J. Phys. Chem. B 2008, 112, 3989-3996
3989
On the Difference between Foams Stabilized by Surfactants and Whole Casein or β-Casein. Comparison of Foams, Foam Films, and Liquid Surfaces Studies Julia Maldonado-Valderrama†,§ and Dominique Langevin*,‡ Grupo de Fı´sica de Fluidos y Biocoloides, UniVersidad de Granada, Facultad de Ciencias, Campus de FuentenueVa, sn, 18071, Granada, Spain, and Laboratoire de Physique des Solides, UMR 8502, CNRS, UniVersite´ Paris-Sud, Bat, 510, 91405, Orsay Cedex, France ReceiVed: NoVember 28, 2007; In Final Form: January 19, 2008
This research work aims to investigate the behavior of a mixed system composed of a commercial protein (whole casein) and a low molecular weight surfactant (Tween 20) in order to understand its foam stability on the basis of fundamental surface quantities such as surface and disjoining pressure. These experiments prove to be extremely useful in the understanding of the processes determining foam stability. The complex behavior of whole casein/Tween 20 mixtures is directly deduced from the surface pressure isotherms. Concretely, the isotherm of the mixed system is displaced to smaller surfactant concentrations as compared to the pure surfactant system. This feature is quantified by a critical aggregation concentration and suggests formation of protein/ surfactant complexes within the surface layer. The disjoining pressure isotherms of the pure and mixed systems provide key information regarding the structure and composition of the mixed whole casein/Tween 20 surface layers. Furthermore, they provide a direct correlation with the foam stability in terms of the thickness of the final foam film; the thinner the film, the less stable the foam. The experimental results are further discussed in terms of literature studies of similar systems, and a final rather accurate description of the system arises. This work investigates the importance of the nature of the protein in the stability of the foams of protein/ surfactant mixtures and highlights the fundamental role of the surface properties in the understanding of such a major phenomenon in colloid science and technology.
1. Introduction Proteins, surfactants, lipids, and other foaming agents coexist in commercial foams and because of their amphiphilic nature compete for the available interfacial area. The competition between different molecules can result in a very complex behavior of the mixed interfaces. Proteins form viscoelastic structures that stabilize dispersions, owing to a strong and immobile interfacial network. Conversely, low molecular weight surfactants form mobile interfacial structures that stabilize the resulting dispersion because of the Gibbs-Marangoni mechanism.1 Once these two kinds of molecules coexist at the interfaces, the two mechanisms also compete and it could happen that none of them can operate correctly. Therefore, understanding the interactions between components and the structure of the mixed interfacial layers is crucial in the correct elaboration of foams. Although there have been great advances recently and aspects of competitive adsorption between low molecular weight surfactants and proteins seem now to be fairly well understood, further work is required with regard to the effects of adsorbed film structural inhomogeneity on the stability of the foam. In particular, as stated by B. S. Murray in a very recent review, the complexity of the mixtures involved in many of the existing studies in the literature sheds little light on the fundamental mechanisms of the formation and stabilization of foams.2 Accordingly the main objective of this research work is to perform a comprehensive analysis of the foam stability of * Corresponding author. E-mail:
[email protected]. † Universidad de Granada. ‡ Universite ´ Paris-Sud. § Present address: Institute of Food Research, Norwich Research Park, NR4 7UA, Norwich, UK.
a mixed protein/surfactant over different length scales with the aim of highlighting the basic mechanisms underlying the stability of foams. A previous work was devoted to the study of a model system in which a theoretical basis of competitive adsorption between Tween 20 and β-casein with respect to foam stability was provided.3 β-Casein is the major component of whole casein, a family of phosphorylated proteins containing different monomers that constitute 80% of milk proteins. The effect of Tween 20 on the stability of foams of whole casein and β-casein appears very different. Accordingly, further investigation of the fundamental features underlying this different behavior seemed worthwhile. On one hand, the properties of whole casein foams might have great interest among the food industry, and on the other hand, the information on β-casein foams should be very valuable for fundamental understanding. However, because of the complex nature of whole casein, the application of the theoretical model used with β-casein’s adsorption data is clearly not feasible. Therefore, in order to understand the behavior of foams made with the mixed whole casein/Tween 20 system, the properties of the thin liquid films formed by such mixtures are evaluated here along with the surface properties. Because of the great amount of free interfacial area present in foams, the surface properties of the components seem extremely important for its understanding. Surface pressure measurements provide some basic information regarding the structural properties of proteins and surfactants at interfaces.4,5 Thus, these measurements constitute the starting point of the work. Subsequently, thin liquid film studies contain important information related to the stability of the foam since the disjoining pressure quantifies the forces acting between bubbles.4
10.1021/jp7112686 CCC: $40.75 © 2008 American Chemical Society Published on Web 03/07/2008
