Protein/surfactant interfacial interactions Part 3. Competitive adsorption of protein + surfactant in emulsions

Colloids and Surfaces A: Physicochemical and Engineering Aspects 100(1995) 267-277

ELSEVIER

Protein/surfactant Part 2. Electrophoretic

A

SURFACES

interfacial interactions

mobility of mixed protein + surfactant systems

Jianshe Chen, Eric Dickinson

*

Procter Department of Food Science, University of Leeds, Leeds LS2 9JT, U.K.

Received 5 February 1995; accepted 19 April 1995

Abstract Protein-protein and protein-surfactant interactions have been investigated in bulk aqueous solution and in oilin-water emulsion systems by electrophoretic mobility measurements. The interaction between oppositely charged P-lactoglobulin and gelatin has been studied under neutral pH conditions. The addition of cationic gelatin to a filactoglobulin-stabilized emulsion induces flocculation and charge neutralization of the emulsion droplets. The procedure of mixing the gelatin solution with the P-lactoglobulin-stabilized emulsion has no significant effect on the observed mobility behaviour of the emulsion droplets. Binding of the anionic surfactant sodium lauryl ether sulphate (SLES 2EO) to P-lactoglobulin and gelatin was observed under neutral pH conditions. The charge neutralization line for gelatin + SLES 2E0 complexes in distilled water appears consistent with its maximum precipitation line. However, the charge neutralization line of gelatin + SLES 2E0 at pH 7.0 occurs at slightly lower surfactant concentrations compared to the maximum precipitation line. The addition of SLES 2E0 to a gelatin-stabilized emulsion causes a change in calculated zeta potential of the emulsion droplets and a partial charge neutralization for the flocculated emulsion droplets. Electrophoretic mobility measured in solutions of fi-lactoglobulin + gelatin + SLES 2E0 shows that the three-component complexes (or precipitates) are negatively charged. Various possible interaction mechanisms are discussed taking account also of results obtained on the same systems by other complementary techniques. Keywords: Anionic surfactant; Electrophoretic

mobility; Gelatin; Interactions;

1. Introduction

Many colloid systems contain a mixture of biopolymers and low-molecular-weight surfactant, and the interactions between these components have a large influence on the colloidal stability. How biopolymers and low-molecular-weight surfactants interact with each other at the oil droplet surface and in the bulk aqueous phase is an

* Corresponding author. 0927-7757/95/$09.500 1995Elsevier Science B.V. All rights reserved SSDI 0927-7757(95)03205-3

j?-Lactoglobulin;

Zeta potential

important factor affecting the formation, stability, and rheology of food oil-in-water emulsions [ 11. The chief driving force for the interaction of polymer and surfactant species in aqueous solution is the association of surfactant molecules with the hydrophobic regions of dissolved polymer molecules. This leads to a reduction in the hydrophobewater contact area of the alkyl chains, whilst at the same time the head-groups of ionic surfactants may be involved in attractive interactions with oppositely charged groups along the polymer chain. This same driving force will also favour

268

J. Chen ef al. JColloidv Surfaces A: Physicochem. Eng. Aspects 100 (1995) 267-277

association of biopolymers. Proteins contain a mixture of hydrophobic groups, polar groups, and electrically charged groups, and so it is not surprising that many small amphiphilic molecules bind strongly to proteins to form protein-surfactant complexes. As a result of these interactions, denaturation and unfolding of the protein both in the bulk phase and at the oil/water interface will occur, and this will, of course, affect the physical properties of emulsions. Much progress has been reported in the literature recently on the subject of polymer-surfactant complexes [ 2-41. Protein-surfactant interactions may result in either co-operative adsorption or competitive adsorption. Protein-surfactant binding affects the adsorption isotherm in two main ways: it changes the adsorption energy of the protein for the interface by affecting the net charge or the overall hydrophobicity, and it reduces the amount of free surfactant in the bulk phase which is available for displacing protein from the interface [3]. Shubin [S] found that the adsorption of cationic polymer onto negatively charged surfaces was affected in the presence of anionic surfactant. It was found that associative binding of surfactant to the polymer results in variety of interfacial behaviour depending on the concentration of added amphiphile. Polymer-surfactant interactions affect both the surface coverage and the conformation of adsorbing macromolecules. In the bulk aqueous phase, a stoichiometric precipitate can be produced from a mixture of protein + ionic surfactant. Goddard and Pethica showed [ 61 that when a protein, such as bovine serum albumin, is in solution on the acid side of its isoelectric point, the addition of an “equivalent” amount of anionic surfactant causes the formation of stoichiometric precipitates, with equal numbers of positive charges (from the protein) and negative charges (from the surfactant). Such precipitates can be resolubilized on addition of excess surfactant, giving rise to the concept that a second layer of bound surfactant ions, with their ionic groups pointing “outwards”, is attached to the first bound layer through association with the hydrocarbon chains of the first layer. A variety of models proposed for the structure of complexes of the anionic surfactant sodium

