- Email: [email protected]
A study of the properties of mixed nonionic surfactants microemulsions by NMR, SAXS, viscosity and conductivity
Journal of Molecular Liquids 142 (2008) 103–110
Contents lists available at ScienceDirect
Journal of Molecular Liquids j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / m o l l i q
A study of the properties of mixed nonionic surfactants microemulsions by NMR, SAXS, viscosity and conductivity Monzer Fanun ⁎ Faculty of Science and Technology, Al-Quds University, P.O. Box 51000 East-Jerusalem, Palestine
A R T I C L E
I N F O
Article history: Received 4 February 2008 Received in revised form 9 May 2008 Accepted 16 May 2008 Available online 25 May 2008 Keywords: Surfactants mixture One-phase microemulsion Percolation threshold Periodicity of domain size Amphiphilicity factor Correlation length Microstructure of microemulsions Relative diffusion coefficients
A B S T R A C T The pseudoternary phase behavior of the water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)limonene systems was investigated for different surfactants mixing ratios (w/w) at 25 °C. The microemulsion boundaries were determined and the surfactants content at the interface of water- R (+)-limonene was estimated. For surfactants mixing ratio (w/w) equals unity, the area of the one phase microemulsion region reaches its maximum. The system with the maximum microemulsion area was investigated using electrical conductivity, dynamic viscosity, small angle X-ray scattering, and nuclear magnetic resonance. Electrical conductivity increases as the water volume fraction increases and a percolation threshold was observed. Dynamic Viscosity varies as function of the water volume fraction in a non-monotonic way giving twopeaked plot. The characteristics of the domain size of the microemulsions called periodicity measured by small angle X-ray scattering increases with the increase in the water volume fraction. The correlation length of the domain size reaches a maximum when plotted against the water volume fraction in the microemulsions. Relative diffusion coefficients of water increase and those of oil decrease with increasing the water volume fractions in the microemulsions indicating structural transitions. © 2008 Elsevier B.V. All rights reserved.
1. Introduction The formulation of nonionic, safe, biodegradable microemulsions with specific properties like high solubilization power and temperature insensitivity was focused in many researches [1,2]. The extensively reported [3,4] microemulsions physicochemical properties, which include thermodynamic stability, high solubilization power, low interfacial tensions, transparency, and low viscosity, are of great importance. Microemulsions are broadly used in food, cosmetic and pharmaceutical applications [5–8]. The surfactants used in the majority of the industrial applications almost consist of a mixture of surfactants. The tendency to form aggregates in systems containing mixed surfactants would be considerably different from that in systems based on single surfactant. The properties of a mixture of homologous surfactants belonging to the same class are simply predictable from the properties of the individual components. On the other hand, mixtures of surfactants formed from different classes usually show some deviation from ideal behavior. The incorporation of another surfactant into the mixture of water, oil and surfactant provides an additional degree of freedom, which enables the adjustment of phase behavior. Mixtures of nonionic surfactants are more effective than the other class of surfactants for the reason that they are more effective at lowering the monomer concentration and
⁎ Tel.: +972 522 40 60 61; fax: +972 22 79 69 60. E-mail addresses: [email protected], [email protected]. 0167-7322/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.molliq.2008.05.006
they are not susceptible to precipitation themselves [9–11]. Synergistic mixing properties of surfactants can only be achieved by understanding how surfactants interact with each other and how these interactions influence performance. Understanding the mixed surfactants behavior in the presence of water and oil is an important issue for the development and the optimal design of mixed surfactants based microemulsions for industrial applications [9]. Techniques used to study how changes in the relative amounts of the microemulsion components affect the microstructure of the solution in singlesurfactant systems also apply to mixed surfactant systems. These methods include electrical conductivity [12,13], dynamic viscosity [3,14], nuclear magnetic resonance [13,15,16], scattering methods [17,18], and imaging techniques [19,20]. In the present work, we evaluate the effect of adding ethoxylated mono-di-glyceride (EMDG) on the phase behavior and properties of the water/sucrose laurate/R (+)-limonene (W/L1695/LIM) system by exploring the following subjects: 1. Phase behavior of the single and mixed nonionic surfactants, in the presence of water and the biologically compatible R (+)-limonene oil, which permits the determination of the microemulsions phase regions. 2. Inter droplet interactions of the formulated microemulsions using electrical conductivity. 3. Microstructure transitions in the formulated microemulsions using dynamic viscosity, small angle X-ray scattering, and nuclear magnetic resonance.