3990 J. Phys. Chem. B, Vol. 112, No. 13, 2008 Thus, further structural information of the system is searched for with this experimental technique. However, thin liquid films made of proteins are complex to study and to interpret5,6 and although there are some recent attempts in the literature,7-9 thin liquid films studies of mixed systems of protein and surfactant are still rather scarce. The experimental study of thin liquid films of proteins and surfactant mixtures is the first goal of our research work. Furthermore, the correlation between thin film studies and foam stability is fairly unclear and is the second goal of our work. It provides very promising results and substantially contributes to the better understanding of fundamental processes governing foam stability. Accordingly, this work has been organized as follows. First, the surface behavior of whole-casein/Tween 20 systems is presented and compared to that of β-casein/ Tween 20. Next, further information on the surface structure of these mixed systems is investigated for the behavior upon confinement in thin liquid films. Finally, the stability of the foams formed by whole casein/Tween 20 mixtures is evaluated, and the results are thoroughly discussed in terms of the adsorption layers and thin liquid films. 2. Experimental Methods 2.1. Set-Up. 2.1.1. Surface Tension. The surface tension measurements were performed in a pendant drop instrument based on axisymmetric drop shape analysis (ADSA), which is described in detail by Cabrerizo-Vilchez et al.10 The solution droplet is formed at the tip of a PTFE (polytetrafluoroethylene) capillary. The whole set up is computer controlled, and the program fits the experimental drop profile to the YoungLaplace equation of capillarity by using ADSA and provides as output the drop volume V, the surface tension γ, and the surface area A. The drop is formed in a glass cuvette (Hellma) that is kept in a thermostated cell at a constant temperature of T ) 23 °C. The adsorption process is measured by recording the change of surface pressure π (π ) γ0 - γ), γ0 being the surface tension of the clean surface. The surface area is kept constant at 28 mm2, and the experiment records the adsorption process for very long periods of time (10 h). All the glassware was cleaned in sulfuric acid and then repeatedly rinsed with ultrapure water. The surface tension (γ0) of the clean surface was measured before each experiment to ensure the absence of surface-active contaminants obtaining values of (72.5 ( 0.5) mJ/m2. 2.1.2. Disjoining Pressure. Disjoining pressures were measured with a modified version of the porous-plate technique.11,12 This experimental setup is presented in Figure 1 and fully described elsewhere.13 Single thin-liquid foam films are formed in a hole drilled through a fritted glass disk, onto which a glass capillary tube is fused. This film holder is cleaned very thoroughly before every experiment with the following protocol. First, it is placed in Micro-90 cleaning solution, heated at 90 °C for 3 h, and carefully rinsed by sucking through 2 L of water. Next, it is boiled in a 1:4 mixture of NH3 and H2O2 for 30 min. Finally, it is rinsed again by sucking through ultrapure water. The foaming solution is placed in the fritted glass holder and then placed in the balance so that the free end of the capillary tube is at atmospheric pressure and the disk remains enclosed in a pressurized cell. The pressure is regulated by a syringe pump, which allows the pressure (∆P) applied to the film to be regulated. Once formed, the film thins because of the applied pressure, it flattens and, eventually it reaches an equilibrium thickness. At equilibrium, ∆P is compensated by the disjoining pressure Π.
Maldonado-Valderrama and Langevin
Figure 1. Schematic diagram of the modified version of the porousplate technique.
The film thickness is measured by using Scheludko’s microinterferometric method.14 As can be seen in Figure 1, a white light beam is focused onto the film. The reflected beam is collected by an objective and divided in two. One is sent to a video camera and the other to an optical fiber and an objective. In this manner, the information extracted from the foam film is double. On one hand, the video camera provides real time images of the thinning process which are recorded in a computer.13,15 On the other hand, the second light beam is filtered at 632 nm, and its intensity is measured with a photomultiplier. A chopper modulates the light beam intensity at 370 Hz, and the photomultiplier signal is analyzed with a lock-in amplifier providing the intensity of the reflected light. The equilibrium value of this intensity, i.e., constant reflected intensity during 10 min, is recorded for each value of the applied pressure and transformed into film thickness as described by Scheludko.14 Different thicknesses coexisting in the film, due to nucleation of dark holes, appear with different gray levels and can be measured by positioning the objective and estimating the reflected light. The applied pressure is gradually increased and the film thickness measured in each step so that the disjoining pressure versus thickness isotherm, Π(h) is drawn.13,15,16 The maximum pressure applied ∆P is limited by film rupture. Protein films were left to equilibrate for 30 min before measuring the foam film thickness. Measurements were repeated three times, and the reproducibility is indicated by the error bars in the plots. 2.1.3. Foam Stability. Foam stability has been studied using a specially designed vertical foam column. The foaming solution is placed in a hollow rectangular prism of section of 2.56 × 10-2 m2 and height of 1 m. The column is made of plexiglass and is connected to a fritted glass filter of porosity 40-100 µm at the bottom. The air flux is blown into the column through the filter by means of a flowmeter. The flow rate is kept constant in all of the experiments at 5 × 10-4 m3/s. The experiments are performed with the following protocol: the column is filled with 200 mL of solution, and subsequently the flowmeter is connected. Once the foam has reached a height of around 0.8 m, the air flow is stopped. The stability of the foam formed is estimated by measuring the lifetime of the foam (t1/2), time taken by the foam to decay to half the original height after the air flow is stopped. These results, though qualitative in character, allow a comparison of the foam stability of the different solutions.3,17,18 2.2. Materials. Whole casein (or caseins) from bovine milk was purchased from Sigma-Aldrich and used without further
Foam Stability
J. Phys. Chem. B, Vol. 112, No. 13, 2008 3991
Figure 2. Surface Pressure isotherms. The concentration of protein in the mixtures is 5 × 10-3mol/m3.
