Influence of Biopolymer Emulsifier Type on Formation and Stability of Rice Bran Oil-in-Water Emulsions: Whey Protein, Gum Arabic, and Modified Starch

Transcription

1 Influence of Biopolymer Emulsifier Type on Formation and Stability of Rice Bran Oil-in-Water Emulsions: Whey Protein, Gum Arabic, and Modified Starch Ratchanee Charoen, Anuvat Jangchud, Kamolwan Jangchud, Thepkunya Harnsilawat, Onanong Naivikul, and David Julian McClements Abstract: Rice bran oil (RBO) is used in foods, cosmetics, and pharmaceuticals due to its desirable health, flavor, and functional attributes. We investigated the effects of biopolymer emulsifier type and environmental stresses on the stability of RBO emulsions. Oil-in-water emulsions (5% RBO, 1 mm citrate buffer) stabilized by whey protein isolate (), gum arabic (), or modified starch () were prepared using high-pressure homogenization. The new used had a higher number of octenyl succinic anhydride (OSA) groups per starch molecule than conventional. The droplet diameters produced by and were considerably smaller (d < 3 nm) than those produced by (d > 1 nm). The influence of ph (3 to 8), ionic strength ( to 5 mm NaCl), and thermal treatment (3 to 9 C) on the physical stability of the emulsions was examined. Extensive droplet aggregation occurred in -stabilized emulsions around their isoelectric point (4 < ph < 6),athighsalt(> 2 mm, ph 7), and at high temperatures (>7 C, ph 7, 15 mm NaCl), which was attributed to changes in electrostatic and hydrophobic interactions between droplets. There was little effect of ph, ionic strength, and temperature on emulsions stabilized by or, which was attributed to strong steric stabilization. In summary: produced small droplets at low concentrations, but they had poor stability to environmental stress; produced large droplets and needed high concentrations, but they had good stability to stress; new produced small droplets at low concentrations, with good stability to stress. Keywords: emulsions, gum arabic, modified starch, rice bran oil, stability, whey protein Practical Application: This study showed that stable rice bran oil-in-water emulsions can be formed using biopolymer emulsifiers. These emulsions could be used to incorporate RBO into a wide range of food products. We compared the relative performance of whey protein,, and a new at forming and stabilizing the emulsions. The new OSA was capable of forming small stable droplets at relatively low concentrations. Introduction Rice is an important food crop with more than half of the world s population eating it as a major part of their diet (Van Hoed and others 26; Naivikul and others 28). Rice bran oil (RBO) is a byproduct of milled rice (Orthoefer and Eastman 24; Van Hoed and others 26). RBO is used as a cooking and salad oil due to its high smoke point and desirable flavor profile (Ghosh 27). It is also rich in vitamin E complex, tocopherols, tocotrienols, phytosterols, polyphenols, and squalene and contains a potent antioxidant known as gamma oryzanol (Vieno and others 2; Juliano and others 25; Van Hoed and others 26). The Submitted 8/5/21, Accepted 1/19/21. Authors Charoen, Jangchud, Jangchud, and Harnsilawat are with Dept. of Product Development, Faculty of Agro-Industry, Kasetsart Univ., Bangkok, Thailand. Author Naivikul is with Dept. of Food Science and Technology, Faculty of Agro-Industry, Kasetsart Univ., Bangkok, Thailand. Author McClements is with Dept. of Food Science, Univ. of Massachusetts Amherst, Amherst, MA, 13, U.S.A. Direct inquiries to author McClements ( mcclements@foodsci.umass.edu) presence of high levels of natural antioxidants means that RBO has a relatively long shelf life compared to other cooking oils. Interest in RBO has also been growing because of its potential health benefits. Studies have shown that RBO reduces bad cholesterol (low density lipoprotein) levels without reducing good cholesterol (high density lipoprotein) levels (Gunstone and others 1994; Moreau and others 22). RBO contains high levels of vitamin E that have been reported to have hypocholesterolemic, anticancer, and neuroprotective properties (Sen and others 27). Recently, it has been shown that gamma oryzanol from RBO can regulate antioxidant and stress genes in rats (Ismail and others 21). Consequently, there is increasing interest in incorporating RBO into a wide variety of food products in order to benefit from its desirable functional and nutritional characteristics. In many food applications, oils are present in the form of lipid droplets dispersed within an aqueous medium (oil-in-water emulsions), rather than as bulk oils (Dickinson 22; McClements 25). In these systems, it is important that the oil droplets remain both physically and chemically stable throughout the shelf life of the product. The relatively high levels of functional and C 21 Institute of Food Technologists R doi: /j x Further reproduction without permission is prohibited Vol. 76, Nr. 1, 211 Journal of Food Science E165

