Index | Search | Home | Table of Contents

Kleiman, R. 1990. Chemistry of new industrial oilseed crops. p. 196-203. In: J. Janick and J.E. Simon (eds.), Advances in new crops. Timber Press, Portland, OR.

Chemistry of New Industrial Oilseed Crops

Robert Kleiman


  1. INTRODUCTION
  2. NEW OILSEED CROPS
    1. Sources of Medium Chain Fatty Acids—Cuphea
    2. Seed Oils of the Apiaceae (Umbelliferae)
    3. Meadowfoam (Limnanthes alba)
    4. Lesquerella
    5. Jojoba (Simmondsia chinensis)
    6. High-Erucic Acid Sources
    7. Epoxy Oils
    8. Other Potential Species
  3. REFERENCES
  4. Table 1
  5. Table 2
  6. Table 3
  7. Table 4
  8. Table 5
  9. Table 6
  10. Fig. 1
  11. Fig. 2
  12. Fig. 3
  13. Fig. 4
  14. Fig. 5
  15. Fig. 6
  16. Fig. 7
  17. Fig. 8
  18. Fig. 9
  19. Fig. 10

INTRODUCTION

The attraction for developing new oilseeds is several-fold. First, a new crop can provide a significant increase in the level of income for the farmer; second, new crops can contribute to the alleviation of negative balance of payments; and third, they can provide materials that are critical to our national interests, especially if we at present depend on imported materials. The reason we select specific plant species for development is the difference in chemical functionality of the proposed new seed oil from the oils we now have domestically available. This paper deals with only a few of the plants available from the wild that have potential as new sources of industrial materials. The pertinent chemical background of some species that are now cultivated experimentally in the United States will be presented.

NEW OILSEED CROPS

Sources of Medium Chain Fatty Acids—Cuphea

The basic source of medium chain fatty acids is tropically-grown coconut and palm kernel oil. The U.S. has no domestic source of these materials and therefore imports about 500,000 tonnes (one billion pounds) annually of "lauric" oils.

However, many species from the genus Cuphea (Lythraceae) have potential as sources of medium chain triglycerides (Wilson et al. 1960, Miller et al. 1964, Wolf et al. 1983, Graham and Kleiman 1985, Graham et al. 1981). These plants are native to the New World, from Southern U.S. to Northern South America. Most are herbaceous annuals that will grow in many locations. However, Cuphea is only a few years from the wild and still has the characteristics of a wild plant. Those characteristics that differ from cultivated plants are its propensity to seed shatter, its indeterminate flowering nature, and its overall stickiness. If these wild traits can be overcome, Cuphea's chemistry, coupled with the annual and therefore renewable nature of the plant, certainly can make it a new crop.

Table 1 illustrates the diversity in fatty acid composition available in Cuphea germplasm. While there is some variation from accession to accession, the table shows species that are rich in specific single fatty acids. Cuphea painteri, for instance, is very rich in caprylic (8:0) acid (73%) while C. carthagenensis has lauric acid (12:0), as its major fatty acid (81%). Cuphea koehneana is probably the best example of a monoacid seed oil, with more than 95% of its acyl groups as capric acid. As a source for lauric acid, Cuphea ssp. have more to offer than coconut oil (Table 1), because the concentration of lauric acid in the oil is potentially much greater. Isolation of single fatty acids should be easily accomplished and tailor-made fatty acid compositions should be possible.

Oil and protein values were determined only on a handful of accessions because only a few seeds have been collected from many of the species. In those species, oil percent varied from 16 to 42%. Protein (%N x 6.25) levels were from 15 to 24% of the whole seed. Table 2 shows the amino acid composition of two Cuphea species.

The triglyceride analysis of C. lanceolata (Litchfield et al. 1967) shows that the combination of acyl groups within triglyceride molecules is not the result of random distribution but the specific combination of certain fatty acids, particularly C10. Gas chromatography of the intact triglycerides show that the C8 was completely contained in the group with carbon number of C28 indicating that the C30 peak was made up of three C10 fatty acids and not a combination of C8, C10, and C12. These combinations have significance when Cuphea oils are to be used in nutritional and medical applications.