TABLE 1: Composition of Whole Casein and Properties of the Fractions19,22 % in molecular isoelectric charge at monomers caseins weight, kD point pH 6.6 R1-casein
38.5
23
4.93
-21.8
R2-casein
10
25
5.34
-14.7
β-casein
35.8
24
5.24
-12.3
κ-casein
12.7
19
5.76
-2.5
structure R-helix, β-sheet R-helix, β-sheet, SS R-helix, β-sheet R-helix, β-sheet, SS
purification. The caseins are a family of phosphorylated proteins, and whole casein powder contains many different monomers. The proportions and some of the basic properties of each monomer in the caseins are displayed in Table 1. Protein concentration in solution is kept in all cases at 0.1 g/L ) 5 × 10-3 mol/m3. Solutions were sonicated to ensure good solubilization, and only fresh solutions were used in all the experiments. Polyethylene glycol sorbitan monolaurate (or polyoxyethylenesorbitan monolaurate) was also purchased from SigmaAldrich and used without further purification. This is a nonionic surfactant, soluble in water with a molecular weight of 1228 g/mol. Its chemical composition is C58H114O26, and its generic name is Tween 20 (or polysorbate 20). The aqueous subphase used in all the solutions is a buffer solution made with 6.61 g/L Trizma hydrochloride and 0.97 g/L Trizmabase (Sigma-Aldrich). In order to prevent bacterial contamination, 0.5 g/L NaN3 was added to the buffered solvent. The buffer solution has a final pH 7.4 and an ionic strength of 16.4 mM. Surfactant solutions were prepared by successive dilution from a concentrated solution. The mixed solutions were formed by adding the desired amount of surfactant to the protein solution (5 × 10-3 mol/m3 whole casein). The final concentration of surfactant in the mixed solutions ranges between 10-5 and 1 mol/m3. All solutions were prepared daily, and 0.054 µS Milli-Q purified water was used for buffer preparation and all other purposes. All experiments were performed at T ) 23 °C. 3. Results 3.1. Surface Properties. We first describe the surface properties of the mixed system formed by Tween 20 and whole casein. With the aim of comparing the differences in surface behavior, Figure 2 shows the surface pressure isotherm of the Tween 20, of whole casein/Tween 20, and of β-casein/Tween 20 taken from a previous work.3 To facilitate comparison, the
surface pressure isotherms of the pure whole casein and pure β-casein are also plotted. The π-c isotherms shown in Figure 2 display the final value (after 10 h of adsorption at constant surface area) of the surface pressure corresponding to each of the concentrations analyzed. The form of the π-c isotherms displayed by the caseins and its major fraction β-casein at the air-water interface is very similar. However, the whole casein shows a slightly lower surface activity (the isotherm is displaced to slightly lower bulk concentrations) and the final surface pressure is also slightly lower for whole casein than for β-casein. Thus, the structural differences between both proteins could yield to differences in the surface structure. Benjamins has indeed reported differences in the surface viscoelastic properties of the two proteins.19 The differences between whole casein and its major component also appear clearly in the surface behavior of their mixtures with Tween 20. As can be seen in Figure 2, the β-casein/Tween 20 mixture converges to that of Tween 20 for c g 10-2 mol/m3. As was discussed in detail in a previous work,3 the surfactant displaces the protein from the surface when the surface pressure becomes larger than that for the protein alone. Also, the critical aggregation concentration (cac) of the β-casein/Tween 20 mixture coincides with the critical micellar concentration (cmc) of Tween 20 (2 × 10-2 mol/m3). Conversely, the whole casein/ Tween 20 mixture does not converge to that of the isotherm of Tween 20. In fact, the surface pressure of the mixed system rises at lower concentration of Tween 20 in the mixture, 10-3 mol/m3, compared to the mixture with β-casein, and the surface pressure isotherm shows a critical aggregation concentration of cac ) 10-2 mol/m3, that is, slightly lower that the cmc of Tween 20. This behavior might indicate the formation of mixed protein/ surfactant layers in the whole casein/Tween 20 system. In view of the different behavior of the β-casein/Tween 20 system, the mixed layers should contain the other fractions contained in the whole casein. Similar interpretations of the surface pressure isotherms have been carried out in the literature with other protein/surfactant systems. Namely, the surface pressure isotherm of lysozyme/non-ionic surfactant converges to that of the sole surfactant whereas the lysozyme/ionic surfactant system displays an enhanced surface activity, suggesting the formation protein/surfactant complexes in the latter case.9 3.2. Drainage of Thin Liquid Films. The analysis of the surface pressure isotherms of mixed whole casein/Tween 20 systems suggests the possible formation of mixed protein/ surfactant layers. With the aim of further investigating this behavior, the confinement in thin liquid films of the same systems was studied. We evaluate first the properties of the drainage of the films formed by each of the individual systems and subsequently that of the mixture. Figure 3 illustrates the drainage of a single foam film stabilized by