2 nutraceutical components present within RBO may mean that it behaves differently in emulsions than other edible oils. The presence of minor components within edible oils can impact emulsion performance, due to their ability to impact the physical or chemical stability of these systems (Khan and Shahidi 2; McClements 25; Arima and others 29). For example, some minor components may be surface active and interfere with emulsion formation and stability by competing with emulsifiers. A major objective of this work was therefore to establish whether RBO can be used to form stable oil-in-water emulsions suitable for application in a variety of food products. Ultimately, it would be desirable to produce RBO emulsions that could be incorporated into final products with consumerfriendly labels, and so we utilized biopolymer emulsifiers (rather than synthetic surfactants) to stabilize them. We selected 3 different surface-active biopolymers that have previously been shown to stabilize oil-in-water emulsions: gum arabic (), modified starch (), and whey protein isolate () (Chanamai and Mc- Clements 22). These biopolymers were selected because they differ in their molecular characteristics: is a mixture of amphoteric globular proteins (Dalgleish 1997; Wilde 2); is a mixture of anionic polysaccharides and protein fractions (Garti and Leser 21; Dickinson 23; Al-Assaf and Phillips 28; Castellani and others 21); consists of starch molecules that have been chemically reacted with octenyl succinic anhydride (OSA) to give them some hydrophobic character (Trubiano 1995; Tan 24; Given 29). These differences in molecular characteristics mean that each type of emulsifier has different interfacial properties (Erni and others 27) and abilities to form and stabilize emulsions (Chanamai and McClements 22). During homogenization, globular proteins tend to adsorb more rapidly to oil droplet surfaces than polysaccharides, and hence are capable of forming smaller droplets (McClements 25). In addition, they can usually be used at much lower emulsifier-to-oil levels than or traditional emulsifiers (Chanamai and Mc- Clements 22). On the other hand, globular proteins tend to stabilize emulsions primarily through electrostatic interactions and are therefore sensitive to changes in ph and salt (Chanamai and McClements 22), whereas and tend to stabilize emulsions primarily through steric interactions and are therefore less influenced by ph and salt (Wilde 2; Dickinson 23; Mc- Clements 25). It should be noted that both and do have some negative charge due to the presence of anionic groups (Mirhosseini and others 28; Nakauma and others 28), and hence electrostatic interactions may play an important role in some of their applications, for example, interaction with cationic species in the surrounding aqueous phase. In addition, globular proteins unfold when they are heated above their thermal denaturation temperature, which may promote emulsion instability through an increase in their surface hydrophobicity (Kim and others 22; Kim and others 25). On the other hand, polysaccharide emulsifiers tend to be less influenced by thermal treatment (Chanamai and McClements 22). We therefore examined the influence of ph, ionic strength, and heating on the stability of the biopolymer-coated RBO droplets, since food emulsions typically experience variations in these parameters in commercial products. A novel aspect of this work was that we tested the performance of a newly developed OSA- as an emulsifier. This starch is reported to have a higher level and more even distribution of OSA groups on each starch molecule. This study shows that this biopolymer emulsifier can be used at relatively low levels to produce small droplets that have good stability to changes in solution and environmental conditions. Materials and Methods Materials RBO was purchased from a food ingredient manufacturer (Thai Edible Oil Co., Ltd., Bangkok, Thailand). This RBO was isolated from various rice bran sources using a solvent extraction method. The fatty acid composition of this product has been analyzed previously: 19.8% palmitic acid, 44.6% oleic acid, 3.2% linoleic acid, and.9% linolenic acid (Van Hoed and others 26). The concentration of the major lipophilic functional components has also been reported: total sterols (1.83 g/1 g); oryzanol (.3 g/1 g); total tocopherol and tocotrienol (.62 g/1 g) (Van Hoed and others 26). (97.7 wt% protein) was donated by Davisco Foods Intl. (Le Sueur, Minn., U.S.A.). was donated by TIC Gums (Belcamp, Md., U.S.A.). was donated by the Natl. Starch LLC (Bridgewater, N.J., U.S.A.). It was reported by the manufacturer that the (PURITY GUM TM Ultra) used in this study was produced using a new approach that leads to a higher level and more uniform distribution of bound OSA on the starch molecules than conventional. Sodium chloride, sodium citrate, and sodium azide were purchased from Sigma Chemical Co. (St. Louis, Mo., U.S.A.). Double-distilled water was used to prepare all solutions and emulsions. It is important to note that the 3 biopolymer emulsifiers used in this study were from individual batches. It is known that the composition, structure, and functionality of biopolymer emulsifiers may vary