dodecyl sulphate (SDS) with water-soluble polymers have been summarized by Ibel et al. [ 71. Surfactant molecules generally bind to a polymer chain in clusters which closely resemble the micelles formed in the absence of polymer. If the polymer is less polar or contains hydrophobic regions or sites, there is an intimate contact between the micelles and the polymer chains. In such a situation the contact between one surfactant aggregate and two polymer segments will be favourable and a network of two or more polymer chains may be formed through a. cross-linking mechanism [S]. A recent review on these aspects of polymersurfactant interactions has been published [ 91. Gelatin is a protein widely used in food, pharmaceutical, photographic, and other industries. The interaction of gelatin with surfactants has received extensive attention in the last decade or so. Greener et al. concluded [lo] that there was a possible micellar binding mechanism and that the interaction was governed by electrostatic forces. NMR studies have confirmed [ 1l] the binding of SDS micelles to gelatin molecules, and it is believed that cationic residues such as arginine and lysine are most likely to be attracted to the micellar surface. A rheological study has shown [ 121 that anionic surfactants, at concentrations above the critical micelle concentration (CMC), lead to a greater increase in the viscosity of gelatin solutions than do cationic surfactants, and that nonionic surfactants have little effect on the viscosity of gelatin solutions. Surface shear viscosity experiments in this laboratory have shown [13] evidence for interfacial interaction between gelatin and anionic surfactant and also possible interfacial complex formation between gelatin and oppositely charged fi-lactoglobulin. In the first part of this work [14], we investigated the flocculation behaviour of emulsion systems containing mixed proteins (/?-lactoglobulin + gelatin) and anionic surfactant. We found bridging flocculation of a /?-lactoglobulin-stabilized emulsion in the presence of gelatin and also a much stronger flocculation of emulsion droplets in the presence of gelatin + anionic surfactant at a concentration higher than the CMC. It appears that the surfactant becomes aggregated in the form of small spherical micelles attached to the gelatin

J. Chen et al./ColloidsSurfaces A: Physicochem.Eng. Aspects100 (1995) 267-277

chains. In this paper we use electrophoretic mobility measurements to probe further the properties of mixed solutions of protein + anionic surfactant (sodium lauryl ether sulphate) and the corresponding emulsion systems. The two proteins involved here, gelatin and /?-lactoglobulin, were investigated under neutral pH conditions, with the gelatin molecules being positively charged and the /3lactoglobulin molecules negatively charged.

2. Materials and methods 2.1. Materials

Acid-processed gelatin (type A, 220 bloom strength, 18 mesh, moisture < 13 wt.%) was obtained from Sanofi Bio-Industry Ltd (Newbury, England). The protein /3-lactoglobulin (99%, from bovine milk) contained both genetic variants A and B and was purchased from Sigma (St. Louis, MO, USA). The anionic surfactant sodium lauryl ether sulphate (SLES 2EO), C12H25-0(CH,CH20)2-S03Na (376.2 Da, 99%) was supplied by Unilever Research, Port Sunlight Laboratory. The CMC of SLES 2E0 is about 2 x 1O-3 M (or 0.075 wt.%). AnalaR-grade n-hexadecane (>99%) was obtained from Sigma Chemicals. The bis-tris-propane buffer solutions were prepared from Sigma AnalaR-grade reagents and doubly-distilled water. All chemicals were used without further treatment. 2.2. Emulsion preparation To make the P-lactoglobulin-stabilized emulsion, a solution of 0.5 wt.% /?-lactoglobulin was prepared by dissolving an accurately weighed amount of protein in the 20 mM bis-tris-propane pH 7.0 buffer. Samples of 3.00 g oil and 12.00 g fi-lactoglobulin solution (20:80 by weight) were mixed a’t room temperature for 30 s. The coarse premix was then homogenized using a small-scale single-stage valve homogenizer operating at a pressure of 300 bar. The fresh emulsion was mixed with an appropriate amount of 5.0 wt.% gelatin solution which had been stored at 40°C in a water bath. The gelatin-stabilized emulsion was also produced with

269

the mini-homogenizer operating at 300 bar, but with a higher protein concentration (0.71 wt.% gelatin solution) and less oil (lQ90 by weight). 2.3. Electrophoretic

mobility

Electrophoretic mobility was determined at room temperature using the Malvern Zetasizer 4. The instrument was aligned and calibrated daily with a standard latex dispersion. The mobility of emulsion droplets was measured at a suitable count rate (about 1500 kcps) by diluting the emulsion extensively in the same buffer solution as used for the emulsion preparation. The mobility of the macromolecular complex in aqueous solution was studied using undiluted samples. The ZET5104 sample cell was used for all the experiments and was carefully cleaned for each experiment. About 10 ml sample solution or diluted emulsion was injected through the sample cell in each experiment, and the quoted result is the average of 4 independent measurements. The zeta potential [ was calculated from the measured mobility [l] and given automatically by the instrument.