Tween 20 and shows its corresponding disjoining pressure isotherm, using a concentration of surfactant well above its cmc. Just after the film formation, an interference pattern appears, accounting for the inhomogeneous thickness of the film. These patterns are very fluid due to the mobile nature of the surfactant. Increasing the surface pressure causes the nucleation of dark holes that gradually grow and provide films with coexistence of different thickness. Smaller thickness domains nucleate and expand until the thinner domain fully covers the film. Drainage then stops, and the pressure corresponding to the equilibrium thickness reached is the disjoining pressure, which is plotted in Figure 3. Further increasing the pressure causes the rupture of the film. This stratification phenomenon is quantified by the disjoining pressure isotherm. In this
3992 J. Phys. Chem. B, Vol. 112, No. 13, 2008
Maldonado-Valderrama and Langevin
Figure 3. Images of the film drainage and disjoining pressure isotherm of Tween 20 (40 mol/m3).
Figure 4. Images of the film drainage and disjoining pressure isotherm of whole casein, 5 × 10-3mol/m3.
isotherm, one can distinguish four different branches separated by approximately 11 nm. This distance agrees with earlier results for Tween 20 solutions and equals the diameter of a Tween 20 micelle.20,21 The stratification observed reflects the sequential expulsion of micelles of Tween 20 trapped in the foam film. Also, the final thickness of the film is of (11 ( 3) nm, suggesting the existence of a Newton black film, covered by two surfactant monolayers. Figure 4 illustrates the drainage of a foam film stabilized by whole casein as well as the disjoining pressure isotherm measured for a concentration of protein of 1 g/L. Foam films stabilized by proteins usually show aggregates that remain immobile within the film and gradually disappear with increasing pressure.18 Aggregates are the small thick regions, i.e., light regions, which can be seen in Figure 4. Contrary to that found for Tween 20 films, the thickness of the whole casein film diminishes in a continuous manner until reaching a finite final thickness. Because of the lack of studies of protein stabilized thin liquid films, the confinement of proteins in thin liquid films is somehow unclear. However, Casc¸ ao-Pereira et al. have recently performed a very complete study of the drainage of thin films of β-casein solutions under comparable conditions to the ones used herein.6 These authors state that protein films might undergo a stepwise drainage depending on the pH and on the ionic strength of the media; hence, on the net charge of the protein. In particular, if the conditions favor the electrostatic repulsion between proteins, the drainage process is continuous. Conversely, at a pH near the isoelectric point of the protein and/or high ionic strength, conditions under which the electrostatic repulsion might be screened in the system, the protein films show a stratified drainage process. In this case, the authors relate the jumps in the disjoining pressure isotherm to the expulsion of a layer of β-casein molecules.6 Accordingly, the continuous drainage shown in Figure 4 for whole casein is
probably due to an electrostatic repulsion between the protein layers at the surfaces of the foam film. In addition, the final thickness of the protein film shown in Figure 4 equals (16 ( 4) nm. According to literature results, the thickness of a monomolecular layer of β-casein is of 10 nm.6,21 Casc¸ ao-Pereira et al. report a value of 26 nm for the β-casein films showing homogeneous drainage, stating that the final film is formed by two layers of β-casein separated by a layer of water, the thickness of which is determined by the electrostatic repulsion between protein layers.6 Accordingly, the final thickness shown by the whole casein film in Figure 4 indicates the existence of a different structure within the foam film. On one hand, our experimental conditions are not exactly the same as that used by Casc¸ ao-Pereira et al. Namely, those authors work at a pH 9 and 10 mM NaCl whereas the foam films displayed in Figure 4 were recorded at pH 7.4 and ionic strength 16.4 mM. In this sense, Grigoriev et al. measure a smaller adsorption layer thickness, using ellipsometry, for β-casein at conditions rather similar to ours (pH 7 and 10 mM NaCl).22 Accordingly, in our case the electrostatic repulsion would be lower and the water layer thinner. On the other hand, the differences between caseins could well play an important role in the behavior in thin liquid films. According to the literature, κ-casein is the fraction that diffuses more rapidly to the surface,23 the less charged (-2.5 e at pH 6.6) and the smallest (9 nm diameter).19 Also, it contains disulfide bridges, and so does the R2-casein fraction. Therefore, these fractions are able to form supramolecular structures.23 Accordingly, the liquid film formed by caseins might present heterogeneities. In summary, the two layers of whole casein present in the film would tend to come nearer than those reported for β-casein at pH 9 and 10 mM salt, and this phenomenon can be explained in terms of a lower electrostatic repulsion between
Foam Stability
Figure 5. Image of a mixed foam film of whole casein (5 × 10-3mol/m3) and Tween 20 (1 mol/m3).
