and ω-unsaturated fatty acids as base chemicals derived from vegetable oils were developed. Interesting applications were opened by the epoxidation of C-C-double bonds giving the possibility of photochemically initiated cationic curing and access to polyetherpolyols. Enantiomerically pure fatty acids as part of the chiral pool of nature can be used for the synthesis of nonracemic building blocks.
Average annual world oil production in the years 1996 to 2000 amounted to 105.0×106 t and will increase in the years 2016 to 2020 to 184.7×106 t (ISTA Mielke GmbH Hamburg 2002). Eighty to eighty-one percent of the produced oils and fats are consumed as human food; 5–6% as feed. Approximately 14%, 15–17 million tonnes are used by industry (Gunstone and Hamilton 2001). In contrast, the world consumption of fossil mineral oil was approximately 4,000×106 t in the year 2002. The chemical share was about 11% in the European Union (EU).
Where “nonfood” uses are concerned, genetic engineering approaches can make a special contribution to the expansion in the wealth of raw materials available to oleochemistry such as increasing the content of individual fatty acids or drastically changing the oil quality by the introduction of a new fatty acid, e.g., the development of high lauric rapeseed (Biermann et al. 2000a).
New plant oils
New plant oils containing new and interesting functionalities are becoming industrially available (Baumann et al. 1988; Gunstone 2001). Fatty acids containing the C-C-double bond in unusual positions of the alkyl chain and containing conjugated double bonds are most interesting from a chemical view point (Fig. 1). Moreover, fatty acids from the natural chiral pool are exciting substrates for stereoselective transformations to give enantiomerically pure products. We can hope that the increasing usage of renewable feedstocks will also eventually enlarge the agricultural biodiversity. It is to be expected that this is further boosted by the development of genetically engineered plants through metabolic engineering, an area (Biermann et al. 2000a) which is outside of the scope of this article. Significant advances were already made to produce polyunsaturated fatty acids such as docosahexenoic acid, eicospentenoic acid, and arachidonic acid in modified oil-seed crops (Singh et al. 2005; Huang et al. 2004).
Petroselinic acid (6) from the seed oil of C. sativum is an 18:1 acid with unsaturation at C6 (Meier zu Beerentrup and Röbbelen 1987). Meadowfoam (Limnanthes alba) oil contains approximately 65% 20:1 acid with double bond at C5. Both fatty acids show some novel reactivities based on the proximity of the double bond to the carboxyl group. Thus, cyclizations to cyclopentane (Metzger and Mahler 1993) and cyclohexanone (Metzger and Biermann 1993) derivatives were reported (Biermann et al. 2000a).
The seed oil of C. officinalis contains up to 60% of calendic acid [(8E,10E,12Z)-octadecatrienoic acid] (7) with a conjugated and stereochemically well-defined hexatriene system (Janssens and Vernooij 2001). α-Eleostearic acid [(9Z,11E,13E)-octadecatrienoic acid] (8) with a conjugated hexatriene as well is obtained from Chinese wood oil (tung oil). Both are drying oils and nteresting applications in alkyd resins were reported. Both hexatriene fatty acids allow highly regioselective and stereoselective Diels–Alder addition of maleic anhydride to the trans, trans-conjugated diene system (Metzger and Biermann 2006).
Santalbic acid (9) is the main fatty acid of the seed oil of sandalwood [S. album (Linn.)]. It contains a unique conjugated enyne system in the alkyl chain, which could be successfully exploited for highly regioselective additions (Biermann et al. 2000b) and for the selective synthesis of halogenated fatty compounds (Lie Ken Jie et al. 2003).
Vernolic acid (10) can be obtained from the seed oil of V. galamensis and of Euphorbia lagascae. Vernolic acid [(12S,13R,9Z)-12,13-epoxy-9-octadecenoic acid] is an enantiomerically pure unsaturated epoxy fatty acid with interesting applications as binder in coatings and preferentially in photocuring coatings (Crivello and Carlson 1996) Vernolic acid is an interesting substrate for the stereoselective synthesis of enantiomerically pure compounds.