3. Results and discussion 3.1. Protein + protein systems

Protein-protein interactions determine the phase behaviour of the biopolymer mixture in aqueous solution and the stability of the corresponding emulsion systems [ 1,151. The electrophoretie mobility of the mixed /?-lactoglobulin + gelatin system has been measured in 20 mM bistris-propane pH 7.0 buffer solution. Fig. 1 shows the calculated zeta potential plotted against the gelatin concentration in a solution containing 0.4 wt.% /3-lactoglobulin. The P-lactoglobulin has its isoelectric point at about pH 5.2. So, the pure /?lactoglobulin molecule at neutral pH condition is negatively charged with a zeta potential of about - 11.5 mV. However, the acid-processed gelatin has an isoelectric point above pH 8, and so it is positively charged at neutral pH. On introduction of gelatin into the P-lactoglobulin solution, there is an attractive interaction between two oppositely

J. Chen et al.JColloids Surfaces A: Physicochem. Emg. Aspects 100 (1995) 267-277

270

5

0

-20

-25 0

0.2

0.4

0.6

0.8

Gelatin concentration wt % Fig. 1. The electrophoretic mobility of mixed proteins (0.4 wt.% /3-lactoglobulin + gelatin) in 20 mM bis-&is-propane pH 7.0 buffer at 25°C. The calculated zeta potential is plotted against gelatin concentration.

charged proteins, and this reduces the effective zeta potential of the complex to a less negative value due to the charge neutralization. The change of zeta potential is strongly related to the gelatin concentration and the zeta potential is reduced to approximately -3.5 mV in the presence of 0.8 wt.% gelatin. The mixed protein complex is still negatively charged at high gelatin concentration, and no visible phase separation could be observed in the mixed /?-lactoglobulin + gelatin solution under the experimental conditions employed. The zeta potential at pH 7 of P-lactoglobulinstabilized emulsions mixed with gelatin is shown in Fig. 2 as a function of gelatin concentration. We can see that pure /?-lactoglobulin-coated oil droplets have a zeta potential of about -22 mV. The addition of oppositely charged gelatin induces an attractive interaction between two kinds of proteins in the aqueous phase and also at the interface. This electrostatic interaction results in further adsorption of gelatin onto the oil droplets as we have found previously [ 141. Initial destabilization followed by restabilization of emulsion droplets with respect to flocculation has also been observed in the presence of gradually increasing amounts of gelatin [ 141. Electrophoretic mobility results

0

0.2

0.4

0.6

0.8

Gelatin concentration wt % Fig. 2. The electrophoretic mobility of a /3-lactoglobulinstabilized emulsion (0.4 wt.% P-lactoglobulin, 20 wt.% oil, 20 mM bis-tris-propane buffer, pH 7.0) mixed with gelatin at 25’ C. The gelatin was added to fresh emulsion after emulsification. The calculated zeta potential is plotted against gelatin concentration.

obtained here show a consistent trend. That is, the addition of a small amount of gelatin (~0.2 wt.%) causes a dramatic decrease of zeta potential, and as a result, the electrostatic repulsion between oil droplets will certainly decrease as well. A charge neutralization point is observed .at a gelatin concentration of about 0.25 wt.%, which is very consistent with the destabilization experiments, where a maximum degree of flocculation was observed at this same gelatin concentration [ 141. It has been found that there are typically 0.96 mM cationic functional groups per gram of gelatin [ 16,171. Nozaki et al. [lS] estimated a net charge of approximately - 10 units for /3-lactoglobulin at pH 7.0, which corresponds to about 0.54 mM anionic functional groups per gram of P-lactoglobulin. So, to reach the point of charge neutralization, 1 g of /?-lactoglobulin theoretically requires approximately 0.56 g gelatin. This is in very good agreement with the experiment result here. At high gelatin concentration (>0.3 wt.%), the emulsion droplets become positively charged with a very small positive mobility, which corresponds to a restabilization of the emulsion [ 141. This is

J. Chen et al./Colloids

Surfaces A: Physicochem.