104
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
2. Experimental 2.1. Materials The sucrose laurate (L1695) was obtained from Mitsubishi–Kasei Food Corp., (Mie, Japan) (see chemical structure in Fig. 1a). The purity of combined Lauric acid equals 95 %, the esters compositions are 80% monoester and 20% di, tri and polyester, its HLB equals 16. Ethoxylated mono-di-glyceride (EMDG) (MAZOL 80 MG KOSHER), its HLB equals 13.5. Ethoxylated mono-di-glyceride was obtained from BASF Corporation (Gurnee, Illinois, USA) (see chemical structure in Fig. 1b). R (+)-limonene (LIM), (98%) (see chemical structure in Fig. 1c) was purchased from Sigma Chemicals Co. (St. Louis, USA). Sodium chloride (NaCl) of analytical grade was purchased from J.T. Baker Inc. (Phillipsburg, USA). All of the components were used as supplied without further purification. Water was double distilled. 2.2. Methods 2.2.1. Pseudoternary phase diagrams at constant temperature The phase behavior of a system consisting of water, oil, surfactant (or mixture of surfactants) may be described on a phase tetrahedron whose apexes respectively represent the pure components. However, it is more convenient to describe the phase behavior on a pseudoternary phase triangles. Obviously, a fixed (weight, volume or mole) ratio must be chosen for any two of the components and one of the triangle vertices represents 100% of this binary mixture. Phase diagrams of such multi-component systems are generally based on constant ratios of surfactant-to-water or cosurfactant-to-surfactant [21–23]. Such a presentation enables us to follow directly and conveniently, the variation of the surfactant amount needed to solubilize a given amount of water. Mixtures at fixed weight ratios of oil phase and the surfactant or mixed surfactants were prepared in culture tubes sealed with Viton lined screw caps. Water was then added dropwise until its solubilization limit was reached. After this
Fig. 2. Schematic phase diagram presenting the solubilization parameters of mixed nonionic surfactants in water and oil at 25 °C. AT (%) is the area of the one phase microemulsions region. 1φ is the W/O, bicontinuous, and O/W microemulsions phase region; II(Wm + O) is the area of the two phase region that signifies water continuous micellar system with excess oil.
point, larger increments of aqueous phase were added. All of the aqueous phase additions were followed by vigorous stirring on a vortex mixer. The time for equilibration between additions of successive aliquots was typically, from a few minutes up to 24 h. Phase transitions detected visually by the appearance of cloudiness or sharply defined separated phases. The completion of this process was hastened by centrifuging the samples. The phase diagrams were determined at 25 ± 0.5°C. 2.2.2. Water solubilization parameter The water solubilization capacity of different amphiphilic systems should be compared at optimal conditions [24,25]. Li et al. [23] have employed as a solubilization parameter the area of the one-phase microemulsion region (AT) that is area limited by the microemulsification failure boundaries. In this study we used the one phase microemulsion region to compare the water solubilization capacity in the studied systems. The relative error in determining the AT (%) was estimated to be ±3% for all of the systems studied. Fig. 2 present a schematic phase diagram where we indicated the one phase region, the microemulsification failure and the multiple phase regions. 2.2.3. Electrical conductivity measurements Conductivity measurements were performed at 25 ± 0.5 °C on samples the compositions of which lie along the one phase channel, using conductivity meter, the conductivity cell used is Tetra Con® 325, the electrode material is graphite and the cell constant is 0.475 cm− 1 ± 1.5%. The range of application is between 1 µS/cm to 2 S/cm with an accuracy of ±0.5%, and the temperature range is from −5 to 100 °C. In the case of nonionic microemulsions, a small amount of an aqueous electrolyte must be added for electrical conduction [26]. Thus, a 0.01 M sodium chloride aqueous solution was used in the preparation of the microemulsion samples in place of pure water. The electrode was dipped in the microemulsion sample until equilibrium was reached and reading becomes stable. Reproducibility was checked for certain samples and no significant differences where observed. The constant of the conductivity cell was calibrated using standard KCl solutions and checked a minimum of three times during the course of the working shift.
Fig. 1. The chemical structures of [a] sucrose laurate [b] ethoxylated mono-di-glyceride (average structure) and [c] R (+)-limonene.
2.2.4. Viscosity measurements Viscosity was measured using a rotational viscometer, model DV1PL spindle from Anton Paar GmbH (Graz, Austria). “Double cylinder” geometry was used. Viscosities at 200 s− 1 shear rate were obtained at
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