depending on the batch used. Consequently, the results reported should be seen as a guide to general behavior of different categories of biopolymer emulsifiers. Methods Preparation of oil-in-water emulsions. Aqueous phases were prepared by dispersing (.2 to 1 wt%), (.1 to 1 wt%), or (.2 to 5 wt%) in aqueous buffer solutions (1. mm sodium citrate,.1 wt% sodium azide, ph 7.) followed by stirring at room temperature overnight to ensure complete dispersion and hydration. Rice bran oil-in-water emulsions were prepared by homogenizing 5. wt% oil phase with 95. wt% aqueous phase at ambient temperature. An emulsion premix was prepared using a high-speed blender (2 min, Biospec Products Inc., Bartlesville, U.S.A.), which was then passed through a high-pressure homogenizer (Model 11, Microfluidics, Newton, Mass., USA) 3 times at 932 psi. A Y-type homogenization interaction chamber was used with an internal channel diameter of 75 μm. The influence of environmental stresses on emulsion stability. The stability of the various emulsions to ph, ionic strength, and temperature was tested. Three 5 wt% oil-in-water emulsions were prepared using different emulsifier types and concentrations determined in preliminary experiments described in Section 3.1: (.45 wt%); (1. wt%); (1. wt%). Stability to ph. Emulsion samples were prepared in aqueous buffer solutions and then the ph was adjusted to the desired final value (ph 3 to 8) using either NaOH and/or HCl solution. Emulsion samples (1 ml) were then transferred into glass test tubes (16 15 mm) and stored at ambient temperature overnight prior to analysis. Stability to ionic strength. Emulsions (ph 7) were diluted with different amounts of NaCl and buffer solution to form a series of samples with the same droplet concentration but different salt E166 Journal of Food Science Vol. 76, Nr. 1, 211

3 concentrations ( to 5 mm NaCl). The emulsions were stirred for 3 min and then transferred into glass test tubes (16 15 mm) and stored at ambient temperature overnight prior to analysis. Stability to heating. Emulsions (ph 7) were prepared containing either or 15 mm NaCl and then 1 ml samples were transferred into glass test tubes (16 15 mm), which were stored in a water bath for 3 min at a fixed temperature ranging from 3 to 9 C. The emulsion samples were then immediately placed at room temperature and stored overnight prior to analysis. Measurement of emulsion stability. Particle size. The particle size distribution (PSD) of the emulsions was measured using a laser light scattering instrument (Master Sizer 2, Malvern Instruments Ltd., Worcestershire, U.K.). This instrument measures the intensity of laser light scattered from a dilute emulsion, and then reports the PSD that gives the closest fit between theoretical calculations (Mie theory) and experimental measurements of intensity versus scattering angle. To avoid multiple scattering effects emulsions were diluted with the same buffer as the continuous phase. Particle size measurements are reported as volume-surface mean diameters d 32 ( = n i d 3 i / n i d 2 i ) or volume-weighted mean diameters d 43 ( = n i d 4 i / n i d 3 i ), where n i is the number of particles with diameter d i. The refractive indices of the dispersed and continuous phases used in the calculations of the PSD were and 1.33, respectively. We assumed that the imaginary part of the refractive index of the RBO was zero, although it did have a slight yellowish color, which may therefore have had some effect on the reported PSDs. It should be noted that dilution and stirring are likely to disrupt any weakly flocculated droplets, and so the particle size data on highly aggregated emulsions should be interpreted with caution. ζ -potential. The electrical charge (ζ -potential) of lipid droplets in the emulsions was determined using a particle electrophoresis instrument (ZEN36, Nano-series, Zetasizer, Malvern Instruments). Emulsions were diluted until they gave an instrument attenuation factor of approximately 6 using buffer solution at the same ph and NaCl concentration as the initial sample. The emulsions were agitated prior to analysis to ensure that they were homogeneous. The ζ -potential of each individual sample was calculated from the average of 2 freshly prepared samples with at least 2 replications per sample. The instrument used the Smoluchowski approximation to calculate the ζ -potential from the measured electrophoretic mobility of the particles. Creaming index. Emulsion samples (1 ml) were placed in glass test tubes (16 15 mm) and then stored at ambient temperature for 7 d before analysis. The susceptibility of the emulsions to creaming was ascertained by measuring the height of the boundary layer between the opaque droplet-rich layer at the top and the transparent or turbid droplet-depleted layer at the bottom of the test tubes. Creaming results are reported as the Creaming Index (CI) = 1 (height of interface)/(height of total emulsion) (Demetriades and McClements 2). (.2 to 5 wt%). The dependence of the mean droplet diameter (d 43 ) of the resulting emulsions on initial emulsifier