Food Dietary Fiber and Carbohydrate Digestibility

having two- to four-unit side chains consisting of (1 3)-b-D-galactopyranosyl units joined to it by (1 6)-linkages. Both the main chain and the numerous side chains have attached a-L-arabinofuranosyl, a-L-rhamnopyranosyl, b-D-glucuronopyranosyl, and 4-O-methyl-b-D-glucuronopyranosyl units. The two uronic acid units occur most often as ends of chains. Gum arabic dissolves easily when stirred in water. It is unique among the food gums because of its high solubility and the low viscosity of its solutions. Solutions of 50% concentration can be made. At this concentration, the dispersion is somewhat gellike. Gum arabic is both a fair emulsifying agent and a very good emulsion stabilizer for flavor oil-in-water emulsions. It is the gum of choice for emulsification of citrus, other essential oils, and limitation flavors used as baker’s emulsions and concentrations for soft drinks. The soft drink industry consumes about 30% of the gum supply as an emulsifier and stabilizer. For a gum to have an emulsion-stabilizing effect, it must have anchoring groups with a strong affinity for the surface of the oil and a molecular size large enough to cover the surfaces of dispersed droplets. Gum arabic has surface activity and forms a thick, sterically stabilizing, macromolecular layer around oil droplets. Emulsions made with flavor oils and gum arabic can be spray-dried to produce dry flavor powders that are nonhygroscopic and in which the flavor oil is protected from oxidation and volatization. Rapid dispersion and release of flavor without affecting product viscosity are other attributes. These stable flavor powders are used in dry package products such as beverage, cake, dessert, pudding, and soup mixes. Another important characteristic is its compatibility with high concentrations of sugar. Therefore, it finds widespread use in confections with a high sugar content and a low water content. More than half the world’s supply of gum arabic is used in confections such as caramels, toffees, jujubes, and pastilles. Its functions in confections are to prevent sucrose crystallization and to emulsify and distribute fatty components. Avoidance of surface accumulation of lipids is important because this occurrence results in a greasy surface whitening, called bloom. Another use is as a component of the glaze or coating of pan-coated candies. 4.12 Dietary Fiber and Carbohydrate Digestibility [7,23,29,45,46,54,57] Dietary fiber is not necessarily fibrous in nature. Dietary fiber is a nutritional term that has nothing to do with its physical or chemical nature, although both chemical and physical properties are involved in its determination. Dietary fiber is actually defined by the method used to measure it, of which there are several. Both insoluble plant cell-wall materials, primarily cellulose and lignin, and nonstarch, water-soluble polysaccharides are components of dietary fiber. The only common feature of these substances is that they are nondigestible polymers. Therefore, not only do natural components of foods contribute dietary fiber, but so also do gums that are added to modify rheological properties, to provide bulk, and/or to provide other functionalities as already described. One natural component of dietary fiber is a water-soluble polysaccharide, commonly known as b-glucan, but more properly bD-glucan, that is present in oat and barley brans. Oat b-glucan has become a commercial food ingredient because it has been shown to be effective in reducing the level of serum cholesterol. Oat b-glucan is a linear chain of b-D-glucopyranosyl units. About 70% are linked (1 4) and about 30% (1 3). The (1 3) linkages occur singly and are separated by sequences of two or three (1 4) linkages. Thus, the molecule is composed of (1 3)-linked b-cellotriosyl [ 3)-bGlcp-(1 4)- bGlcp-(1 4)- Pag e 219 FIGURE 49 Representative structure of a seg ment of oat and barley b-g lucans where n usually is 1 or 2, but occasionally may be larg er (shorthand notation). bGlcp-(1 ] and b-cellotetraosyl units (Fig. 49). Such (1 4,1 3)-b-glucans are often called mixed-linkage b-glucans. When taken orally in foods, b-glucans reduce postprandial serum glucose levels and the insulin response—that is, they moderate the glycemic response—in both normal and diabetic human subjects. This effect seems to be correlated with viscosity. They also reduce serum cholesterol concentrations in rats, chicks, and humans. These physiological effects are typical of those of soluble dietary fiber. Other soluble polysaccharides have similar effects but to differing degrees. The mechanism(s) of action remains to be determined. Carbohydrates have always been the principal source of metabolic energy for humans and the means for maintaining health of the human gastrointestinal tract. Carbohydrates are the principal providers of the bulk and body of food products. The higher saccharides may be digestible (most starch-based products), partially digestible (retrograded amylose, the so-called resistant starch), or nondigestible (essentially all other polysaccharides). When digestive hydrolysis to monosaccharides occurs, the products of digestion are absorbed and catabolized. [Only D-glucose is produced by digestion of polysaccharides (starch) in humans.] Those carbohydrates not digested to monosaccharides by human enzymes in the small intestine (all others except sucrose, lactose, and those related to starch) may be metabolized by microorganisms in the large intestine, producing substances that are absorbed and catabolized for energy. Therefore, carbohydrates may be caloric, partially caloric, or essentially noncaloric. They may be soluble or insoluble, and they may produce high or low viscosities. Naturally occurring plant carbohydrates are nontoxic. The most common bulking agents in natural food are remnants of plant cells resistant to hydrolysis by enzymes in the alimentary tract. This material includes cellulose, hemicelluloses, pectin, and lignin. Dietary fiber bulking agents are important in human nutrition because they maintain normal functioning of the gastrointestinal tract. They increase intestinal and fecal bulk, which lowers intestinal transit time and helps prevent constipation. Their presence in foods induces satiety at meal time. Nutritionists set requirements of dietary fiber at 25–50 g/day. Insoluble fiber bulking agents are claimed to decrease blood cholesterol levels, lessening the chance of heart disease. They also reduce the chances of colonic cancer, probably due to their sweeping action. Soluble gums other than b-glucans have similar effects in the gastrointestinal tract and on the level of cholesterol in blood, but to different extents. Some gums that have been specifically examined in this regard are pectin, guar gum, xanthan, and hemicelluloses. [For example, guar gum ingested at a rate of 5 g/day results in an improved glycemic index, a 13% lowering of serum cholesterol, and no decrease in the high-density lipoprotein (HDL) fraction, the beneficial cholesterol carrier.] In addition to cereal brans, kidney and navy beans are especially good sources of dietary fiber. A product based on psyllium seed hulls has high water-binding properties, leading to rapid transit time in the gastrointestinal tract, and is used to prevent constipation. A product with a methylcellulose base is sold for the same purpose. Pag e 220 The starch polysaccharides are the only polysaccharides that can be hydrolyzed by human digestive enzymes. They, of course, provide D-glucose, which is absorbed by microvilli of the small intestine to supply the principal metabolic energy of humans. Other polysaccharides consumed normally as natural components of edible vegetables, fruits, and other plant materials, and those food gums added to prepared food products, are not digested in the upper digestive tract of humans, but pass into the large intestine (colon) with little or no change. (The acidity of the stomach is not strong enough, nor is the residence time of polysaccharides in the stomach sufficiently long, to cause significant chemical cleavage.) When the undigested polysaccharides reach the large intestine, they come into contact with normal intestinal microorganisms, some of which produce enzymes that catalyze hydrolysis of certain polysaccharides or certain parts of polysaccharide molecules. The consequence of this is that polysaccharides not cleaved in the upper intestinal tract may undergo cleavage and microbial metabolism within the large intestine. Sugars that are split from the polysaccharide chain are used by the microorganisms of the large intestine as energy sources in anaerobic fermentation pathways that produce lactic, propionic, butyric, and valeric acids. These short-chain acids can be absorbed through the intestinal wall and metabolized, primarily in the liver. In addition, a small, though significant in some cases, fraction of the released sugars can be taken up by the intestinal wall and transported to the portal bloodstream where they are conveyed to the liver and metabolized. It is calculated that, on average, 7% of human energy is derived from sugars split from polysaccharides by microorganisms in the large intestine and/or from the acid by-products produced from them by these microorganisms via anaerobic fermentation. The extent of polysaccharide cleavage depends on the abundance of the particular organism(s) producing the specific enzymes required. Thus, when changes occur in the type of polysaccharide consumed, utilization of the polysaccharide by colonic microorganisms may be temporarily reduced until organisms capable of splitting the new polysaccharide proliferate. Some polysaccharides survive almost intact during their transit through the entire gastrointestinal tract. These, plus larger segments of other polysaccharides, give bulk to the intestinal contents and lower transit time. They can be a positive factor in health through a lowering of blood cholesterol concentration, perhaps by sweeping out bile salts and reducing their chances for reabsorption from the intestine. In addition, the presence of large amounts of hydrophilic molecules maintains a water content of the intestinal contents that results in softness and consequent easier passage through the large intestine. Bibliography Dey, P. M., and R. A. Dixon, eds. (1985). Biochemistry of Storage Carbohydrates in Green Plants, Academic Press, London. Eliason, A.