because, at higher gelatin concentration, a second adsorption layer of protein is formed around the oil droplets, and this makes the oil droplets behave as gelatin-coated ones. Although the further increase of gelatin concentration causes restabilization of emulsion [ 141, no significant change in mobility could be observed with this further increasing of gelatin concentration. The electrophoretic mobility of a gelatin-stabilized emulsion has also been determined separately in this laboratory [19] and the small electrophoretic mobility value seems very consistent with the results here. We have also measured the electrophoretic mobility of emulsion droplets made with 0.4 wt.% fi-lactoglobulin + gelatin, both present before emulsification, in order to see the effect of the mixing procedure on the protein surface charge density. The plot of zeta potential against gelatin concentration is shown in Fig. 3. We can see that the behaviour is very similar to that obtained with gelatin added after emulsification (Figure 2). The independence of the mobility on the mixing procedure suggests that the gelatin molecules adsorb as a secondary outer adsorption layer via electrostatic interaction with P-lactoglobulin at the sur-

Eng. Aspects 100 (1995) 267-277

211

face, irrespective of whether or not the gelatin is mixed with the globular protein before or after the emulsification. In other words, the gelatin is less surface active than the fl-lactoglobulin, and is acting as a protective colloid around the /I-lactoglobulin droplets. This is in agreement with some complementary flocculation experiments [ 141, where we found that the addition of gelatin before or after the emulsification stage made no significant difference to the emulsion flocculation behaviour. This result is also consistent with surface shear viscosity results obtained in this laboratory [ 13,201, where the p-lactoglobulin was found to be the dominant component at the oil-water interface in the mixed P-lactoglobulingelatin film. It is noteworthy that there is a significant qualitative difference between the electrophoretic mobility behaviour of the mixed proteins in bulk solution and the mobility behaviour of droplets in the emulsion system. This may reflect a different interaction mechanism of the proteins in the bulk aqueous phase as compared with that in the emulsion. Partial unfolding of the globular protein at the surface could be a contributory explanation. 3.2. Protein

+ sur-uctant systems

5

0

-20

-25 0

0.2

0.4

0.6

0.8

Gelatin concentration wt %

Fig. 3. The electrophoretic mobility of a mixed protein stabilized emulsion at 25°C (0.4 wt.% P-lactoglobulin, gelatin, 20 wt.% oil, 20mM bis-tris-propane buffer, pH 7.0). The gelatin was present before emulsification. The calculated zeta potential is plotted against gelatin concentration.

The interactions between each protein and the anionic surfactant SLES 2E0 have been studied by monitoring the change in electrophoretic mobility on addition of surfactant. Fig. 4 shows the zeta potential of the P-lactoglobulin-surfactant complex plotted against the surfactant concentration in 20 mM bis-tris-propane pH 7.0 buffer solution. The increase in magnitude of the negative zeta potential in the presence of SLES 2E0 suggests binding of surfactant molecules to j-lactoglobulin molecules. We know that /?-lactoglobulin and SLES 2E0 are both negatively charged here. So, the binding of SLES 2E0 to fi-lactoglobulin is most probably via hydrophobic interactions. The results confirm recent interfacial shear viscosity experiments suggesting an interfacial complex between P-lactoglobulin and SLES 2E0 at the oil/water interface [ 131. The binding of anionic surfactant SDS to P-lactoglobulin has been confirmed in the pH range 357.0 [21]. It was also inferred that the binding is initially of an ionic

272

J. Chen et al./Colloids Surfaces A: Physicochem. Eng. Aspects 100 (1995) 267-277 0

-15

-5 -25

S g s -30 i

-35

-40

45

-25

-50 0

0.2 SLES

0.4

0.6

0.8

1

ZEO concentretion wt %

Fig. 4. The electrophoretic mobility of mixed solution of 0.4 wt.% /?-lactoglobulin + anionic surfactant SLES 2E0 in 20 mM bis-t&-propane pH 7.0 buffer. The mobility was measured at 25°C. The calculated zeta potential is plotted against SLES 2E0 concentration.

nature; but at higher binding amounts the co-operative nature of binding is more characteristic of hydrophobic bonding [Zl]. Separate surface shear viscosity studies in this laboratory [ 221 show that the addition of the anionic surfactant DATEM to an aged /I-lactoglobulin film can induce a dramatic decrease in surface viscosity followed by a slow recovery. All these results are consistent with interfacial complex formation between P-lactoglobulin and anionic surfactant. Fig. 5 shows zeta potential data for j?-lactoglobulin-coated emulsion droplets mixed with surfactant at pH 7.0. A dramatic change of zeta potential is observed at a very low SLES 2E0 concentration. This could be attributed to two possible reasons: the displacement of /I-lactoglobulin from the interface by the surfactant, or the further binding of the surfactant to the interface in the form of complex. With increase in SLES 2E0 concentration, the absolute magnitude of the zeta potential of emulsion droplets also shows a gradual increase (5 becomes more negative) over the whole range of experimental surfactant concentrations. The interaction between gelatin and anionic surfactant has received extensive attention in the