layers and a higher intermolecular interaction between some fractions of the caseins. Finally, we describe the drainage of thin films formed by mixed systems. Mixtures were made with a fixed concentration of protein and varying that of Tween 20, as was done in the surface characterization. As a general observation, the drainage behavior appears to be very dependent on the concentration of surfactant in the mixture. Figure 5 illustrates a single foam film in which the coexistence of the two species within the film can be easily observed. In fact, the films made from mixed solutions show an intermediate behavior compared to that of pure systems. The drainage of the mixed film more resembles that of the sole surfactant, as the concentration of Tween 20 increases in the mixture. Hence, nucleation of dark holes and a stratified drainage appears at sufficiently high concentrations of surfactant in the foam film. This intermediate behavior as well as the transition to the behavior of a pure surfactant foam film has been also observed by Wilde et al. in mixed films of BSA/ Tween 20.20 We now evaluate in further detail the drainage of thin films made from mixtures by analyzing the disjoining pressure isotherms. Figure 6 shows the disjoining pressure isotherms recorded for two representative mixed systems with the same amount of protein (5 × 10-3 mol/m3): one with a concentration of Tween 20 (10-2 mol/m3) below but of the same order of magnitude of its cmc and one with a concentration of surfactant (1 mol/m3) well above its cmc. Figure 6 shows how, in the first case, the film undergoes a monotonous drainage, and the final thickness of the film has a smaller value than that of the pure protein film, (10 ( 4) nm, but close to that of the pure surfactant film. At this concentration of surfactant, there are no micelles in the solution. Since the mixed monolayers contain both surfactant and protein, this suggests that the film is made of a double layer of surfactant molecules with incorporated proteins. In order to obtain a better understanding of the electrostatic interactions in these films, we have fitted the disjoining pressure Π curves with exponential functions of the film thickness h: Πelec ∼ exp(-κh), κ-1 being the Debye length. As it is described in detail in ref 16, one may obtain the value of an exponential decay length from the disjoining pressure isotherms which can be compared with the Debye length calculated with the actual ionic strength. Hence, the experimental data in Figure 4 provides a decay length of 6.3 nm for the whole casein foam film. This value is larger than the Debye length which is 2.5 nm. Accordingly, there should be an additional force that reduces the pure electrostatic repulsion and makes the repulsion weaker than that corresponding to pure electrostatic forces. An additional attractive interaction, possibly due to bridging of the
J. Phys. Chem. B, Vol. 112, No. 13, 2008 3993 protein molecules across the foam film, could produce this effect. An additional repulsive steric force, dominating over electrostatic repulsion and longer ranged, could also be present. In the presence of surfactant (Figure 6 (left)), the calculated decay length is 2.9 nm, quite similar to that indicated by the ionic strength. Accordingly, the presence of Tween 20 within the foam film seems to compact the protein, possibly preventing bridging between protein molecules or suppressing steric repulsive forces. We now evaluate the effect of increasing the concentration of Tween 20 in the mixed foam film. Figure 6 (right) shows the disjoining pressure isotherm of the mixed systems containing a high concentration of Tween 20 (above its cmc). Figure 6 (right) shows the appearance of oscillations in the isotherms, quantifying the stratification observed in Figure 5. Taking into account the absence of oscillations in the caseins films and the fact that the jumps coincide with the diameter of a Tween 20 micelle, the stratification phenomena seen in these mixed films should be due to the sequential expulsion of layers of micelles. Besides, the final thickness of these mixtures is (15 ( 2) nm. This value is larger than that obtained at lower surfactant concentration, as well as for pure Tween 20. The similarity of this value with that corresponding to pure whole casein films might indicate that the film consists mainly of a bilayer of protein, but in view of the surface tension data, surfactant is certainly also present in the film. In summary, increasing the concentration of surfactant in the whole casein/Tween 20 system substantially increases the final thickness of the foam film. This feature might well have an effect on the stability of such films, and this is analyzed in detail below. 3.3. Foam Stability. This final section is dedicated to study the behavior of the foam formed by whole casein, by Tween 20, and by the mixtures of composition equal to those used in the study of surface properties and of thin liquid films. The ultimate aim is to correlate the previous fundamental properties with the stability of the foam. The stability of the foams formed by each of the systems studied in this work was characterized by their half-lifetime. Figure 7 shows the half-lifetime of the foam formed by the pure Tween 20 compared to that of the foam of the mixture with whole casein and an increasing concentration of Tween 20. The effect of Tween 20 on the stability of the foam formed by whole casein’s major fraction, β-casein, is also shown. The stability of the foam formed by pure Tween 20 under these experimental conditions is described in detail in ref 3. In short, Tween 20 produces poorly stable foam only above its cmc. Regarding the whole casein, it was evaluated only at the fixed concentration also used in the mixtures, similar to that of the rest of the experiments performed. Although the lifetime of the whole casein foam is larger than that of Tween 20, it appears much lower that that reported for its major fraction in previous work.3 Namely, β-casein foam shows a half-lifetime of ∼2000 s whereas whole casein’s foam half-lifetime is only 850 s. This difference might have its origin in the difference in film thickness reported in the previous section. Accordingly, the larger thickness reported for β-casein films would definitely protect the foam bubbles against collapse, playing a crucial role in the stability of the subsequent foam. Moreover, the slightly lower surface activity displayed by whole casein as compared with β-casein (Figure 1) also accounts for this feature. Regarding the effect of Tween 20 on the foaming properties of whole casein, Figure 7 shows the evolution of the half-lifetime of the foam of the mixed system as the concentration of Tween
3994 J. Phys. Chem. B, Vol. 112, No. 13, 2008
Maldonado-Valderrama and Langevin
Figure 6. Disjoining pressure isotherms of mixed systems composed of whole casein (5 × 10-3mol/m3) and Tween 20.