-C., ed. (1996). Carbohydrates in Food, Marcel Dekker, New York. El Khadem, H. S. (1988). Carbohydrate Chemistry, Academic Press, San Diego. Glicksman, M., ed. (1982–1984). Food Hydrocolloids, vols. I–III, CRC Press, Boca Raton, FL. Kennedy, J. F., ed. (1988). Carbohydrate Chemistry, University Press, New York. Lineback, D. R., and G. E. Inglett, eds. (1982). Food Carbohydrates, AVI, Westport, CT. Mathlouthi, M., and P. Reiser, eds. (1995). Sucrose: Properties and Applications, Blackie, Glasgow. Stephen, A. M., ed. (1995). Food Polysaccharides and Their Applications, Marcel Dekker, New York. Whistler, R. L., and J. N. BeMiller, eds. (1993). Industrial Gums, 3rd ed., Academic Press, San Diego. Whistler, R. L., J. N. BeMiller, and E. F. Paschall, eds. (1994). Starch: Chemistry and Technology, 2nd ed., Academic Press, New York. Wurzburg, O. B., ed. (1986). Modified Starches: Properties and Uses, CRC Press, Boca Raton, FL. Pag e 221 References 1. BeMiller, J. N. (1993). Starch-based gums, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 579–600. 2. Biliaderis, C. G. (1990). Thermal analysis of food carbohydrates, in Thermal Analysis of Foods (V. R. Harwalkar and C. – Y. Ma, eds.), Elsevier Science, New York, pp. 168–220. 3. Biliaderis, C. G. (1991). The structure and interactions of starch with food constituents. Can. J. Physiol. Pharmacol. 69:60–78. 4. Christensen, S. H. (1984). Pectins, in Food Hydrocolloids, vol. III (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 205–230. 5. Clare, K. (1993). Algin, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 105–143. 6. Dickinson, E., ed. (1991). Food Polymers, Gels, and Colloids, Royal Society of Chemistry, London. 7. Dixon, R. A. (1985). b-(1 3)-Linked glucans from higher plants, in Biochemistry of Storage Carbohydrates in Green Plants (P. M. Dey and R. A. Dixon, eds.), Academic Press, London, pp. 229–264. 8. El Khadem, H. S. (1988). Carbohydrate Chemistry, Academic Press, San Diego. 9. Emodi, A. (1982). Polyols: chemistry and application, in Food Carbohydrates (D. R. Lineback and G. E. Inglett, eds.), AVI, Westport, CT, pp. 49–61. 10. Feather, M. S. (1982). Sugar dehydration reactions, in Food Carbohydrates (D. R. Lineback and G. E. Inglett, eds.), AVI, Westport, CT, pp. 113–133. 11. Feddersen, R. L., and S. N. Thorp (1993). Sodium carboxymethylcellulose, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 537–578. 12. Fujimaki, M., M. Namiki, and H. Kato, eds. (1986). Amino-Carbonyl Reactions in Food and Biological Systems, Elsevier, Amsterdam. 13. Glicksman, M. (1982). Functional properties of hydrocolloids, in Food Hydrocolloids, Vol. I (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 47–99. 14. Glicksman, M. (1983). Gum arabic (gum acacia), in Food Hydrocolloids, vol. II (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 7–29. 15. Glicksman, M. (1983). Red seaweed extracts (agar, carrageenan, furcellaran), in Food Hydrocolloids, vol. II (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 73–113. 16. Grover, J. A. (1984). Methylcellulose (MC) and hydroxypropylmethylcellulose (HPMC), in Food Hydrocolloids, vol. III (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 121–154. 17. Grover, J. A. (1993). Methylcellulose and its derivatives, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 475–504. 18. Harris, P., ed. (1990). Food Gels, Elsevier Applied Science, London. 19. Herald, C. T. (1984). Locust/carbo bean gum, in Food Hydrocolloids, vol. III (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 161–170. 20. Herald, C. T. (1984). Guar gum, in Food Hydrocolloids, vol. III (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 171–184. 21. Hizukuri, S. (1985). Relationship between the distribution of the chain length of amylopectin and the crystalline structure of starch granules. Carbohydr. Res. 141:295–306. 22. Huzukuri, S. (1986). Polymodal distribution of the chain lengths of amylopectins and its significance. Carbohydr. Res. 147:342–347. 23. Johnson, I. T., and Southgate, D. A. T. (1994). Dietary Fibre and Related Substances, Chapman & Hall, London. 24. Kainuma, K. (1988). Structure and chemistry of the starch granule, in The Biochemistry of Plants (J. Preiss, ed.), Academic Press, San Diego, pp. 141–180. 25. Kang, K. S., and D. J. Pettitt (1993). Xanthan, gellan, welan, and rhamsan, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 341–397. 26. Keller, J. D. (1984). Sodium carboxymethylcellulose (CMC), in Food Hydrocolloids, vol. III (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 43–109. Pag e 222 27. Kennedy, J. F., ed. (1988). Carbohydrate Chemistry, University Press, New York. 28. King, A. H. (1983). Brown seaweed extracts (alginates), in Food Hydrocolloids, vol. II (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 115–188. 29. Kritchevsky, D., and Bonfield, C., eds. (1995). Dietary Fiber in Health and Disease, Eagan Press, St. Paul, MN. 30. Labuza, T. P., Reineccius, G. A., Monnier, V., O’Brien, J., and Baynes, J., eds. (1995). Maillard Reactions in Chemistry, Food, and Health, CRC Press, Boca Raton, FL. 31. Maier, H., M. Anderson, C. Karl, K. Magnuson, and R. L. Whistler (1993). Guar, locust bean, tara and fenugreek gums, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 181–226. 32. Manners, D. J. (1985). Some aspects of the structure of starch. Cereal Foods World 30461–467. 33. Manners, D. J. (1985). Starch, in Biochemistry of Storage Carbohydrates in Green Plants (P. M. Dey and R. A. Dixon, eds.), Academic Press, London, pp. 149–203. 34. Manners, D. J. (1989). Recent developments in our understanding of amylopectin structure. Carbohydr. Polym. 11:87–112. 35. Mathlouthi, M., and Reiser, P., eds. (1995). Sucrose: Properties and Applications, Blackie, Glasgow. 36. Miles, M. J., V. J. Morris, P. D. Orford, and S. G. Ring (1985). The roles of amylose and amylopectin in the gelation and retrogradation of starch. Carbohydr. Res. 135:271–281. 37. Morris, V. J. (1994). Starch gelation and retrogradation. Trends Food Sci. Technol. 1:2–6. 38. Morrison, W. R., and J. Karkaas (1990). Starch, in Methods in Plant Biochemistry, vol. 2, Carbohydrates (P. M. Dey, ed.), Academic Press, San Diego, pp. 323–352. 39. Pennington, N. L., and C. W. Baker, eds. (1990). Sugar, A User’s Guide to Sucrose, Van Nostrand Reinhold, New York. 40. Pettitt, D. (1982). Xanthan gum, in Food Hydrocolloids, vol. I (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 127–149. 41. Rolin, C. (1993). Pectin, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 257–293. 42. Sand, R. E. (1982). Nomenclature and structure of carbohydrate hydrocolloids, in Food Hydrocolloids, vol. I (M. Glicskman, ed.), CRC Press, Boca Raton, FL, pp. 19–46. 43. Schenck, F. W., and R. E. Hebeda, eds. (1992). Starch Hydrolysis Products, VCH, New York. 44. Slade, L., and H. Levine (1989). A food polymer science approach to selected aspects of starch gelatinization and retrogradation, in Frontiers in Carbohydrate Research—1, (R. P. Millane, J. N. BeMiller, and R. Chandrasekaran, eds.), Elsevier Applied Science, London, pp. 215–270. 45. Spiller, G. A. (1993). CRC Handbook of Dietary Fiber in Human Nutrition, 2nd ed., CRC Press, Boca Raton, FL. 46. Staub, H. W., and R. Ali (1982). Nutritional and physiological value of gums, in Food Hydrocolloids, vol. I (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 101–121. 47. Stephen, A. M., ed. (1995). Food Polysaccharides and Their Applications, Marcel Dekker, New York. 48. Strickler, A. J. (1982). Corn syrup selections in food applications, in Food Carbohydrates (D. R. Lineback and G. E. Inglett, eds.), AVI, Westport, CT, pp. 12–24. 49. Takeda, Y., T. Shitaozono, and S. Hizukuri (1990). Structures of subfractions of corn amylose. Carbohydr. Res. 199:207–214. 50. Therkelsen, G. H. (1993). Carrageenan, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 145–180. 51. Thomas, W. R. (1984). Microcrystalline cellulose (MCC or cellulose gel), in Food Hydrocolloids, vol. III (M. Glicksman, ed.), CRC Press, Boca Raton, FL, pp. 9–42. 52. Van Beynum, G. M. A., and J. A. Roels, eds. (1985). Starch Conversion Technology, Marcel Dekker, New York. 53. Whistler, R. L. (1993). Exudate gums, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 309–339. 54. Whistler, R. L. (1993). Hemicelluloses, in Industrial Gums (R. L. Whistler and J. N. BeMiller, eds.), Academic Press, San Diego, pp. 295–308. Pag e 223 55. Whistler, R. L., and J. N. BeMiller, eds. (1993). Industrial Gums, 3rd ed., Academic Press, San Diego. 56. Whistler, R. L., J. N. BeMiller, and E. F. Paschall, eds. (1984). Starch: Chemistry and Technology, 2nd ed., Academic Press, New York. 57. Wood, P. J., ed. (1993). Oat Bran, American Association of Cereal Chemists, St. Paul, MN. 58. Wurzburg, O. B., ed. (1986). Modified Starches: Properties and Uses, CRC Press, Boca Raton, FL. 59. Yaylayan, V. A., and A. Huyghues-Despointes (1994). Chemistry of Amadori rearrangement products: Analysis, synthesis, kinetics, reactions, and spectroscopic properties. Crit. Rev. Food Sci. Nutr. 34:321–369. 