- Email: [email protected]
The effect of temperature on food emulsifiers at fluid–fluid interfaces
Colloids and Surfaces B: Biointerfaces 21 (2001) 87 – 99 www.elsevier.nl/locate/colsurfb
The effect of temperature on food emulsifiers at fluid–fluid interfaces Juan M. Rodrı´guez Patino *, M. Rosario Rodrı´guez Nin˜o, Cecilio Carrera Sa´nchez, Jose´ M. Navarro Garcı´a, Germa´n Rodrı´guez Rodrı´guez Mateo, Marta Cejudo Ferna´ndez Departamento de Ingenierı´a Quı´mı´ca, Uni6ersidad de Se6illa, C/. Professor Garcı´a Gonza´lez, s/Nu´m, 41012 -Se6illa, Se6illa, Spain
Abstract Heat-induced interfacial aggregation of a whey protein isolate (WPI) with a high content of b-lactoglobulin (\92%), previously adsorbed at the oil–water interface, was studied by means of interfacial dynamic characteristics performed in an automatic drop tensiometer. Protein concentration in aqueous bulk phase ranging between 1 ×10 − 1 and 1×10 − 5 % wt/wt was studied as a variable. The experiments were carried out at temperatures ranging from 20–80°C with different thermal regimes. During the heating period, competition exists between the effect of temperature on the film fluidity and the increase in mechanical properties associated with the interfacial gelation process. Interfacial crystallisation of food polar lipids (monopalmitin, monoolein, and monolaurin) previously adsorbed at the oil–water interface, was studied by interfacial dynamic characteristics (interfacial tension and surface dilational properties). The temperature, ranging between 40 and 2°C, and the lipid concentration in aqueous oil phase, ranging between 1 ×10 − 2 and 1 ×10 − 4 % wt/wt, were studied as variables. Significant changes in interfacial dynamic characteristics associated with interfacial lipid crystallisation were observed as a function of lipid concentration in the bulk phase. Interfacial crystallisation of food polar lipids (monopalmitin, monoolein, and monolaurin) at the air–water interface, was studied by y-A isotherms performed in a Langmuir trough coupled with Brewster angle microscopy (BAM). A condensation in monoglyceride monolayers towards lower molecular area was observed as the temperature decreased. This effect was attributed to lipid crystallisation at lower temperatures. BAM images corroborated the effect of temperature on the monolayer structure, as a function of the monoglyceride type. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Interfacial crystallisation; Emulsifier; Interfacial gelation; Proteins; Monoglycerides
1. Introduction
* Corresponding author. Tel.: + 34-95-4557183; fax: +3495-4557134. E-mail address: [email protected] (J.M. Rodrı´guez Patino).
The most important food processing operation that contributes to emulsifier (proteins and lipids) functionality involves heat treatment. Several reviews on protein heat-denaturation [1,2]
0927-7765/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 7 7 6 5 ( 0 1 ) 0 0 1 8 7 - 4
88
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
or lipid crystallisation [3,4] have been published. Gelation is the property of forming a structural network that maintains shape, has mechanical strength, viscoelasticity, and entrapped water with minimum synersis [5]. Gelation is an important functional attribute of functional proteins in many applications [5,6]. Protein solutions when heated above a critical temperature undergo conformational changes and on cooling may set to a viscous, soft, opaque coagula or clear viscoelastic gel depending on the type of protein, concentration, heating rate, and environmental conditions, especially pH and the presence of calcium [2,5,6]. The formation of a thermally induced gel or coagulum from proteins involves the following three sequential events [5]: DENATURATION AGGREGATION CROS-LINKING. Protein aggregation involves the formation of higher molecular weight complexes from the denatured protein, which can then cross-link by specific bonding at specific sites on the protein strands or by non-specific bonding occurring along the protein chains. Denaturation is therefore a prerequisite for the formation of protein aggregates or gel. On the other hand, food emulsions — such as margarine, ice cream and whippable emulsions — are often prepared by mixing the components at elevated temperature, and the final product is then obtained by cooling during some homogenisation process. That is, industrial emulsification procedures for food products usually involve stirring under cooling. Emulsifiers have been used in ice cream mix manufacture for many years [8]. Emulsifiers used in ice cream manufacture today are of two main types: the mono- and diglycerides and the sorbitan esters. Their mechanism of action is to lower the fat– water interfacial tension in the mix, resulting in protein displacement from the fat globule surface, which in turn reduces the stability of the fat globule to the partial coalescence that occurs during the whipping and freezing process. This fact lead to the formation of a fat structure in the frozen product that contributes greatly to texture and meltdown properties [9,10]. It is obvious that thermal lipid phase transitions at fluid – fluid interfaces can occur during such processing conditions. Thus, the effect of cooling on interfa-
cial characteristics of polar lipid at fluid –fluid interfaces is of practical importance. The aim of this study was to extend the investigation of gelation of WPI adsorbed films at the oil–water interface. In a preliminary study, we have analysed for the first time the viscoelastic characteristics of WPI heat induced gels at the oil– water interface as a function of temperature and heating conditions [11]. The interfacial gelation of globular proteins may possess great technological importance [6], even though it is not well known [12,13], because many protein-stabilized emulsions undergo some degree of thermal exposure during their processing, storage or usage. The stability and physicochemical properties of oil-in water emulsions stabilized by WPI — protein surface coverage, surface shear viscosity, and stability [14–17] — are particularly sensitive to the thermal history [14] as observed in a preliminary study [11] for WPI adsorbed films at the oil–water interface. In this work we have investigated the effect of protein concentration in solution, including the effect of temperature, on the viscoelastic characteristics of adsorbed protein films at the oil-water interface. In addition, in this work we present preliminary results on the effect of temperature changes on interfacial characteristics of monoglyceride films adsorbed or spread on fluid –fluid-interfaces. The behaviour of liquid interfaces near the freezing point or under cooling treatment has not been investigated in detail. Although the subject is fundamental and of considerable intrinsic importance, both from a basic and applied point of view, further systematic research is necessary. As is well known the surface tension of all liquids decreases with a rise in temperature. On the other hand, it was also observed that the surface or interfacial tension of a supercooled liquid passes continuously through the freezing point. From these studies it was concluded that a transition in the monolayer structure was produced a few degrees above the crystallisation of the liquid phase [18]. That is, fluids crystallise in the bulk phase in a different way than in the interface, with the crystallisation being initiated at the interface. In recent studies, the effect of additives — including proteins [18], monoglyceride [19] and diglycerol
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
esters [20] — to the aqueous or oil phase on the interfacial tension versus temperature curves were reported.
2. Experimental
2.1. Chemicals Whey protein isolate (WPI), a native whey protein with a very high content of b-lactoglobulin [21] obtained by fractionation, was supplied by Danisco Ingredients (Denmark). The sample was stored below 0°C and all is work was done without further purification. Samples for interfacial characteristics of WPI films were prepared using Milli-Q ultrapure water and were buffered at pH 5.0. Trisun oil (fatty acid composition, C16: 4%, C18: 4%, C18:1: 80%, C18:2: 9%, C18:3: traces, C20: 0.5%, and C22: 1%) supplied by Danisco Ingredients, was Florisil 60–100 mesh (Aldrich)treated to remove any surface-active impurities. Synthetic 1-monohexadecanoyl-rac-glycerol (monopalmitin, DIMODAN® PA 90), 1-mono(cis-9-ctadecanoyl)glycerol (monoolein, RYLO™ MG 19), and 1-monododecanoyl-rac-glycerol (monolaurin, DIMODAN® ML 90) were supplied by Danisco Ingredients with over 95– 98% of purity. To form the surface film, monoglyceride was spread in the form of a solution, using hexane:ethanol (9:1, v:v) as a spreading solvent. Analytical grade hexane (Merck, 99%) and ethanol (Merck, \ 99.8%) were used without further purification. The water used as subphase was purified by means of a Millipore filtration device (Milli-Q). To adjust subphase pH, buffer solutions were used. Acetic acid/sodium acetate aqueous solution (CH3COOH/CH3COONa) was used to achieve pH 5, and a commercial buffer solution called trizma ((CH2OH)3CNH2/ (CH2OH)3CNH3Cl) for pH 7. All these products were supplied by Sigma (\ 99.5%). Ionic strength was 0.05 M in all the experiments. The absence of active surface contaminants in the aqueous buffered solutions was checked by interfacial tension measurements before sample preparation.
