Interest in, and consumption of, polyunsaturated fatty acids (PUFA) has greatly increased in recent years due to their health benefits, particularly related to brain and heart function. Urea inclusion compounds (UICs), clanthrates formed between linear fatty acyl groups and urea, have been shown to be efficient in selectively removing long-chain fatty acyl groups possessing low degrees of saturation. Their formation enables the robust isolation of PUFA from fatty acid mixtures produced from fish oils and other oils encountered in nature. UICs present many advantages compared to other separation approaches, including enhanced eco-friendliness, mild operating conditions and low materials costs. This article reviews the underlying science of UIC formation, the practical aspects of fatty acid fractionation and presents a prospectus for future areas of activity.
by Dr Douglas G. Hayes
The need for eco-friendly fatty acid purification
The concern for improved health throughout the world has led to an increased scrutiny of food products. It is known that a diet rich in unsaturated fats and poor in saturated fats reduces obesity, a major contributor to cardiovascular disease among other health problems. In particular, polyunsaturated fatty acids (PUFA) have received great attention as dietary supplements because of their benefit for the function of the heart and the brain, and their ability to lower blood triglyceride and LDL cholesterol levels and increase the level of HDL cholesterol. PUFA are particularly important for proper brain development in children, and are important for patients recovering from angioplasty. PUFA such as eicosapentaenoic (20:5-5c,8c,11c,14c,17c), and docosahexaenoic (22:6-4c,7c,10c,13c,16c,19c) acids (EPA and DHA, respectively) have been shown to be beneficial in the treatment of heart disease, diabetes, attention deficit/hyperactivity disorder (ADHD), hyperactivity disorder, cystic fibrosis and Alzheimer's disease. Arachidonic (20:4-5c,8c,11c,14c) acid, (AA), is reported to benefit the performance of athletes. Gamma linolenic (18:3-6c, 9c, 12c) acid (GLA), is effective in the treatment of arthritis, obesity, vascular disease, eczema, ADHD and high blood pressure.
A simple, safe and inexpensive method for separating saturated and polyunsaturated free fatty acids (FFA) is thus much desired. Molecular distillation, a process that involves high temperature exposure for a significant duration (typically > 200oC for several minutes, unless extensive and expensive vacuum pressure is applied), is the most common means of FFA fractionation. The extreme operating conditions can lead to formation of trans fatty acids and peroxides in the product, chemicals which are linked to obesity asthma, heart diseases and Alzheimer’s disease (particularly for PUFA). In addition the use of high temperatures requires high energy input, which increases process costs and increases carbon dioxide emissions, thus decreasing process sustainability.
Urea inclusion compound (UIC)-based fractionation may be a safer and less expensive approach for the downstream separations of natural and food-related materials, such as lipids (e.g., fatty acids) and biopolymers [e.g., poly(lactic acid)]. The technique involves primarily abundant, inexpensive, biocompatible and natural materials, and uses low temperatures and ambient pressure, making it an excellent candidate for the sustainable purification of food-related substances.
Urea Inclusion Compounds: an efficient and sustainable separation approach
The use of UIC-based fractionation to separate hydrocarbons, fatty acids and other molecules with long alkyl chains was developed 70 years ago. In the late 1930s, M. F. Bengen of Germany discovered UICs serendipitously when investigating the effects of urea on protein behaviour in milk. Investigations of UIC-based fractionation during the 1940s led to several patents, held mostly by petrochemical companies. Due to the patents, UICs have received little attention for use in separation methods since the early 1950s. The expiration of patent rights during the 1990s and the increased interest in sustainability has led to an increased interest in UIC-based fractionation.
Reviewed elsewhere, UICs consist of
spiral, hydrogen-bonded networks of urea molecules surrounding narrow, linear molecular chains that result in a distinct, stable solid phase [Figure 1] [1,2]. Branched, bulky, double bond-containing, isomerically less linearly shaped, and smaller size (e.g. lower molecular weight) molecules are less likely (i.e. less energetically favoured) to form UICs, i.e., to serve as UIC “guests.” It is notable that UICs will not form in the absence of “guests.” A feedstock containing a mixture of linear versus nonlinear molecules can thus be separated into highly-concentrated linear and nonlinear products. UIC formation also occurs more readily for molecules with long chain length and trans (rather than cis) unsaturation, and is less likely to occur for alkane chains that possess branching, unless the branched moiety occurs near the a or w position in an alkyl chain. However poor “guests”, such as branched or polyunsaturated molecules, can be incorporated into UICs in the presence of good “guests.” UICs form needle-like hexagonal clathrates (or adducts), which are readily distinguishable from the tetratagonal crystals formed by pure urea. Inclusion complexes can also be formed using urea homologues (thiourea and selenourea), cyclodextrins, crown ethers and others. Urea forms inclusion compounds with other small molecules, such as diacids and hydrogen peroxide, that possess a quite different lattice structure on a nanoscale level.