Other members of the Lythraceae have been examined to see if fatty acid compositions are similar to Cuphea (Graham and Kleiman 1987). However, only small amounts of medium chain fatty acids were found.

Seed Oils of the Apiaceae (Umbelliferae)

While the seeds oils from the plants of the Apiaceae (Umbelliferae) do not contain medium chain fatty acids per se, most do contain large amounts of petroselinic (cis-6-octadecenoic) acid. This positional isomer of the common oleic (cis-9-octadecenoic) acid can be cleaved at the double bond to form two industrially useful materials, adipic and lauric acids (Fig. 1). The method of choice to perform this type of cleavage is ozonolysis. For example, the production of pelargonic and azeleic acids is from the ozonolysis of oleic acid. However, the production of lauric acid by this means will have to compete with lauric acid from a much cheaper starting material, coconut oil.

With the possible exception of dill, most Umbelliferae are not produced in the U.S. for their seed oil. Production of such crops as celery, dill, carrots is for food and or condiments. However, seed yields for common fennel of over 2800 kg of seed per hectare have been reported. With agronomic and breeding research, higher seed yields and oil content should be achieved. Wild species of the Apiaceae have been reported that have as much as 40% seed oil containing 80% petroselenic acid (Kleiman and Spencer 1982).

Meadowfoam (Limnanthes alba)

The oil composition from meadowfoam seed (Table 3, Fig. 2) is unique in several ways. First, over 95% of its acyl groups are longer than C18; secondly, about 90% of these fatty acids have double bonds in the delta-5 position; and thirdly, there are essentially no polyunsaturates (Smith et al. 1960, Bagby et al. 1961, Phillips et al. 1971). Along with the 60% cis-5-eicosenoic acid, the 8% cis-5-docosenoic acid, and the 10% erucic acid one finds 15-30% of 22:25,13. The latter is a polyunsaturated dienoic acid, but six carbon atoms separate the double bonds, thus making the acid react essentially as a monoenoic fatty acid in terms of its oxidative stability.

The oil should be oxidatively stable, not only because of the lack of polyenoic fatty acids and its long chain nature but also because the delta-5 bond is more stable than olefins with the double bond in the center of the fatty acid molecule (Kaneniwa et al. 1988). The cis-5 bond also allows lactonization to proceed easily, especially after epoxidation (Fore and Sumrell 1966) (Fig. 3 and 4). Oxidative cleavage of the 20:15 at the double bond results is glutaric and pentadecanoic acids. The use of the latter 15:0 acid should be evaluated because it is not now industrially available. The 22:25,13 has been successfully epoxidized (Carlson et al. 1989). The reactive diepoxide should be a useful industrial intermediate.

Natural meadowfoam oil is in the form of triglycerides. However, they have been converted to liquid wax-esters by reduction of fatty acids to alcohols and reesterification with the unreacted fatty acids (Miwa 1972, Nieschlag et al. 1977). These C40 and C42 molecules should be useful, both sulfurized and unreacted, in lubricants.

Meadowfoam (Limnanthes alba) is a member of the Limnanthaceae. Though the presence of the sulfur containing glucosinolates is usually associated with the Brassicaceae (Cruciferae), meadowfoam seed contains significant levels of glucosinolates that upon hydrolysis yield both a volatile mustard oil, m-methoxybenzyl isothiocyanate, and the nonvolatile 5,5-dimethyloxazolidine-2-thione (Daxenbichler and VanEtten 1974) (Fig. 5). These products must be considered when meadowfoam defatted meal is used as animal feed. Meadowfoam is grown as a winter annual principally in the Willamette Valley of Oregon where yields of over 1.1 t/ha are routine (Jolliff, pers. commun.). Breeding and agronomic research are ongoing at Oregon State University and at other locations. The genus is native to southern Oregon and northern California.

Lesquerella

Plants of the genus Lesquerella are members of the Brassicaceae (Cruciferae) and are native to North America. We became interested in these plants when we found that their seed oils are rich in hydroxy fatty acids (Smith et al. 1961, Mikolajczak et al. 1962, Smith et al. 1962, Kleiman et al. 1972).