89
2.2. Methods 2.2.1. Automatic drop tensiometer For interfacial tension and surface dilational property measurements of adsorbed protein films at the oil–water interface an automatic drop tensiometer as described elsewhere was used [21,22]. The average standard accuracy of the interfacial tension is roughly 0.1 mN/m. However, the reproducibility of the results (for at least two measurements) range between 0.5 and 1.5%, with the minimum reproducibility corresponding to higher temperatures. The surface viscoelastic parameters — such as surface dilational modulus, E, and its elastic, Ed, and viscous, Ev, components were measured as a function of time, t, amplitude, A, and angular frequency, . The method involved a periodic automatically-controlled, sinusoidal interfacial compression and expansion performed by decreasing and increasing the drop volume, at the desired amplitude (DA). The surface dilational modulus derived from the change in interfacial tension (dilational stress), |, (Eq. (1)), resulting from a small change in surface area (dilational stress), A, (Eq. (2)), may be described by the Eq. (3) [23]. |= |o − | sin( t + l),
(1)
A= Ao sin( t),
(2)
E=
d| dy =− , dA/A d ln A
(3)
where |o and Ao are the strain and stress amplitudes, respectively, l is the phase angle between stress and strain, y= | o − | is the interfacial pressure, and | o is the subphase interfacial tension. The dilational modulus is a complex quantity and is composed of real and imaginary parts (Eq. (4)). The real part of the dilational modulus or storage component is the dilational elasticity, Ed= E cos l. The imaginary part of the dilational modulus or loss component is the surface dilational viscosity Ev= E sin l. The ratio (|o/ Ao) is the absolute modulus) E , a measure of the total unit material dilational resistance to deformation (elastic+ viscous). For a perfectly elastic material the stress and strain are in phase (l=0)
90
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
and the imaginary term is zero. In the case of a perfectly viscous material l = 90° and the real part is zero. The loss angle tangent, tan l, can be defined by Eq. (5). If the film is purely elastic, the loss angle tangent is zero. E =(|o/Ao)(cos l +i sin l) =Ed + i Ev,
(4)
tan l =Ed/Ev
(5)
For the analysis of WPI gelation at the oil–water interface, the experiments were carried out at temperatures ranging from 20 to 80°C. The temperature of the system was maintained constant within 0.1°C at t B 40°C and within 90.3°C at T \40°C by circulating water from a thermostat. The pH and the ionic strength were maintained constant at 5.0 and 0.05 M, respectively, by using a citric–citrate buffer. WPI solutions ranging between 1×10 − 1 and 1 ×10 − 5 % wt/wt were studied as variables. The protein solution was placed in the syringe and the oil in the cuvette. Thus, the heat fluxed from the oil to the interface where a previously adsorbed WPI film gelled as a function of time. A drop of protein solution was delivered into the oil phase and allowed to stand for 120 min at 20°C to allow protein adsorption to take place at the oil– water interface. Afterwards, the temperature was changed to 40, 60 and 80°C, maintaining at each level the gelation temperature constant (40, 60, or 80°C). During this process the WPI gelation was monitored by observing the changes both in the interfacial tension and in the film viscoelastic characteristics with time. After 60 min of gelation at 80°C, the system was cooled back to 20°C and then the interfacial tension and viscoelastic characteristics of the protein film were monitored during 15 min. Finally, the drop was allowed to stand overnight and then (after 12 h) the interfacial tension and viscoelastic characteristics of the WPI film were monitored for 120 min at 20°C. For the analysis of the effect of temperature on monoglyceride adsorbed films at the oil –water interface, experiments were carried out at temperatures ranging from 40– 5°C. The pH and the ionic strength were maintained constant at 7.0 and 0.05 M, respectively, by using a trizma solution. A drop of monoglyceride oil solution was
delivered into the aqueous phase and allowed to stand for 60 min at 40°C to achieve the monoglyceride adsorption at the oil–water interface. Afterwards, the temperature was changed from 40 to 5°C, in intervals of 5°C, maintaining at each level the cooling-down temperature constant. During this process the heat-treated monoglyceride film was monitored by observing the changes both in the interfacial tension and in the film viscoelastic characteristics with time. Microscopic observation (shape and is opacity) of the drop during the cooling-down, coupled with image analysis gave complementary information about the effect of low temperatures on monoglyceride adsorbed films. The opacity of the drop was measured by means of the grey level of the drop in relation to that of the continuous phase. The shape of the drop after a sudden compression-expansion cycle provides qualitative information about the texture of the monoglyceride-cooled film at the oil–water interface. The materials in contact with the oil phase and protein or monoglyceride solution must be clean in order to prevent any contamination by surfaceactive compounds. The reproducibility of the results (for at least two measurements) was better than 5%.
2.2.2. Surface film balance The monolayer structural characteristics were studied by measuring y-A isotherms at air– water interface on a computer-controlled Langmuir film balance, as described elsewhere [24,25]. The monoglyceride solutions were spread on the subphase by means of a micrometric syringe at 20°C. Aliquots of 250 ml (6.1×1016 –7.63×1016 molecules) were spread in each experiment. The experiments were carried out at temperatures ranging from 20 to 5°C. The same precautions as in previous studies were taken to allow for the evaporation of the spreading solvent (a 15 min wait before beginning the isotherm recording), and for the choice of compression rate (0.062 nm2 molecule/min). Some experiments were repeated (at least twice). In these cases, the mean deviation was within 9 0.1 is mN/m for surface pressure and 9 0.005 nm2/molecule for area.
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
2.2.3. Brewster angle miroscope (BAM) A commercial Brewster angle microscope (BAM), BAM2, manufactured by NFT (Go¨ ttingen, Germany) was used to study the morphology of the monolayer. Further characteristics of the device and operational conditions were described elsewhere [26,27]. The BAM was positioned over the film balance on a specially designed frame structure that allows easy movement of the BAM along the length of the film balance. The location of BAM along the film balance allows any inhomogeneity in the overall film to be seen. The surface pressure measurements, area, and grey level as a function of time were carried out simultaneously by means of a device connected between the film balance and BAM. The frequency was fixed at one measurement every 5 s in order to reduce the noise in the grey level signal not related with the optical properties of the monolayer. These measurements were performed during continuous compression and expansion of the monolayer at constant rate with different shutter speeds ranging from 1/50 to 1/500 s. To measure the relative thickness of the film a previously camera calibration is necessary in order to determine the relationship between the grey level (GL) and the relative reflectivity (I), according to a procedure described previously [26,27]. In previous papers the relationship between grey level (GL) and incident angle () was determined, as well as the relative reflectivity dependence of the grey level as a function of the shutter speed [26,27].
91
the preferred method for characterizing viscoelastic foods [2,28]. Dynamic measurements allow coagulation and gelation to be monitored since the induced deformations are usually so small that their effect on structure is negligible. Fig. 1 gives the time-dependent interfacial tension and surface dilational properties of WPI adsorbed films on the oil – water interface as a consequence of thermal treatment, for concentrations of protein in the bulk phase of 1× 10 − 1% wt/wt, as an example. Overall, WPI adsorbed films behaved qualitatively in a similar manner after similar heat-treatment no matter what the protein concentration in the bulk phase, ranging between 1×10 − 1 and 1 ×10 − 5 % wt/wt. Briefly, E decreased during heating, passed through a minimum and then increased as the heating progressed and tended to a plateau value just at the
3. Results and discussion
3.1. Gelation of WPI adsorbed films at the oil– water interface Intermolecular interactions resulting from attractions between adjacent protein molecules with the formation of weak transient networks can exhibit non-newtonian behaviour and show viscoelastic properties upon protein gelation. Thus, rheological measurements provide a valuable tool for the characterization of gel networks. Dynamic rheological measurements in which sinusoidal oscillating stress or strain is applied to the sample is
Fig. 1. Time evolution of ( — ) temperature, (*) interfacial tension, and surface rheological properties — (O) surface dilational modulus E and ( ) phase angle — for WPI adsorbed film during the thermal treatment at 80°C. Protein concentration in solution: 1 × 10 − 1 %; pH= 5; I =0.05 M. Frequency of oscillation: 100 mHz. Amplitude of sinusoidal oscillation: 15%.
92
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
Table 1 Surface dilational modulus, E (mN/m), and interfacial tension, | (mN/m), of WPI adsorbed films at the oil–water interface at (a) 20°C after 120 min of adsorption time, at 20°C after heat treatment at 80, and at 20°C at 12 h after previous heat treatment at 80°C WPI concentration in solution (% wt/wt)
1×10−1 1×10−2 1×10−3 1×10−5
T= 20°C after 120 min of adsorption time
T= 20°C after heat-treatment at 80°C
T =20°C after 12 h of previous heat-treatment at 80°C
E (mN/m)
| (mN/m)
E (mN/m)
| (mN/m)
E (mN/m)
| (mN/m)
25.9 23.8 30.5 9.3
10.95 17.8 20.2 18.7
41.8 46.0 42.7 44.7
7.7 4.0 4.0 3.3
– 46.0 66.1 44.6
– 8.95 5.95 4.8
end of the heating period before the isothermal treatment. During the isothermal treatment (at 40, 60 and 80°C), E tended to increase to a plateau, especially during the first period at 40°C. The E value at the plateau decreased after thermal treatment at 60 and 80°C due to the effect of temperature on rheological parameters [11]. On the other hand, a sharp jump in the interfacial tension was observed during the heating period — from 20 to 40, 40 to 60°C, or 60 to 80°C due to the fact that some protein desorption takes place because the solution in the drop must be removed in order to maintain constant the drop volume. However, during the isothermal treatment at 40, 60, and 80°C the interfacial tension decreased. That is, the surface activity was increased with the heat treatment because the conformational changes of molecules during heating may include further unfolding, reorganization, and aggregation of the molecules to bring more hydrophobic segments from the interior of the molecule to the oil– water interface. Table 1 gives the evolution of the surface dilational modulus and interfacial tension as a consequence of the heat treatment (after cooling back to 20°C for a previously heat-treated WPI film, at 80°C) and after ageing overnight at room temperature in relation to an unheated WPI adsorbed film at the same temperature (20°C). The main differences were observed in the degree of changes in molecular conformation and aggregation during the heat treatment at 40, 60, and 80°C, as a function of protein concentration in solution. In fact, the film strength during the
heat treatment, as detected by the surface dilational modulus, was higher for the more concentrated protein solution (Fig. 1) and decreased when the protein concentration in solution was of 1 × 10 − 5 % (data not shown). However, the effect of protein concentration on the surface activity and viscoelastic characteristics of WPI adsorbed films at the end of the heat treatment is complex (Table 1). The interfacial tension and phase angle decreased to less than 9 mN/m (Table 1) and 1° (data not shown), respectively, after the heattreatment, no matter what the protein concentration in solution. These results suggest that the molecular changes that take place during thermal treatment produced high E values in WPI adsorbed films and increased the elastic component and surface activity. At every WPI concentration in solution, the heat-treatment produced irreversible changes in WPI adsorbed films because the interfacial characteristics (surface dilational modulus and interfacial tension) did not return to original values after cooling back towards the original temperature (20°C) (Table 1). The rate of thermal changes in WPI adsorbed films at the oil–water interface also increased with protein concentration in solution. Data in Table 2, especially the time dependence of E and phase angle can be used for a quantitative kinetic analysis of the gelation process at interface. If we assume that during heating of WPI at 40°C the protein is practically denatured, as a consequence of the first adsorption period of 120 min [22] and the pre-heating period up to 40°C, the time de-
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
pendence of E could be used to quantify the aggregation and cross-linking associated with the gelation process. If the interfacial gelation follows consecutive first-order steps, as for protein in solution [1], we propound the use of a first order kinetics equation (Eq. (6)) to monitor the time dependence of E during heating. ln [(E160 − Eq )/(E160 −E0) = −ki q,
(6)
where E160, Eu, and E0, are the surface dilational moduli at a heating time of 160 min at 40°C, at any time, 0, and at the beginning of holding at 40°C, respectively, and ki are the first-order kinetic constants associated with the gelation process. By processing the data in accordance with Eq. (6) it is possible to obtain two linear regressions (Fig. 2) which account for the aggregation and/or cross-linking steps occurring consecutively and/or concurrently. Unfortunately it is not possible to identify what the mechanism(s) is (are) for the rate changes which take place during the interfacial WPI gelation from surface dilational experiments alone. It seems certain that aggregation-crosslinking occurred fairly simultaneously with denaturation and one cannot easily separate these processes. The first-order rate constants are included in Table 2. It can be seen that the main changes as a function of protein concentration in solution are produced during the first
Fig. 2. Fits of experimental data during isothermal treatment at 40°C by a first order kinetics equation (Eq. (6)). Protein concentration in solution: 1 × 10 − 1 %; pH=5; I =0.05 M. Frequency of oscillation: 100 mHz. Amplitude of sinusoidal oscillation: 15%.