concentration was then measured 6 h after homogenization. The mean droplet diameter tended to decrease as the emulsifier concentration was increased (Figure 1), which can be attributed to the fact that there was more emulsifier available to cover the newly formed oil-water interfaces created during homogenization, as well as to the fact that the interfaces become saturated more rapidly at higher emulsifier concentrations (Walstra 1993; Walstra 23; Jafari and others 28). The mean droplet diameter decreased steeply with emulsifier concentration initially up to a certain emulsifier level (Region I), but then changed gradually when the emulsifier concentration was increased further (Region II). In Region I, the droplet size is limited by the amount of emulsifier available to cover the droplets formed, but in Region II, the droplet size is mainly limited by the maximum disruptive forces generated by the homogenizer (Tcholakova and others 23; Tcholakova and others 24). The emulsifier concentration demarking Region I and II was around.25% for,.1% for, and.5% for (Figure 1). The minimum droplet diameter that could be produced also depended on emulsifier type and concentration, for example, at 4 wt% emulsifier, d 43 =.28,.38, and 1.1 μm,,, and, respectively. In the remainder of the experiments, we prepared RBO emulsions using emulsifier concentrations that were capable of producing small droplet sizes without having too much excess emulsifier present, that is,.45%, 4.5%, and 1.%. It should be noted that these values were determined under a specific set of homogenization conditions, for example, oil concentration, oil phase composition, aqueous phase composition, temperature, operating pressure, and number of passes. In commercial applications, the optimum amount of a specific emulsifier required would have to be determined for the particular set of product compositions and homogenization conditions utilized. Indeed, recent studies have shown that the molecular characteristics of are altered during homogenization by an amount that depends on the homogenization d 43 ( m) 2 Results and Discussion RBO emulsion formation The purpose of our initial experiments was to establish the minimum amount of each type of biopolymer emulsifier that could be used to prepare stable emulsions with small mean droplet diameters. Emulsions were prepared by homogenizing 5% RBO with 95% aqueous phases containing different emulsifier types and concentrations: (.2 to 1 wt%); (.1 to 1 wt%); Emulsifier Concentration (wt%) Figure 1 Influence of emulsifier concentration on the mean particle diameter of diluted 5% rice bran oil-in-water emulsions stabilized by whey protein isolate (), gum arabic (), or modified starch (). Vol. 76, Nr. 1, 211 Journal of Food Science E167

4 pressure, which may affect its performance as an emulsifier (Al- Assaf and others 29). Clearly, further study is required to identify the influence of varying homogenization conditions on the molecular characteristics and performances of other biopolymer emulsifiers. Influence of ph on RBO emulsion stability In this section, we examined the influence of ph on the physicochemical properties of RBO emulsions stabilized by,, and. The mean droplet diameter (d 43 ) of the -stabilized emulsions was around.53 μm at relatively low (ph 3) and high (ph 6 to 8) ph values (Figure 2). Nevertheless, a large increase in mean particle diameter was observed around the isoelectric point (pi) (4 < ph < 6) of the. On the other hand, the mean droplet diameter of -stabilized emulsions (d approximately 1.2 to 1.8 μm) and -stabilized emulsions (d approximately.5 to.69 μm) remained relatively constant from ph 3 to 8 (Figure 2). Light scattering measurements indicated that there was little ph dependence of the PSD of the - or -stabilized emulsions, but that extensive aggregation occurred in the -stabilized emulsions at ph 5 (Figure 3). The particle size measurements were supported by creaming stability measurements, which indicated that stabilized emulsions were highly unstable to creaming at ph values around their pi but stable at higher and lower ph values, whereas - and -stabilized emulsions were stable across the entire ph range studied (Figure 4). These results are in agreement with earlier studies of the ph-stability of biopolymer-stabilized oil-inwater emulsions (Demetriades and others 1997; Kulmyrzaev and others 2; Chanamai and McClements 22). The poor stability of the -stabilized emulsions around the protein s pi can be attributed to a reduction in the electrostatic repulsion between the oil droplets, which leads to droplet flocculation (McClements 25). The good ph-stability of the - and -stabilized emulsions can be accounted for by the fact that the lipid droplets are coated by a relatively thick layer of hydrophilic polysaccharide d 43 (μm) 3 3 molecules, and hence they are largely stabilized by steric repulsion rather than electrostatic repulsion (Chanamai and McClements 22; McClements 25). However, the fact that both and tend to have some negative charge means that there may also be some electrostatic contribution to the