25 ± 0.5 °C. Reproducibility (triplicate) was checked for the samples and no significant differences (±SD) where observed. 2.2.5. Small angle X-ray scattering (saxs) Scattering experiments were performed using Ni-filtered Cu Kα radiation (0.154 nm) from an Eliott GX6 rotating X-ray generator that operated at a power rating up to 1.2 kW. X-radiation was further monochromated and collimated by a single Franks mirror and a series of slits and height limits and measured by a linear position-sensitive detector. The sample was inserted into 1–1.5 mm quartz or lithium glass capillaries, which were then flame-sealed. Each sample was checked before and after the experiment to verify that, no fluid had been lost during the time of exposure, approximately 3 h. The temperature was maintained at 25 ± 1 °C. The sample-to-detector distance was 0.46 m, and the scattering patterns were measured using the Lake procedure [27]. 2.2.6. X-ray data analysis In this case, the scattering patterns after background subtraction were fit to the expression due to Teubner and Strey [18]: IðqÞ ¼ 1=a2 þ c1 q2 þ c2 q4 þ b
ð1Þ
With the constants a2, c1, c2 and b obtained by using the Levenburg–Marquardt procedure [28]. Such a functional form is simple and convenient for the fitting of spectra. Eq. (2) corresponds to a real space correlation function of the form
γ ðrÞ ¼ ðsin kr=kr Þe−r=
ð2Þ
The correlation function describes a structure with periodicity d = (2π/k) damped as a function of correlation length ξ. This formalism also predicts the surface to volume ratio, but because this ratio is inversely related to the correlation length and therefore must go to zero for a perfectly ordered system, calculated values are frequently found to be too low [29]. d and ξ are related to the constants in Eq. (1) by [18]: h i−1=2 d ¼ ð1=2Þðða2 =c2 ÞÞ1=2 −ðc1 =4c2 Þ
ð3Þ
h i−1=2 ¼ ð1=2Þðða2 =c2 ÞÞ1=2 þðc1 =4c2 Þ
ð4Þ
A third parameter, which can also be defined, is f a, the amphiphilicity factor [18,30–34], which relates to the behavior of the correlation function γ(r) and reflects the ability of the surfactant to impose order on the microemulsion: fa ¼ c1 =½4a2c2 1=2
105
gradient pulse length, Δ is the time between the two gradients in the pulse sequence (and hence defines the diffusion time). Typically, we use Δ = 100 ms, δ = 8 ms, and vary G from 1.7 to 32.3 G cm− 1 in 32 steps. 3. Results and discussion 3.1. Phase behavior A number of studies [37–39] reported on the investigation of sugar based surfactants microemulsions indicate the need of adding a cosurfactant or another surfactant to the sucrose esters in order to obtain better mutual solubilization of oil and water. In our previously published work [40] we studied the phase behavior of the water/ sucrose laurate/ethoxylated mono-diglyceride/R(+)-limonene systems as function of temperature where we specified the phase regions with labels. Two phase regions were observed designated by II (Wm + O) or (W + Om) and the three phase regions designated by III (W + D + O). The II phase (Wm + O) signifies water continuous micellar system with excess oil and (W + Om) means excess water with oil continuous micellar system. III phase (W + D + O) means a water-rich (W, lower), amphiphile-rich (D, middle), and an oil-rich (O, upper) phases. In this study, the pseudoternary phase diagrams at constant temperature equal to 25 °C of the water/sucrose laurate/ethoxylated-mono-diglyceride/R (+)-limonene system was explored. It indicates the presence of an isotropic and low-viscosity area which is a microemulsion one phase region (1φ), the remainder of the phase diagram represents a two phase region composed of water continuous micellar solution with excess oil designated by (Wm + O). A small microemulsion phase region was observed (i.e. the onephase microemulsion area (AT) equals 5%) in the ternary system water/sucrose laurate/R (+)-limonene. Adding ethoxylated-mono-diglyceride to the ternary system enhances the formation of the one phase microemulsion. For ethoxylated-mono-di-glyceride content in the mixture of (ethoxylated-mono-di-glyceride + sucrose laurate) equals 25 wt.%, the total one phase microemulsion area (AT) rises to 22%. At equal amounts of ethoxylated-mono-di-glyceride and sucrose laurate in the surfactants mixture (i.e. surfactants mixing ratio (w/w) equals unity), the one phase microemulsion region appears from the first drop of water added. This one phase region extends overall the water contents range along the dilution line N60 and the its area (AT) equals 62% (see Fig. 3). The one-phase microemulsion area (AT) decreases to the value of 47% by increasing the ethoxylated-mono-di-glyceride weight ratio in
ð5Þ
2.2.7. Pulsed gradient spin echo-nuclear magnetic resonance (PGSE-NMR) NMR measurements was performed on Bruker DRX-400 spectrometer with a BGU II [35,36] gradient amplifier unit and a 5-mm BBI probe equipped with a z-gradient coil, providing a z-gradient strength (G) of up to 55 G cm− 1. The self-diffusion coefficients were determined using bipolar-pulsed field gradient stimulated spin-echo (BPFG-SSE). In this work, we used bipolar gradient pulses as described by Wu et al. [35] to reduce the eddy-current effects. Experiments were carried out by varying the gradient strength and keeping all other timing parameters constant. The self-diffusion coefficient (D) is given by δ I ¼ Io exp γ2 G2 δ2 Δ− D 3
ð6Þ
Where I is the measured signal intensity, Io is the signal intensity for G = 0, γ is the gyro magnetic ratio for the 1H nucleus, δ is the
Fig. 3. Pseudoternary phase behavior of the system water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity. The one phase region is designated by 1ϕ, and the two phase region consisted of water continuous micellar solution with excess oil is designated by (Wm + O). N60 is the water dilution line where the weight ratios of sucrose laurate/ethoxylated mono-di-glyceride/R(+)-limonene equal to 3/3/4.