With a production of 8.9×106 t in 1990, soaps still ranked first in worldwide statistics for industrial use of fats and oils and for surfactants (Schumann and Siekmann 2002). In 2005, the global production of oleochemicals—excluding soaps and biodiesel—was estimated to amount to 6.7×106 t/a. The basic oleochemicals—the production in 2000 is given in brackets—are free fatty acids (3.05×106 t/a), methyl esters (0.66×106 t/a), fatty alcohols (1.44×106 t/a), and amines (0.57×106 t/a) and glycerol (0.75×106 t/a) as a by-product (Gunstone 2001). The free fatty acids are obtained by hydrolysis of triglycerides with water in a continuous process at 20–60 bar and 250°C. The fatty esters are produced by transesterification of the triglycerides with the respective alcohol, mostly methanol. Hydrolysis and transesterification can be performed enzymatically at ambient temperature and normal pressure. However, economics restrict up to now the use of this technology (Bühler and Wandrey 1987). It was claimed that by applying modified technologies for fat splitting and direct transesterification of triglycerides, it would become possible to lower the enzyme concentration dramatically, resulting in an even more economic process compared to the classical methods. In addition, the quality of the products becomes better and even process units of around 2,000 t/a of feed become economic (Noweck et al. 2004).
The production of long-chain fatty alcohols is an important industrial process. Catalytic hydrogenation of fatty acid methyl esters gives long-chain fatty alcohols at approximately 200°C and 250–300 bar. Long-chain fatty alcohols are also produced from petrochemical feedstocks by the Ziegler Alfol process from ethylene and by hydroformylation of olefins (Noweck 2002). It is remarkable to note that the share of natural sources is rising. It is most important that fat alcohols derived from fats and oils as renewable feedstocks show a more favorable life cycle assessment (LCA) than petrochemical alcohols (Hirsinger 2001) and show that they are an important example that base chemicals derived from renewable feedstocks can be commercially competitive. It is possible to perform the hydrogenation of the ester group with retention of the C-C-double bond making oleyl alcohol easily available by hydrogenation of the methyl ester of high oleic sunflower or rapeseed oil.
Fatty amines are produced from fatty acids in a multistep process via nitriles followed by hydrogenation to give the primary amines, which are converted to tertiary amines and to quaternary ammonium compounds (quats) (Franklin et al. 2001).
The production of fatty acids and esters from triglycerides gives as a by-product about 10wt% of glycerol. New uses of glycerol and new chemical transformations to interesting products are of increasing importance. Fortunately, glycerol has a melting point of 20°C and could possibly be a suitable compound to be used as interior render with latent heat stores.
Fatty acid methyl esters have found an important new application as biofuel. The biodiesel production in the EU added up in 2003 to 1.4×106 t with steadily increasing tendency. Agenda 21 calls for “criteria and methodologies for the assessment of environmental impacts and resource requirements throughout the full life cycle of products and processes.” A simple metric for the production of biofuels is the overall energy efficiency that is the heating value of biofuel divided by the energy required to produce the biofuel. The biodiesel production in Germany from rapeseed has—without credits for the coproduct glycerol—an overall energy efficiency of 1.9 (Kraus et al. 1999) and the respective from soybean in the USA of about 3 (Sheehan et al. 1998).
Vegetable oils are increasingly used as environmentally friendly industrial fluids and lubricants because of their biodegradability and their favorable water pollution class (Erhan and Perez 2002). The usage as insulating fluid for electric utilities such as transformers (Fields 2004; Lewand 2004) and as additive to asphalt to improve the surface properties (Carmen 2004) are interesting new industrial applications.
Current state of chemical conversion of oleochemicals
It was said that “more than 90% of oleochemical reactions were those occurring at the fatty acid carboxy group, while less than 10% have involved transformations of the alkyl chain. However, future progress will be along the lines of these latter types of reactions with their potential for considerably extending the range of compounds obtainable from oils and fats” (Baumann et al. 1988). Modern methods of synthetic organic chemistry including enzymatic and microbial transformations were applied extensively to fatty compounds for the selective functionalization of the alkyl chain (Biermann et al. 2000a; Biermann and Metzger 2004b).
Diacids, ω-hydroxy fatty acids, and ω-unsaturated fatty acids
Splitting ricinoleic acid with caustic soda in a ratio of 1:1 at 180–200°C gives as major products 2-octanone and 10-hydroxydecanoic acid. Using a ratio of 2:1 and 250–275°C, 2-octanol and sebacic acid (decane diacid) are obtained (Gunstone 2001).