60. Young, R. A., and R. M. Rowell, eds. (1986). Cellulose, John Wiley, New York. 61. Zobel, H. F. (1988). Molecules to granules: A comprehensive starch review. Starch/Stärke 40:44–50. Pag e 225 5 Lipids WASSEF W. NAWAR University of Massachusetts, Amherst, Massachusetts 5.1. Introduction 226 5.2. Nomenclature 226 5.2.1. Fatty Acids 226 5.2.2. Acylglycerols 228 5.2.3. Phospholipids 231 5.3. Classification 232 5.3.1. Milk Fats 232 5.3.2. Lauric Acids 232 5.3.3. Vegetable Butters 232 5.3.4. Oleic-Linoleic Acids 233 5.3.5. Linolenic Acids 233 5.3.6. Animal Fats 233 5.3.7. Marine Oils 233 5.4. Physical Aspects 233 5.4.1. Theories of Triacylglycerol Distribution Patterns 233 5.4.2. Positional Distribution of Fatty Acids in Natural Fats 237 5.4.3. Crystallization and Consistency 239 5.5. Chemical Aspects 254 5.5.1. Lipolysis 254 5.5.2. Autoxidation 255 5.5.3. Thermal Decomposition 288 5.5.4. Chemistry of Frying 292 5.5.5. Effects of Ionizing Radiation on Fats 295 5.5.6. Concluding Thoughts on Lipid Oxidation in Food 299 5.6. Chemistry of Fat and Oil Processing 299 5.6.1. Refining 299 5.6.2. Hydrogenation 300 5.6.3. Interesterification 304 Pag e 226 5.7. Role of Food Lipids in Flavor 307 5.7.1. Physical Effects 307 5.7.2. Lipids as Flavor Precursors 308 5.8. Physiological Effects of Lipids 310 5.8.1. Consumption and Trends 310 5.8.2. Nutritional Functions 310 5.8.3. Safety of Lipids Exposed to Frying Conditions 310 5.8.4. Safety of Hydrogenated Fats 311 5.8.5. Safety of Irradiated Food 312 5.8.6. Dietary Lipids, Coronary Heart Disease, and Cancer 312 5.8.7. Concluding Thoughts on Dietary Lipids and Health 313 Acknowledgments 313 Bibliography 313 References 314 5.1 Introduction Lipids consist of a broad group of compounds that are generally soluble in organic solvents but only sparingly soluble in water. They are major components of adipose tissue, and together with proteins and carbohydrates, they constitute the principal structural components of all living cells. Glycerol esters of fatty acids, which make up to 99% of the lipids of plant and animal origin, have been traditionally called fats and oils. This distinction, based solely on whether the material is solid or liquid at room temperature, is of little practical importance and the two terms are often used interchangeably. Food lipids are either consumed in the form of “visible” fats, which have been separated from the original plant or animal sources, such as butter, lard, and shortening, or as constituents of basic foods, such as milk, cheese, and meat. The largest supply of vegetable oil comes from the seeds of soybean, cottonseed, and peanut, and the oil-bearing trees of palm, coconut, and olive. Lipids in food exhibit unique physical and chemical properties. Their composition, crystalline structure, melting properties, and ability to associate with water and other non lipid molecules are especially important to their functional properties in many foods. During the processing, storage, and handling of foods, lipids undergo complex chemical changes and react with other food constituents, producing numerous compounds both desirable and deleterious to food quality. Dietary lipids play an important role in nutrition. They supply calories and essential fatty acids, act as vitamin carriers, and increase the palatability of food, but for decades they have been at the center of controversy with respect to toxicity, obesity, and disease. 5.2 Nomenclature Lipid nomenclature can be understood more readily if simple nomenclature of the various classes of organic compounds is reviewed first. International recommendations for the nomenclature of lipids have been published [66]. 5.2.1 Fatty Acids This term refers to any aliphatic monocarboxylic acid that can be liberated by hydrolysis from naturally occurring fats. Pag e 227 5.2.1.1 Saturated Fatty Acids Five accepted naming systems exist for fatty acids. 1. The acids can be named after hydrocarbons with the same number of carbon atoms (CH3 replaced by COOH). The terminal letter e in the name of the parent hydrocarbon is replaced with oic. If the acid contains two carboxyl groups the suffix becomes dioic (e.g., hexanedioic). When a single carboxyl group is present it is regarded as carbon number 1, as shown here. 2. The acids can be called carboxylic acids with the prefix being the hydrocarbon to which the carboxyl group is attached. Thus, when a carboxyl group is attached to pentane, the fatty acid is 1-pentanecarboxylic acid. In this system carbon number 1 is the carbon atom adjacent to the terminal carboxyl group. This convention corresponds to the long-practiced use of Greek letters a, b g, d, and so on, in which the a-carbon atom is that adjacent to the carboxyl carbon. 3. A common name can be used, such as butyric, stearic, or oleic. 4. Fatty acids can be represented by a simple numerical expression consisting of two terms separated by a colon, with the first term depicting the number of carbon atoms and the second the number of double bonds; for example, 4:0, 18:1, and 18:3 represent butyric, oleic, and linolenic acids, respectively. 5. In abbreviated designations for triacylglycerols, each acid can be given a standard letter abbreviation such as P for palmitic and L for linoleic. Thus, the fatty acid CH3CH2CH2COOH can be referred to as 4:0, n-butanoic, 1-propanecarboxylic, or butyric acid. Similarly, the compound here is named 3-methylbutanoic, 2-methyl-1-propanecarboxylic, or b-methylbutyric acid. 5.2.1.2 Unsaturated Fatty Acids As in the case of the saturated fatty acids, unsaturated acids can be named after the parent unsaturated hydrocarbons. Replacement of the terminal anoic by enoic indicates unsaturation, and the di, tri, and so on represent the number of double bonds present. Hence, we have hexadecenoic for 16:1, octadecatrienoic for 18:3, and so on. The simplest way to specify the location of double bonds is to put, before the name of the acid, one number for each unsaturated linkage. Oleic acid, for example, with one double bond between carbons 9 and 10, is named 9-octadecenoic acid. In certain cases it is convenient to distinguish unsaturated fatty acids by the location of the first double bond from the methyl end Pag e 228 of the molecule, that is, the omega carbon. Linoleic acid (9,12-octadecadienoic acid) is therefore an 18:2w6 (or n-6) acid. The geometric configuration of double bonds is usually designated by the use of cis (Latin, on this side), and trans (Latin, across), indicating whether the alkyl groups are on the same or opposite sides of the molecule. The cis configuration is the naturally occurring form, but the trans configuration is thermodynamically favored. Linoleic acid, with both double bonds in the cis configuration, is named cis-9, cis-12-octadecadienoic acid. Difficulty arises, however, if the four attached groups are all different as in this compound: In this case the two atoms or groups attached to each carbon are assigned “priorities” in accord with the Cahn-Ingold-Prelog procedure (described in Sec. 5.2.2.1, under R/S System). If the high-priority groups (greater atomic number) lie on the same side of both carbons, the letter Z (German, zusammen) is used to designate the configuration. If the two high-priority groups are on opposite sides, the letter E (German, entgegen) is used. Table 1 gives a list of some of the fatty acids commonly found in natural fats, with various designations for each. Table 2 lists some uncommon, naturally occurring, polyunsaturated fatty acids that are claimed to offer attractive applications in nutrition, cosmetics, and pharmacology. 5.2.2 Acylglycerols Neutral fats are mono-, di-, and triesters of glycerol with fatty acids, and are termed monoacyl-glycerols, diacylglycerols, and triacylglycerols, respectively. Use of the old terms mono-, di-, and triglycerides is discouraged. The compound shown here can be named any of the following: tristearoylglycerol, glycerol tristearate, tristearin, or StStSt. Although glycerol by itself is a completely symmetrical molecule, the central carbon atom acquires chirality (asymmetry) if one of the primary hydroxyl groups (on carbons 1 and 3) is esterified, or if the two primary hydroxyls) are esterified to different acids. Several methods have been used to specify the absolute configuration of glycerol derivatives. Pag e 229 TABLE 1 Nomenclature of Some Common Fatty Acids Abbreviation Systematic name Common name Symbol 4:0 Butanoic Butyric B 6.0 Hexanoic Caproic H 8:0 Octanoic Caprylic Oc 10:0 Decanoic Capric D 12:0 Dodecanoic Lauric La 14:0 Tetradecanoic Myristic M 16:0 Hexadecanoic Palmitic P 16:1 (n-7) 9-Hexadecenoic Palmitoleic Po 18:0 Octadecanoic Stearic Sta 18:1 (n-9) 9-Octadecenoic Oleic O 18:2 (n-6) 9, 12-Octadecadienoic Linoleic L 18:3 (n-3) 9, 12,15-Octadecatrienoic Linolenic Ln 20:0 Arachidic Eicosanoic Ad 20:4 (n-6) 5,8,11,14-Eicosatetraenoic Arachidonic An 20:5 (n-3) 5,8,11,14,17-Eicosapentaenoic EPA 22:1 (n-9) 13-Docosenoic Erucic E 22:5 (n-3) 7,10,13,16,19-Docosapentaenoic 22:6 (n-6) 4,7,10,13,16,19-Docosahexaenoic DHA aSome authors use S for stearic, but this can be confusing , since S is also used for “saturated” whenever triaclycerol composition is expressed in terms of saturated (S) and unsaturated (U) fatty acids. For example, S 3 or SSS=all three fatty acids saturated, SU 2 or SUU = diunsaturatedmonosaturated, and so on. 5.2.2.1 R/S System Use of the prefixes R and S was proposed by Cahn et al. [17]. A sequence of priority is assigned to the four atoms or groups of atoms attached to a chiral carbon, with the atoms of greatest atomic number assigned the highest priority. The molecule is oriented so that the group of lowest priority is directed straight away from the viewer, and the remaining groups are directed toward the viewer in a tripodal fashion. If the direction of decrease in order of priority is clockwise, the configuration is R (Latin, rectus); if counterclockwise, it is