Carotenoids as End-Cap-Active Agents in Lecithin Cylindrical Micelles...
0 downloads
77 Views
64KB Size
3026
Langmuir 1999, 15, 3026-3028
Carotenoids as End-Cap-Active Agents in Lecithin Cylindrical Micelles P. A. Cirkel, M. Fontana, and G. J. M. Koper* Leiden Institute of Chemistry, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlands Received June 25, 1998. In Final Form: November 10, 1998 Lecithin is able to form water-in-oil microemulsions with polymer-like behavior. In most experiments on this system lecithin from natural sources (soybean, egg yolk) has been used that contains trace amounts of carotenoids. In samples in which the polymer-like phase coexists with one or two other phases, these carotenoids are only present in the polymer-like phase. The addition of small amounts (≈1%) of β-carotene to a sample in which the lecithin is disposed of carotenoids has a drastic effect on the viscoelastic properties. These observations lead to the conclusion that in this system carotenoids are likely to specifically adsorb at the end caps and work as “end-cap-active agents”, which decrease the end-cap energy.
Introduction Mixtures of lecithin and water are able to form cylindrical micelles in various organic solvents. Although most experiments described in the literature involved soybean lecithin, there are some reports of experiments with lecithin from egg yolk and synthetic lecithin, indicating that the source of the lecithin is an important factor determining the formation and structure of cylindrical micelles.1 This is not surprising considering the fact that natural lecithin consists of a mixture of surfactants with different chain lengths and degrees of saturation. Although usually experiments are performed with lecithin of high purity (96%), impurities might play an important role as well. This is particularly the case when such impurities preferentially adsorb at the end caps of the micelles. When, in such a situation, the amount of impurity is comparable to the amount of surfactant in the end caps (approximately 1% of the total amount), the endcap energy is significantly affected, which in turn influences the length distribution of cylindrical micelles.2 This idea is in close analogy with the Gibbs adsorption equation for the change in surface tension by surfactants: If a species is adsorbed at the end cap (surface), it gives rise to a decrease in end-cap energy (surface tension). In the case of cylindrical micelles the species can be called an end-cap-active agent instead of a surface-active agent (surfactant). In this paper we will show that carotenoids, which are normally present in commercially available soybean and egg-yolk lecithin, are likely to be such end-cap-active agents. The first indication for this is found in the fact that samples in which a cylindrical micellar phase coexists with another phase, only the cylindrical micellar phase displays the yellow color that is typical for the carotenoid impurity. This inspired us to purify the lecithin and add controlled amounts of β-carotene, upon which we found a strong effect on the viscosity. Experimental Section Materials. Isooctane was of analytical grade and purchased from Fluka. Water was filtered by a Millipore apparatus before it was used to make gels. Synthetic trans-β-carotene was purchased from Aldrich. Soybean lecithin was purchased from (1) Schurtenberger, P.; Peng, Q.; Leser, M. E.; Luisi, P. L. J. Colloid Interfaces Sci. 1993, 156, 43. (2) Israelachvili, J.; Mitchel, D. J.; Ninham, B. H. J. Chem. Soc. Faraday 2 1976, 72, 1525; Mitchell, D. J.; Ninham, B. H. J. Chem. Soc. Faraday 2 1981, 77, 601.
Lucas Mayer, type Epikuron 200. According to their gascromatographic analysis, it has the following distribution of chain lengths, n, and degrees of saturation, m, to be denoted by (n, m): (16, 0) 14.24%; (18, 0) 3.74%; (18, 1) 10.45%; (18, 2) 63.66%; (18, 3) 7.08%; (24, 1) 0.33%. In order remove the carotenoids the following steps were performed in a cold room (-15 °C): 10 g of the waxy yellow lecithin was mixed with 100 mL of isooctane and allowed to stand overnight. The milky substance was centrifuged and the residue-containing lecithin was kept. This procedure was repeated three times and the white residue was freeze-dried and used as purified lecithin. The purification was checked by TLC (thin layer chromatography). Thin layer chromatograms were run on a silica gel on plastic roll, type 60F254 of Merck. The eluents used consisted of chloroform-methanol-water, 65:25:4 (v/ v). The spots were made visible with an UV lamp. The thin layer chromatogram shows that the purification did not seem to have a substantial impact on the lecitin composition except for the fact that the carotenoids were removed. That the carotenoids are removed is also very easily observed from the color, which was yellow for commercial lecithin and white for the purified sample. This process was also followed by UV/vis spectroscopy. Before purification the spectrum showed 3 peaks at 425, 450, and 476 nm, typical for the carotenoid impurity,3 whereas the purified sample did not display these peaks. Sample Preparation. The samples were prepared by adding controlled amounts of water to the appropriate lecithin isooctane mixture. The samples where identified by two parameters: The lecithin volume fraction, φ, and the molar ratio of added waterto-surfactant w0. From IR spectroscopy it was estimated that the natural lecithin already contained 0.7 mol of water/mol of lecithin, whereas the purified sample contained 0.4 mol of water/ mol of lecithin. For the samples to which β-carotene was added an extra parameter is introduced: δ, the molar ratio of β-carotene-tosurfactant. It is important to point out that it takes a long time for the β-carotene to dissolve. Reaching equilibrium typically takes several months. Photos of the Phase Behavior. The samples on the photographs were given 2 months to equilibrate. For the samples with two phases no change was observed after the first week, whereas for the samples with three phases more than 1 month was needed. The photos were taken against a blue background in order to get a high contrast with a yellow or white phase, respectively. Viscosity Measurements. Viscosity measurements were performed on a Haake RV20 rotoviscometer with a RC20 rheocontroller and a CV 100 measuring system. This is a couettetype rotational viscometer. The lowest shear rate at which measurements could be taken with this system was 0.01 s-1. (3) Young, A.; Britton, G. Carotenoids in Photosynthesis; Chapman & Hall: London, 1993; p 462.