UICs produce a unique environment for the inclusion guests. Typically, alkanes and polymers must be in an extended planar zig-zag arrangement to reside in a UIC channel. The mobility of C-H bonds of the host’s hydrocarbon chain is very limited inside of the UICs, except near the termini. “Guest” molecules within UICs are protected from autooxidation, which is important for the integrity of unsaturated fatty acids “guests.”
The first step for UIC formation is co-solubilisation of urea and the guest, requiring a solvent (or a solvent mixture) and/or elevated temperature. The best solvents are polar and lack the ability to serve as a UIC guest (e.g., methanol and ethanol). UIC formation, an exothermic process, occurs upon the lowering of the mixture's temperature. Most previous studies applied a slow cooling process, as would occur for crystallisation. However, my group at the University of Tennessee has demonstrated that for low-molecular weight molecules such as fatty acids, a rapid cooling process is quite efficient, selective and repeatable. The solid-phase UIC clathrates are readily isolated from the liquid phase via filtration or sedimentation. In the absence of solvent, UICs can decompose at higher temperatures than some high-melting fatty acyl inclusion guests (e.g., palmitic acid) to form complex phase behaviour with urea, including miscibility gaps.
UICs decompose through contact with an extractant that selectively removes either the inclusion host or urea. Alternatively, UICs decompose upon increasing temperature. Although UICs are quite stable, decomposition typically occurs below the melting point for urea, 135oC (e. g., >100 oC), and increases linearly with increasing molecular weight of inclusion guest. Employment of such a high temperature is not practical and can degrade the double bonds of PUFA. In addition, UIC-forming mixtures of free fatty acid and urea yield complex phase diagrams. In summary, UIC-based fractionation is an attractive separation approach due to the easy separation of liquid- and solid-phase products, and the simple isolation of products from the two phases.
UIC-based fractionation of fatty acids
UIC-based fractionation has been frequently used to fractionate mixtures of free fatty acids, FFA, and their methyl or ethyl esters (FAME and FAEE, respectively) obtained through hydrolysis or transesterification of seed oils, respectively [Table 1]. Fatty alcohols can be fractionated as well. The most successful and useful application of UIC-based fractionation is for the isolation of PUFA, particularly DHA, EPA and AA, from fish oils, since PUFA are poor UIC guests due to their multiple double bonds. In addition, GLA is efficiently isolated via UIC-based fractionation from borage or evening primrose oil, not only due to its 3 double bonds, but also to the relatively close proximity of its double bond at C6 to the -COO terminus of the acyl chain. Alternatively, UICs have been employed for the selective removal of saturated fatty acids, particularly palmitic and stearic acids, from FFA derived from canola oil, as a means of improving their nutritional value. The palmitic acid-rich by-product may be useful as a feedstock for surfactants and detergents. Urea fractionation was not successful in isolating hydroxy or epoxy fatty acids such as ricinoleic acid or its homologues, or vernolic acid, since the molecular size of the oxygenated groups was not sufficiently large to inhibit UIC formation. Similarly, for branching to deter molecules serving as UIC guests, branch lengths must be at least 3 methylene units. The use of UICs for the removal of polyunsaturated FAMEs in the purification of biodiesel is intriguing because of its robust incorporation into a biorefinery. Moreover, the typical solvents employed for UIC-based fractionation, ethanol and methanol, are abundant biorefinery biofuels.