106
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
the mixed surfactants to 75 wt.%. In the ternary system water/ ethoxylated mono-di-glyceride/R (+)-limonene, the one-phase microemulsion area (AT) decreases to the value of 30%. It is considered that surfactant monolayers at the interface of water and R (+)-limonene domains inside the microemulsions are directly related to the solubilization of water and R (+)-limonene. The monodisperse solubilities of sucrose laurate and ethoxylated mono-di-glyceride in R (+)-limonene are very small [37]. This means that the R (+)-limonene domains in the microemulsion phase are almost the same as bulk R (+)limonene phase when solubilization is large. It can be assumed that the monomeric solubilities of sucrose laurate (SL1695) and that of ethoxylated mono-di-glyceride (SEMDG) in the water phase forming the microemulsions are similar to their respective critical micelle concentrations (CMC) which equal 3.4 ⁎ 10− 4 and 1.1 ⁎ 10− 5 M for sucrose laurate and ethoxylated mono-di-glyceride, respectively. Since surfactant molecules at the water-R(+)-limonene interface inside microemulsions are directly related to the solubilization, it is important to estimate the mixing fraction of each surfactant. The surfactant content at interface could be obtained by simple mass balance equations as follows:
min CEMDG ¼ X min XEMDG −
1−X min SEMDG 2ð1−SL1695 −SEMDG Þ
ð7Þ
and 1−X min S L1695 min − CL1695 ¼ X min 1−XEMDG 2ð1−SL1695 −SEMDG Þ
ð8Þ
where CL1695 and CEMDG indicate the weight of sucrose laurate and ethoxylated mono-di-glyceride at the water-R (+)-limonene interface, Xmin is the minimum weight fraction of mixed surfactants capable of solubilizing equal amounts of water and R(+)-limonene in the microemulsions, Xmin EMDG is the minimum weight fraction of the lipophilic surfactant (in our case the ethoxylated mono-di-glyceride) corresponding to Xmin. CL1695 + CEMDG is the weight fraction of total surfactants in surfactants monolayer at the water-R(+)-limonene interface inside the mixed surfactants microemulsion system and is directly related to the net maximum solubilizing power of the mixed surfactants. The value of CL1695 + CEMDG is much smaller than the CL1695 and CEMDG in single surfactant based system, which indicates that the mutual solubilization of water and R (+)-limonene increases due to the mixing of surfactants. At equal amounts of water and R (+)-limonene the values of CEMDG/(CL1695 + CEMDG) depend on the surfactants mixing ratio. Table 1 shows the values of the mixing fractions of ethoxylated mono-di-glyceride surfactant at the water-R (+)-limonene interface. From Table 1 we can see that a minimum value of CEMDG/(CL1695 + CEMDG) is obtained at surfactants mixing ratio equals unity which corresponds to maximum water solubilization. Another way to determine the surfactants content at the interface of water-R (+)limonene in the microemulsion systems is by calculating the mixing weight fraction of the surfactant at the interface using the equation: XEMDG ¼ SSEMDG þ
SEMDG SSL1695 −SL1695 SSEMDG 1 −1 ROW X 1−SL1695 −SEMDG
ð9Þ
Where XEMDG represents the weight fraction of ethoxylated monodi-glyceride (the lipophilic surfactant) in the total mixed surfactants, SSL1695 and SSEMDG represents the mixing weight fraction at the waterTable 1 Mixing fractions of ethoxylated mono-di-glyceride at the water-R (+)-limonene interfaces Surfactants mixing ratio
CEMDG/(CL1695 + CEMDG)
SSEMDG
1/3 1/1 3/1
0.150 0.120 0.150
0.148 0.122 0.147
R (+)-limonene interface of sucrose laurate and ethoxylated mono-diglyceride respectively, Row is the weight fraction of R (+)-limonene in water + R (+)-limonene, and X is the weight fraction of mixed surfactants in the microemulsions. In order to do the calculations we estimate that the water and R (+)-limonene are pure and do not dissolve in each other. By plotting XEMDG versus 1x −1 a straight line is S obtained. SEMDG is the intercept and should be equal to the value of CEMDG/(CL1695 + CEMDG). The obtained values of SSEMDG are in good agreement with the values obtained for CEMDG/(CL1695 + CEMDG) at equal amounts of water and R (+)-limonene in the microemulsions as shown in Table 1. In other words, it is assumed that the sucrose laurate molecules are present only at the surfactant layers inside the microemulsion phase. The ethoxylated mono-di-glyceride molecules are distributed between the micro-water domains and the interface inside the microemulsion phase in a one phase microemulsions. Kuneida et al. [1,41–45] reported on similar results obtained with mixtures of sucrose monolaurate and polyethylene glycol alkyl ether systems in the presence of heptane, decane and hexadecane oils. The solubilization capability increases with mixing of surfactants in particular when surfactants with different hydrophilic–lipophilic balances are mixed [21]. The monomeric solubility of lipophilic surfactant (ethoxylated mono-di-glyceride) in R (+)-limonene is low as was reported in our previous study [40] and its mixing with sucrose laurate enables us to obtain large solubilization capacity of water and R (+)-limonene. 3.2. Transport properties 3.2.1. Electrical conductivity (σ) The electrical conductivity was measured on the (water + sodium chloride)/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene ((W + NaCl)/L1695/EMDG/LIM) where the mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity along the dilution line N60 (see phase diagram Fig. 3). The concentration of sodium chloride in water is 0.01 M. All samples appeared transparent and isotropic. Similar independence of phase behavior in the presence of a small amount of electrolyte is reported in the literature [13,26,46,47]. Fig. 4a displays the influence of water volume fraction on the electrical conductivity (σ). As the volume fraction of water increases the electrical conductivity increases exponentially. The increase in the electrical conductivity as function of water volume fraction is due to the increase in the fraction of sodium chloride ions that are not enclosed in the core of the microemulsions. The high values of electrical conductivity at high water volume fractions are explained by the fact that the sodium chloride ions are present in the external phase, which