10.1021/la980762w CCC: $18.00 © 1999 American Chemical Society Published on Web 04/01/1999
Letters
Langmuir, Vol. 15, No. 9, 1999 3027
Figure 1. Samples of lecithin solutions demonstrating the phase behavior at 6 °C and constant lecithin concentration of 60 mM. The two tubes on the right respectively contain a β-carotene solution in isooctane and a lecithin/isooctane solution without added water.
Figure 2. Samples of lecithin solutions demonstrating the phase behavior at 6 °C, a constant lecithin concentration of 30 mM, and a varying water-to-lecithin ratio w0. The arrows indicate the location of the phase boundary.
Results A phase diagram of soybean lecithin before purification at 25 °C has been published before.4 Some typical samples showing the phase behavior at 6 °C are shown in Figure 1. If no water is added to a sample of lecithin in isooctane, the system separates into a fluffy white powder and a clear yellow solution (Figure 1, tube 5). Upon the addition of water to the sample, a highly viscous clear yellow solution is obtained (Figure 1, tube 1). The further addition of water results in phase separation into a highly viscous clear yellow solution on the bottom and a clear inviscid colorless solution on the top (Figure 1, tubes 2 and 3). In Figure 2 the ratio of added water-to-surfactant (w0) is varied systematically. At the highest w0 the system separates into three phases and in between the two phases (4) Schurtenberger, P.; Scartazzini, R.; Magid, L. J.; Leser, M. E.; Luisi, P.L. J. Phys. Chem. 1990, 94, 3695.
Figure 3. Flow curves for samples containing 150 mM lecithin at a water-to-surfactant ratio of 1.5. The molar β-carotene-tolecithin ratios are δ ) 0 (2), δ ) 0.0315 (9), and δ ) 0.124 (b).
already discussed a white opaque phase of intermediate viscosity is formed. At a higher lecithin concentration all transitions are shifted to higher w0. Higher temperatures roughly induce the same phase behavior. The results of the viscosity measurements are shown in Figures 3 and 4. In Figure 3 the flow curves are displayed, showing the stress as a function of strain rate for different amounts of β-carotene added to a sample with fixed lecithin and water concentrations. It turns out that all samples containing β-carotene show Newtonian behavior, whereas in the sample without β-carotene tremendous shear thinning occurs. For the samples with Newtonian behavior the viscosity as a function of β-carotene content is shown in Figure 4. Apparently, the viscosity is inversely proportional to the β-carotene content. Discussion The results shown in this paper were all obtained at surfactant volume fractions which are far above the overlap threshold, such that in all cases the micelles build
3028 Langmuir, Vol. 15, No. 9, 1999
Figure 4. Zero-shear viscosity for samples containing 150 mM lecithin at a water-to-surfactant ratio of 1.5 with a varying molar β-carotene-to-lecithin ratio δ.