Castor oil is the only commercial source of hydroxy acids, and the fatty acid in the oil is ricinoleic, 12-hydroxy cis-9-octadecenoic, acid, found at the 85% level. Though most Lesquerella species have small amounts of this acid, large concentrations of other hydroxy acids are found. The structures of these compounds are shown in Fig. 6. Though most species in this genus have several hydroxy fatty acids, they are generally rich in only one. Table 4 illustrates the different types of Lesquerella compositions. The dienoic hydroxy fatty acids, densipolic and auricolic, have potential as fatty intermediates and tung oil replacement. However, we are presently concerned with the high lesquerolic acid types. At the USDA's Water Conservation Laboratory, Phoenix, Arizona, agronomic and breeding research is now underway on one of these, Lesquerella fendleri. Yields of better than 1500 kg/ha have been achieved in just a few years (A.E. Thompson, pers. commun.). With continued work in this area increases in seed yields, lesquerolic acid, and oil content are expected.

The structure of lesquerolic acid is homologous to ricinoleic acid. It has two additional carbon atoms at the carboxyl end of the molecule. This similarity allows many of the same reactions and, presumably, the same uses as for the castor-based fatty acid. This includes the production of 2-octanol and 2-octanone, depending on reaction temperature during alkali cleavage of the acid. However, this same reaction produces dodecanedioic and hydroxydodecanoic acids from lesquerolic acid instead of the analogous acids from ricinoleic that are two carbons shorter. The C12 dibasic acid, dodecanedioic acid, is the basis for nylon 1212 (Fig. 7). The current raw material for production of dodecanedioic acid is petroleum. Without alkali, a major product from thermal fragmentation of ricinoleic acid is 10-undecenoic acid (Naughton et al. 1979) (Fig. 8). This material, after bromination and subsequent amination, is the starting material for nylon-11. The analogous 12-tridecenoic acid is produced from lesquerolic acid (Fig. 8) and could be used in the production of the monomer for nylon-13. The saturated, via hydrogenation, hydroxy acids, as their lithium soaps, could be useful in the production of greases. The intact triglycerides from Lesquerella should be suitable for many uses. For example, they could substitute for dehydrated and oxidized castor oil. Many of the properties may be enhanced over those of castor oil because of the increased chainlength of this new crop oil.

Like most of its cruciferous relatives, Lesquerella's seed meal contains glucosinolates. There are several different glucosinolates identified in Lesquerella spp. (Daxenbichler et al. 1962). Many of these can be removed easily from the meal if converted to the volatile isothiocyanate (Daxenbichler et al. 1961). The amino acid composition of the seed meal has been reported (Miller et al. 1962a). Its especially high lysine content makes it attractive as a protein supplement.

Jojoba (Simmondsia chinensis)

Made up almost exclusively of liquid wax-esters, the oil of jojoba seed is unique in the plant kingdom (Wisniak 1987). The general structure of these lipids is shown in Fig. 9. Jojoba is often touted as a whale oil substitute, while in fact it has features that, in most applications, make it superior. First of all, whale oil has a significant (about 30%) amount of polyunsaturated triglycerides as part of its total lipids. These make it oxidatively less stable than jojoba, which has essentially no triglycerides, polyunsaturated or otherwise. Secondly, jojoba wax-esters are considerably longer in chainlength, ranging from C40 to C44. The bulk of sperm whale oil has chainlengths of C32 to C36. The fatty acid and alcohol compositions, making up the wax-esters of jojoba oil, are listed in Table 5. The specific combinations of alcohols and acids have been established by mass spectrometry (Spencer et al. 1977).

Jojoba oil's current use is centered on the cosmetic and personal care industry This is due primarily to its ability to lubricate without the sense of greasiness. There are about 6,000 productive hectares (15,000 acres) of jojoba in the southwest U.S. As these areas become more fruitful, and additional areas come on-line, the price of the oil is expected to decrease. As this comes about, more industrially oriented uses for this material will emerge in the market place. A number of derivatives have been made. These include sulfurized and halogenated jojoba oil, for high-pressure applications such as in automobile transmissions, and hydrogenated jojoba as a wax.