is the water. These results permit to distinguish between water-in-oil and oil-in-water microemulsions. Similar results and interpretations were provided by a number of authors [14,48,49] who studied electrical conductivity variations as function of water volume fraction as a method of characterization of microemulsions interdroplet interactions. Fig. 4b shows a plot of d log (σ)/d(ϕ) as function of water volume fraction for the system presented in Fig. 4a. The changes observed in the curve presented in Fig. 4b have been attributed to the occurrence of a percolation transition. Some authors [48,50] who studied the electrical conductivity of microemulsions as function of water volume fraction suggested a model for the explanation of the behavior of d log (σ)/d(ϕ) as function of water volume fraction. In this percolation model, the conductivity remains low up to a certain volume fraction (ϕc) of water. It must be emphasized that these conducting water-in-oil droplets, below ϕc are isolated from each other embedded in non-conducting continuum oil phase and hence contribute very little to the conductance. However, as the volume fraction of water reaches the percolation threshold ϕc = 0.17, some of these conductive droplets begin to contact each other and form clusters which are sufficiently close to each other. The number of such clusters increases very rapidly above the
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
107
the prefactors A, B by numerical analysis with adjustment by the least squares method using simultaneously the Eqs. (10) and (11). The computed s and t values are 0.16 and 1.0, respectively. The resulting ϕc = 0.17 obtained in this manner is close to the values obtained by the numerical estimate of the maximum of (d logσ/dϕ) versus ϕ (Fig. 4b). There is a reasonable agreement between calculated (by Eqs. (10) and (11)) and experimental values within prescribed range of composition with a mean deviation of 6%. The above equations are valid only near ϕc and cannot be extrapolated to infinite dilution and unit concentration. In addition, these are not applicable at the immediate vicinity of ϕc, where there is a continuous variation within a narrow interval around the percolation threshold.
Fig. 4. [a] Variation of the electrical conductivity (σ) of (water + sodium chloride)/ sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene system as function of water volume fraction along the dilution line N60 at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity. Sodium chloride concentration in water is 0.01 M. The phase diagram is presented in Fig. 3. [b] Plot of d log (σ)/ d(ϕ) as function of water volume fraction for the system presented in a.
percolation threshold (between ϕc = 0.17 and 0.5), giving rise to the observed changes of properties, in particular to the increase of electrical conductivity. The increase in electrical conductivity for the volume fractions ϕ above ϕc has been attributed to the transfer of counter ions from one droplet to another through water channels opening between droplets during sticky collisions through transient merging of droplets [3]. The existence and position of this threshold depends on the interactions between droplets, which control the duration of the collision and the degree of the interface overlapping, hence the probability of merging. Building up of conductivity needs attractive interactions and ϕc decreases when the strength of these interdroplet interactions increases as predicted by recent theoretical calculations [48,49]. In the present study, a theoretical model of Safran et al. [50], which is based on the dynamical picture of percolation, has been utilized to analyze the conductivity results of the system. According to the theory σ ¼ Að/c −ϕÞ−s σ ¼ Bð/−ϕc Þt
if / b /c if / N /c
3.2.2. Dynamic viscosity (η) The formation of microemulsions is a dynamic self-organizing phenomenon where aggregating disaggregating processes operate in conjunction. In their process dynamics, exchange of matters between different phases continuously occur resulting in an overall equilibrium. The flow of microemulsions under stress and the transport of molecules through them are also of potential importance in the study of microemulsion dynamics. The study of the dynamic viscosity of microemulsions can provide information on the intrinsic and derived processes in the microemulsion system, as well as furnish knowledge on the overall geometry of the particles of the dispersed phase [3,48,51,52]. Viscosity measurements evidenced the dependence of the size and shape of microemulsions droplets on the amount of solubilized water [53–56]. The understanding of structural consistencies in microemulsions has also been attempted from viscosity measurements by others [57,58]. Fig. 5 shows the variation in the dynamic viscosity as function of water volume fraction in the water/sucrose laurate/ethoxylated monodi-glyceride/R (+)-limonene system along the dilution line N60, the mixing ratio of sucrose laurate/ethoxylated mono-di-glyceride equals unity (see phase diagram in Fig. 3). The existence of two peaks in viscosity values with increasing water volume fraction implying the existence of at least three microstructural regions: one before, one after and the other in between two peaks. The maxima are indicative of the transition from a mono- to a bicontinuous structure. The gradual increase of viscosity in the range of 0–0.2 water volume fraction suggests that the dispersed phase is present in the form of droplets in the microemulsion systems. These results are interpreted considering clustering of water droplets with the mixed surfactants by the formation of a mixed monomolecular layer of the surfactants couple. For water volume fraction above 0.20, the viscosity of the system decreased and continues to
ð10Þ ð11Þ
Where s and t are, scaling exponents express the dynamic aspect of the percolation phenomenon. We have determined ϕc, s and t and
Fig. 5. Variation of the dynamic viscosity (η) of the water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene system as function of water volume fraction along the dilution line N60 at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/ sucrose laurate equals unity. The phase diagram is presented in Fig. 3.