93
Table 2 Characteristic parameters for time-dependent surface dilational modulus during thermal treatment of WPI adsorbed films at the oil–water interface at 40°C WPI concentration in solution (% wt/wt)
k1×100 (1/min) k1×100 (1/min)
1×10−3 1×10−2 1×10−1
2.82 3.78 6.42
1.66 1.61 1.54
heating period. The first-order rate constant increased significantly from 0.0282 up to 0.0642/(1/ min) as the protein concentration in solution increased from 1× 10 − 3 up to 1 × 10 − 1 % wt/wt. That is, conformational changes and protein aggregation increase with protein–protein interactions, which increase with protein concentration in solution. Thus, it can be concluded that the gelling time, quantified by the first-order constants for E-time dependence, is a useful parameter for detecting transition points in heat-treated protein films at the oil–water interface, such as for gelation in solution [29–31]. These transitions are also accompanied by changes in the viscoelastic characteristics of the film — it should be remembered that the same E-time dependence according to Eq. (6) exists for its elastic component and the phase angle — typical behaviour for protein gelation in solution [4]. As for protein gelation in solution [31], the transition in viscoelastic characteristics of thermal treated WPI films is concentration dependent (Table 2). For WPI solutions it was observed [32,33] that the elastic modulus mainly indicates the total amount of protein in the gel. Thus, differences observed in interfacial WPI gelation here may be due to a higher amount of aggregated protein and stronger interparticle forces with increasing temperature or protein concentration. An important characteristic of WPI interfacial gelation, both from a theoretical and practical point of view, is the low protein concentration in solution necessary for interfacial gelation, in comparison with that necessary for WPI gelation in solution. In this study we have observed the existence of significant changes in interfacial dynamic properties associated with WPI gelation in ad-
94
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
sorbed films at the oil – water interface at protein concentration in is solution as low as 1× 10 − 5 % wt/wt, even at temperatures of 40°C. These values are far below those between 1 and 2.5% required for gelation in solution under similar conditions of pH [2,7,28,34]. The reason for this characteristic behaviour for interfacial gelation must be associated with the fact the interfacial protein concentration is far above its concentration in solution. As observed in the experimental data (Fig. 1 and Table 1) the interfacial tension decreases steadily with heat-treatment, which means that the interfacial activity of WPI films increases in the same way. At each protein concentration in solution, the minimum interfacial tension coincided with the maximum E value. This phenomenon was observed after interfacial gelation at 80°C with cooling back to 20°C for an aged film (Table 1). The trend observed in interfacial characteristics (interfacial tension and viscoelastic characteristics) of WPI adsorbed films with the heat-treatment is similar to the influence of temperature on the stability of WPI-stabilized oil-in-water emulsions [14,17]. In particular it has been shown that there is a maximum in droplet aggregation at about 75°C, with a decrease at higher temperatures. Hydrophobic interactions have been suggested to be important in determining the stability of heated emulsions [17,35,36]. Above a certain temperature whey proteins unfold and expose nonpolar amino acids which have been located in their hydrophobic interior, thus giving the surface of emulsion droplets some hydrophobic properties. When protein molecules unfold they expose reactive amino-acid residues, which leads to enhanced protein– protein interactions via hydrophobic interactions [17,37– 39]. These interactions may occur between molecules adsorbed to the same droplet (intradroplet) or between those adsorbed to different droplets (interdroplet). Intradroplet interactions lead to an increase in viscoelasticity of the surface layer [37], as was observed in this work, whereas interdroplet interactions lead to an increased tendency to flocculate [14,38]. It has been observed [14] that the emulsions tend to become more fluid with increasing temperature from 30 to 70°C. Above 70°C the
surface denaturation of the adsorbed proteins would lead to enhanced interdroplet attraction because of the increase in surface hydrophobicity of emulsion droplets. The increase in strength of droplet interactions would lead to the observed increase in gel strength and decrease in phase angle (the emulsions became less fluid), as was observed in viscoelastic characteristics of WPI adsorbed on a single drop (Table 1).
3.2. Interfacial crystallisation of monoglycerides at the air–water interface Interfacial crystallisation of food polar lipids (monopalmitin, monoolein, and monolaurin) at the air–water interface, was studied by y-A isotherms performed in a Langmuir trough coupled with Brewster angle microscopy (BAM).
Fig. 3. y-A isotherm for (A) monopalmitin and (B) monoolein monolayers spread on the air – water interface. Temperature (°): (— —) 20, (…) 10, and ( — )5; pH =5.l= 0.05 M.
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
Fig. 3 shows the y-A isotherm during the monopalmitin and monoolein monolayer compression, at pH=5 and I =0.05 M. From the y-A isotherm (Fig. 3A), different structures can be deduced for the monopalmitin monolayer as a function of surface pressure and temperature. At 20°C, the liquid expanded (LE) structure (at yB 5 mN/m), a degenerate first order phase transition between liquid-expanded (LE) and liquid-condensed (LC) structures (at 5B yB20 mN/m), the liquid-condensed structure (at 20ByB 43 N/m), the solid (S) structure (at y\ 44 mN/m), and, finally, the collapse phase at a surface pressure of about 53 mN/m were observed. However, at lower temperatures: (i) only the liquid condensed and solid structures were deduced; and (ii) a condensation in the monolayer structure was deduced from the translation of the y-A isotherms towards the y-axis as the temperature decreased. This effect can be attributed to lipid crystallisation at lower temperatures. Monoolein monolayers behave quite differently (Fig. 3B). Monoolein monolayers present only the liquid expanded structure and the collapse phase at the equilibrium surface pressure (ye $45.7 mN/ m). In addition, the monoolein structural characteristics did not depend on the temperature during the cooling-down treatment. BAM images corroborated the effect of temperature on the monolayer structure, as a function of the monoglyceride type. Monopalmitin monolayer at 20°C and at 43 mN/m (Fig. 4B) shows typical circular LC domains closely packed. At 5°C and at 42.5 mN/m (Fig. 4A), the contrast vanishes and the presence of solid microdomains can be observed at the interface. In fact, the crystallisation of monopalmitin with a solid structure was clearly distinguished by the dynamic formation/destruction of microdomains as the temperature decreased to 5°C. BAM images (Fig. 4C– D) corroborated that only the homogeneous LE phase is present during the cooling-down treatment of monoolein monolayers. The uniform intensity and the independence of this intensity on the analyser angle [26] confirmed the absence of crystalline domains in monoolein monolayer, at any surface pressure or temperature. Clearly, the effect of low tempera-
95
tures on monopalmitin and monoolein monolayers is quite different. Monolaurin spread monolayers behave in a different way to monopalmitin and monoolein monolayers during the cooling-drown treatment. BAM images (Fig. 4E– F) corroborated the existence of condensed domains at any temperature. But these domains do not adopt circular crystalline-like shapes typical of LC structures as for monopalmitin. However, the number of microdomains grew as the temperature decreased, but with their size remained the same. A high surface mobility was observed due to the formation/destruction of monolaurin microdomains as the temperature decreased to 5°C.