The whole jojoba seed contains about 15% protein (N x 6.25). When considering the large amount of oil in the seed (50%), the defatted meal has about 30% protein. However, the seed contains also about 11% antinutritional compounds. These are simmondsin, simmondsin 2'-ferulate, 5-desmethylsimmondsin, and didesmethylsimmondsin (Wisniak 1987) as illustrated in Fig. 10. Several methods are now under development to eliminate these materials in order to use the meal as a nutritional animal feed.

High-Erucic Acid Sources

Crambe (White and Higgins 1966) (Crambe abyssinica) and industrial rape (Ackman 1983) (either Brassica napus or B. campestris) are potential sources for erucic acid in the U.S. While European rape has long been imported into this country, it is recently being introduced as a new crop for the U.S. Crambe has been suggested for many years as an erucic acid source and has been grown intermittently on small areas. All three species are members of the Brassicaceae (Cruciferae).

Erucic (cis-13-docosenoic) acid is now converted to erucamide for use as a slip agent in the manufacture of polyethylene sheets. The acid also has been cleaved to form brassylic and pelargonic acids. The brassylic acid has been used to produce nylon 1313. Both erucic acid itself and its cleavage product, brassylic acid, have potential in other commercial products such as plasticizers, lubricants, and surfactants.

The seed meals of these crops have high protein levels and good amino acid compositions (Miller et al. 1962) (See Table 3). While the glucosinolates in rapeseed can be eliminated through breeding, no germplasm is yet available to lower the 8% glucosinolates in crambe seed. However, crambe meal has been cleared for use at the 5% level in animal feeds.

Epoxy Oils

Many plants produce seed oils with epoxy fatty acids (Earle 1970). A few produce as much as 80% of one epoxy acid, vernolic (12,13-epoxy-cis-9-octadecenoic) acid. Those with potential as crops are Vernonia galamensis (Perdue et al. 1986), Stokes Aster (Earle 1970), and Euphorbia lagascae (Kleiman et al. 1965). The fatty acid compositions of the seed oils of these species are given in Table 6. None of these species are now grown in the U.S. However, Vernonia galamensis is grown experimentally in its native equatorial Africa. Until recently, the long daylight hours of the U.S. did not allow flowering of this species. Workers at the Water Conservation Laboratory have now grown a few lines whose flowering pattern looks promising for that area. Stokes aster (Stokesia laevis), a southeast U.S. herbaceous perennial, has been grown experimentally with promising results (T.A. Campbell, pers. commun.). Euphorbia lagascae is naive to southern Europe and is being considered as a crop there.

These epoxy oils have an advantage over commercially epoxidized oils in that the location, number, and configuration of epoxy and olefinic groups are rigorously known. The oil of Vernonia galamensis forms excellent baked coatings on steel (Carlson et al. 1981) and interpenetrating polymer networks with other polymers (Sperling et al. 1983). These oils have potential in the production of plastics, paints, and lubricants.

Other Potential Species

I have tried to summarize the chemical aspects of the new industrial oil seed crops currently under exploration in the United States. There are a number of other proposed crops that are not now being looked at here but should be mentioned. Dimorphotheca pluvialis produces an oil rich in dimorphecolic acid (Earle et al. 1962), a C18 conjugated dienol, and Crepis alpina (Spencer et al. 1969) is rich in crepenynic acid, an acetylenic acid. Hopefully, in the future we can take greater advantage of the diversity of nature and use American produced renewable raw materials for our industrial needs.

REFERENCES


Table 1. Fatty acid composition of some Cuphea seed oils.

Distribution (% of total fatty acids)
Species 8:0 caprylic 10:0 capric 12:0 lauric 14:0 myristic Others
C. painteri 73.0 20.4 0.2 0.3 6.1
C. hookeriana 65.1 23.7 0.1 0.2 10.9
C. koehneana 0.2 95.3 1.0 0.3 3.2
C. lanceolata 87.5 2.1 1.4 9.0
C. viscosissima 9.1 75.5 3.0 1.3 11.1
C. carthagenensis 5.3 81.4 4.7 8.6
C. laminuligera 17.1 62.6 9.5 10.8
C. wrightii 29.4 53.9 5.1 11.6
C. lutea 0.4 29.4 37.7 11.1 21.4
C. epilobiifolia 0.3 19.6 67.9 12.2
C. stigulosa 0.9 18.3 13.8 45.2 21.8
Coconut 8 7 48 18 19


Table 2. Amino acid compositions of some potential oil seed species.