0

0.2 SLES

0.4 2E0

0.6

0.8

concentration wt %

Fig. 5. The electrophoretic mobility of a /I-lactoglobulinstabilized emulsion (0.4 wt.% p-lactoglobulin, 20 wt.% oil, 20 mM bis-t&propane buffer, pH 7.0) with anionic surfactant SLES 2E0 added after emulsification. The mobility was measured at 25°C. The calculated zeta potential is plotted against SLES 2E0 concentration.

literature [23-251. We have studied precipitation and resolubilization behaviour of gelatin + SLES 2E0 complexes in bulk aqueous solution [14]. Fig. 6 shows the zeta potential change of the gelatin-SLES 2E0 complex in distilled water in the presence of gradually increasing amounts of anionic surfactant. The zeta potential value starts at about 13 mV for a pure 0.3 wt.% gelatin solution and then ends at approximately -25 mV in the presence of 0.1 wt.% SLES 2E0. A charge neutralization point is clearly observed at an SLES 2E0 concentration of about 0.0275 wt.%, which corresponds very well to the maximum degree of precipitation observed previously [14]. This means that the precipitation of gelatin-SLES 2E0 complexes is most probably based on the charge neutralization mechanism. At higher SLES 2E0 concentration (>0.0275 wt.%), the complex becomes negatively charged (Fig. 6) which corresponds to a gradual resolubilization of the precipitate [14]. This phenomenon agrees quite well with Pankhurst’s postulate [ 261 that the solubilization of insoluble gelatin-SDS complexes in excess of SDS is achieved by the physical adsorption of SDS

213

J. Chen et al./Colloids Surfaces A: Physicochem. Eng. Aspects 100 (1995) 267-277 15

1

10

F g

5

ae

O

z

g i

2 3 5

-5

e 8 c

d -10 a

0.1

0.01

s 3

-15

d

0.001

-20

3

-25 -

0

0.02

0.04

0.06

0.08

0.1

SLES 2E0 concentration wt %

Fig. 6. The electrophoretic mobility of a 0.3 wt.% gelatin + SLES 2E0 mixture in distilled water. The mobility was measured at 25°C. The calculated zeta potential is plotted against SLES 2E0 concentration.

on to insoluble gelatin-SDS complexes, with the physically adsorbed SDS oriented toward the water. Knox and Parshall investigated [24] the surface tension of gelatin-SDS complexes and found a correlation with different precipitation stages. However, the smooth zeta potential change found here does not give any discernible indication of the starting point of precipitation or the point of complete resolubilization of the precipitate. This suggests a gradual binding of the SLES 2E0 molecules to the gelatin molecules. Fig. 7 gives the charge neutralization line of gelatin + SLES 2E0 in unbuffered distilled water. The line appears at exactly the same position as the maximum precipitation line obtained previously [ 141. This again confirms charge neutralmechanism of ization as the predominant gelatin-SLES 2E0 precipitation in distilled water. This finding is consistent with the observation of Goddard and co-workers [27,28] that the position of maximum precipitation in the case of the Polymer JR (a quaternary nitrogen substituted cellulose ether) + SDS system corresponds to a stoichiometric 1:l complex, and is based on the mechanism of charge neutralization. From Fig. 7, we can also see that, at sufficiently low gelatin

0.0001 0.001

0.01

0.1

SLES 2EO concentration wt 56

Fig. 7. The charge neutralization lines of gelatin + SLES 2E0 in distilled water ( ?? ) and in 20 mM bis-tris-propane pH 7.0 buffer solution (0). The electrophoretic mobility was measured at 25°C. The mobility of gelatin-surfactant complex was measured over a wide range of gelatin and surfactant concentration. The charge neutralization line was drawn through the points of zero mobility or zero zeta potential.

concentration (6 0.1 wt.%), the charge neutralization line becomes rather independent of the gelatin concentration. This is because in this concentration region most of the added surfactant is required to maintain equilibrium with the precipitated stoichiometric complex [ 91. Turbidity measurements have recently shown [ 141 a different phase diagram for gelatin + SLES 2E0 in 20 mM bis-tris-propane pH 7.0 buffer solution compared to that in distilled water. It was found [ 141 that precipitation in the former system occurs at a higher gelatin concentration. A qualitatively similar trend is also observed in the electrophoretic mobility investigation here. Fig. 7 also gives the charge neutralization line of gelatin + SLES 2E0 in 20 mM bis-tris-propane pH 7.0 buffer solution. We can see that the line moves to higher gelatin concentrations, but lower surfactant concentrations, compared to that in the distilled water, and it becomes impossible to determine at low gelatin concentration, say smaller than about 0.1 wt.%. The pH of the solution determines the