3.3. Interfacial crystallisation of monoglycerides at the oil–water interface Interfacial crystallisation of food polar lipids (monopalmitin, monoolein, and monolaurin) previously adsorbed at the oil–water interface, was studied by interfacial dynamic characteristics (interfacial tension and surface dilational properties) coupled with microscopic observation and image analysis of the drop after heat-treatment. The interfacial tension (|), surface dilational properties (E and ) and the opacity of the oil drop containing varying concentrations of monoolein (as an example), in relation to that of the aqueous continuous phase (I/Io), is shown in Fig. 5. It can be seen how on cooling: (i) the interfacial tension is reduced, especially at the higher monoolein concentration in the oil drop (1× 10 − 2 %, wt/wt); (ii) the surface dilational modulus increased, especially at the lower monoolein concentrations in the oil drop; (iii) the viscoelastic characteristics of the interface increased — with an increase in the loss angle as the cooling-down treatment progressed —, especially at the lower concentrations of monoolein in the drop, and, finally; (iv) significant changes in drop image (relative opacity) associated with interfacial lipid crystallisation were observed, with higher drop opacity at lower temperatures, irrespective of the monoolein concentration in the oil drop. In addition, a deformation is produced in the drop surface during the compression–expan-
96
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
Fig. 4. Visualisation by Brewster angle microscopy of monoglyceride monolayers spread on the air – water interface at pH =7 and at I = 0.005 M. (A) monopalmitin at 5°C and y= 42.5 mN/m; (B) monopalmitin at 20°C and y= 43 mN/m; (C) monoolein at 5°C and y= 41 mN/m; (D) monoolein at 20°C and y= 44 mN/m; (E) monolaurin at 5°C and y= 29 mN/m; and (F) monolaurin at 20°C and y= 32 mN/m.
sion cycle giving a rigid film as the drop is cooled at 5°C. That is, the drop loses the original laplacian shape during a compression/expansion cycle at the working frequency (100 mHz). Thus, at the lowest temperature (5°C) the surface dilational data of the monoolein adsorbed film are only apparent data. The data discussed in the previous section are typical of monoglyceride adsorbed films at the oil– water interface. The interfacial tension, surface dilational characteristics, and opacity of the oil– water interface containing varying concentra-
tions of monolaurin in the oil phase as a function of decreasing temperature, from 40 to 5°C, is shown in Fig. 6. Briefly, the interfacial tension and the opacity decreased and the surface dilational modulus and viscoelasticity increased during the cooling-down treatment no matter what the monolaurin concentration in the oil drop. Monopalmitin adsorbed films (data not shown) behave in a similar way to monoolein and monolaurin, with the only exception that the interfacial tension of the oil– water interface containing high concentrations of monopalmitin in the oil phase is
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
97
close to zero. Thus the drop burst during the compression/expansion cycle, giving no reproducible results. All these phenomena can be associated with interfacial monoglyceride crystallisation, as a function of lipid concentration in the bulk phase. The interfacial crystallisation of food emulsifiers seems to be a rather common phenomenon. In
Fig. 6. Surface dilational modulus (E, mN/m) interfacial tension (|, mN/m), loss angle (, degrees), and relative light intensity (I/Io) for monolaurin adsorbed films at the oil –water interface. pH =7, I =0.05 M. Amplitude of sinusoidal oscillation: 15%. Frequency: 100 mHz. Monolaurin concentration in the oil-bulk phase (%, wt/wt): ( ) 1 ×10 − 2, (O) 1 ×10 − 3, and (D) 1 × 10 − 4.
Fig. 5. Surface dilational modulus (E, mN/m), interfacial tension (|, mN/m), loss angle (, degrees), and relative light intensity (I/Io) for monoolein adsorbed films at the oil – water interface. pH = 7, I= 0.05 M. Amplitude of sinusoidal oscillation: 15%. Frequency: 100 mHz, Monoolein concentration in the oil-bulk phase (%, wt/wt): ( ) 1× 10 − 2, (O) 1 × 10 − 3, and (D) 1 × 10 − 4.
fact, the reduction of interfacial tension on cooling of adsorbed monoglycerides [19], diglycerol esters [20], and their mixtures with proteins [18– 20] was explained by the formation of surface-active crystals at the interface. According to Krog et al. [19], crystallisation at the interface (including the adsorption of the monoglyceride from the bulk of the oil phase) can be expected to be faster
98
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
than nucleation and crystal growth in the oil — this hypothesis can be supported by the reduced drop opacity at the lower temperatures in relation to the opacity of the oil bulk phase (Fig. 5D and Fig. 6D). As the emulsifier crystals will expose the methyl end group towards the oil phase and the polar head to towards water, it is natural to call them surface active crystals. That is, emulsifier crystallisation starts at the interface [40] because of the accumulation of molecules in this region, and the more condensed monolayer structure at lower temperatures (see Fig. 3), as occurs in the crystal. These results have been found to correlate well with the polymorphic behaviour of emulsifier crystals in the emulsifier-oil bulk phase [20,41,42]. The interfacial crystallisation explains the mechanisms of numerous food emulsion phenomena, such as interfacial desorption of proteins by monoglycerides in the preparation of ice-cream [19] and the rheology and particle size of food emulsions [20].
Acknowledgements This research was supported by the European Community through Grant FAIR-CT96-1216, by CICYT through Grant AL197-1274-CE, and by DGICYT through Grant PB97-0734.
References [1] J.I. Boye, C.-Y. Ma, V.R. Harwalkar, in: S. Damodaran, A. Paraf (Eds.), Food Proteins and their Applications, Marcel Dekker, NY, 1977 Chapter 2. [2] G.R. Ziegler, E.A. Foegeding, Adv. Food Nutr. Res. 34 (1990) 203. [3] F.G. Gunstone, F.B. Padley (Eds.), Lipid Technologies and Applications, Marcel Dekker, NY, 1997. [4] K. Larsson, Lipids-Molecular Organization, Physical Functions and Technical Applications, The Oily Press, Dunde (Scotland), 1994. [5] J.E. Kinsella, D.M. Whitehead, Adv. Food Nutr. Res. 33 (1989) 343. [6] E. Dickinson, An Introduction to Food Colloids, Oxford University Press, Oxford, 1992. [7] P. Cayot, D. Lorient, in: S. Damodaran, A. Paraf (Eds.), Food Proteins and their Applications, Marcel Dekker, New York, 1997 Chapter 8. [8] W.H. Knightly, Ice Cream Trade J. 55 (1959) 24.
[9] H.D. Goff, W.K Jordan, J. Dairy Sci. 72 (1989) 18. [10] H.D. Goff, in: F.D. Gunstone, F.B. Padley (Eds.), Lipid Technologies and Applications, Marcel Dekker, New York, 1997 Chapter 12. [11] J.M. Rodrı´guez Patino, M.R. Rodrı´guez Nino, C. Carrera Sanchez, J. Agric. Food Chem. 47 (1999) 3640. [12] A. Ball, R.A.L. Jones, Langmuir 11 (1995) 3542. [13] R.J. Green, I. Hopkinson, R.A.L. Jones, in: E. Dickinson, J.M. Rodriguez Patino (Eds.), Food Emulsions and Foams: Interfaces, Interactions and Stability, The Royal Society of Chemistry, Cambridge, 1999, p. 285. [14] K. Demetriades, J.N. Coupland, D.J. McClements, J. Food Sci. 62 (1997) 462. [15] E. Dickinson, S.-T. Hong, J. Agric. Food Chem. 42 (1994) 1602. [16] E. Dickinson, D.J. McClements, Advances in Food Colloids, Blackie Academic & Professional, London, 1995. [17] F.J. Monahan, D.J. McClements, J.B. German, J. Food Sci. 61 (1996) 504. [18] K.S. Birdi, Handbook of Surface and Colloid Chemistry, CRC Press, New York, 1997. [19] N. Krog, K. Larsson, Fat Sci. Technol. 94 (1992) 55. [20] J. Holstborg, B.V. Pedersen, N. Krog, S.K. Olesen, Colloids Surfaces B 12 (1999) 383. [21] J.M. Rodrı´guez Patino, M.R. Rodrı´guez Nin˜ o, C. Carrera Sanchez, J. Agric. Food Chem. 47 (1999) 3640. [22] J.M. Rodrı´guez Patino, M.R. Rodrı´guez Nin˜ o, C. Carrera Sanchez, J. Agric. Food Chem. 47 (1999) 2241. [23] J. Lucassen, M. van Den Tempel, Chem. Eng. Sci. 27 (1972) 1283. [24] M.R. Rodrı´guez Nin˜ o, C. Carrera, J.M. Rodrı´guez Patino, Colloids Surfaces B 12 (1999) 161. [25] C. Carrera, M.R. Rodrı´guez Nin˜ o, J.M. Rodrı´guez Patino, Colloids Surfaces B 12 (1999) 175. [26] J.M. Rodrı´guez Patino, C. Carrera, M.R. Rodrı´guez Nin˜ o, Langmuir 15 (1999) 2484. [27] J.M. Rodrı´guez Patino, C. Carrera, M.R. Rodrı´guez Nin˜ o, Food Hydrocolloids 13 (1999) 401. [28] D. Oakenfull, J. Pearce, R.W. Burley, in: S. Damodaran, A. Paraf (Eds.), Food Proteins and their Applications, Marcel Dekker, New York, 1997 Chapter 4. [29] T. Beveridge, L. Jones, M.A. Tung, J. Agric. Food Chem. 32 (1984) 307. [30] J.E. Kinsella, D.J. Rector, L.G. Phillips, in: R.Y. Yada, R.L. Jackman, J.L. Smith (Eds.), Protein Structure — Function Relationships in Food, Blackie Academic and Professional, London, 1994. [31] R.M. Hillier, R.L.J. Lyster, G.C. Cheeseman, J. Sci. Food Agric. 31 (1980) 1152. [32] M. Paulsson, P.O. Hegg, H.B. Castberg, J. Food Sci. 51 (1986) 87. [33] M. Verheul, S.P.F. Roefs, K.G. Kruif, J. Agric. Food Chem. 46 (1998) 896. [34] M. Verheul, S.F. Roets, J. Mellema, K.G. Kruif, Langmuir 14 (1998) 2263. [35] J.A. Hunt, D.G. Dalgleish, J. Food Sci. 42 (1994) 2131.
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99 [36] J.A. Hunt, D.G. Dalgleish, J. Food Sci. 50 (1995) 1120. [37] E. Dickinson, Y. Matsumura, Int. J. Biol. Macromol. 13 (1991) 26. [38] D.J. McClements, F.J. Monohan, J.E. Kinsella, J. Food Sci. 58 (1993) 1036. [39] F.J. Monahan, D.J. McClements, J.E. Kinsella, J. Agric. Food Chem. 41 (1993) 1826.
.
99
[40] E. Dickinson, F.-J. Kruizenga, M.J.W. Povey, M. Van der Molen, Colloids Surfaces A 81 (1993) 273. [41] N. Krog, J.B. Lauridsen, in: S.E. Friberg (Ed.), Food Emulsions, Marcel Dekker, New York, 1976 Chapter 3. [42] N. Krog, in: S.E. Friberg, K. Larsson (Eds.), Food Emulsions, 3rd, Marcel Dekker, New York, 1997 Chapter 4.