S (Latin, sinister). To apply this system to acylglycerols, it is helpful to consider the structures in Figure 1. In both instances, the H atom on asymmetric carbon 2 is the substituent of lowest priority and thus is depicted as being beneath the plane of the paper. Among the substituents of carbon 2, oxygen has the highest rank. The remaining two substituents are -CO- (hydrogens on carbon 1 and 3 are disregarded since O TABLE 2 Polyunsaturated Fatty Acids of Biolog ical Sig nificance in Oils from Various Plant Sources (%) Fatty acid Black currant Borag e Evening primrose 18:2 n-6 (LA) 48 38 72 18:3 n-6 (GLA) 17 23 9 18:3 n-3 (a) 13 — — 18:4 n-3 (SA) 4 — — Note: LA, linoleic; GLA, g amma-linolenic; a,alpha-linolenic; SA, stearidonic. Pag e 230 FIGURE 1 Application of the R/S system of nomenclature for triacylg lycerols. has a higher priority). Thus, a comparison must be made of the atoms attached to these -CO- groups. Long saturated acyls have higher priority than short. Unsaturated chains outrank saturated chains, a double bond outranks single branching, two double bonds outrank one, cis outranks trans, and a branched chain outranks an unbranched chain. Accordingly, the groupings in configuration I have a clockwise arrangement in order of decreasing priority, and thus configuration I is regarded as R. Similarly, configuration II is S. Although the R/S system does convey the stereochemical configuration of acylglycerols, it is obvious that the designated structure depends on the nature of the acyl groups in positions 1 and 3 of the triacylglycerol molecule. Consequently, this system cannot be applied in situations in which these positions contain mixtures of fatty acids, as in the case of natural fats, unless individual triacylglycerols are separated. 5.2.2.2 Stereospecific Numbering The sn system, as proposed by Hirschmann [60], is simple, is applicable to both synthetic and natural fats and has now been universally adopted. The usual Fischer planar projection of glycerol is utilized with the middle hydroxyl group positioned on the left side of the central carbon. The carbon atoms are numbered 1 to 3 in the conventional top-to-bottom sequence, as shown: If, for example, stearic acid is esterified at the sn-1 position, oleic at sn-2, and myristic at sn-3, the triacylglycerol would appear as: Pag e 231 and can be designated any one of the following: 1-stearoyl-2-oleoyl-3-myristoyl-sn-glycerol, sn-glycerol-1-stearate-2-oleate-3- myristate, sn-StOM, or sn-18:0-18:1-16:0. The following prefixes are now widely used with abbreviations to designate the positional distribution of fatty acids within triacylglycerol molecules. sn: Used immediately preceding the term “glycerol,” indicates that the sn-1, sn-2, and sn-3 positions are listed in that order. rac: Racemic mixture of two enantiomers. This specifies that the middle acid in the abbreviation is attached at the sn-2 position, and that the remaining two acids are equally divided between sn-1 and sn-3 (e.g., rac-StOM indicates equal amounts of snStOM and sn-MOSt). b-: Middle acid in the abbreviation is at the sn-2 position but positioning of the remaining two is unknown (e.g., b-StOM indicates a mixture of sn-StOM and sn-MOSt in any proportion). No prefix is given in case of monoacid acylglycerols (e.g., MMM) or if the positional distribution of the acids is unknown, and hence any mixture of isomers is possible (e.g., StOM indicates a possible mixture of sn-StOM, sn-MOSt, sn-OStM, sn-MStO, sn-StMO, and sn-OMST in any proportion). 5.2.3 Phospholipids The term “phospholipid” may be used for any lipid containing phosphoric acid as a mono- or diester. “Glycerophospholipid” signifies any derivative of glycerophosphoric acid that contains an O-acyl, O-alkyl, or O-alkenyl group attached to the glycerol residue. Thus all phosphoglycerols contain a polar head (hence the term polar lipids) and two hydrocarbon tails. These compounds differ from one another in the size, shape, and polarity of the alcohol component of their polar head. The two fatty acid substituents also vary. Usually one is saturated while the other unsaturated and mainly located in the sn-2 position. The common glycerophospholipids are named as derivatives of phosphatidic acid, such as 3-sn-phosphatidylcholine (common name, lecithin), or by their systematic name, similar to the system for triacylglycerols. The term “phospho” is used to indicate the phosphodiester bridge; for example, 1-stearoyl-2-linoleoyl-sn-glycero-3-phosphocholine is the designation for this compound: In addition to phosphatidycholine, phosphatidylethanolamine, phosphatidylserine, and phosphatidylinositol are commonly found 3 Classification A general classification of lipids based on their structural components is presented in Table 3. Such a classification, however, is possibly too rigid for a group of compounds so diverse as lipids, and should be used only as a guide. It should also be recognized that other classifications may sometimes be more useful. For example, the sphingomyelins can be classed as phospholipids because of the presence of phosphate. The cerebrosides and the gangliosides can also be classed as glycolipids because of the presence of carbohydrate, and the sphingomyelins and the glycolipids can be classed as sphingolipids because of the presence of sphingosine. The most abundant class of food lipids is the acylglycerols, which dominate the composition of depot fats in animals and plants. The polar lipids are found almost entirely in cellular membranes (phospholipids being the main components of the bilayer) with only very small amounts in depot fats. In some plants, glycolipids constitute the major polar lipids in cell membranes. Waxes are found as protective coatings on skin, leaves, and fruits. Edible fats are traditionally classified into the following subgroups. 5.3.1 Milk Fats (See also Chap. 14) Fats of this group are derived from the milk of ruminants, particularly dairy cows. Although the major fatty acids of milk fat are palmitic, oleic, and stearic, this fat is unique among animal fats in that it contains appreciable amounts of the shorter chain acids C4 to C12, small amounts of branched and odd-numbered acids, and trans-double bonds. 5.3.2 Lauric Acids (See also Chap. 16) Fats of this group are derived from certain species of palm, such as coconut and babasu. The fats are characterized by their high content of lauric acid (40–50%), moderate amounts of C6, C8, and C10 fatty acids, low content of unsaturated acids, and low melting points. 5.3.3 Vegetable Butters (See also Chap. 16) Fats of this group are derived from the seeds of various tropical trees and are distinguished by their narrow melting range, which is due mainly to the arrangement of fatty acids in the TABLE 3 Classification of Lipids Major class Subclass Description Simple lipids Acylg lycerols Waxes Glycerol + fatty acids Long -chain alcohol + long -chain fatty acid Compound lipids Phosphoacylg lycerols (or g lycerophospholipids) Glycerol + fatty acids + phosphate + another g roup usually containing nitrog en Sphing omyelins Sphing osine + fatty acid + phosphate + choline Cerebrosides Sphing osine + fatty acid + simple sug ar Gang liosides Sphing osine + fatty acid + complex carbohydrate moiety that includes sialic acid Derived lipids Materials that meet the definition of a lipid but are not simple or compound lipids Examples: carotenoids, steroids, fat-soluble vitamins Pag e 233 triacylglycerol molecules. In spite of their large ratio of saturated to unsaturated fatty acids, trisaturated acylglycerols are not present. The vegetable butters are extensively used in the manufacture of confections, with cocoa butter being the most important member of the group. 5.3.4 Oleic-Linoleic Acids (See also Chap. 16) Fats in this group are the most abundant. The oils are all of vegetable origin and contain large amounts of oleic and linoleic acids, and less than 20% saturated fatty acids. The most important members of this group are cottonseed, corn, peanut, sunflower, safflower, olive, palm, and sesame oils. 5.3.5 Linolenic Acids (See also Chap. 16) Fats in this group contain substantial amounts of linolenic acid. Examples are soybean, rapeseed and flaxseed, wheat germ, hempseed, and perilla oils, with soybean being the most important. The abundance of linolenic acid in soybean oil is responsible for the development of an off-flavor problem known as flavor reversion. 5.3.6 Animal Fats (See also Chap. 15) This group consists of depot fats from domestic land animals (e.g., lard and tallow), all containing large amounts of C16 and C18 fatty acids, medium amounts of unsaturated acids, mostly oleic and linoleic, and small amounts of odd-numbered acids. These fats also contain appreciable amounts of fully saturated triacylglycerols and exhibit relatively high melting points. Egg lipids are of particular importance because of their emulsifying properties and their high content of cholesterol. The lipid content of whole eggs is approximately 12%, almost exclusively present in the yolk, which contains 32–36% lipid. The major fatty acids in egg yolks are 18:1 (38%), 16:0 (23%), and 18:2 (16%). Yolk lipids consist of about 66% triacylglycerols, 28% phospholipids, and 5% cholesterol. The major phopholipids of egg yolk are phosphatidylcholine (73%) and phosphatidylethanolamine (15%). 