Figure 7. Half-life of the foam formed by Tween 20 (solid line), whole casein+Tween 20 (dashed line), and β-casein+Tween 20 (dotted line).
20 is increased in the mixture. Figure 7 also shows the effect of Tween 20 on the stability of β-casein foam which decreases with increasing concentration of surfactant. Conversely, the effect of addition of Tween 20 to the whole casein solution decreases the stability of the foam at low concentrations only. Above the cmc of the surfactant in the mixture, the half-lifetime of the foam increases and reaches a value practically equal to that corresponding to the pure caseins solution. 4. Discussion The set of experiments performed in this work provide understanding of the mechanisms underlying the stability of foams. It can be clearly seen in Figure 7 that the effect of Tween 20 in the foaming behavior of whole casein and its major fraction β-casein differ significantly. The surface pressure isotherms of the whole casein/Tween 20 and β-casein/Tween 20 mixtures reported in Figure 2 already indicate different surface behaviors. Moreover, the displacement of β-casein by Tween 20 inferred from its surface behavior is perfectly reproduced in the foaming behavior as was discussed in detail by Maldonado-Valderrama et al.3 The surface behavior of the whole casein/Tween 20 mixture suggests a different kind of interaction occurring between these two species (Figure 2). Furthermore, the different foaming behavior of the same mixture reported in Figure 7 clearly corroborates the existence of different interactions. However, the surface behavior proves insufficient to understand the foam behavior. The study of thin liquid films of the pure and mixed systems proves to be extremely useful in the further understanding of the foam stability. The values of the Debye length obtained in the presence and absence of Tween 20 within the protein foam film already predicted a major effect in the interactions between molecules. The disappearance of the bridging between protein molecules due to the presence of Tween 20 might well cause a decrease
in the stability of the foam in agreement with Figure 7. Furthermore, Figure 6 shows how the thickness of the film formed by whole casein and Tween 20 at a concentration below its cmc decreases substantially as compared to that of films of pure caseins. Moreover, further increasing the concentration of Tween 20 in the mixture yields an increase in the final film thickness that even approaches the value of pure caseins’ single foam film. The evolution of the stability of the foam with increasing concentrations of Tween 20 in the whole casein solutions shows a completely similar tendency. Furthermore, although such connection is not clearly stated, literature studies combining thin liquid films and foam stability measurements show parallel increases of film thickness and foam stability. Hence, Rippner et al. show thicker films of β-lactoglobulin with an increasing amount of polymeric surfactant together with a slight increase in foam stability.7 Similarly, Alahverdjiev et al. show increasing thickness of β-lactoglobulin-ionic surfactant thin films along with increasing foam stability.9 Accordingly, the thickness of the foam films seems to play a fundamental role in foam stability, and a direct correlation can be extracted from the analysis of the data and from literature results. Namely, the thinner the final foam film, the less stable the foam. Many literature studies point toward the key role also played by the viscoelastic properties of surfaces as regards the foaming behavior.4,24 In this sense, analysis of the surface rheology of the system formed by whole casein and Tween 20 would certainly aid in the understanding of the foaming behavior; however, no results have been found in the literature so far. However, there is literature regarding the dilatational viscoelasticity of BSA/Tween 20 systems under similar experimental conditions.24 Besides, Benjamins et al. report several similarities between bovine serum albumin (BSA) and whole casein with respect to their dilatational viscoelasticity, and there also appear some similarities between the drainage behavior of whole casein/ Tween 20 systems and that reported by Wilde et al. for BSA/ Tween 20 systems. Accordingly, the surface viscoelasticity of BSA/Tween 20 mixtures might highlight some important issues on the system considered herein. In this sense, Rodriguez-Nino et al. report a minimum in the surface elasticity of the mixed system at a concentration of Tween 20 in the mixture just below its cmc.25 At large protein/surfactant ratio, the protein-protein interactions are predominant in the mixed films, and hence the surface viscolasticy is large and equals that of the protein. At smaller protein/surfactant ratio, but in the absence of micelles in the bulk, the behavior upon surface expansion changes: the surfactant molecules might rapidly adsorb at the surface and disrupt the protein surface layer, significantly reducing its surface viscoelasticity. Hence, the viscoelasticity shows a minimum when the concentration of single surfactant molecules is maximum (Tween 20 concentration on the same order of