up some kind of network. Without the addition of water unpurified lecithin does not dissolve in isooctane. Although there is already 0.7 mol of water present, it takes a w0 ) 0.5 for a clear solution to be formed. This clear solution is highly viscous and this has been attributed to the presence of a network of entangled cylindrical micelles.1 As a function of w0 the system becomes more viscous, as a result of water-induced micellar growth. Several experiments suggest that above w0 ) 2 the micelles get branched.5,6 In the zero-shear viscosity this can be seen as a change in the exponent by which it scales with the surfactant concentration. At w0 > 3 it even affects the viscosity itself which starts to decrease as a function of w0. Above w0 ) 2 one also observes separation into two phases at low temperatures and/or low surfactant concentrations. Separation into a connected network phase and an excess solvent phase at low temperature and/or surfactant concentration has indeed been predicted by theory.7 The carotenoids are only present in the network phase. This means that there must be a very strong affinity of the carotenoids toward the network. At higher water content a third phase appears. This phase only appears after a rather long time, at least several days. Reaching full equilibrium takes more than a month. Before this time this phase is mixed with the yellow bottom phase, which is then opaque and becomes gradually more transparent as the white opaque phase is formed above it. This indicates that these two phases are very similar in density. It has been suggested that this middle phase is liquid crystalline since it would be birefringent.4 We were not able to confirm this because of the high scattering intensity. The addition of only small amounts of β-carotene has a marked effect on the viscoelastic properties of a solution of lecithin cylindrical micelles. If β-carotene is added, the solution behaves Newtonian and the viscosity is inversely proportional to the β-carotene content. This is confirmed by the fact that without the addition of β-carotene the zero-shear viscosity diverges and the solution does not behave Newtonian anymore. The stress in this case seems to reach a plateau value which already sets in at very low strain rates. It has been suggested for other viscoelastic surfactant solutions that this is caused by shear banding, that is, the coexistence of bands with a different structure (5) Cavaco, C.; Schurtenberger, P. Helv. Phys. Acta 1994, 67, 227. (6) Cirkel, P. A.; van der Ploeg, J. P. M.; Koper, G. J. M. Phys. Rev. E 1998, 57, 6875. (7) Drye, T. J.; Cates, M. E. J. Chem. Phys. 1992, 96, 1367.
Letters
at different distances from the wall.8 In this particular case this could be due to the formation of a connected network, which has to break up in order to flow. At this stage however it seems too ambigious to decide on a quantitative model for both this effect and the linear proportionality of the viscosity with the carotenoid content. A key to the understanding of the strong impact of trace amounts of carotenoids on these lecithin microemulsions lies in the partitioning of this impurity. Since β-carotene has roughly a factor of 1000 higher affinity for organic solvents than for water,9 the possibility that the carotenoids are inside the cylindrical micelles can be safely excluded. Also, it is highly improbable that the carotenoids reside in the continuous oil phase because in those instances where the microemulsion is in equilibrium with an excess oil phase the carotenoids appear to be only in the microemulsion phase. The conclusion must be that the carotenoids reside in the lecithin layer. The relative amount of β-carotene, however, is too small to drastically change the overall curvature of the oil/water interface. The observed effects are therefore most likely due to a partitioning of the carotenoids in the surfactant layer. In principle, there are two ways in which this could happen. One possibility is that there would be a structural transition such as that observed on the addition of trace amounts of vitamin K1 to the monolein/water system.10 In our case such transitions might occur at lower carotenoid additions as studied here and could explain the qualitatively different viscoelastic behavior in the system without any carotenoids. However, in this study we observe a gradual decrease in the viscosity upon the addition of β-carotene. Under conditions similar to those here, the existence of a worm-like micellar phase without branch points has been firmly established.11 For such a structure the observed decrease in the viscosity could be explained by a second possibility, that is, a specific adsorption of carotenoids at the end caps. This certainly will affect the end-cap energy and hence the size distribution of the cylindrical micelles. Also, the kinetics of micelle breaking and recombination may be affected by the locally adsorbed carotenoids. Both effects could explain the decrease in viscosity with β-carotene content. One of the remaining questions would be how this affects experimental results on natural lecithin without purification. Since the carotenoids in soybean lecithin probably mainly consist of luthein12 (carotene with two hydroxy groups on the ends), the ability to decrease the end-cap energy in lecithin cylindrical micelles is apparently general for carotenoids. Of course, to test the effect of all components individually, it would be desirable to have experimental data on mixtures with both synthetic lecithin and carotenoids. Acknowledgment. The authors thank Jan Groenewold for stimulating discussions. LA980762W (8) Cates, M. E. In Fundamental Problems in Statistical Mechanics VII; van Beijeren, H., Ernst, M. H., Eds.; Elsevier Science BV: Amsterdam, 1994. (9) Treszczanowicz, T.; Kasprzycka-Guttman, T.; Cieslak, D.; Treszczanowicz, A. J. J. Chem. Eng. Data 1998, 43, 632. (10) Caboi, F.; Nylander, T.; Razumas, V.; Talaikyte, Z.; Monduzzi, M.; Larsson, K. Langmuir 1997, 13, 5476. (11) Schurtenberger, P.; Cavaco, C. Langmuir 1994, 10, 100. (12) Monma, M.; Terao, J.; Ito, M.; Saito, M.; Chikuni Biosci, K. Biotechnol. Biochem. 1994, 58, 926.