5.3.7 Marine Oils (See also Chap. 15) These oils typically contain large amounts of long-chain omega-3-polyunsaturated fatty acids, with up to six double bonds, and they are usually rich in vitamins A and D. Because of their high degree of unsaturation, they are less resistant to oxidation than other animal or vegetable oils. 5.4 Physical Aspects 5.4.1 Theories of Triacylglycerol Distribution Patterns 5.4.1.1 Even or Widest Distribution This theory resulted from a series of systematic studies initiated by Hilditch and Williams on the triacylglycerol composition of natural fats [59]. Component fatty acids in the triacylglycerol molecules of natural fats tend to be distributed as broadly as possible. If this is conformed to, when an acid, S, forms less than one-third of the total fatty acids present, it should not appear more than once in any triacylglycerol. If X refers to the other acids present only XXX and SXX species should be found. If an acid forms between one-third and two-thirds of the total fatty acids, it should, according to this theory, occur at least once but never three times in any one Pag e 234 molecule; that is, only SXX and SSX should be present. If an acid forms more than two-thirds of the total acids, it should occur at least twice in every molecule; that is, only SSX and SSS should be present. Shortcomings of the even distribution theory were soon realized. Analysis of many natural fats, especially those of animal origin, revealed marked deviation from theory. Trisaturated acylglycerols were found in fats containing less than 67% saturated fatty acids. Also, the theory can be applied only to two component systems and does not take into account positional isomers. Thus, this theory is no longer considered valid. 5.4.1.2 Random (21,2,3-Random) Distribution According to this theory, fatty acids are distributed randomly both within each triacylglycerol molecule and among all the triacylglycerols. Thus, the fatty acid composition of all three positions should be the same and also equivalent to the fatty acid composition of the total fat. The proportion of any given fatty acid species expected on the basis of this theory can be calculated according to the equation. %sn-XYZ = (mol% X in total fat) × (mol% Y in total fat) × (mol% Z in total fat) × 10-4 where X, Y, and Z are the component fatty acids at positions 1, 2, and 3, respectively, of the acylglycerol. For example, if a fat contains 8% palmitic acid, 2% stearic, 30% oleic, and 60% linoleic, 64% triacylglycerol species (n = 4, n 3 = 64) would be predicted. The following examples illustrate calculations for such species. %sn-OOO = 30 × 30 × 30 × 10-4 = 2.7 %sn-PLSt = 8 × 60 × 2 × 10-4 = 0.096 %sn-LOL = 60 × 30 × 60 × 10-4 = 10.8 Most fats do not conform to a completely random distribution pattern. For example, the proportion of fully saturated triacylycerols expected on the basis of random distribution exceeds, in many cases, that found experimentally. Modern techniques of analysis have revealed that in natural fats the fatty acid composition of the sn-2 position is always different from that of the combined 1,3 positions. Calculations of random distribution are, however, useful for understanding other hypotheses and for predicting distribution patterns of fatty acids in fats randomized by interesterification. 5.4.1.3 Restricted Random According to this hypothesis, first proposed by Kartha [77], saturated and unsaturated fatty acids in animal fats are distributed randomly. However, fully saturated triacylglycerols (SSS) can be present, but only to the extent that the fat can remain fluid in vivo. Excess SSS, according to this theory, can be exchanged with UUS and UUU to form SSU and SUU. Kartha’s calculations do not account for positional isomers or positioning of individual acids. 5.4.1.4 1,3-Random-2-Random The fatty acid composition at the 2 position is known to be different from that of the 1 or 3 positions. This theory assumes that two different pools of fatty acids are separately and randomly esterified to the 2 and 1,3 positions. Thus, composition at positions 1 and 3 will, presumably, be identical. On the basis of this hypothesis, the amount of a given triacylglycerol can be computed as %sn-XYZ = (mol% X at 1,3) × (mol% Y at 2) × (mol% Z at 1,3) × 10-4 Pag e 235 Composition of the sn-2 and/or the combined 1,3 positions can be obtained by analysis of the mono- or diacylglycerols derived from partial deacylation by chemical or enzymic methods [27,98]. Chemical Deacylation Representative diacylglycerols are prepared by deacylation with a Grignard reagent (e.g., ethyl magnesium bromide). The reagent reacts randomly with one of the ester linkages of the triacylglycerol molecules to produce a diacylglycerol and a tertiary alcohol: The reaction is stopped with acetic acid at a point of maximum diacylglycerol yield, and the sn-1,2(2,3)- and sn-1,3- diacylglycerols are separated by thin-layer chromatography. Fatty acid composition of the combined 1 and 3 positions can be obtained by direct analysis of the 1,3-diacylglycerols, while that of the sn-2 position can be deduced by difference. Alternatively, the compositions of the sn-1,3 position can be calculated as follows: Enzymatic Deacylation Pancreatic lipase selectively hydrolyzes the primary ester bonds of triacylglycerols. To achieve rapid hydrolysis and minimum acyl migration, the reaction is carried out at 37°C and pH 8 with vigorous agitation. Compositional requirements include the presence of calcium, an emulsifier, and a large enzyme:sample ratio [98]. Fatty acid composition of the monoacylglycerols produced gives the composition at the sn-2 position, allowing composition at the combined 1,3 position to be calculated. Pag e 236 5.4.1.5 1-Random-2-Random-3-Random According to this theory, three different pools of fatty acids are separately but randomly distributed at each of the three positions of the triacylglycerol molecules of a natural fat. Thus, the possibility of a given fatty acid appearing in each sn position would likely be different. The content of a given triacylglycerol species can be calculated as %sn-XYZ = (mol% X at sn-1) × (mol% Y at sn-2) × (mol% Z at sn-3) × 10-4 To calculate the range of molecular species expected to be present in a natural fat on the basis of this theory, the fatty acid compositions of the sn-1 and the sn-3 position must be distinguished [27]. Two techniques are available: The Method of Brockeroff [15] The sn-1,2(2,3) diacylglycerols, prepared using pancreatic lipase, are reacted with phenyl dichlorophenol to produce a mixture of sn-1,2-diacyl-3-phosphatidylphenol and sn-2,3-diacyl-1-phosphatidylphenol. Incubation with phospholipase A liberates fatty acids from the 2 position of the sn-3-phosphatide but leaves the 2 position of the sn-1 phosphatide intact. Fatty acid analysis of the lysophosphatide gives composition of the sn-1-position. Analysis of the free acids liberated from phospholipase hydrolysis gives composition of the sn-2 position. Composition of the sn-3 position can be computed: sn-3 = 2(unhydrolyzed phosphatide) – monoglyceride Pag e 237 The Method of Lands [93] The sn-1,2(2,3)-diacylglycerols, prepared by reaction with pancreatic lipase or a Grignard reagent, are incubated with diacylglycerol kinase, which phosphorylates sn-1,2-diacylglycerols but not sn-2,3-diacylglycerols. Fatty acid analysis of each of the three triacylglycerol positions can be computed from analysis of the resulting phosphatides, the monoacylglycerols (from lipase hydrolysis), and the total fat, according to the following equations: sn-1 = 2(phosphatide) – (monoglyceride) sn-2 = monoglyceride sn-3 = 3(total fat) – 2(phosphatide) 5.4.2 Positional Distribution of Fatty Acids in Natural Fats In earlier studies, major classes of triacylglycerols were separated on the basis of unsaturation (i.e., trisaturated, disaturated, diunsaturated, and triunsaturated) via fractional crystallization and oxidation-isolation methods. More recently, the techniques of stereospecific analysis made possible the detailed determinations of individual fatty acid distribution in each of the three positions of the triacylglycerols of many fats. The data listed in Table 4 clearly indicate the differences in distribution patterns among plant and animal fats. 5.4.2.1 Plant Triacylglycerols In general, seed oils containing common fatty acids show preferential placement of unsaturated fatty acids at the sn-2 position. Linoleic acid is especially concentrated at this position. The saturated acids occur almost exclusively at the 1,3 positions. In most cases, the individual saturated or unsaturated acids are distributed in approximately equal quantities between the sn-1 and the sn-3 positions. The more saturated fats of plant origin show a different distribution pattern. Approximately 80% of the triacylglycerols in cocoa butter are disaturated, with 18:1 concentrated in the Pag e 238 TABLE 4 Positional Distribution of Individual Fatty Acids in Triacylg lycerols of Some Natural Fats Fatty acid (mol%) Source Position 4:0 6:0 8:0 10:0 12:0 14:0 16:0 18:0 18:1 18:2 18:3 20:0 20:1 22:0 24:0 Cow’s milk 1 5 3 1 3 3 11 36 15 21 1 2 3 5 2 6 6 20 33 6 14 3 3 43 11 2 4 3 7 10 4 15 0.5 Coconut 1 1 4 4 39 29 16 3 4 2 0.3 2 5 78 8 1 0.5 3 2 3 3 32 13 38 8 1 0.5 3 2 Cocoa butter 1 34 50 12 1 2 2 2 87 9 3 37 53 9 Corn 1 18 3 28 50 2 2 27 70 3 14 31 52 1 Soybean 1 14 6 23 48 9 2 1 22 70 7 3 13 6 28 45 8 Olive 1 13 3 72 10 0.6 2 1 83 14 0.8 3 17 4 74 5 1 Peanut 1 14 5 59 19 1 1 1 2 2 59 39 0.5 3 11 5 57 10 4 3 6 3 Beef (depot) 1 4 41 17 20 4 1 2 9 17 9 41 5 1 3 1 22 24 37 5 1 Pig (outer back) 1 1 10 30 51 6 2 4 72 2 13 3 3 7 73 18 Pag e 239 2 position and saturated acids almost exclusively located in the primary positions (b-POSt constitutes the major species). There is approximately 1.5 times more oleic in the sn-1 than in the sn-2 position. Approximately 80% of the triacylglycerols in coconut oil are trisaturated, with lauric acid concentrated at the sn-2 positions, octanoic at the sn-3, and myristic and palmitic at the sn-1 positions. Plants containing erucic acid, such as rapeseed oil, show considerable positional selectivity in placement of their fatty acids. Erucic acid is preferentially located at the 1,3 position, but more of it is present at the sn-3 position than at the sn-1 position. 