Foam Stability magnitude but lower than the cmc). Above the cmc of the surfactant, the micelles must be broken into the monomers for adsorption upon surface area deformation.26 If the micelles in solution are very stable, they cannot rapidly provide surfactant monomers to the newly created surface. Therefore, the increase in viscoelasticity once the concentration of Tween 20 in the mixture overcomes the cmc from ref 25 would be due to the protein network. It is interesting to note the correlation between surface viscoelasticity and foam stability. Differences between whole casein and β-casein have been generally attributed to the presence of κ-casein among the former. κ-Casein is a more compact protein that was believed to further resist the action of Tween 20. Recently, Woodward et al. have provided evidence of such a claim using atomic force microscopy imaging.27 Precisely, the authors image the displacement of all the fractions composing whole casein by the non-ionic surfactant Tween 20, showing how the minor component (κ-casein) is responsible for the final failure of the sodium caseinate interfacial film when displaced by Tween 20. This result is completely consistent with the findings reported in this work and could account for the encountered phenomena. Hence, in view of the stability of foam of whole casein as compared to β-casein, the latter does not seem to contribute to the overall stability of the caseins that seems to be mainly governed by κ-casein. This could be the reason why, although Tween 20 easily displaces β-casein from the surface, this does not affect the overall stability of whole casein since κ-casein is more resistant to its displacement and remains in control of the stability of the system in the presence of surfactant. 5. Conclusions The ultimate aim of this research work is to understand the effect of a low molecular weight surfactant (Tween 20) on the stability of the foam formed by a commercial protein (whole casein) on the basis of the structural properties of the mixed system at the air-water interface. For this purpose, its surface behavior upon adsorption and its confinement in thin liquid films is investigated along with the stability of the foam. The surface properties are analyzed by means of the surface pressure isotherms of the individual and mixed systems. The experimental data already foresee a complex behavior by comparing the results for whole casein/ Tween 20 with that of β-casein/Tween 20 reported in a previous study. Concretely, in the latter system, the surfactant was found to displace the protein, and the cac of the mixture equals the cmc of the surfactant. The whole casein/ Tween 20 mixture displays a different behavior characterized by a cac lower than the cmc of the surfactant. Accordingly, the protein/surfactant interaction involves the different fractions composing the whole casein. In order to further characterize the structural properties of the whole casein/Tween 20 mixture, its confinement in thin liquid films was studied by looking into the drainage of pure protein foam films, pure surfactant films, and films made with different mixtures. These experiments provide valuable structural information on the surface layers. As regards the whole casein, the thin films appear substantially thinner as compared with β-casein foam films. Interestingly, this thinner structure provided by whole casein probably accounts for the significantly lower half-lifetime encountered for the foam formed by whole casein as compared to that of its major fraction. As regards the single foam films made by whole casein/Tween 20 mixtures, the structures are strongly dependent upon surfactant concentration. Thus, the equilibrium thickness and the resulting stability of the film are significantly diminished at lower concentrations but remain similar to surfactant foams