Microbial ω-oxidation of fatty acids, which leads via ω-hydroxy fatty acids to diacids, is of great interest (Biermann et al. 2000b). Cognis developed a metabolically engineered strain of Candida tropicalis to oxidize a terminal methyl group of an alkyl chain. The reaction of oleic acid gives the respective unsaturated ω-hydroxy C18-acid and finally the C18-diacid (Craft et al. 2003).
Very special diacids produced on a commercial scale are dimer fatty acids obtained by heating of unsaturated fatty acids from tall oil at around 230°C with a montmorillonite. Distillation gives a monomer and a dimer fraction containing some trimers. These fractions are very complex mixtures, which are not fully identified (Brutting and Spiteller 1994). Most important, the dimer acids are C36-dibasic acids. They are used mainly for polyamides.
Unsaturated fatty compounds are preferably epoxidized on an industrial scale by in situ performic acid procedure (Baumann et al. 1988). Numerous new epoxidation methods were applied to oleic acid (Biermann et al. 2000a). Chemoenzymatic epoxidation is of considerable interest because this method totally suppresses undesirable ring opening of the epoxide. Initially, the unsaturated fatty acid or ester is converted into an unsaturated percarboxylic acid by a lipase-catalyzed reaction with H2O2 and is then self-epoxidized in an essentially intermolecular reaction (Rüsch gen. Klaas and Warwel 1997).
In the industry, vegetable oil epoxides are currently used mainly as PVC stabilizers. Interesting applications were opened by the possibility of photochemically initiated cationic curing (Crivello and Narayan 1992). The comparison of UV-curable coatings with linseed oil epoxide as binder to a binder produced on a petrochemical basis derived from propylene oxide by LCA showed clear advantages for the renewable raw material linseed oil (Eissen et al. 2002).
Polyether polyols for polyurethanes are mostly produced from propylene oxide. Propylene oxide is one of the top 50 chemicals in terms of production: in 1997, 1.9×106 t were produced in the USA, about 1×106 t were produced in Germany, and approximately 4×106 t were produced worldwide. It is the top 50 chemical with the highest gross energy requirement of about 105 GJ/t (Eissen et al. 2002). Eventually, polyetherpolyols derived from epoxidized fatty compounds may substitute the petrochemical compounds in various applications.
It is also a challenge to synthesize suitable diisocyanates via diamino compounds derived from vegetable oils, making the production of polyurethanes possible completely from renewables.
Chiral and enantiomerically pure oleochemicals
It seems to be remarkable that enantiomerically pure fatty acids as part of the chiral pool of nature were not used very extensively for the synthesis of nonracemic compounds. In contrast, the stereochemistry is lost in most oleochemical transformations performed with, e.g., ricinoleic acid.
Chemical epoxidation of ricinoleic acid gives, with low diastereoselectivity, two diastereomers of epoxidized ricinoleic acid that were transformed in two steps in the enantiomerically pure hydroxy aziridines. The hydroxyaziridines were tested for cytostatic/cytotoxic activity with regard to different tumor cell lines. It is interesting to note that the minor diastereomer in all cases showed stronger activity than the major diastereomer. Similar results were obtained for antimicrobial activity tested on different microorganisms. Vernolic acid was reacted quite analogously to give the respective aziridine (Fürmeier and Metzger 2003).