214

J. Chen et al.IColloids Surfaces A: Physicochem. Eng. Aspects 100 (1995) 267-277

effective charge on the protein and therefore its ability to attract ionic surfactant. The charge density of gelatin becomes smaller in pH 7.0 buffer than in distilled water and, therefore, the precipitation moves to higher gelatin concentration, and consequently less anionic surfactant is required for the charge neutralization. Another noteworthy point is that the charge neutralization line of gelatin-SLES 2E0 complexes in 20 mM bis-trispropane pH 7.0 buffer solution does not coincide with the maximum precipitation line of the same system as determined previously [ 141. This means that simple charge neutralization is not the only cause of the precipitation. Ananthapadmanabhan concluded [29] that, at low charge density values, in addition to electrostatic factors, conformational aspects of the polymer appear to play a role in determining the onset of binding and the shift in the isotherms to low surfactant concentrations. So, the position of maximum precipitation of macromolecules of low charge density may be based on a combination of both charge neutralization and hydrophobic or hydrogen bonding interactions. We turn now from the bulk aqueous phase behaviour to the emulsion systems. Fig. 8 shows the results obtained from electrophoretic mobility measurements of a gelatin-stabilized emulsion mixed with the anionic surfactant SLES 2E0 at pH 7.0. We can see that the inferred zeta potential of the emulsion droplets is strongly dependent on the concentration of SLES 2E0. Charge neutralization of the oil droplets is observed at a very low SLES 2E0 concentration. However, the particle size measurements show [ 141 that the emulsion is still quite stable toward flocculation at this charge neutralization point. The steric stabilization by adsorbed protein is probably the predominant factor here. It is interesting that the partial charge neutralization of the emulsion droplets starts at an SLES 2E0 concentration of about 0.1 wt.%. The maximum partial charge neutralization is observed at about 0.3 wt.% SLES 2E0, and this is then followed by a dramatic increase in the value of the (negative) zeta potential at higher surfactant concentrations. Separate emulsion stability experiments have shown [14] that the gelatin-stabilized emulsion is unflocculated in the presence of less

-5

s

-10

g -15 I E -20 t I?

b

,P

‘! \

/

‘b,

-25

,,’

‘\k,’

I,/

-30

1, ?

-35 f

I

-40 451

0.001

t 0.01

0.1

1

SLES 2EO concentration wt %

Fig. 8. The electrophoretic mobility of a gelatin-stabilized emulsion (0.64 wt.% gelatin, 10 wt.% oil, 20mM bis-trispropane buffer, pH 7.0) mixed with anionic surfactant SLES 2E0. The mobility was measured at 25°C. The calculated zeta potential is plotted against SLES 2E0 concentration.

than 0.1 wt.% surfactant. Flocculation begins when the SLES 2E0 concentration exceeds 0.1 wt.%, and the maximum degree of flocculation occurs at a surfactant concentration of about 0.3 wt.% [ 141. The stability results are very consistent with the zeta potential results reported here. The competitive adsorption experiments of gelatin stabilized emulsions mixed with SLES 2E0 [30] show also a consistent trend: the presence of anionic surfactant induces further adsorption of gelatin rather than the displacement of gelatin from the oil droplet surface. The maximum gelatin surface concentration agrees very well with the peak of partial charge neutralization determined here. The coincidence of the maximum degree of flocculation with the peak of partial charge neutralization and the maximum gelatin surface concentration strongly suggests that charge neutralization is one of the main reasons causing emulsion destabilization. Comparing these electrophoretic results with those of gelatin + SLES 2E0 in aqueous solution, we find that the zeta potential change for gelatin + SLES 2E0 in aqueous phase and that for the corresponding emulsion system show very different trends as a function of SLES 2E0 concentration.