5.4.2.2 Animal Triacylglycerols Distribution patterns of fatty acids in triacylglycerols differ among animals and vary among parts of the same animal. Depot fat can be altered by changing dietary fat. In general, however, the saturated acid content of the sn-2 position in animal fats is greater than that in plant fats, and the difference in composition between the sn-1 and sn-2 positions is also greater. In most animal fats the 16:0 acid is preferentially esterified at the sn-1 position and the 14:0 at the sn-2 position. Short-chain acids in milk fat are selectively associated with the sn-3 position. The major triacylglycerols of beef fat are of the SUS type. Pig fat is unique among animal fats. The 16:0 acid is significantly concentrated at the central position, the 18:0 acid is primarily located at the sn-1 position, 18:2 at the sn-3 position, and a large amount of oleic acid occurs at positions 3 and 1. Major triacylglycerol species in lard are sn-StPSt, OPO, and POSt. The long polyunsaturated fatty acids, characteristic of marine oils, are preferentially located at the sn-2 position. 5.4.3 Crystallization and Consistency 5.4.3.1 Nucleation and Crystal Growth The formation of a solid from a solution or a melt is a complicated process in which molecules must first come in contact, orient, and then interact to form highly ordered structures known as nuclei. Nucleation can be encouraged by stirring, or the nucleation process can be circumvented by seeding the supercooled liquid with tiny crystals of the type ultimately desired. As with chemical reactions, an energy barrier exists to hinder nucleation. The more complex and more stable polymorphic forms (i.e., highly ordered, compact, of high melting point) are more difficult to nucleate. Thus, stable crystals generally do not form at temperatures just below their melting points; rather, the liquid persists for some time in a supercooled state. On the other hand, the least stable, least ordered form (a) nucleates and crystallizes readily at a temperature just below its melting point. Although the magnitude of the energy barrier to nucleation decreases as the temperature is lowered, the rate of nucleation does not increase indefinitely with decrease in temperature. At some point, decreasing the temperature increases the viscosity of the lipid sufficiently to slow the rate of nucleation. Following nucleation, enlargement of these nuclei (crystal growth) progresses at a rate dependent mainly on the temperature. 5.4.3.2 Crystal Structure [8, 99, 131] Most of our present knowledge regarding the crystalline structure and behavior of fats has resulted from x-ray diffraction studies. However, significant insights also have been gained from Pag e 240 the application of other techniques, such as nuclear magnetic resonance, infrared spectroscopy, calorimetry, dilatometry, microscopy, and differential thermal analysis. Indeed recent advances in magnetic resonance imaging (MRI) will allow greater insight into the actual dynamics of the crystallization process [130]. In the crystalline state, atoms or molecules assume rigid positions forming a repeatable, highly ordered, three-dimensional pattern. If reference points representing this regularity of structure are chosen (e.g., the center of a certain atom or a convenient point in a molecule), the resulting three-dimensional arrangement in space is known as a space lattice. This network of points embodies all the symmetry properties of the crystal. If the points of a space lattice are joined, a series of parallel-sided “unit cells” is produced, each of which contains all elements of the lattice. A complete crystal can thus be regarded as unit cells packed side by side in space. In the example of the simple space lattice given in Figure 2, each of the 18 unit cells has an atom or molecule at each of its corners. However, since each corner is shared by eight other adjacent cells, there is only one atom (or molecule) per unit cell. It can be seen that each point in the space lattice is equivalent in its environment to all other points. The ratios a:b:c (called axial ratios), as well as the angles between the crystallographic axes OX, OY, and OZ, are constant values used to distinguish different lattice arrangements. Organic compounds having long chains pack side by side in the crystal to obtain maximum van der Waals interaction. In the unit cell, three spacings can be identified, two short and one long. Thus, the long spacing of normal alkanes increases steadily with an increase in carbon number, but the short spacings remain constant. The molecular end groups (e.g., methyl or carboxyl) associate with each other to form planes. If the chains are tilted with respect to the base of the unit cell, the long spacing will be smaller to an extent dependent on the angle of tilt. Fatty acids tend to form double molecules oriented head to head by sharing hydrogen bonds between carboxyl groups. Consequently, the long spacings of fatty acids are almost twice as great as those of hydrocarbons of equal carbon number (Fig. 3). When lipid mixtures of different but similar compounds are present, crystals containing more than one kind of molecule can be formed. In the case of medium-or low-molecular-weight fatty acids differing in chain length by one carbon atom, compound crystals are formed in which dissimilar pairs are bound carboxyl to carboxyl but are otherwise arranged as in crystals containing only one acid. Also common are solid solutions in which component molecules of one type are distributed at random into the crystal lattice of another. Under certain conditions slow cooling can result in formation of layer crystals in which layers of one type of crystal deposit on the surfaces of other crystals. FIGURE 2 Crystal l 5.4.3.3 Polymorphism [8, 33, 53] Polymorphic forms are crystalline phases of the same chemical composition that differ among themselves in structure but yield identical liquid phases upon melting. Each polymorphic form, sometimes termed polymorphic modification, is characterized by specific properties, such as x-ray spacings, specific volume, and melting point. Several factors determine the polymorphic form assumed upon crystallization of a given compound. These include purity, temperature, rate of cooling, presence of crystalline nuclei, and type of solvent. Depending on their particular stabilities, transformation of one polymorphic form into another can take place in the solid state without melting. Two crystalline forms are said to be “monotropic” if one is stable and the other metastable throughout their existence and regardless of temperature change. Transformation will take place only in the direction of the more stable form. Two crystalline forms are “enantiotropic” when each has a definite range of stability. Either modification may be the stable one, and transformation in the solid state can go in either direction depending upon the temperature. The temperature at which their relative stability changes is known as the transition point. Natural fats are invariably monotropic, although enantiotropism is known to occur with some fatty acid derivatives. With long-chain compounds, polymorphism is associated with different packing arrangements of the hydrocarbon chains or different angles of tilt. The mode of packing can be described using the subcell concept. Subcells A subcell is the smallest spatial unit of repetition along the chain axis within the unit cell. The schematic in Figure 4 represents a “subcell lattice” in a fatty acid crystal. In this case, each subcell contains one ethylene group and the height of the subcell is equivalent to the distance between alternate carbon atoms in the hydrocarbon chain, that is, 2.54 Å. The methyl and acid groups are not part of the subcell lattice. Seven packing types (crystal systems) for hydrocarbon subcells are known to exist. Pag e 242 FIGURE 4 Subcell lattice in a fatty acid crystal. (From Ref. 131.) The most common types are the three shown in Figure 5. In the triclinic (T//) system, called b, the two methylene units together make up the ethylene repeat unit of which there is one per subcell, and all zigzag planes are parallel. This subcell packing occurs in n-hydrocarbons, fatty acids, and triacylglycerols. It is the most stable polymorphic form. In the common orthorhombic (O^) system, also called b there are two ethylene units in each subcell, and alternate chain planes are perpendicular to their adjacent planes. This packing occurs with n-paraffins and with fatty acids and their esters. The b form has intermediate stability. The hexagonal (H) system, generally called a, often occurs when hydrocarbons are rapidly cooled to just below their melting points. The chains are randomly oriented and exhibit rotation about their long vertical axes. This type of packing is observed with hydrocarbons, alcohols, and ethyl esters. The a form is the least stable polymorphic form. Fatty Acids Even-numbered saturated fatty acids can be crystallized in any of the polymorphic forms. In decreasing length of long spacings (or increasing tilt of the chains), these forms are designated A, B, and C. Similarly, the acids with odd numbers of carbon atoms are designated A’, B’, and FIGURE 5 Common types of hydrocarbon subcell packing . (From Ref. 131.) Pag e 243 C’. The A and A’ forms have triclinic subcell chain packing (T//); the remaining forms are packed in the common orthorhombic (O^) manner. The b form of stearic acid has been studied in detail. The unit cell is monoclinic and contains four molecules. Its axial dimensions are a = 5.54 Å, b = 7.38 Å, and c = 48.84 Å. However, the c axis is inclined at an angle of 63°38′ from the a axis, which results in a long spacing of 43.76 Å (Fig. 3). In the