overall droplet droplet repulsion in these systems. The presence of a thick polysaccharide layer may also decrease the magnitude of the attractive van der Waals forces acting between the droplets, which also increases emulsion stability to flocculation (Guzey and McClements 27). The ph dependence of the droplet ζ -potential for the 3 types of emulsion is shown in Figure 5. The ζ -potential of the stabilized droplets went from highly positive at low ph to highly negative at high ph, with a point of zero charge around ph 4.6. This ph dependence of the droplet charge is due to the fact that the pi of adsorbed layer of molecules is around ph 5 (Demetriades and others 1997; Demetriades and McClements 1998; Kulmyrzaev and others 2; Chanamai and McClements 22). At relatively high H + concentrations (ph << pi), the amino groups are positively charged (-NH 3 + )andthecarboxyl groups are neutral (-COOH), so the net protein charge is positive. At relatively low H + concentrations (that is, ph >> pi), the carboxyl groups are negatively charged (-COO ) and the amino groups are neutral (-NH 2 ), so the net protein charge is negative. At the pi, the number of positively and negatively charged groups on the protein is balanced and so the protein has no net charge. The interfacial layers formed by globular proteins tend to be relatively thin (a few nanometer thick), so this type of biopolymer emulsifier tends to stabilize emulsions mainly by electrostatic (rather than steric) repulsion. Hence, when the protein loses its net charge around the pi the stability of the droplets to aggregation is greatly reduced since the attractive van der Waals forces then dominate. The ζ -potentials of the lipid droplets coated by and were negative at all ph values (Figure 5), which can be attributed to the presence of some negatively charged side groups (-COO ) on these polysaccharide molecules (Ray and others 1995; Tan 1997; Padala and others 29). Interestingly, the had a much higher negative charge than the at all ph values, which suggests that the linear charge density of the was higher than the. This may have important consequences for the interactions of biopolymer-coated lipid droplets with other charged species in food and beverage systems, such as transition metals that promote lipid oxidation. For example, it has been shown that negatively charged droplets attract positively charged transition metals to lipid droplet surfaces, which promote lipid oxidation (McClements and Decker 2; Hu and others 23). There was a slight reduction in the negative charge on the - and -coated lipid droplets when the ph was reduced below about 5, which can be attributed to the fact that this solution ph moved around and below the pk a values of the carboxyl groups so that they lost some of their negative charge. The fact that both and were electrically charged, suggests that there may also have been some electrostatic repulsion between the droplets that helped stabilize them Figure 2 ph dependence of the mean droplet diameter (d 43 )ofdiluted5% rice bran oil-in-water emulsions stabilized by.45% whey protein isolate (), 1.% gum arabic (), or 1.% modified starch (). ph Influence of ionic strength on RBO emulsion stability In this section, the influence of ionic strength ( to 5 mm NaCl) on the stability of RBO emulsions stabilized by the 3 different kinds of biopolymer emulsifier was examined. In the absence of salt, the mean droplet diameters (d 43 ) of the emulsions were initially approximately.68, 1.5, and.54 μm for,, and, respectively. There was an appreciable increase in the mean particle diameter of the -stabilized emulsions at ionic strengths of 2 mm and E168 Journal of Food Science Vol. 76, Nr. 1, 211

5 higher (Figure 6). This increase in droplet aggregation at higher salt concentrations is due to screening of the electrostatic repulsion between the protein-coated droplets (McClements 25; Guzey and McClements 27). Above a critical salt level, the electrostatic repulsion is no longer strong enough to overcome the attractive interactions (van der Waals and hydrophobic) acting between the droplets. On the other hand, there was little change in the particle size of the emulsions stabilized by or with increasing ionic strength (Figure 6), which can be attributed to the fact that these emulsions are stabilized primarily by steric repulsion (with some electrostatic contribution), rather than primarily by electrostatic interactions (with some steric contribution) as is the case for globular proteins (Chanamai and McClements 22; Dickinson 23; McClements 25). Visual observations of the emulsions containing different salt levels indicated that a distinct cream layer formed on top of the -stabilized emulsions at higher salt levels ( 2 mm), but that the rest of the emulsions were relatively stable to gravitational separation (data not shown). For all 3 emulsifiers, there was a decrease in the magnitude of the negative ζ -potential with increasing salt concentration (Figure 7), which can be attributed to electrostatic screening effects A ph 3 B ph 5 Particle Volume (%) Particle Volume (%) C Particle Diameter ( m) ph Particle Diameter ( m) 12 Particle Volume (%) Figure 3 Particle size distributions of diluted 5% rice bran oil-in-water emulsions stabilized by.45% whey protein isolate (), 1.% gum arabic (), or 1.% modified starch () at (A) ph 3; (B) ph 5; (C) ph Particle Diameter ( m) Vol. 76, Nr. 1, 211 Journal of Food Science E169