275

J. Chen et al,lColloids Surfaces A: Physicochem. Eng. Aspects 100 (1995) 267-277

This suggests the presence of different types of molecular interactions in gelatin-surfactant complexes at the oil/water interface from those in the bulk aqueous solution. The partial charge neutralization and flocculation of emulsion droplets at moderately high SLES 2E0 concentration (more than 0.1 wt.%) is most probably associated with the formation of surfactant micelles. We know that SLES 2E0 has a CMC of about 0.075 wt.%. When the SLES 2E0 concentration is higher than 0.1 wt.% in the emulsion, some of the surfactant will therefore probably exist in the form of surfactant micelles. Binding of these surfactant micelles to the tails and loops of adsorbed gelatin molecules is thus quite likely; then the emulsion may start to become flocculated via a surfactant cluster crosslinking mechanism [S]. 3.3. Mixed protein + surfactant systems The last set of experimental results to be reported in this paper relate to systems containing mixed proteins (fi-lactoglobulin + gelatin) as well as anionic surfactant. As already reported above, interactions amongst these three components are very likely to occur both in aqueous solution and at the oil/water interface. We first present electrophoretic mobility results for mixed protein (0.4 wt.% gelatin + 0.4 wt.% P-lactoglobulin) + surfactant in 20 mM bis-tris-propane pH 7.0 buffer solution. The calculated zeta potential is plotted against the surfactant concentration as shown in Fig. 9. We see that the protein complex particles are negatively charged over the whole range of SLES 2E0 concentration. The zeta potential becomes gradually more negative with an initial small addition of SLES 2E0, but a more significant increase of zeta potential is observed at surfactant concentrations higher than about 0.25 wt.%. The turbidity observation shows that phase separation starts to occur at a SLES 2E0 concentration of about 0.03 wt.% and that the precipitate becomes completely resolubilized at approximately 0.5 wt.% SLES 2E0 [14]. Obviously, the mechanism of complex formation here is different from that found with the single protein + surfactant system, where charge neutralization plays a determining role in the phase separation process as discussed above.

0 ,

i\ ?

-16

i -10 -20

i I

0.01

+

0.1

1

SLES 2E0 concentration wt %

Fig. 9. The electrophoretic mobility of 0.4 wt.% fi-lactoglobulin + 0.4 wt.% gelatin + SLES 2E0 in 20 mM bis-tris-propane pH 7.0 buffer solution. The mobility was measured at 25°C. The calculated zeta potential is plotted against surfactant concentration.

From the results presented here we must conclude that phase separation in this three components system is not based on a simple charge neutralization mechanism. The interaction between the two oppositely charged proteins may induce structural denaturation and unfolding of the globular protein. Thus, the added anionic surfactant molecules will tend to interact with the proteins in two possible ways: the negatively charged heads of surfactant molecules may interact with the positively charged sites on the gelatin and ,@lactoglobulin via electrostatic interactions, and leave the hydrophobic chains pointing outwards towards the aqueous phase; or, the chains of the surfactant molecules may interact with the globular protein via hydrophobic interactions and leave the heads pointing outwards into the aqueous phase. The former arrangement will favour phase separation, whilst the latter will tend to make the complex more soluble. The actual phase behaviour probably will be determined by a combination of these two factors. At high SLES 2E0 concentrations, surfactant micelles become available to interact with the protein side chains. This could be the reason why the precipitates are resolublized at high surfactant

J. Chen et al./CoNoids Surfaces A: Physicochem. Eng. Aspects 100 (1995) 267-277

276

concentration, together with the large negative zeta potential. Finally, we show the electrophoretic results of emulsions containing mixed proteins + surfactant. Fig. 10 shows the change in calculated zeta potential of a fi-lactoglobulin-stabilized emulsion mixed with 0.3 wt.% gelatin and different amounts of anionic surfactant after emulsification. A very dramatic change of zeta potential occurs at very low SLES 2E0 concentration. However, further increase in SLES 2E0 concentration leads only to a gradual change in zeta potential. We can see that the electrophoretic mobility behaviour of the mixed protein + surfactant emulsion system is quite different from that in bulk aqueous solution. This would again seem to show a different interaction mechanism for the protein-surfactant complex at the oil/water interface from that in bulk aqueous solution. In our complementary emulsion stability investigation, we have found [ 141 strong flocculation of /3-lactoglobulin-stabilized emulsions in the presence of gelatin and an appropriate amount of

SLES 2E0. However, in the present experiments, we find that no corresponding change in zeta potential is observed in the flocculation region. The emulsion containing 0.8 wt.% gelatin + surfactant shows almost same trend of zeta potential change (see also Fig. 10) even though there is a large difference in the quantity of gelatin present. This suggests that the presence of anionic surfactant has a predominant role in determining the mobility of the emulsion droplets coated with the mixed proteins.

Acknowledgement Unilever Research (Port Sunlight Laboratory) is acknowledged for financial support of this project.

References Cl1 E. Dickinson and Cl. Stainsby, Colloids in Food, Applied

Science, London, 1982.

PI E.D. Goddard, Colloids Surfaces, 19 (1986) 255, 301. c31 E.