108
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
decrease as the water volume fraction increases. At 0.7 water volume fraction the viscosity increases again and reaches a second maximum at water volume fraction equals 0.75. The increase of dynamic viscosity for water volume fractions below 0.20 indicates attractive interaction and aggregation of droplets of water phase including molecular reorganization on the interface where the water-in-oil microemulsions are present. The slow decrease in dynamic viscosity for water volume fractions between 0.20 and 0.70 indicates a transition from water-in-oil microemulsions droplets to a bicontinuous structure. The increase in dynamic viscosity for water contents between 0.70 and 0.75 indicates a structural transition from bicontinuous structure to an oil-in-water microemulsions microstructure. The sharp decrease in dynamic viscosity for water volume fractions above 0.80 indicates that the water, which is the least viscous component of the microemulsion system, becomes the outer phase and oil-in-water microemulsions are formed. The percolation phenomenon in microemulsion essentially involves droplet association i.e. clustering and fusion. It must, therefore, have a direct influence on the internal structure and hence viscosity. The study of viscosity confirmed the presence of two percolation processes for this microemulsion system. The structural inversion of the water-in-oil microemulsion to the oil-in-water type without any phase separation takes place in two stages. With increasing water volume fraction, the oil-continuous microemulsion transforms into the bicontinuous form at roughly 0.2 water volume fraction (water percolation threshold), and then at roughly 0.75 water volume fraction the bicontinuous form transforms into the water continuous structure (oil percolation threshold). Other authors [48,49,51,52] reported about similar behavior of dynamic viscosity of microemulsions based on nonionic surfactant in the presence of cosurfactants, oil and water. Djordjevic et al. [52] demonstrated also similar effect of increasing water volume fraction on the dynamic viscosity based on nonionic surfactants in the presence of water and isopropylmyristate microemulsion system. Like electrical conductivity, viscosity may also follow scaling type equations in the water-in-oil microemulsion region [59–61]. η ¼ Að/−ϕc Þ−μ if / N /c
ð12Þ
η ¼ Bð/c −ϕÞ−s if / /c
ð13Þ
where ϕ is the volume fraction of water, ϕc, is the percolation threshold, A and B are parameters, and µ and s are scaling exponents. The slopes of the log η versus log (ϕc − ϕ) plot for ϕc N ϕ and the log η versus log (ϕ − ϕc) for ϕ N ϕc plot, yield s and µ parameters. The average values of μ and s in Eqs. (12) and (13) are 1.04 and 0.18, respectively. These values are fairly near to the t and s values 0.90 and 0.16, respectively, obtained for conductivity percolation. The s value corresponds to the dynamic value of 0.16 obtained for static conductance percolation. The estimated scaling parameters which are in good agreement with the experimental values obtained with both electrical conductivity and dynamic viscosity signifies that these microemulsion system show an interdependence of the viscosity– conductivity, especially at the stage of water percolation.
and so all data are presented in arbitrary units. Only the polar head groups and water regions are visible with SAXS experiments because their electron densities are higher than the electron density of the surrounding oil. In this section, we used the SAXS technique to investigate the microemulsion system water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene microemulsion system as a function of the water volume fraction (ϕ) along the dilution line N60. The mixing ratio of ethoxylated mono-di-glyceride/sucrose laurate equals unity (see phase diagram in Fig. 3). Characteristic profiles of the system are presented in Fig. 6 as an example of what happens in these systems. In Fig. 6, we observe that in each case, the scattering profile exhibits a single intensity maximum at q ≠ 0, followed by a high-angle tail. With increasing water volume fraction, the position of the maximum moves to a lower angle. By fitting all the scattering curves (Fig. 6) to the Teubner–Strey equation [18] (Eq. (1)) we were able to derive from the values of the periodicity, d, correlation length, ξ, and the amphiphilicity factor, fa, as described in the experimental section (Eqs. (3)–(5)). The dependence of the d and ξ parameters on the volume fraction of water, ϕ was plotted in Fig. 7. In Fig. 7a we find that the periodicity, d, increases from 58 to 217 Å. Plots of d versus ϕ can probe the dimentionality of swelling along the dilution lines. Eq. (3) was used to determine the values of the correlation length, ξ. Initially, the growth of ξ (Fig. 7b) parallels that of d, with the former being smaller than the latter. ξ reaches maximum at ϕ equals 0.3, while d increases in a monotonic fashion over the whole range of water dilution. The increase of the correlation length, ξ , as function of water volume fraction is explained as follows: when the water is the dispersed phase, increasing the water volume fraction increases the size of the scattering units and the correlation length, ξ, whereas when water is in the bulk, increasing the water volume fraction dilutes the scattering units and ξ decreases. Eq. (5) was used to calculate the amphiphilicity factor, fa, (see values of fa in Table 2). In all cases, fa is negative with values that lie between −0.94 and −0.72, consistent with oscillatory behavior of the correlation function γ(r) and the appearance of a well-defined scattering peak at q ≠ 0. Within this range, for more negative fa, the microemulsion is more ordered. Other authors reported about similar results on the behavior of the microstructural parameters of nonionic microemulsions based on ethoxylated nonionic surfactants [31,63,64]. Recently, other authors [65,66] studied the formation of glass microemulsions using anhydrous powders of sugars or aqueous concentrated sugar solutions and sucrose ester surfactants in R (+)-limonene. Small-angle neutron scattering experiments confirm the presence of well-structured microemulsions with domain sizes ranging from ~35 to 60 nm in the
3.3. Structural properties 3.3.1. Periodicity, correlation length and amphiphilicity factor by saxs Small-angle X-ray scattering (SAXS) probes the pertinent colloidal length scales of 1–100 nm and therefore is the method of choice for determining size of colloidal particles. The scattering intensity depends on the different scattering length densities of the particles and the solvent. For SAXS, the scattering length density is proportional to the electron density, which is a linear function of the number region of aggregates composed of surfactant molecules [62]. However, at the time of these experiments, SAXS data were not measured on absolute scale
Fig. 6. Small angle X-ray scattering curves for samples whose compositions lie along the N60 dilution line for the system water/sucrose laurate/ethoxylated mono-di-glyceride/ R (+)-limonene at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/ sucrose laurate equals unity. The phase diagram is presented in Fig. 3.