Content (g/16 g of protein)
Amino acids Cuphea painteri Limnanthes douglasii
(Meadowfoam) VanEtten et al. 1961
Lesquerella fendleri
Miller et al. 1962a
Simmondsia chinensis
(Jojoba) Wisniak 1987
Aspartic acid 8.0 8.0 7.2 10.0
Threonine 3.0 4.3 4.5 6.0
Serine 4.9 4.4 4.6 6.9
Glutamic acid 15.3 16.3 13.7 10.4
Proline 3.5 4.2 6.7 6.0
Glycine 4.5 6.1 5.9 16.2
Alanine 3.9 4.4 4.5 6.3
Valine 4.6 5.0 4.8 5.4
Cystine 1.2 1.4 1.8 3.5
Methionine 1.7 1.4 1.3 0.9
Isoleucine 3.9 3.8 3.6 3.1
Leucine 6.2 6.9 5.8 6.8
Tyrosine 2.7 2.2 3.0 3.6
Phenylalanine 3.8 3.7 3.8 3.3
Lysine 3.9 6.9 6.6 4.3
Histidine 2.2 2.2 2.5 1.7
Arginine 10.2 7.5 7.9 5.6


Table 3. Seed characteristics of Limnanthes alba.

Seed characteristic Content
Oil (%) 17-29
20:15 (%) 50-65
22:l5 + 22:113 (%) 10-29
22:25,13 (%) 15-30
Protein (%) 11-28
Glucosinolates (%) 3-10
Seed wt (g/1000) 4.2-9.8


Table 4. Fatty acid composition of Lesquerella spp.

Content (% of total fatty acids)
Speciespalmitic 16:0 palmitoleic 16:1 stearic 18:0 oleic 18:1 linoleic 18:2 linolenic 18:3 ricinoleic 18:1-OH densipolic 18:2-OH lesquerolic 20:1-OH auricolic 20:2-OH Others
L. lindheimeri 1.5 0.5 1.5 11.8 5.8 0.9 0.9 tr 72.6 tr 4.0
L. fendleri 1.5 1.4 2.4 15.2 7.6 13.1 0.3 0.2 53.2 3.8 0.0
L. densipilia 5.8 1.2 2.6 22.1 3.0 10.1 2.0 50.7 1.3
L. auriculata 5.8 1.4 5.4 27.0 3.0 6.9 5.3 2.1 9.8 32.0 1.9


Table 5. Typical jojoba oil composition.

ChainlengthFatty acids (%) Alcohols (%)
16 2 1
18 13 1
20 71 51
22 13 42
24 1 5


Table 6. Fatty acid composition of epoxy oils.

Content (% of total fatty acids)
Fatty acids Vernonia
galamensis
Stokesia
laevis
Euphorbia
lagascae
16:0 (palmitic) 2.8 2.6 4.1
18:0 (stearic) 2.6 1.1 2.0
18:1 (oleic) 4.0 6.9 20.7
18:2 (linoleic) 12.8 13.7 9.2
18:3 (linolenic) 0.1 0.2 0.4
Vernolic acid 76.4 74.1 61.8
Others 1.3 1.4 1.8


Fig. 1. Ozonolysis of petroselinic acid.


Fig. 2. Delta-5 acids of meadowfoam (Limnanthes) seed oil


Fig. 3. Production of 4-eicosanolactone.


Fig. 4. Production of 6-hydroxy-5-eicosanolactone.


Fig. 5. Reaction of glucosinolates from meadowfoam seed.


Fig. 6. Hydroxy acids from Lesquerella species.


Fig. 7. Thermal-alkali reactions of hydroxy-monoene fatty acids.


Fig. 8. Thermal reaction of hydroxy-monoene fatty acids.


Fig. 9. Hydrolysis of jojoba wax-esters.


Fig. 10. Antinutritional compounds from jojoba seed.

Last update August 26, 1997 by aw