The effect of temperature on food emulsifiers at fluid–fluid interfaces Juan M. Rodrı´guez Patino *, M. Rosario Rodrı´guez Nin˜o, Cecilio Carrera Sa´nchez, Jose´ M. Navarro Garcı´a, Germa´n Rodrı´guez Rodrı´guez Mateo, Marta Cejudo Ferna´ndez Departamento de Ingenierı´a Quı´mı´ca, Uni6ersidad de Se6illa, C/. Professor Garcı´a Gonza´lez, s/Nu´m, 41012 -Se6illa, Se6illa, Spain
Abstract Heat-induced interfacial aggregation of a whey protein isolate (WPI) with a high content of b-lactoglobulin (\92%), previously adsorbed at the oil–water interface, was studied by means of interfacial dynamic characteristics performed in an automatic drop tensiometer. Protein concentration in aqueous bulk phase ranging between 1 ×10 − 1 and 1×10 − 5 % wt/wt was studied as a variable. The experiments were carried out at temperatures ranging from 20–80°C with different thermal regimes. During the heating period, competition exists between the effect of temperature on the film fluidity and the increase in mechanical properties associated with the interfacial gelation process. Interfacial crystallisation of food polar lipids (monopalmitin, monoolein, and monolaurin) previously adsorbed at the oil–water interface, was studied by interfacial dynamic characteristics (interfacial tension and surface dilational properties). The temperature, ranging between 40 and 2°C, and the lipid concentration in aqueous oil phase, ranging between 1 ×10 − 2 and 1 ×10 − 4 % wt/wt, were studied as variables. Significant changes in interfacial dynamic characteristics associated with interfacial lipid crystallisation were observed as a function of lipid concentration in the bulk phase. Interfacial crystallisation of food polar lipids (monopalmitin, monoolein, and monolaurin) at the air–water interface, was studied by y-A isotherms performed in a Langmuir trough coupled with Brewster angle microscopy (BAM). A condensation in monoglyceride monolayers towards lower molecular area was observed as the temperature decreased. This effect was attributed to lipid crystallisation at lower temperatures. BAM images corroborated the effect of temperature on the monolayer structure, as a function of the monoglyceride type. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Interfacial crystallisation; Emulsifier; Interfacial gelation; Proteins; Monoglycerides
1. Introduction
* Corresponding author. Tel.: + 34-95-4557183; fax: +3495-4557134. E-mail address: [email protected] (J.M. Rodrı´guez Patino).
The most important food processing operation that contributes to emulsifier (proteins and lipids) functionality involves heat treatment. Several reviews on protein heat-denaturation [1,2]
0927-7765/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 7 7 6 5 ( 0 1 ) 0 0 1 8 7 - 4
88
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
or lipid crystallisation [3,4] have been published. Gelation is the property of forming a structural network that maintains shape, has mechanical strength, viscoelasticity, and entrapped water with minimum synersis [5]. Gelation is an important functional attribute of functional proteins in many applications [5,6]. Protein solutions when heated above a critical temperature undergo conformational changes and on cooling may set to a viscous, soft, opaque coagula or clear viscoelastic gel depending on the type of protein, concentration, heating rate, and environmental conditions, especially pH and the presence of calcium [2,5,6]. The formation of a thermally induced gel or coagulum from proteins involves the following three sequential events [5]: DENATURATION AGGREGATION CROS-LINKING. Protein aggregation involves the formation of higher molecular weight complexes from the denatured protein, which can then cross-link by specific bonding at specific sites on the protein strands or by non-specific bonding occurring along the protein chains. Denaturation is therefore a prerequisite for the formation of protein aggregates or gel. On the other hand, food emulsions — such as margarine, ice cream and whippable emulsions — are often prepared by mixing the components at elevated temperature, and the final product is then obtained by cooling during some homogenisation process. That is, industrial emulsification procedures for food products usually involve stirring under cooling. Emulsifiers have been used in ice cream mix manufacture for many years [8]. Emulsifiers used in ice cream manufacture today are of two main types: the mono- and diglycerides and the sorbitan esters. Their mechanism of action is to lower the fat– water interfacial tension in the mix, resulting in protein displacement from the fat globule surface, which in turn reduces the stability of the fat globule to the partial coalescence that occurs during the whipping and freezing process. This fact lead to the formation of a fat structure in the frozen product that contributes greatly to texture and meltdown properties [9,10]. It is obvious that thermal lipid phase transitions at fluid – fluid interfaces can occur during such processing conditions. Thus, the effect of cooling on interfa-
cial characteristics of polar lipid at fluid –fluid interfaces is of practical importance. The aim of this study was to extend the investigation of gelation of WPI adsorbed films at the oil–water interface. In a preliminary study, we have analysed for the first time the viscoelastic characteristics of WPI heat induced gels at the oil– water interface as a function of temperature and heating conditions [11]. The interfacial gelation of globular proteins may possess great technological importance [6], even though it is not well known [12,13], because many protein-stabilized emulsions undergo some degree of thermal exposure during their processing, storage or usage. The stability and physicochemical properties of oil-in water emulsions stabilized by WPI — protein surface coverage, surface shear viscosity, and stability [14–17] — are particularly sensitive to the thermal history [14] as observed in a preliminary study [11] for WPI adsorbed films at the oil–water interface. In this work we have investigated the effect of protein concentration in solution, including the effect of temperature, on the viscoelastic characteristics of adsorbed protein films at the oil-water interface. In addition, in this work we present preliminary results on the effect of temperature changes on interfacial characteristics of monoglyceride films adsorbed or spread on fluid –fluid-interfaces. The behaviour of liquid interfaces near the freezing point or under cooling treatment has not been investigated in detail. Although the subject is fundamental and of considerable intrinsic importance, both from a basic and applied point of view, further systematic research is necessary. As is well known the surface tension of all liquids decreases with a rise in temperature. On the other hand, it was also observed that the surface or interfacial tension of a supercooled liquid passes continuously through the freezing point. From these studies it was concluded that a transition in the monolayer structure was produced a few degrees above the crystallisation of the liquid phase [18]. That is, fluids crystallise in the bulk phase in a different way than in the interface, with the crystallisation being initiated at the interface. In recent studies, the effect of additives — including proteins [18], monoglyceride [19] and diglycerol
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
esters [20] — to the aqueous or oil phase on the interfacial tension versus temperature curves were reported.
2. Experimental
2.1. Chemicals Whey protein isolate (WPI), a native whey protein with a very high content of b-lactoglobulin [21] obtained by fractionation, was supplied by Danisco Ingredients (Denmark). The sample was stored below 0°C and all is work was done without further purification. Samples for interfacial characteristics of WPI films were prepared using Milli-Q ultrapure water and were buffered at pH 5.0. Trisun oil (fatty acid composition, C16: 4%, C18: 4%, C18:1: 80%, C18:2: 9%, C18:3: traces, C20: 0.5%, and C22: 1%) supplied by Danisco Ingredients, was Florisil 60–100 mesh (Aldrich)treated to remove any surface-active impurities. Synthetic 1-monohexadecanoyl-rac-glycerol (monopalmitin, DIMODAN® PA 90), 1-mono(cis-9-ctadecanoyl)glycerol (monoolein, RYLO™ MG 19), and 1-monododecanoyl-rac-glycerol (monolaurin, DIMODAN® ML 90) were supplied by Danisco Ingredients with over 95– 98% of purity. To form the surface film, monoglyceride was spread in the form of a solution, using hexane:ethanol (9:1, v:v) as a spreading solvent. Analytical grade hexane (Merck, 99%) and ethanol (Merck, \ 99.8%) were used without further purification. The water used as subphase was purified by means of a Millipore filtration device (Milli-Q). To adjust subphase pH, buffer solutions were used. Acetic acid/sodium acetate aqueous solution (CH3COOH/CH3COONa) was used to achieve pH 5, and a commercial buffer solution called trizma ((CH2OH)3CNH2/ (CH2OH)3CNH3Cl) for pH 7. All these products were supplied by Sigma (\ 99.5%). Ionic strength was 0.05 M in all the experiments. The absence of active surface contaminants in the aqueous buffered solutions was checked by interfacial tension measurements before sample preparation.