-60 1 0

0.2

0.4

0.6

0.8

1

SLES2EOconcentmtionwt%

Fig. 10. The electrophoretic mobility of a /?-lactoglobulinstabilized emulsion (0.4 wt.% j3-lactoglobulin, 20 wt.% oil, 20 mM bis-tris-propane buffer, pH 7.0) mixed with gelatin + SLES 2EO: ?? , fresh emulsion + 0.3 wt.% gelatin + SLES 2EO; 0, fresh emulsion + 0.8 wt.% gelatin + SLES 2E0. The mobility was measured at 25°C. The calculated zeta potential is plotted against SLES 2E0 concentration.

Dickinson and C.M. Woskett, in R.D. Bee, P. Richmond and J. Mingins (Eds.), Food Colloids, Royal Society of Chemistry, Cambridge, 1989, p.74. 141 M.N. Jones, Chem. Sot. Rev., 21 (1992) 127. c51 V. Shubin, Langmuir, 10 (1994) 1093. C61 E.D. Goddard and B.A. Pethica, J. Chem. Sot., (1951) 2659. c71 K. Ibel, R.P. May, K. Kirschner, H. Szadkowski, E. Mascher and P. Lundahl, Eur. J. Biochem., 190 (1990) 311. PI B. Lindman, A. Carlsson, S. Gerdes, G. Karlstron, L. Piculell, K. Thalberg and K. Zhang, in E. Dickinson and P. Walstra (Eds.), Food Colloids and Polymers: Stability and Mechanical Properties, Royal Society of Chemistry, Cambridge, 1993, p. 113. (Eds.), c91 E.D. Goddard and K.P. Ananthapadmanabhan Interactions of Surfactant with Polymers and Proteins, CRC Press, Boca Raton, FL, 1993. B.A. Constestable and M.D. Bale, Cl01 J. Greener, Macromolecules, 20 (1987) 2490. Cl11 D.D. Miller, W. Lenhart, B.J. Antalek, A.J. Williams and J.M. Hewitt, Langmuir, 10 (1994) 68. t-121 A.M. Howe, A.G. Wilkins and J.W. Goodwin, J. Photogr. Sci., 40 (1992) 234. Cl31 J. Chen and E. Dickinson, Food Hydrocolloids, 9 (1995). Cl41 J. Chen and E. Dickinson, Colloids Surfaces A: Physiochem. Eng. Aspects, 100 (1995) 253-265. Cl51 B. Bergenstahl, P. FXldt and B. Lassen, in J. Sjiiblom

J. Chen et aL/Colloids Surfaces A: Physicochem. Eng. Aspects 100 (1995) 267-277

[ 161 [17] [lg] [ 191 [20]

1211 1221 [23]

(Ed), Emulsions - A Fundamental and Practical Approach, Dordrecht, Kluwer, 1992, p.51. A. Kenchington and A. Ward, Biochem. J., 58 (1954) 202. W. Ames, J. Sci. Food Agric., 3 (1952) 379. Y. Nozaki, L.G. Bunville and C. Tanford, J. Am. Chem. Sot., 81 (1959) 5523. M. Goller, PhD Thesis, University of Leeds, 1994. J. Castle, E. Dickinson, B.S. Murray and G. Stainsby, ACS Symp. Ser., 343, Proteins at Interfaces: Physicochemical and Biochemical Studies, J.L. Brash and T.A. Horbett (Eds.), American Chemical Society, Washington, D.C., 1987, p.118. M.N. Jones and A. Wilkinson, Biochem. J., 153 (1976) 713. E. Dickinson and S.T. Hong, J. Agric. Food Chem., 42 (1994) 1602. W.J. Knox, Jr., and J.F. Wright, J. Colloid Interface Sci., 20 (1965) 177.

[24]

211

W.J. Knox, Jr., and T.O. Parshall, J. Colloid Interface Sci., 33 (1970) 16. [25] H. Fruhner and G. Kretzschmar, Colloid Polym. Sci., 270 (1992) 177. 1261 K.G.A. Pankhurst, Surface Chem. Research Suppl., (1949) 109. 1271 E.D. Goddard, T.S. Phillips and R.B. Hannan, J. Sot. Cosmet. Chem., 26 (1975) 461. [28] E.D. Goddard and R.B. Hannan, J. Colloid Interface Sci., 55 (1976) 73. [29] K.P. Ananthapadmanabhan, in E.D. Goddard and K.P. Ananthapadmanabhan, (Eds.), Interactions of Surfactants with Polymers and Proteins, CRC Press, Boca Raton, FI, 1993, p.319. [30] J. Chen and E. Dickinson, Colloids Surfaces A: Physicochem. Eng. Aspects, in press.