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
Fig. 7. [a] Microemulsion periodicity, d as a function of volume fraction of water, ϕ, for samples along the dilution line N60 of water/sucrose laurate/ethoxylated mono-diglyceride/R (+)-limonene system at 25 °C. The mixing ratio (w/w) of ethoxylated monodi-glyceride/sucrose laurate equals unity. The phase diagram is presented in Fig. 3. [b] Microemulsion correlation length, ξ, as a function of volume fraction of water for the system presented in b.
109
3.3.2. Diffusion coefficients by nuclear magnetic resonance We used the diffusion coefficient obtained by nuclear magnetic resonance analysis to characterize the microstructure changes in the microemulsions system water/sucrose laurate/ethoxylated mono-diglyceride along the dilution line N60 for surfactants mixing ratio (w/w) equals unity (see phase diagram in Fig. 3). To evaluate the selfdiffusion data in terms of microstructure, the calculation of the relative diffusion coefficient, D/D0, of the different components of the microemulsion especially oil and water is needed [67]. Relative diffusion coefficients were obtained by dividing water (DWater) and oil (DOil) diffusion coefficients in the microemulsion by the diffusion coefficient of water in the pure water phase (D0Water) and oil in the neat phase (DOil 0 ). It is well-documented [15,16,67] that if the D/D0 values of water and oil differ by more than 1 order of magnitude, discrete particles of the slowly diffusing solvent are implied, whereas if the D/D0 values of water and oil are of the same order of magnitude, a bicontinuous structure is suggested. Fig. 8 shows the relative diffusion coefficients of water and R (+)-limonene as a function of the water volume fraction. One can clearly see that at the two extremes of aqueous-phase concentrations (up to 0.2 and above 0.7 wt water volume fractions), the DWater/DWater values are easily interpreted, while the in between 0 regions are somewhat more difficult to explain since gradual changes take place. As Fig. 8 indicates that microemulsions containing up to 0.2 water volume fraction, have a confined water molecules microstructure, since the relative diffusion coefficients of water and R (+)limonene differ by more than 1 order of magnitude. Microemulsions with 0.20 to 0.70 water volume fraction have a microstructure where the nor water neither oil can be distinguished as confined phases, as the diffusion coefficients of water and R (+)-limonene are of the same order of magnitude. Relative diffusion coefficients results for water volume fractions above 0.70 indicates that the oil phase is the confined one since the relative diffusion coefficients of water and R (+)-limonene differ by more than 1 order of magnitude. Comparing these results to the NMR results presented in our pervious published work [46], one can see that the relative diffusion coefficients of both water and oil are lower compared to the diffusion coefficients obtained for both oil and water in the previous study. As we suggested in our previous work the mixed surfactants content in the microemulsions samples affects the diffusion coefficients of both oil and water. This can be one of the causes for the difference in diffusion coefficients for both oil and water. The second reason that can cause the difference in the diffusion coefficients values is the ratio of mixed
case of the aqueous concentrated sugar solutions/sucrose ester surfactants/R (+)-limonene systems. With few exceptions, the patterns of microemulsion with concentrated sugar solutions are very similar to that of aqueous systems. The mixing of ethoxylated mono-di-glyceride with sucrose laurate in our systems induces high water solubilization and larger periodicity values compared to the reported systems [65,66] that are composed of single surfactant the sucrose ester.
Table 2 Values of the Amphiphilicity factor (fa) for the microemulsion system water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene as a function of the water volume fraction (ϕ) along the dilution line N60 Water volume fraction (ϕ)
Amphiphilicity factor (fa)
0 0.11 0.21 0.32 0.42 0.52 0.62
−0.79 −0.94 −0.94 −0.84 −0.72 −0.72 −0.72
The mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity. Values of fa calculated from Eq. (5) from data obtained at 25 °C.