Current state of enzymatic conversion of fats and oils
Enzymes useful for lipid modification
Synthesis of structured triglycerides
Cocoa-butter equivalent, Betapol
Enrichment of specific fatty acids
PUFA from fish oils
Incorporation of specific fatty acids
PUFA into plant oils
Synthesis of fatty acid derived products
Removal of fatty acids in sn1- or sn2-position (PLA1 or PLA2)
Degumming of oils
Removal of phosphate group (PLC)
Head group exchange (PLD)
Hydroxylation of fatty acids
Precursor for polyesters/lactones
Epoxidation of double bonds
Synthesis of FA-hydroperoxides
By far the most often used biocatalysts are lipases (EC 18.104.22.168, triacylglycerol hydrolases) for which fats and oils are their natural substrates. These enzymes do not require cofactors, many of them are available from commercial suppliers and they exhibit high activity and stability, even in nonaqueous systems. A plethora of publications on the use of lipases has appeared in the last two decades and only the most important and recent examples are highlighted here. Because lipases show chemo-, regio-, and stereoselectivity, they can be used for the tailoring of natural lipids to meet nutritional properties, especially for humans. The most prominent example is the synthesis of cocoa butter equivalents (Quinlan and Moore 1993). Cocoa butter is predominantly a 1,3-disaturated-2-oleyl-glyceride where palmitic, stearic, and oleic acids account for more than 95% of the total fatty acids. Cocoa butter is crystalline and melts between 25 and 35°C providing the desirable “mouth feel.” Unilever (Coleman and Macrae 1977) and Fuji Oil (Matsuo et al. 1981) filed the first patents for the lipase-catalyzed synthesis of cocoa butter equivalent from palm oil and stearic acid. Both companies currently manufacture it using 1,3-selective lipases to replace palmitic acid with stearic acid at the sn1- and sn3-positions. Reactions are usually performed as transesterification or acidolysis of cheap oils using tristearin or stearic acid as acyl donors and a 1,3-specific lipase.
Recent examples for successfully industrialized processes include lipase-catalyzed production of zero-trans margarines (ADM and Novozymes) and diglyceride-based cooking and frying oils (Kao Corp. and ADM) (Watanabe et al. 2004). The zero-trans and reduced trans oils and fats are produced on industrial scale by transesterification using lipase from Thermomyces lanuginosa (TL IM) in combination with a cost-effective immobilization technology (Anonymous 2005). Thus, the use of isolated technical enzymes, in contrast to conventional chemical means, provided cost-effective, simple, and straightforward methods to obtain the desired products. In addition, these biocatalytic processes are environmentally friendly and can thus be seen as a result of sustainable developments. The technology developed by ADM and Novozymes received the Presidential Green Chemistry Challenge Award in 2005.
Phospholipase (PLA1 and PLA2) are used on large-scale for degumming—the removal of phospholipids—of natural fats and oils. Whereas the earlier process used a mammalian phospholipase from porcine pancreas specific for the sn2-position (PLA2), this method was recently changed to the application of a microbial biocatalyst obtained from Fusarium oxysporum, which exhibits sn1-selectivity (PLA1) (Clausen 2001). In both cases, lysophospholipids are formed, which are easily hydratable and therefore allow for the reduction of the phospholipid content below 10 ppm. The microbial enzyme is safer (with respect to enzyme supply and a lowered risk of contamination) and shows better performance in the process, and the products obtained from them are considered kosher and halal.
Whereas phospholipase C has no application in biocatalysis yet, phospholipase D can be used in the head group exchange. This allows for the synthesis of nonnatural phospholipids and the synthesis of compounds bearing natural head groups, i.e., phosphatidyl serine (Skolaut et al. 2005).
To our knowledge, the other enzymes listed in Table 1 do not have large-scale applications yet. One exception is the use of whole-cell systems, i.e., Candida yeast for the biohydroxylation to yield dicarboxylic acids as mentioned above. However, the progress made in gene technology, metabolic engineering, and biocatalysis should make the application of these oxidative enzymes feasible in the near future.
Future perspectives of chemical and biotechnological conversion
Most products obtainable from renewable raw materials may at present not be able to compete with the products of the petrochemical industry, but this will change as oil becomes scarcer and oil prices rise. It is very important to note that fat alcohols derived from fats and oils are an example that shows that base chemicals derived from renewable feedstocks can be commercially competitive with petrochemical products. At present, it can be observed that petrochemical fat alcohols will increasingly be substituted. More examples will come up in the next few years. Middle- and long-chain diacids, ω-hydroxy fatty acids, ω-unsaturated fatty acids, and epoxidized vegetable oils may be the next candidates. The use of modern synthetic methods together with enzymatic and microbiological methods has lead to an extraordinary expansion in the potential for the synthesis of novel fatty compounds. With the breeding of new oil plants, numerous fatty compounds of adequate purity are now available, which makes them attractive for chemical synthesis and industrial use. However, numerous synthetic problems have to be solved and solutions have to be found in the coming years.
U. Bornscheuer thanks the Fonds der Chemischen Industrie (Frankfurt, Germany) for financial support.