6 (Israelachvili 1992). Counter-ions (Na + ) in the aqueous phase accumulate around the negatively charged groups (-COO )on the protein surface due to electrostatic attraction, thereby reducing their net charge (McClements 25). Influence of thermal processing on RBO emulsion stability In this section, we investigated the influence of heat treatment (3 to 9 C, 3 min) and salt concentration ( or 15 mm NaCl) on the properties of RBO emulsions stabilized by,, and Creaming index (%) Mean diameter (μm) ph Figure 4 ph dependence of the creaming index of 5% rice bran oil-in-water emulsions stabilized with.45% whey protein isolate (), 1.% gum arabic (), or 1.% modified starch (). 5 NaCl (mm) Figure 6 Ionic strength dependence of the mean particle diameter (d 43 ) of diluted 5% rice bran oil-in-water emulsions stabilized by.45% whey protein isolate (), 1.% gum arabic (), or 1.% modified starch () ζ-potential (mv) ζ-potential (mv) Figure 5 ph dependence of particle electrical charge (ζ -potential) of diluted 5% rice bran oil-in-water emulsions stabilized by.45% whey protein isolate (), 1.% gum arabic (), or 1.% modified starch (). ph NaCl (mm) Figure 7 Dependence of particle electrical charge (ζ -potential) of diluted 5% rice bran oil in water emulsions stabilized by.45% whey protein isolate (), 1.% gum arabic (), or 1.% modified starch () on salt concentration (ph 7). E17 Journal of Food Science Vol. 76, Nr. 1, 211

7 at ph 7. The NaCl was added to the emulsions before they were subjected to heat treatment, since this has previously been shown to have the biggest negative impact on emulsion stability for globular protein stabilized emulsions (Kim and others 22; Kim and others 25). The particle size was measured after the emulsions had been stored at room temperature overnight, whereas the creaming stability was measured after they had been stored at room temperature for 7 d. In the absence of added salt, all the emulsions were relatively stable to droplet aggregation and creaming after heat treatments with little change in mean particle diameter and no visible evidence of phase separation (data not shown), with the exception of the sample at high temperatures. For example, after heat treatments rangingfrom3to9 C, the mean droplet diameters (d 43 )ofthe emulsions were approximately.58 ±.3, 1.7 ±.1, and.58 ±.8 μm for,, and, respectively. The only sample that showed an appreciable increase in mean particle diameter in the absence of salt was for the -stabilized emulsion heated at 9 C for 3 min: d 43 increased from.53 μm before heating to.78 μm after heating. In the presence of added salt (15 mm), the -stabilized emulsions became unstable to droplet aggregation when they were heated above 6 C, as demonstrated by an appreciable increase in mean particle diameter (Figure 8) and some visible evidence of creaming (data not shown). The instability of the -emulsions to heating in the presence of salt can be attributed to thermal denaturation of the globular proteins adsorbed to the lipid droplet surfaces (Kim and others 22; Kim and others 24; Kim and others 25). When the molecules unfold, they expose nonpolar groups to the surrounding aqueous phase, which increases the surface hydrophobicity of the droplets and promotes aggregation through hydrophobic attraction (Kim and others 25). d 43 (μm) Temperature (ºC) In addition, sulfhydryl groups are also exposed when the protein is heated above its thermal denaturation temperature, which promotes droplet droplet aggregation through covalent disulfide bonds (Monahan and others 1996). In the absence of salt, the electrostatic repulsion is strong enough to overcome the hydrophobic and van der Waals attraction, but in the presence of salt, the additional hydrophobic attraction associated with protein unfolding promotes droplet aggregation (McClements 25). The -and -stabilized RBO emulsions did not exhibit extensive droplet aggregation at either or 15 mm NaCl (Figure 8), which can be attributed to the fact that they are stabilized primarily by polysaccharides that do not unfold to expose nonpolar groups at higher temperatures. There was a slight increase in the mean particle diameter of the -stabilized emulsions above about 6 C (Figure 8), but no evidence of creaming (data not shown). This effect may be attributed to other changes in the molecular characteristics of starch upon heating. Martinez and others (23) examined the influence of heating on the PSDs of emulsions stabilized by. They found that heating starch-stabilized emulsions around 72 to 83 C, which was reported to be the swelling region of starch granules, led to an increase in particle aggregation (Martinez and others 23). The electrical characteristics of all of the emulsions were unchanged by heating (data