89
2.2. Methods 2.2.1. Automatic drop tensiometer For interfacial tension and surface dilational property measurements of adsorbed protein films at the oil–water interface an automatic drop tensiometer as described elsewhere was used [21,22]. The average standard accuracy of the interfacial tension is roughly 0.1 mN/m. However, the reproducibility of the results (for at least two measurements) range between 0.5 and 1.5%, with the minimum reproducibility corresponding to higher temperatures. The surface viscoelastic parameters — such as surface dilational modulus, E, and its elastic, Ed, and viscous, Ev, components were measured as a function of time, t, amplitude, A, and angular frequency, . The method involved a periodic automatically-controlled, sinusoidal interfacial compression and expansion performed by decreasing and increasing the drop volume, at the desired amplitude (DA). The surface dilational modulus derived from the change in interfacial tension (dilational stress), |, (Eq. (1)), resulting from a small change in surface area (dilational stress), A, (Eq. (2)), may be described by the Eq. (3) [23]. |= |o − | sin( t + l),
(1)
A= Ao sin( t),
(2)
E=
d| dy =− , dA/A d ln A
(3)
where |o and Ao are the strain and stress amplitudes, respectively, l is the phase angle between stress and strain, y= | o − | is the interfacial pressure, and | o is the subphase interfacial tension. The dilational modulus is a complex quantity and is composed of real and imaginary parts (Eq. (4)). The real part of the dilational modulus or storage component is the dilational elasticity, Ed= E cos l. The imaginary part of the dilational modulus or loss component is the surface dilational viscosity Ev= E sin l. The ratio (|o/ Ao) is the absolute modulus) E , a measure of the total unit material dilational resistance to deformation (elastic+ viscous). For a perfectly elastic material the stress and strain are in phase (l=0)
90
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
and the imaginary term is zero. In the case of a perfectly viscous material l = 90° and the real part is zero. The loss angle tangent, tan l, can be defined by Eq. (5). If the film is purely elastic, the loss angle tangent is zero. E =(|o/Ao)(cos l +i sin l) =Ed + i Ev,
(4)
tan l =Ed/Ev
(5)
For the analysis of WPI gelation at the oil–water interface, the experiments were carried out at temperatures ranging from 20 to 80°C. The temperature of the system was maintained constant within 0.1°C at t B 40°C and within 90.3°C at T \40°C by circulating water from a thermostat. The pH and the ionic strength were maintained constant at 5.0 and 0.05 M, respectively, by using a citric–citrate buffer. WPI solutions ranging between 1×10 − 1 and 1 ×10 − 5 % wt/wt were studied as variables. The protein solution was placed in the syringe and the oil in the cuvette. Thus, the heat fluxed from the oil to the interface where a previously adsorbed WPI film gelled as a function of time. A drop of protein solution was delivered into the oil phase and allowed to stand for 120 min at 20°C to allow protein adsorption to take place at the oil– water interface. Afterwards, the temperature was changed to 40, 60 and 80°C, maintaining at each level the gelation temperature constant (40, 60, or 80°C). During this process the WPI gelation was monitored by observing the changes both in the interfacial tension and in the film viscoelastic characteristics with time. After 60 min of gelation at 80°C, the system was cooled back to 20°C and then the interfacial tension and viscoelastic characteristics of the protein film were monitored during 15 min. Finally, the drop was allowed to stand overnight and then (after 12 h) the interfacial tension and viscoelastic characteristics of the WPI film were monitored for 120 min at 20°C. For the analysis of the effect of temperature on monoglyceride adsorbed films at the oil –water interface, experiments were carried out at temperatures ranging from 40– 5°C. The pH and the ionic strength were maintained constant at 7.0 and 0.05 M, respectively, by using a trizma solution. A drop of monoglyceride oil solution was
delivered into the aqueous phase and allowed to stand for 60 min at 40°C to achieve the monoglyceride adsorption at the oil–water interface. Afterwards, the temperature was changed from 40 to 5°C, in intervals of 5°C, maintaining at each level the cooling-down temperature constant. During this process the heat-treated monoglyceride film was monitored by observing the changes both in the interfacial tension and in the film viscoelastic characteristics with time. Microscopic observation (shape and is opacity) of the drop during the cooling-down, coupled with image analysis gave complementary information about the effect of low temperatures on monoglyceride adsorbed films. The opacity of the drop was measured by means of the grey level of the drop in relation to that of the continuous phase. The shape of the drop after a sudden compression-expansion cycle provides qualitative information about the texture of the monoglyceride-cooled film at the oil–water interface. The materials in contact with the oil phase and protein or monoglyceride solution must be clean in order to prevent any contamination by surfaceactive compounds. The reproducibility of the results (for at least two measurements) was better than 5%.
2.2.2. Surface film balance The monolayer structural characteristics were studied by measuring y-A isotherms at air– water interface on a computer-controlled Langmuir film balance, as described elsewhere [24,25]. The monoglyceride solutions were spread on the subphase by means of a micrometric syringe at 20°C. Aliquots of 250 ml (6.1×1016 –7.63×1016 molecules) were spread in each experiment. The experiments were carried out at temperatures ranging from 20 to 5°C. The same precautions as in previous studies were taken to allow for the evaporation of the spreading solvent (a 15 min wait before beginning the isotherm recording), and for the choice of compression rate (0.062 nm2 molecule/min). Some experiments were repeated (at least twice). In these cases, the mean deviation was within 9 0.1 is mN/m for surface pressure and 9 0.005 nm2/molecule for area.
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
2.2.3. Brewster angle miroscope (BAM) A commercial Brewster angle microscope (BAM), BAM2, manufactured by NFT (Go¨ ttingen, Germany) was used to study the morphology of the monolayer. Further characteristics of the device and operational conditions were described elsewhere [26,27]. The BAM was positioned over the film balance on a specially designed frame structure that allows easy movement of the BAM along the length of the film balance. The location of BAM along the film balance allows any inhomogeneity in the overall film to be seen. The surface pressure measurements, area, and grey level as a function of time were carried out simultaneously by means of a device connected between the film balance and BAM. The frequency was fixed at one measurement every 5 s in order to reduce the noise in the grey level signal not related with the optical properties of the monolayer. These measurements were performed during continuous compression and expansion of the monolayer at constant rate with different shutter speeds ranging from 1/50 to 1/500 s. To measure the relative thickness of the film a previously camera calibration is necessary in order to determine the relationship between the grey level (GL) and the relative reflectivity (I), according to a procedure described previously [26,27]. In previous papers the relationship between grey level (GL) and incident angle () was determined, as well as the relative reflectivity dependence of the grey level as a function of the shutter speed [26,27].
91
the preferred method for characterizing viscoelastic foods [2,28]. Dynamic measurements allow coagulation and gelation to be monitored since the induced deformations are usually so small that their effect on structure is negligible. Fig. 1 gives the time-dependent interfacial tension and surface dilational properties of WPI adsorbed films on the oil – water interface as a consequence of thermal treatment, for concentrations of protein in the bulk phase of 1× 10 − 1% wt/wt, as an example. Overall, WPI adsorbed films behaved qualitatively in a similar manner after similar heat-treatment no matter what the protein concentration in the bulk phase, ranging between 1×10 − 1 and 1 ×10 − 5 % wt/wt. Briefly, E decreased during heating, passed through a minimum and then increased as the heating progressed and tended to a plateau value just at the
3. Results and discussion
3.1. Gelation of WPI adsorbed films at the oil– water interface Intermolecular interactions resulting from attractions between adjacent protein molecules with the formation of weak transient networks can exhibit non-newtonian behaviour and show viscoelastic properties upon protein gelation. Thus, rheological measurements provide a valuable tool for the characterization of gel networks. Dynamic rheological measurements in which sinusoidal oscillating stress or strain is applied to the sample is
Fig. 1. Time evolution of ( — ) temperature, (*) interfacial tension, and surface rheological properties — (O) surface dilational modulus E and ( ) phase angle — for WPI adsorbed film during the thermal treatment at 80°C. Protein concentration in solution: 1 × 10 − 1 %; pH= 5; I =0.05 M. Frequency of oscillation: 100 mHz. Amplitude of sinusoidal oscillation: 15%.
92
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
Table 1 Surface dilational modulus, E (mN/m), and interfacial tension, | (mN/m), of WPI adsorbed films at the oil–water interface at (a) 20°C after 120 min of adsorption time, at 20°C after heat treatment at 80, and at 20°C at 12 h after previous heat treatment at 80°C WPI concentration in solution (% wt/wt)
1×10−1 1×10−2 1×10−3 1×10−5
T= 20°C after 120 min of adsorption time
T= 20°C after heat-treatment at 80°C
T =20°C after 12 h of previous heat-treatment at 80°C
E (mN/m)
| (mN/m)
E (mN/m)
| (mN/m)
E (mN/m)
| (mN/m)
25.9 23.8 30.5 9.3
10.95 17.8 20.2 18.7
41.8 46.0 42.7 44.7
7.7 4.0 4.0 3.3
– 46.0 66.1 44.6
– 8.95 5.95 4.8
end of the heating period before the isothermal treatment. During the isothermal treatment (at 40, 60 and 80°C), E tended to increase to a plateau, especially during the first period at 40°C. The E value at the plateau decreased after thermal treatment at 60 and 80°C due to the effect of temperature on rheological parameters [11]. On the other hand, a sharp jump in the interfacial tension was observed during the heating period — from 20 to 40, 40 to 60°C, or 60 to 80°C due to the fact that some protein desorption takes place because the solution in the drop must be removed in order to maintain constant the drop volume. However, during the isothermal treatment at 40, 60, and 80°C the interfacial tension decreased. That is, the surface activity was increased with the heat treatment because the conformational changes of molecules during heating may include further unfolding, reorganization, and aggregation of the molecules to bring more hydrophobic segments from the interior of the molecule to the oil– water interface. Table 1 gives the evolution of the surface dilational modulus and interfacial tension as a consequence of the heat treatment (after cooling back to 20°C for a previously heat-treated WPI film, at 80°C) and after ageing overnight at room temperature in relation to an unheated WPI adsorbed film at the same temperature (20°C). The main differences were observed in the degree of changes in molecular conformation and aggregation during the heat treatment at 40, 60, and 80°C, as a function of protein concentration in solution. In fact, the film strength during the
heat treatment, as detected by the surface dilational modulus, was higher for the more concentrated protein solution (Fig. 1) and decreased when the protein concentration in solution was of 1 × 10 − 5 % (data not shown). However, the effect of protein concentration on the surface activity and viscoelastic characteristics of WPI adsorbed films at the end of the heat treatment is complex (Table 1). The interfacial tension and phase angle decreased to less than 9 mN/m (Table 1) and 1° (data not shown), respectively, after the heattreatment, no matter what the protein concentration in solution. These results suggest that the molecular changes that take place during thermal treatment produced high E values in WPI adsorbed films and increased the elastic component and surface activity. At every WPI concentration in solution, the heat-treatment produced irreversible changes in WPI adsorbed films because the interfacial characteristics (surface dilational modulus and interfacial tension) did not return to original values after cooling back towards the original temperature (20°C) (Table 1). The rate of thermal changes in WPI adsorbed films at the oil–water interface also increased with protein concentration in solution. Data in Table 2, especially the time dependence of E and phase angle can be used for a quantitative kinetic analysis of the gelation process at interface. If we assume that during heating of WPI at 40°C the protein is practically denatured, as a consequence of the first adsorption period of 120 min [22] and the pre-heating period up to 40°C, the time de-
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
pendence of E could be used to quantify the aggregation and cross-linking associated with the gelation process. If the interfacial gelation follows consecutive first-order steps, as for protein in solution [1], we propound the use of a first order kinetics equation (Eq. (6)) to monitor the time dependence of E during heating. ln [(E160 − Eq )/(E160 −E0) = −ki q,
(6)
where E160, Eu, and E0, are the surface dilational moduli at a heating time of 160 min at 40°C, at any time, 0, and at the beginning of holding at 40°C, respectively, and ki are the first-order kinetic constants associated with the gelation process. By processing the data in accordance with Eq. (6) it is possible to obtain two linear regressions (Fig. 2) which account for the aggregation and/or cross-linking steps occurring consecutively and/or concurrently. Unfortunately it is not possible to identify what the mechanism(s) is (are) for the rate changes which take place during the interfacial WPI gelation from surface dilational experiments alone. It seems certain that aggregation-crosslinking occurred fairly simultaneously with denaturation and one cannot easily separate these processes. The first-order rate constants are included in Table 2. It can be seen that the main changes as a function of protein concentration in solution are produced during the first
Fig. 2. Fits of experimental data during isothermal treatment at 40°C by a first order kinetics equation (Eq. (6)). Protein concentration in solution: 1 × 10 − 1 %; pH=5; I =0.05 M. Frequency of oscillation: 100 mHz. Amplitude of sinusoidal oscillation: 15%.