case of oleic acid, the low-melting form has double molecules per unit cell length, and the hydrocarbon portions around the cis double bond are tilted in opposite directions in the plane of molecules (Fig. 6). Triacylglycerols The nomenclature used in earlier literature to designate the different polymorphic forms of triacylglycerols is extremely confusing. Different authors used different criteria, such as melting points or x-ray spacings, as the basis for their nomenclature. Often the same symbols were used by different investigators to designate different polymorphic forms, and much debate took place regarding the number of forms exhibited by certain triacylglycerol species. Much of the disagreement was finally resolved based on results from infrared spectroscopy. In general, triacylglycerols, due to their relatively long chains, take on many of the features of hydrocarbons. They exhibit, with some exceptions, three principal polymorphic forms: a, b’,and b. Characteristics typical of each form are summarized in Table 5. If a monoacid triacylglycerol, such as StStSt, is cooled from the melt, it crystallizes in the least dense, lowest melting form (a). On further cooling of the a form, the chains associate more compactly, and gradual transition into the b form takes place. If the a form is heated to its melting point, a transformation into the most stable b form occurs rapidly. The b form can also be obtained directly by cooling the melt and maintaining the temperature a few degrees above the melting point of the a form. On heating the b form to its melting point, some melting takes place and transition to the stable b form occurs. The general molecular arrangement in the lattice of monoacid triacylglycerols is a double-chainlength modified tuning fork, or chair structure, as shown in Figure 7 for trilaurin. The chains in the 1 and 3 positions are oriented in a direction opposite that of the chain in the 2 position. Because natural triacylglycerols contain a variety of fatty acids, departure from the simple polymorphic classification outlined earlier is bound to occur. In the case of triacylglycerols containing different fatty acids, some polymorphic forms are more difficult to obtain than others; for example, b’ forms have been observed that have higher melting points than the b’ forms. PStP glycerides, as found in cottonseed hardstock, tend to crystallize in a b form of relatively high density. This form results in greater stiffening power when added to oils than does the expanded (snowlike) b form commonly obtained from the StStSt glycerides of soybean hardstock. The polymorphic structure of mixed triacylglycerols is further complicated by a tendency for carbon chains to segregate according to length or degree of unsaturation to form structures in which the long spacing is made up of triple chain lengths. Various tuning-fork structures have been proposed for mixed triacylglycerols containing fatty acids of different chain lengths. If the middle chain is shorter or longer than the other two by four or more carbons, there may be a segregation of chains, as in Figure 8a. In the case of unsymmetrical triacylglycerols, a chair-type arrangement similar to Figure 8b may result. Sorting of chains may also arise on the basis of unsaturation, as in Figure 8c. Such structures are designated by a number following the Greek letter. For example, b-3 would indicate the b modification with a triple chain length (Fig. 9). In the liquid state lamellar units exist in which the triacylglycerols are arranged in a chair conformation with their hydrocarbon chains disordered. A proposed model based on x-ray and Pag e 244 FIGURE 6 Crystal structure of oleic acid. (From Ref. 1.) Raman spectroscopy [58] is shown in Figure 10. The size of the lamellar units decreases with increase in temperature. Upon cooling of a melt, lamellar units increase in size until crystallization occurs. Polymorphic Behavior in Commercial Fats [26, 31, 49, 124] It is evident from the foregoing that the polymorphic behavior of a fat is largely influenced by the composition of its fatty acids and their positional distribution in the acylglycerols. In general, Pag e 245 TABLE 5 Characteristics of the Polymorphic Forms of Monoacid Triacylg lycerols Characteristic a Form b’ Form b Form Chain packing Hexagonal Orthorhombic Triclinic Short spacing (Å) 4.15 4.2 and 3.8 4.6, 3.9, 3.7 Characteristic infrared spectrum Sing le band at 720 cm1 Doublet at 727 and 719 cm-1 Sing le band at and 717 cm-1 Density Least dense Intermediate Most dense Melting point Lowest Medium Hig hest fats that consist of relatively few closely related triacylglycerol species tend to transform rapidly to stable b forms. Conversely, heterogeneous fats tend to transform more slowly to stable forms. For example, highly randomized fats exhibit b’ forms that transform slowly. Fats that tend to crystallize in b forms include soybean, peanut, corn, olive, coconut, and safflower oils, as well as cocoa butter and lard. On the other hand, cottonseed, palm, and rapeseed oils, milk fat, tallow, and modified lard tend to produce b crystals that tend to persist for long periods. The b’ crystals are desirable in the preparation of shortenings, margarine, and baked products since they aid in the incorporation of a large amount of air in the form of small air bubbles, giving rise to products of better plastic and creaming properties. In the food industry, emulsifiers are frequently added to stabilize certain metastable forms. It is believed that the action of such agents, such as sorbitan esters, is not due to their surface activity, but rather to their unique chemical structure, which allows them to fit into the crystallographic structure with the hydrophobic parts of the surfactant molecules parallel to the triacylglycerol hydrocarbon chains, with their hydrophilic moieties forming hydrogen bonds with the carbonyl groups of the triacylglycerols—a phenomenon called the “Button syndrome” [6]. Transformation of the fat crystals into other polymorphic forms is thus prevented or delayed without altering the crystal lattice. In the case of cocoa butter, in which the three main glycerides are POSt (40%), STOSt (30%), and POP (15%), six polymorphic forms (I–VI) have been recognized [102, 106, 124, 145]. Form I is the least stable and has the lowest melting point. Form V is the most stable form that can be crystallized from the melted fat and is the desired structure since it produces the bright glossy appearance of chocolate coatings. Form VI has a higher melting point than form V but cannot be crystallized from the melt; it only forms by very slow transformation of form V. The V–VI transformation that occurs during storage is of particular importance since it usually coincides with the appearance of a defect called “chocolate bloom.” This defect usually involves a loss of desirable chocolate loss and the development of a dull appearance with white or grayish spots on the surface. In addition to theories involving polymorphic transformation as a cause of chocolate bloom, it is believed that melted chocolate fat migrates to the surface and recrystallizes upon recooling, causing the undesirable appearance. Although it is clear that bloom does not depend on the appearance of a specific polymorphic form, polymorphic behavior of the cocoa fat is believed to play an important role in bloom. To delay the appearance of bloom, proper solidification of the chocolate is necessary. This is achieved by a tempering process that involves heating the cocoa butter-sugar-cocoa powder mixture to 50°C, seeding with the stable crystal “seeds,” slowly crystallizing it with continuous stirring as the temperature is reduced to 26–29°C, and then slowly heating it to 32°C. Without the stable seeds, unstable crystal forms will form initially and these are likely to melt, migrate, and transform to more stable forms (bloom). Pag e 246 FIGURE 7 Molecular arrang ement in trilaurin lattice. (From Ref. 95.) Pag e 247 FIGURE 8 Arrang ement of molecules in triacylg lycerol crystals. FIGURE 9 Double and triple chair arrang ements of b form. Pag e 248 FIGURE 10 Lamellar units in the liquid. (From Ref. 58.) Furthermore, emulsifiers have been successfully used to retard undesirable polymorphic transformation and/or migration of melted fat to the surface. 5.4.3.4 Melting [8, 53] Shown in Figure 11 are schematic heat content (enthalpy) curves for the stable b form and the metastable a form of a simple triacylglycerol. Heat is absorbed during melting. Curve ABC represents the increase in heat content of the b form with increasing temperature. At the melting point, heat is absorbed with no rise in temperature (heat of fusion) until all solid is transformed FIGURE 11 Heat content (H) melting curves of stable ( b) and unstable (a) polymorphic forms. Pag e 249 into liquid (final melting point, B). On the other hand, transformation from an unstable to a stable polymorphic form (beginning at point E in Fig. 11 and intersecting with curve AB) involves evolution of heat. Fats expand upon melting and contract upon polymorphic transformation. Consequently, if the change in specific volume (dilation) is plotted against temperature, dilatometric curves very similar to calorimetric curves are obtained, with melting dilation corresponding to specific heat. Since dilatometric measurements involve very simple instruments, they are often used in preference to calorimetric methods. Dilatometry has been used widely to determine the melting behavior of fats. If several components of different melting points are present, melting takes place over a wide range of temperature, and dilatometric or calorimetric curves similar to the schematic in Figure 12 are obtained. Point X represents the beginning of melting; below this point the system is