not shown). For example, the ζ -potentials were 48.8 ±.6, 31.2 ±.6, and 3.5 ±.2 mv for,, and, respectively, after thermal treatments ranging from 3 to 9 C. Conclusions This study has characterized the influence of biopolymer emulsifier type (,, ) on the formation and stability of RBO emulsions. At relatively low emulsifier concentrations (emulsifierto-oil ratio <.1), was able to produce the smallest droplets during homogenization, then, and then. On the other hand, at relatively high emulsifier concentrations (emulsifier-tooil ratio > 1), was able to produce the smallest droplets, then, and then. Emulsions formed using the polysaccharide emulsifiers ( and ) had much better stability to environmental stresses (ph, salt, and thermal processing) than those formed using the globular protein emulsifier (), which was attributed to differences in the nature of the colloidal interactions operating between oil droplets. Globular proteins only form a thin polymeric coating around lipid droplets, and so the primary stabilization mechanism is electrostatic repulsion. Consequently, they tend to become unstable to aggregation when the repulsive electrostatic interactions are weakened (for example, near the proteins pi or at high salt concentrations) or when the attractive hydrophobic interactions are increased (for example, when the proteins are heated above their thermal denaturation temperature and unfold to expose nonpolar groups). On the other hand, polysaccharides form thick hydrophilic polymeric coatings around lipid droplets, and so the primary stabilization mechanism for this type of system is steric repulsion. Consequently, polysaccharide-coated lipid droplets are much less sensitive to changes in ph, salt content, and temperature than globular protein-coated ones. The newly developed was capable of producing small droplets during homogenization at relatively low emulsifier concentrations, as well as forming emulsions that were stable to a wide range of environmental stresses after homogenization. Figure 8 Dependence of the mean droplet diameter (d 43 ) of diluted 5% rice bran oil-in-water emulsions stabilized by.45% whey protein isolate (), 1.% gum arabic (), or 1.% modified starch () on heat treatment (3 to 9 C, 3 min, 15 mm NaCl, ph 7). Acknowledgments This study was financially supported by Natl. Science and Technology Development Agency (NSTDA), Ministry of Science and Vol. 76, Nr. 1, 211 Journal of Food Science E171

8 Technology, Thailand. The authors thank Jason Li and Afaf Makarious (National Starch LLC) for supplying the modified starch and for useful discussions. References Al-Assaf S, Phillips GO. 28. Characterization and prediction of emulsification performance of acacia gums. Lc Gc North America 26(2): Al-Assaf S, Sakata M, McKenna C, Aoki H, Phillips GO. 29. Molecular associations in acacia gums. Struct Chem 2(2): Arima S, Ogawa A, Ueno S, Sato K. 29. Sucrose fatty acid esters on emulsion stability at chilled temperatures of oil-in-water emulsion. J Jpn Soc Food Sci Technol Nippon Shokuhin Kagaku Kogaku Kaishi 56(4): Castellani O, Guibert D, Al-Assaf S, Axelos M, Phillips GO, Anton M. 21. Hydrocolloids with emulsifying capacity. Part 1-Emulsifying properties and interfacial characteristics of conventional (Acacia senegal (L.) Willd. var. senegal) and matured (Acacia (sen) SUPER GUM (TM)) Acacia senegal. Food Hydrocolloids 24(2 3): Chanamai R, McClements DJ. 22. Comparison of gum arabic, modified starch, and whey protein isolate as emulsifiers: influence of ph, CaCl2 and temperature. J Food Sci 67(1):12 5. Dalgleish DG Adsorption of protein and the stability of emulsions. Trends Food Sci Technol 8(1):1 6. Demetriades K, Coupland JN, McClements DJ Physical properties of whey protein stabilized emulsions as related to ph and NaCl. J Food Sci 62(2): Demetriades K, McClements DJ Influence of ph and heating on physicochemical properties of whey protein-stabilized emulsions containing a nonionic surfactant. J Agric Food Chem 46(1): Demetriades K, McClements DJ. 2. Influence of sodium dodecyl sulfate on the physicochemical properties of whey protein-stabilized emulsions. Colloids Surf A Physicochem Eng Asp 161(3): Dickinson E. 22. Food colloids. Curr Opin Colloid Interface Sci 7(5 6): Dickinson E. 23. Hydrocolloids at interfaces and the influence on the properties of dispersed systems. Food Hydrocolloids 17(1): Erni P, Windhab EJ, Gunde R, Graber M, Pfister B, Parker A, Fischer P. 27. Interfacial rheology of surface-active blopolymers: acacia senegal gum versus hydrophobically modifed starch. Biomacromolecules 8(11): Garti N, Leser ME. 21. Emulsification properties of hydrocolloids. Polym Adv Technol 12(1 2): Ghosh M. 27. Review on recent trends in rice bran oil processing. J Am Oil Chem Society 84(4): Given PS. 29. Encapsulation of flavors in emulsions for beverages. Curr Opin Colloid Interface Sci 14(1):43 7. Gunstone FD, Harwood JL, Padley FB The lipid handbook, 2nd ed. London: Chapman & Hall. Guzey D, McClements DJ. 27. Impact of electrostatic interactions on formation and stability of emulsions containing oil droplets coated