93
Table 2 Characteristic parameters for time-dependent surface dilational modulus during thermal treatment of WPI adsorbed films at the oil–water interface at 40°C WPI concentration in solution (% wt/wt)
k1×100 (1/min) k1×100 (1/min)
1×10−3 1×10−2 1×10−1
2.82 3.78 6.42
1.66 1.61 1.54
heating period. The first-order rate constant increased significantly from 0.0282 up to 0.0642/(1/ min) as the protein concentration in solution increased from 1× 10 − 3 up to 1 × 10 − 1 % wt/wt. That is, conformational changes and protein aggregation increase with protein–protein interactions, which increase with protein concentration in solution. Thus, it can be concluded that the gelling time, quantified by the first-order constants for E-time dependence, is a useful parameter for detecting transition points in heat-treated protein films at the oil–water interface, such as for gelation in solution [29–31]. These transitions are also accompanied by changes in the viscoelastic characteristics of the film — it should be remembered that the same E-time dependence according to Eq. (6) exists for its elastic component and the phase angle — typical behaviour for protein gelation in solution [4]. As for protein gelation in solution [31], the transition in viscoelastic characteristics of thermal treated WPI films is concentration dependent (Table 2). For WPI solutions it was observed [32,33] that the elastic modulus mainly indicates the total amount of protein in the gel. Thus, differences observed in interfacial WPI gelation here may be due to a higher amount of aggregated protein and stronger interparticle forces with increasing temperature or protein concentration. An important characteristic of WPI interfacial gelation, both from a theoretical and practical point of view, is the low protein concentration in solution necessary for interfacial gelation, in comparison with that necessary for WPI gelation in solution. In this study we have observed the existence of significant changes in interfacial dynamic properties associated with WPI gelation in ad-
94
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
sorbed films at the oil – water interface at protein concentration in is solution as low as 1× 10 − 5 % wt/wt, even at temperatures of 40°C. These values are far below those between 1 and 2.5% required for gelation in solution under similar conditions of pH [2,7,28,34]. The reason for this characteristic behaviour for interfacial gelation must be associated with the fact the interfacial protein concentration is far above its concentration in solution. As observed in the experimental data (Fig. 1 and Table 1) the interfacial tension decreases steadily with heat-treatment, which means that the interfacial activity of WPI films increases in the same way. At each protein concentration in solution, the minimum interfacial tension coincided with the maximum E value. This phenomenon was observed after interfacial gelation at 80°C with cooling back to 20°C for an aged film (Table 1). The trend observed in interfacial characteristics (interfacial tension and viscoelastic characteristics) of WPI adsorbed films with the heat-treatment is similar to the influence of temperature on the stability of WPI-stabilized oil-in-water emulsions [14,17]. In particular it has been shown that there is a maximum in droplet aggregation at about 75°C, with a decrease at higher temperatures. Hydrophobic interactions have been suggested to be important in determining the stability of heated emulsions [17,35,36]. Above a certain temperature whey proteins unfold and expose nonpolar amino acids which have been located in their hydrophobic interior, thus giving the surface of emulsion droplets some hydrophobic properties. When protein molecules unfold they expose reactive amino-acid residues, which leads to enhanced protein– protein interactions via hydrophobic interactions [17,37– 39]. These interactions may occur between molecules adsorbed to the same droplet (intradroplet) or between those adsorbed to different droplets (interdroplet). Intradroplet interactions lead to an increase in viscoelasticity of the surface layer [37], as was observed in this work, whereas interdroplet interactions lead to an increased tendency to flocculate [14,38]. It has been observed [14] that the emulsions tend to become more fluid with increasing temperature from 30 to 70°C. Above 70°C the
surface denaturation of the adsorbed proteins would lead to enhanced interdroplet attraction because of the increase in surface hydrophobicity of emulsion droplets. The increase in strength of droplet interactions would lead to the observed increase in gel strength and decrease in phase angle (the emulsions became less fluid), as was observed in viscoelastic characteristics of WPI adsorbed on a single drop (Table 1).
3.2. Interfacial crystallisation of monoglycerides at the air–water interface Interfacial crystallisation of food polar lipids (monopalmitin, monoolein, and monolaurin) at the air–water interface, was studied by y-A isotherms performed in a Langmuir trough coupled with Brewster angle microscopy (BAM).
Fig. 3. y-A isotherm for (A) monopalmitin and (B) monoolein monolayers spread on the air – water interface. Temperature (°): (— —) 20, (…) 10, and ( — )5; pH =5.l= 0.05 M.
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
Fig. 3 shows the y-A isotherm during the monopalmitin and monoolein monolayer compression, at pH=5 and I =0.05 M. From the y-A isotherm (Fig. 3A), different structures can be deduced for the monopalmitin monolayer as a function of surface pressure and temperature. At 20°C, the liquid expanded (LE) structure (at yB 5 mN/m), a degenerate first order phase transition between liquid-expanded (LE) and liquid-condensed (LC) structures (at 5B yB20 mN/m), the liquid-condensed structure (at 20ByB 43 N/m), the solid (S) structure (at y\ 44 mN/m), and, finally, the collapse phase at a surface pressure of about 53 mN/m were observed. However, at lower temperatures: (i) only the liquid condensed and solid structures were deduced; and (ii) a condensation in the monolayer structure was deduced from the translation of the y-A isotherms towards the y-axis as the temperature decreased. This effect can be attributed to lipid crystallisation at lower temperatures. Monoolein monolayers behave quite differently (Fig. 3B). Monoolein monolayers present only the liquid expanded structure and the collapse phase at the equilibrium surface pressure (ye $45.7 mN/ m). In addition, the monoolein structural characteristics did not depend on the temperature during the cooling-down treatment. BAM images corroborated the effect of temperature on the monolayer structure, as a function of the monoglyceride type. Monopalmitin monolayer at 20°C and at 43 mN/m (Fig. 4B) shows typical circular LC domains closely packed. At 5°C and at 42.5 mN/m (Fig. 4A), the contrast vanishes and the presence of solid microdomains can be observed at the interface. In fact, the crystallisation of monopalmitin with a solid structure was clearly distinguished by the dynamic formation/destruction of microdomains as the temperature decreased to 5°C. BAM images (Fig. 4C– D) corroborated that only the homogeneous LE phase is present during the cooling-down treatment of monoolein monolayers. The uniform intensity and the independence of this intensity on the analyser angle [26] confirmed the absence of crystalline domains in monoolein monolayer, at any surface pressure or temperature. Clearly, the effect of low tempera-
95
tures on monopalmitin and monoolein monolayers is quite different. Monolaurin spread monolayers behave in a different way to monopalmitin and monoolein monolayers during the cooling-drown treatment. BAM images (Fig. 4E– F) corroborated the existence of condensed domains at any temperature. But these domains do not adopt circular crystalline-like shapes typical of LC structures as for monopalmitin. However, the number of microdomains grew as the temperature decreased, but with their size remained the same. A high surface mobility was observed due to the formation/destruction of monolaurin microdomains as the temperature decreased to 5°C.