completely solid. Point Y represents the end of melting; above this point the fat is completely liquid. Curve XY represents the gradual melting of the solid components in the system. If the fat melts over a narrow range of temperature, the slope of the melting curve is steep. Conversely, a fat is said to have a “wide plastic range” if the difference in temperature between the beginning and the end of melting is large. Thus, the plastic range of a fat can be extended by adding a relatively high-melting and/or a low-melting component. Solid Fat Index [67, 96, 100] The proportion of solids and liquids in a plastic fat at different temperatures can be estimated by constructing calorimetric or melting dilation curves similar to that of Figure 12, or preferably by using nuclear magnetic resonance. Dilatometric measurements can be made starting at temperatures sufficiently low to establish a solid line, at other temperatures high enough to establish a liquid line, and at intervals between these to determine the melting curve. The solid and liquid lines can then be extrapolated, and the solid or liquid fraction at any temperature can be calculated. As shown in Figure 12, ab/ac represents the solid fraction, and bc/ac is the liquid fraction at temperature t. The ratio of solid to liquid is known as the solid fat index (SFI), and this has relevance to the functional properties of fats in foods Dilatometric methods for SFI determination are relatively accurate and precise but are time-consuming and applicable only below 50% SFI. Dilatometry has been largely replaced by wide-line nuclear magnetic resonance (NMR) methods, which presumably measure the fat solids directly since they involve measurement of the ratio of the number of hydrogen nuclei in the solid (which produce faster decaying signals than do those in the liquid) to the total number of hydrogen nuclei in the sample. This ratio is expressed as NMR solids percent. At present, automated pulsed NMR procedures are commonly used and are believed to offer accuracy and precision superior to wide-line NMR techniques. However, low-resolution NMR is even more appropriate for industrial use in view of its reliability and lower cost. An ultrasonic technique has also been recently proposed as an alternative or adjunct to pulsed NMR. It is based on the observation that the velocity of ultrasound is greater in solid fat than in liquid oil [64, 103, 104]. It has the advantages of lower capital cost, faster measurements, higher sensitivity to low concentrations of solid fat, and easy adaptability for online measurements. 5.4.3.5 Consistency of Commercial Fats Although natural fats and products derived from them contain exceedingly complex mixtures of large numbers of individual acylglycerols of different composition and structure, they show a remarkable tendency to behave as simple mixtures of only a few components. Each group of similar compounds appears to act as a single component, so only the distinctly different groups are apparent in melting behavior. This tendency for simplification in the melting behavior of complex mixtures is indeed fortunate since it permits the application of rules governing simple mixtures to the more complex natural or processed fats. An initial reduction in the solidification point of certain mixtures of commercial fats has been observed upon hydrogenation or the addition of a high-melting component. Such systems give rise to eutecticFIGURE 13 Dilatometric curves of (a) a typical all-hydrog enated shortening and (b) a typical American marg arine. (From Ref. 8.) Pag e 251 FIGURE 14 Upper portions of dilatometric curves of cottonseed oil mixed with 0–15% hig hly hardened cottonseed oil. (From Ref. 8.) type behavior and solid-liquid equilibrium curves of striking resemblance to those typical of simple binary systems. The basic concepts involving the melting behavior of commercial fats can be simply illustrated by dilatometric curves such as those in Figures 13–15. However, as already indicated, NMR techniques give comparable information more quickly and are now the preferred methods. The dilatometric curves of plastic fats do not show a smooth melting line, but rather they exhibit a series of somewhat linear segments with visible changes in slope (Fig. 13). The inflection points between the segments are sometimes designated K points, with K being the temperature of final melting. The K points obviously correspond to specific boundaries in the phase diagrams of the complex fat; that commercial fats give rise to only a few K points FIGURE 15 Dilatometric curves of lard and milk fat. (From Ref. 8.) Pag e 252 emphasizes the tendency for different individual components in a narrow melting range to behave like a single component. If a high-melting fraction is added to a natural fat, melting of the “hard” component is clearly reflected in the dilatometric curve. The vertical distance (d in Fig. 14) provides a relative measure of the amount of the hard component. It can be seen from Figures 13, 14, and 15 that much useful information regarding the melting characteristics of plastic fats can be obtained from dilatometric curves. Hard “butters” melt over a relatively narrow range of temperature because their triacylglycerols are mainly POSt, StOSt, and POP. This abrupt melting, which occurs at the temperature of the mouth, makes such fats particularly suitable for confectionary coatings. A typical dilatometric curve for milk fat, showing almost complete melting at the temperature of the mouth, is quite different from that of lard, which exhibits a more gradual course of melting (Fig. 15). The difference in solid content of the two fats at any given temperature is clearly evident. 5.4.3.6 Mesomorphic Phase (Liquid Crystals) As discussed earlier, crystalline lipids are highly ordered structures with regular three-dimensional arrangements of their molecules. In the liquid state, the intermolecular forces are weakened and the molecules acquire freedom of movement and assume a state of disorder. In addition, phases with properties intermediate between those of the liquid and the crystalline states are known to occur, and these mesomorphic phases consist of so-called liquid crystals. Typically, amphiphilic compounds give rise to mesomorphic phases. For example, when an amphiphilic crystalline compound is heated, the hydrocarbon region may melt before the final melting point is reached. This occurs because relatively weak van der Waals forces exist among the hydrocarbon chains as compared with the somewhat stronger hydrogen bonding that exists among polar groups. Pure crystals that form liquid crystals upon heating are said to be thermotropic. In the presence of water, at temperatures above the melting point of the hydrocarbon region (the so-called Krafft temperature), hydrocarbon chains of triacylglycerols transform into a disordered state and water penetrates among the ordered polar groups. Liquid crystals formed in this manner, that is, with the aid of a solvent, are said to be lyotropic. Mesomorphic structure depends on factors such as concentration and chemical structure of the amphiphilic compound, water content, temperature, and the presence of other components in the mixture. The principal kinds of mesomorphic structures are lamellar, hexagonal, and cubic. Lamellar or Neat This structure corresponds to that existing in biological bilayer membranes. It is made up of double layers of lipid molecules separated by water (Fig. 16a). Structures of this kind are usually less viscous and less transparent than the other mesomorphic structures. The capacity of a lamellar phase to retain water depends on the nature of the lipid constituents. Lamellar liquid crystals of monoacylglycerols, for example, can accommodate up to approximately 30% water, corresponding to a water layer thickness of about 16 Å between the lipid bilayers. However, almost infinite swelling of the lamellar phase can occur if a small amount of an ionic surface-active substance is added to the water. If the water content is increased above the swelling limit of the lamellar phase, a dispersion of spherical aggregates consisting of concentric, alternating layers of lipid and water gradually forms. In general, a lamellar liquid crystalline phase tends to transform upon heating into hexagonal II or cubic mesophases. On the other hand, if the lamellar liquid-crystal phase is cooled below the Kraftt temperature, a metastable “gel” will form in which water remains between the lipid bilayers and the hydrocarbon chains recrystallize. During extended holding Pag e 253 FIGURE 16 Mesomorphic structures of lipids: (a) lamellar, (b) hexag onal l, (c) hexag onal II, (d) cubic. (From Ref. 95.) the water is expelled and the gel phase transforms into a microcrystalline suspension in water, called a coagel. Hexagonal In this structure the lipids form cylinders arranged in a hexagonal array. The liquid hydrocarbon chains fill the interior of the cylinders, and the space between the cylinders is taken up by water (Fig. 16b). This type of liquid crystal is termed “hexagonal I” or “middle.” A reversed hexagonal structure, “hexagonal II,” is also possible in which water fills the interior of the cylinders and is surrounded by the polar groups of the amphiphile. The hydrocarbon chains extend outward making up the continuous phase between the cylinders (Fig. 16c). If hexagonal I liquid crystals are diluted with water, spherical micelles form. However, dilution of hexagonal II liquid crystals with water is not possible. Cubic or Viscous Isotropic Although this state is encountered with many long-chain compounds, it is not as well characterized as are the lamellar and hexagonal liquid crystals. Cubic-phase structures were studied in

Leave a Reply

Your email address will not be published.