by beta-lactoglobulin-pectin complexes. J Agric Food Chem 55(2): Hu M, McClements DJ, Decker EA. 23. Impact of whey protein emulsifiers on the oxidative stability of salmon oil-in-water emulsions. J Agric Food Chem 51(5): Ismail M, Al-Naqeeb G, bin Mamat WAA, Ahmad Z. 21. Gamma-oryzanol rich fraction regulates the expression of antioxidant and oxidative stress related genes in stressed rat s liver. Nutr Metab 7(23):1 13. Israelachvili J Intermolecular and surface forces, 2nd ed. London, U.K.: Academic Press. Jafari SM, Assadpoor E, He YH, Bhandari B. 28. Re-coalescence of emulsion droplets during high-energy emulsification. Food Hydrocolloids 22(7): Juliano C, Cossu M, Alamanni MC, Piu L. 25. Antioxidant activity of gamma-oryzanol: mechanism of action and its effect on oxidative stability of pharmaceutical oils. Int J Pharm 299(1 2): Khan MA, Shahidi F. 2. Oxidative stability of stripped and nonstripped borage and evening primrose oils and their emulsions in water. J Am Oil Chem Soc 77(9): Kim HJ, Decker EA, McClements DJ. 22. Role of postadsorption conformation changes of beta-lactoglobulin on its ability to stabilize oil droplets against flocculation during heating at neutral ph. Langmuir 18(2): Kim HJ, Decker EA, McClements DJ. 24. Comparison of droplet flocculation in hexadecane oil-in-water emulsions stabilized by beta-lactoglobulin at ph 3 and 7. Langmuir 2(14): Kim HJ, Decker EA, McClements DJ. 25. Influence of protein concentration and order of addition on thermal stability of beta-lactoglobulin stabilized n-hexadecane oil-in-water emulsions at neutral ph. Langmuir 21(1): Kulmyrzaev A, Chanamai R, McClements DJ. 2. Influence of ph and CaCl2 on the stability of dilute whey protein stabilized emulsions. Food Res Intl 33(1):15 2. Martinez I, Partal P, Munoz J, Gallegos C. 23. Influence of thermal treatment on the flow of starch-based food emulsions. Eur Food Res Technol 217(1): McClements DJ. 25. Food emulsions: principles, practice, and techniques, 2nd ed. Boca Raton: CRC Press. McClements DJ, Decker EA. 2. Lipid oxidation in oil-in-water emulsions: impact of molecular environment on chemical reactions in heterogeneous food systems. J Food Sci 65(8): Mirhosseini H, Tan CP, Taherian AR. 28. Effect of glycerol and vegetable oil on physicochemical properties of Arabic gum-based beverage emulsion. Eur Food Res Technol 228(1): Monahan FJ, McClements DJ, German JB Disulfide-mediated polymerization reactions and physical properties of heated -stabilized emulsions. J Food Sci 61(3):54 9. Moreau RA, Whitaker BD, Hicks KB. 22. Phytosterols, phytostanols, and their conjugates in foods: structural diversity, quantitative analysis, and health-promoting uses. Prog Lipid Res 41(6): Naivikul O, Klinkesorn U, Santiwattana P. 28. Rice bran oil, 1st ed. Bangkok, Thailand: The Natl. Innovation Agency (NIA). Nakauma M, Funami T, Noda S, Ishihara S, Al-Assaf S, Nishinari K, Phillips GO. 28. Comparison of sugar beet pectin, soybean soluble polysaccharide, and gum arabic as food emulsifiers. 1. Effect of concentration, ph, and salts on the emulsifying properties. Food Hydrocolloids 22(7): Orthoefer FT, Eastman J. 24. Rice bran oil. In: Champagne, ET, editor. Rice: chemistry and technology. 2nd ed. Minnesota: American Assn. of Cereal Chemists, Inc. p Padala SR, Williams PA, Phillips GO. 29. Adsorption of gum arabic, egg white protein, and their mixtures at the oil-water interface in limonene oil-in-water emulsions. J Agric Food Chem 57(11): Ray AK, Bird PB, Iacobucci, Clark Jr BC Functionality of gum arabic. Fractionation, characterization and evaluation of gum fractions in citrus oil emulsions and model beverages. Food Hydrocolloids 9(2): Sen CK, Khanna S, Roy S. 27. Tocotrienols in health and disease: the other half of the natural vitamin E family. Mol Asp Med 28(5 6): Tan C. 24. Beverage emulsions. In Friberg SE, Larsson K, Sjoblom J, editors. Food emulsions. 4th ed. New York: Marcel Decker. p Tan CT Beverage emulsions. In: Friberg SE, editor. Food emulsions. 3rd ed. New York: Marcel Dekker. p Tcholakova S, Denkov ND, Danner T. 24. Role of surfactant type and concentration for the mean drop size during emulsification in turbulent flow. Langmuir 2(18): Tcholakova S, Denkov ND, Sidzhakova D, Ivanov IB, Campbell B. 23. Interrelation between drop size and protein adsorption at various emulsification conditions. Langmuir 19(14): Trubiano PC The role of specialty food starches in flavor encapsulation. Flavor Technol 61: Van Hoed V, Depaemelaere G, Ayala J, Santiwattana P, Verhe R, De Greyt W. 26. Influence of chemical refining on the major and minor components of rice brain oil. J Am Oil Chem Soc 83(4): Vieno P, David GL, Tatu AM, Jari T, Anna-Maija L. 2. Plant sterols: biosynthesis, biological function and their importance to human nutrition. J Sci Food Agric 8(7): Walstra P Principles of emulsion formation. Chem Eng Sci 48: Walstra P. 23. Physical chemistry of foods. New York, N.Y..: Marcel Decker. Wilde PJ. 2. Interfaces: their role in foam and emulsion behaviour. Curr Opin Colloid Interface Sci 5(3 4): E172 Journal of Food Science Vol. 76, Nr. 1, 211