When selecting operating conditions, it is important to realise that as the amount of UICs formed increases (by increasing the urea concentration, decreasing the amount or its polarity of the solvent system, or decreasing the operating temperature for UIC formation), the extent of purification increases at the cost of a decreased recovery. For example, for the isolation of DHA and EPA from fish oil, conditions were used in our laboratory which produced UIC yields of 0.32, 0.55, and 0.62 grams per gram of urea + FFA. The purity of DHA in the FFA that was not incorporated into the UICs increased from 10.8% in the original mixture to 11.5% to 16.2% to 23.9% as the UIC yield increased. However, for the same experiments, as the UIC yield increased, the recovery of PUFA-enriched FFA product (i.e., FFA dissolved in the solvent) decreased from 87.5% to 74.4% to 39.9%.
Urea fractionation is often combined with lipase-catalysed esterification or alcoholysis for FFA using a medium-chain n-alkanol as acyl acceptor since the two approaches provide complementary selectivities. Using the separation of fish oil FFA as an example, lipases will selectively esterify common mono- and di-unsaturated fatty acyl groups, such as palmitoleyl (16:1-9c), oleyl (18:1-9c), linoleyl (18:2-9,c,12c), and linolenyl (18:3-9c,12c,15c). The non-esterified FFA, isolated through saponification, will be enriched in saturated fatty acids and PUFA such as AA, EPA, and DHA.
The process employed for UIC fractionation of FFA is illustrated in Figure 2. The first step for UIC formation is co-solubilisation of urea and guest, requiring a solvent (or a solvent mixture) and/or elevated temperature. In practice, 95% ethanol has been employed efficiently in my laboratory. The process involved only water, ethanol, urea and fatty acids, all of which are “Generally Regarded as Safe” (GRAS) by the FDA, so the UIC fractionation process is acceptable for the processing of lipids for foods and cosmetics. It is important that the FFA source be highly pure, since minor impurities such as surfactants, phospholipids and triacylglycerols, which are poor UIC “guests” that will reside in the solvent-rich phase, will enhance the dissolution of FFA in the solvent and thus deter UIC formation.
Typically, the major product is the FFA mixture retained in the solvent-rich phase. One would typically isolate the FFA via the evaporative removal of the solvent. However, a significant fraction of urea remains dissolved in the solvent. Therefore an additional purification step is required. The addition of warm water to the solvent-rich phase, as depicted in Figure 2, dissolves both urea and solvent and liberates the FFA. However, for some systems, a small but significant fraction of the FFA co-dissolves with the solvent in warm water. For those cases, solvent is first evaporated; then, urea is removed from the resultant solid-phase mixture through the addition of warm water. (Alternatively, a lipophilic solvent that does not form UICs, such as isooctane, can be used to extract away the FFA.) If the FFA UIC “guests” are a valuable by-product, they can be recovered from the UICs through the addition of warm water.
Future prospective of urea inclusion compound-based fatty acid fractionation
Due to its effectiveness, low material costs, and sustainability, UIC-based fractionation has received recent attention in the food and pharmaceutical industry for the isolation of PUFA. It is anticipated that future work will focus upon process design and scale-up, employing process units common to the oleochemical industry, such as crystallisers, heat exchangers, centrifuges and evaporators. Such a system can be easily integrated into an existing industrial process, or used in a biorefinery located in close proximity to the cultivation and collection sites, in which UIC formation can be used as a pre-fractionation technique to prepare feedstocks useful for industrial processing. However, a major economic input to the process cost must be reconsidered, namely the recovery of solvent. The typical approach, evaporation, is very expensive. Mathematical modeling will need to be further developed and integrated with economic assessment algorithms to determine optimal processing conditions. Ecological studies will need to be carried out to determine if the aqueous urea solutions can be used as fertiliser and/or released into the environment.
References
1. Hollingsworth MD and Harris KDM. Urea, thiourea, and selenourea, in Solid-State Supramolecular Chemistry: Crystal Engineering, MacNicol DD, Toda F and Bishop R, Eds., Pergamon Press, Oxford, UK, 1996, pp. 177–237.
2. Hayes DG. Purification of Free Fatty Acids via Urea Inclusion Compounds”, in Handbook of Functional Lipids, C. C. Akoh, Ed., CRC Press, Boca Raton, FL USA, 2005, pp. 77-88.
The author
Douglas G. Hayes, Ph.D.,
Department of Biosystems Engineering and Soil Science,
University of Tennessee
2506 E.J. Chapman Drive
Knoxville,
TN 37996-4531
USA
Tel +1 865-974-7991
Fax +1 865-974-4514
email: dhayes1@utk.edu