J. Phys. Chem. B, Vol. 112, No. 13, 2008 3995 above cmc. Conversely, as the concentration of surfactant in the mixture increases, the film thickness and stability appears higher than that found for the sole surfactant at concentrations above its cmc. Once more, the final thickness of the film correlates with the foam stability found for the same systems. Accordingly, the half-lifetime of the mixed foam displays a minimum located precisely at the concentration that provides the thinner foam film, giving a unique direct correlation between film thickness and foam stability. Finally, the overall behavior of the system is also discussed on the basis of literature results, and an explanation of the phenomena is finally proposed. Tween 20 displaces β-casein from the surface, and this feature governs the stability of the foam of β-casein/Tween 20 mixtures. But whole casein appears to be more resistant to the displacement, and this feature is clearly reflected in the foam stability of whole casein/Tween 20 mixtures. The reason for this might be related to the more compact structure of κ-casein and its resistance against displacement, also suggesting that this fraction governs the foam stability of whole casein. This work highlights the important differences of surface behavior of mixed systems depending of the nature of the components and its consequences on foam stability. The conclusions were reached by linking the behavior of the same system over various length scales, a procedure which appears as a valuable approach for the understanding of complex systems with important applications. Acknowledgment. Financial support from “Plan Propio de la Universidad de Granada” and “Ministerio de educacio´n y Ciencia, plan nacional de Investigacio´n cientı´fica, desarrollo e Innovacio´n Tecnolo´gica (I+D+I); project MAT2007-66662C02-01”. Dr. Francisco Monroy and Dr. Arnaud Saint-Jalmes are gratefully acknowledged for their support and fruitful discussions. References and Notes (1) Mackie, A.; Wilde, P. AdV. Colloid Interface Sci. 2005, 117, 3-13. (2) Murray, B. S. Curr. Opin. Colloid Interface Sci. 2007, 12, 232241. (3) Maldonado-Valderrama, J.; Martı´n-Molina, A.; Martı´n-Rodrı´guez, A.; Cabrerizo-Vı´lchez, M. A.; Ga´lvez-Ruiz, M. J.; Langevin, D. J. Phys. Chem. C 2007, 111, 2715-2723. (4) Langevin, D. AdV. Colloid Interface Sci. 2000, 88, 209-222. (5) Yampolskaya, G.; Platikanov, D. AdV. Colloid Interface Sci. 2006, 128-130, 159-183. (6) Casc¸ ao-Pereira, L. G.; Johanson, C.; Radke, C. J.; Blanch, H. W. Langmuir 2003, 19, 7503-7513. (7) Rippner Blomqvist, B.; Ridout, M. J.; Mackie, A. R.; Warnheim, T.; Claesson, P. M.; Wilde, P. J. Langmuir 2004, 20, 10150-10158. (8) Kharlov, A. E.; Filatova, L. Y.; Zadymova, N. M.; Yampol’skaya, G. P. Colloid J. 2007, 69, 117-123-130. (9) Alahverdjiev, V. S.; Khristov, K.; Exerowa, D.; Miller, R. Colloids Surf., A: 2007, doi:10.1016/j.colsurfa.2007.09.026. (10) Cabrerizo-Vı´lchez, M. A.; Wege, H. A; Holgado-Terriza, J. A.; Neumann, A. W. ReV. Sci. Instrum. 1999, 70, 2438-2443. (11) Exerowa, D.; Kolarov, T.; Khristov, Khr. Colloids Surf. 1987, 22, 161-169. (12) Mysels, K. J.; Jones, M. N. Discuss. Faraday. Soc. 1966, 42, 4250. (13) Bergeron, V.; Langevin, D.; Asnacios, A. Langmuir 1996, 12, 1550-1556. (14) Scheludko, A. AdV. Colloid Interface Sci. 1967, 1, 391-464. (15) Ma´rquez-Beltra´n, C.; Guillot, S.; Langevin, D. Macromolecules 2003, 36, 8506-8512. (16) Ma´rquez-Beltra´n, C.; Langevin, D. J. Colloid Interface Sci. 2007, 312, 47-51. (17) Bhattacharyya, A.; Monroy, F.; Langevin, D.; Argillier, J. F. Langmuir 2000, 16, 8727-8732. (18) Saint-Jalmes, A.; Peugeot, M. L.; Ferraz, H.; Langevin, D. Colloids Surf., A: 2005, 263, 219-225. (19) Benjamins, J. Ph.D. Thesis, Wageningen University, The Netherlands 2001.
3996 J. Phys. Chem. B, Vol. 112, No. 13, 2008 (20) Wilde, P.; Rodrı´guez-Nin˜o, M. R.; Clark, Rodrı´guez-Patino, J. M. Langmuir 1997, 13, 7151-7157. (21) Dimitrova, T. D.; Leal-Caldero´n, F.; Gurkov, T. D.; Campbell, B. Langmuir 2001, 17, 8069-8077. (22) Grigoriev, D. O.; Fainerman, V. B.; Makievski, A. V.; Kragel, J.; Wustneck, R.; Miller, R. J. Colloid Interface Sci. 2002, 253, 257-264. (23) Turhan, K. N.; Barbano, D. M.; Etzel, M. R. J. Food Sci. 2003, 68, 1578-1583.
Maldonado-Valderrama and Langevin (24) Maldonado-Valderrama, J.; Martı´n-Rodriguez, A.; Ga´lvez-Ruiz, M. J.; Miller, R.; Langevin, D.; Cabrerizo-Vı´lchez, M. A. Colloids Surf., A: 2007, doi:10.1016/j.colsurfa.2007.11.003. (25) Rodriguez-Nin˜o, M. R; Wilde, P.; Clark, D. C.; Rodriguez-Patino, J. M. J. Agric. Food Chem. 1998, 46, 2177-2184. (26) Oh, S. G; Shah, D. O. Langmuir 1991, 7, 1316-1318. (27) Woodward, N.; Gunning, A. P.; Mackie, A. R.; Wilde, P. J.; Morris, V. C., submitted.