3.3. Interfacial crystallisation of monoglycerides at the oil–water interface Interfacial crystallisation of food polar lipids (monopalmitin, monoolein, and monolaurin) previously adsorbed at the oil–water interface, was studied by interfacial dynamic characteristics (interfacial tension and surface dilational properties) coupled with microscopic observation and image analysis of the drop after heat-treatment. The interfacial tension (|), surface dilational properties (E and ) and the opacity of the oil drop containing varying concentrations of monoolein (as an example), in relation to that of the aqueous continuous phase (I/Io), is shown in Fig. 5. It can be seen how on cooling: (i) the interfacial tension is reduced, especially at the higher monoolein concentration in the oil drop (1× 10 − 2 %, wt/wt); (ii) the surface dilational modulus increased, especially at the lower monoolein concentrations in the oil drop; (iii) the viscoelastic characteristics of the interface increased — with an increase in the loss angle as the cooling-down treatment progressed —, especially at the lower concentrations of monoolein in the drop, and, finally; (iv) significant changes in drop image (relative opacity) associated with interfacial lipid crystallisation were observed, with higher drop opacity at lower temperatures, irrespective of the monoolein concentration in the oil drop. In addition, a deformation is produced in the drop surface during the compression–expan-
96
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
Fig. 4. Visualisation by Brewster angle microscopy of monoglyceride monolayers spread on the air – water interface at pH =7 and at I = 0.005 M. (A) monopalmitin at 5°C and y= 42.5 mN/m; (B) monopalmitin at 20°C and y= 43 mN/m; (C) monoolein at 5°C and y= 41 mN/m; (D) monoolein at 20°C and y= 44 mN/m; (E) monolaurin at 5°C and y= 29 mN/m; and (F) monolaurin at 20°C and y= 32 mN/m.
sion cycle giving a rigid film as the drop is cooled at 5°C. That is, the drop loses the original laplacian shape during a compression/expansion cycle at the working frequency (100 mHz). Thus, at the lowest temperature (5°C) the surface dilational data of the monoolein adsorbed film are only apparent data. The data discussed in the previous section are typical of monoglyceride adsorbed films at the oil– water interface. The interfacial tension, surface dilational characteristics, and opacity of the oil– water interface containing varying concentra-
tions of monolaurin in the oil phase as a function of decreasing temperature, from 40 to 5°C, is shown in Fig. 6. Briefly, the interfacial tension and the opacity decreased and the surface dilational modulus and viscoelasticity increased during the cooling-down treatment no matter what the monolaurin concentration in the oil drop. Monopalmitin adsorbed films (data not shown) behave in a similar way to monoolein and monolaurin, with the only exception that the interfacial tension of the oil– water interface containing high concentrations of monopalmitin in the oil phase is
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
97
close to zero. Thus the drop burst during the compression/expansion cycle, giving no reproducible results. All these phenomena can be associated with interfacial monoglyceride crystallisation, as a function of lipid concentration in the bulk phase. The interfacial crystallisation of food emulsifiers seems to be a rather common phenomenon. In
Fig. 6. Surface dilational modulus (E, mN/m) interfacial tension (|, mN/m), loss angle (, degrees), and relative light intensity (I/Io) for monolaurin adsorbed films at the oil –water interface. pH =7, I =0.05 M. Amplitude of sinusoidal oscillation: 15%. Frequency: 100 mHz. Monolaurin concentration in the oil-bulk phase (%, wt/wt): ( ) 1 ×10 − 2, (O) 1 ×10 − 3, and (D) 1 × 10 − 4.
Fig. 5. Surface dilational modulus (E, mN/m), interfacial tension (|, mN/m), loss angle (, degrees), and relative light intensity (I/Io) for monoolein adsorbed films at the oil – water interface. pH = 7, I= 0.05 M. Amplitude of sinusoidal oscillation: 15%. Frequency: 100 mHz, Monoolein concentration in the oil-bulk phase (%, wt/wt): ( ) 1× 10 − 2, (O) 1 × 10 − 3, and (D) 1 × 10 − 4.
fact, the reduction of interfacial tension on cooling of adsorbed monoglycerides [19], diglycerol esters [20], and their mixtures with proteins [18– 20] was explained by the formation of surface-active crystals at the interface. According to Krog et al. [19], crystallisation at the interface (including the adsorption of the monoglyceride from the bulk of the oil phase) can be expected to be faster
98
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99
than nucleation and crystal growth in the oil — this hypothesis can be supported by the reduced drop opacity at the lower temperatures in relation to the opacity of the oil bulk phase (Fig. 5D and Fig. 6D). As the emulsifier crystals will expose the methyl end group towards the oil phase and the polar head to towards water, it is natural to call them surface active crystals. That is, emulsifier crystallisation starts at the interface [40] because of the accumulation of molecules in this region, and the more condensed monolayer structure at lower temperatures (see Fig. 3), as occurs in the crystal. These results have been found to correlate well with the polymorphic behaviour of emulsifier crystals in the emulsifier-oil bulk phase [20,41,42]. The interfacial crystallisation explains the mechanisms of numerous food emulsion phenomena, such as interfacial desorption of proteins by monoglycerides in the preparation of ice-cream [19] and the rheology and particle size of food emulsions [20].
Acknowledgements This research was supported by the European Community through Grant FAIR-CT96-1216, by CICYT through Grant AL197-1274-CE, and by DGICYT through Grant PB97-0734.
References [1] J.I. Boye, C.-Y. Ma, V.R. Harwalkar, in: S. Damodaran, A. Paraf (Eds.), Food Proteins and their Applications, Marcel Dekker, NY, 1977 Chapter 2. [2] G.R. Ziegler, E.A. Foegeding, Adv. Food Nutr. Res. 34 (1990) 203. [3] F.G. Gunstone, F.B. Padley (Eds.), Lipid Technologies and Applications, Marcel Dekker, NY, 1997. [4] K. Larsson, Lipids-Molecular Organization, Physical Functions and Technical Applications, The Oily Press, Dunde (Scotland), 1994. [5] J.E. Kinsella, D.M. Whitehead, Adv. Food Nutr. Res. 33 (1989) 343. [6] E. Dickinson, An Introduction to Food Colloids, Oxford University Press, Oxford, 1992. [7] P. Cayot, D. Lorient, in: S. Damodaran, A. Paraf (Eds.), Food Proteins and their Applications, Marcel Dekker, New York, 1997 Chapter 8. [8] W.H. Knightly, Ice Cream Trade J. 55 (1959) 24.
[9] H.D. Goff, W.K Jordan, J. Dairy Sci. 72 (1989) 18. [10] H.D. Goff, in: F.D. Gunstone, F.B. Padley (Eds.), Lipid Technologies and Applications, Marcel Dekker, New York, 1997 Chapter 12. [11] J.M. Rodrı´guez Patino, M.R. Rodrı´guez Nino, C. Carrera Sanchez, J. Agric. Food Chem. 47 (1999) 3640. [12] A. Ball, R.A.L. Jones, Langmuir 11 (1995) 3542. [13] R.J. Green, I. Hopkinson, R.A.L. Jones, in: E. Dickinson, J.M. Rodriguez Patino (Eds.), Food Emulsions and Foams: Interfaces, Interactions and Stability, The Royal Society of Chemistry, Cambridge, 1999, p. 285. [14] K. Demetriades, J.N. Coupland, D.J. McClements, J. Food Sci. 62 (1997) 462. [15] E. Dickinson, S.-T. Hong, J. Agric. Food Chem. 42 (1994) 1602. [16] E. Dickinson, D.J. McClements, Advances in Food Colloids, Blackie Academic & Professional, London, 1995. [17] F.J. Monahan, D.J. McClements, J.B. German, J. Food Sci. 61 (1996) 504. [18] K.S. Birdi, Handbook of Surface and Colloid Chemistry, CRC Press, New York, 1997. [19] N. Krog, K. Larsson, Fat Sci. Technol. 94 (1992) 55. [20] J. Holstborg, B.V. Pedersen, N. Krog, S.K. Olesen, Colloids Surfaces B 12 (1999) 383. [21] J.M. Rodrı´guez Patino, M.R. Rodrı´guez Nin˜ o, C. Carrera Sanchez, J. Agric. Food Chem. 47 (1999) 3640. [22] J.M. Rodrı´guez Patino, M.R. Rodrı´guez Nin˜ o, C. Carrera Sanchez, J. Agric. Food Chem. 47 (1999) 2241. [23] J. Lucassen, M. van Den Tempel, Chem. Eng. Sci. 27 (1972) 1283. [24] M.R. Rodrı´guez Nin˜ o, C. Carrera, J.M. Rodrı´guez Patino, Colloids Surfaces B 12 (1999) 161. [25] C. Carrera, M.R. Rodrı´guez Nin˜ o, J.M. Rodrı´guez Patino, Colloids Surfaces B 12 (1999) 175. [26] J.M. Rodrı´guez Patino, C. Carrera, M.R. Rodrı´guez Nin˜ o, Langmuir 15 (1999) 2484. [27] J.M. Rodrı´guez Patino, C. Carrera, M.R. Rodrı´guez Nin˜ o, Food Hydrocolloids 13 (1999) 401. [28] D. Oakenfull, J. Pearce, R.W. Burley, in: S. Damodaran, A. Paraf (Eds.), Food Proteins and their Applications, Marcel Dekker, New York, 1997 Chapter 4. [29] T. Beveridge, L. Jones, M.A. Tung, J. Agric. Food Chem. 32 (1984) 307. [30] J.E. Kinsella, D.J. Rector, L.G. Phillips, in: R.Y. Yada, R.L. Jackman, J.L. Smith (Eds.), Protein Structure — Function Relationships in Food, Blackie Academic and Professional, London, 1994. [31] R.M. Hillier, R.L.J. Lyster, G.C. Cheeseman, J. Sci. Food Agric. 31 (1980) 1152. [32] M. Paulsson, P.O. Hegg, H.B. Castberg, J. Food Sci. 51 (1986) 87. [33] M. Verheul, S.P.F. Roefs, K.G. Kruif, J. Agric. Food Chem. 46 (1998) 896. [34] M. Verheul, S.F. Roets, J. Mellema, K.G. Kruif, Langmuir 14 (1998) 2263. [35] J.A. Hunt, D.G. Dalgleish, J. Food Sci. 42 (1994) 2131.
J.M. Rodrı´guez Patino et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 87–99 [36] J.A. Hunt, D.G. Dalgleish, J. Food Sci. 50 (1995) 1120. [37] E. Dickinson, Y. Matsumura, Int. J. Biol. Macromol. 13 (1991) 26. [38] D.J. McClements, F.J. Monohan, J.E. Kinsella, J. Food Sci. 58 (1993) 1036. [39] F.J. Monahan, D.J. McClements, J.E. Kinsella, J. Agric. Food Chem. 41 (1993) 1826.
.
99
[40] E. Dickinson, F.-J. Kruizenga, M.J.W. Povey, M. Van der Molen, Colloids Surfaces A 81 (1993) 273. [41] N. Krog, J.B. Lauridsen, in: S.E. Friberg (Ed.), Food Emulsions, Marcel Dekker, New York, 1976 Chapter 3. [42] N. Krog, in: S.E. Friberg, K. Larsson (Eds.), Food Emulsions, 3rd, Marcel Dekker, New York, 1997 Chapter 4.