Fig. 8. Relative diffusion coefficients of water (Δ) and R (+)-limonene (O) as function of water volume fraction for samples along the dilution line N60 of water/sucrose laurate/ ethoxylated mono-di-glyceride/R (+)-limonene system at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity. The phase diagram is presented in Fig. 3.
110
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
surfactants to oil, which was variable in our previous study, and in this study the ratio is constant for all of the samples investigated. The diffusion coefficients of sucrose laurate and ethoxylated mono-diglyceride at low water contents are very low. The values of their diffusion coefficients at 0.10 water volume fraction are 0.013 × 10− 5 and 0.011 × 10− 5 cm2/s for sucrose laurate and ethoxylated mono-diglyceride, respectively. For high water volume fractions the diffusion coefficient of sucrose laurate increases and equals 0.063 × 10− 5 cm2/s at water volume fraction equals 0.90. On the other hand, the diffusion coefficient of ethoxylated mono-di-glyceride at water volume fraction equals 0.90 equals the same value as for 0.10 water volume fraction (i.e. 0.011 × 10− 5 cm2/s). These results evidenced the interpretations presented in the phase behavior section that ethoxylated mono-diglyceride is there in the center of a hexagonal arrangement bordered by the more mobile molecules of sucrose laurate. Nuclear magnetic resonance results provide a clear picture of structural transitions along the N60 dilution line studied. As in the dynamic viscosity results three different regions are observed. The first region indicates the presence of water-in-oil droplets microstructure. The second region indicates the transition from water-in-oil microstructure to a bicontinuous microemulsion. In the third region a transition from a bicontinuous microemulsion to an oil-in-water microstructure occurs and a discrete oil-in-water microstructure can be present. 4. Conclusions • The mixed surfactants used in this study enable the formation of alcohol free safe U-type microemulsions. • The values of the weight fraction of total surfactants (CL1695 + CEMDG) in surfactants monolayer at the water-R (+)-limonene interface inside the mixed surfactant microemulsion system is much smaller than the CL1695 and CEMDG in single surfactant based system which indicates that the mutual solubilization of water and R (+)-limonene increases due to the mixing of surfactants. • Conductivity results indicate that these systems are percolated. The estimated scaling parameters indicate that the percolation is static. • Viscosity variation as function of water volume fraction demonstrates the presence of two percolation thresholds (water and oil). The placement of these thresholds supports the occurrence of structural transitions from water-in-oil to bicontinuous to oil-inwater microemulsions. • Upon increasing the volume fraction of water in the microemulsions there is an increase in the periodicity of the microemulsions that indicates an increase in the domain size in one dimension with increasing water content. • Structural transitions from water-in-oil to bicontinuous to oil-inwater microemulsions were also evidenced by the use of NMR technique. References [1] H. Kunieda, C. Solans, How to prepare microemulsions: temperature insensitive microemulsions, in: H. Kunieda, C. Solans (Eds.), Industrial Applications of Microemulsions, Marcel Dekker, Inc., New York, 1996, pp. 21–45. [2] M. Kahlweit, G. Busse, B. Faulhaber, H. Ebil, Langmuir 11 (1995) 4185–4187. [3] S.P. Moulik, B.K. Paul, Adv. Colloid Interface Sci. 78 (1998) 99. [4] T. Sottmann, R. Strey, in: J. Lyklema (Ed.), Soft Colloids V — Fundamentals in Interface and Colloid Science, Elsevier, Amsterdam, 2005, chap. 5. [5] N. Patel, U. Schmid, M.J. Lawrence, J. Agric. Food Chem. 54 (2006) 7817. [6] J. Flanagan, H. Singh, Crit. rev. food sci. nutr. 46 (2006) 221. [7] M.J. Lawrence, G.D. Rees, Adv. drug deliv. rev. 45 (2000) 89. [8] M. Kreilgaard, Adv. drug deliv. rev. 54 (2002) S77–S98 (SUPPL.). [9] K. Ogino, M. Abe (Eds.), Mixed Surfactant Systems, Surfactant Science Series, 46, Marcel Dekker, Inc., New York, 1992. [10] P.M. Holland, D.N. Rubingh, Mixed Surfactant Systems, American Chemical Society, Washington, 1992.
[11] J.F. Scamehorn, in: J.F. Scamehorn (Ed.), Phenomena in Mixed Surfactants Systems, ACS Symposium Series, vol. 311, American Chemical Society, Washington, D.C., 1986 [12] A. Bumajdad, J. Eastoe, J. Colloid Interf. Sci. 274 (2004) 268. [13] M. Fanun, W.S. Al-Diyn, Colloids surf., A Physicochem. eng. asp. 277 (2006) 83. [14] R.K. Mitra, B.K. Paul, J. Colloid Interface Sci. 283 (2005) 565. [15] O. El-Seoud, J. Mol. Liquids 72 (1997) 85. [16] O. Soderman, M. Nyden, Colloids Surfaces. A 158 (1999) 273. [17] O. Glatter, R. Strey, K.-V. Schubert, E.W. Kaler, Ber. Bunsenges. Phys. Chem. 100 (1996) 323. [18] M. Teubner, R. Strey, J. Chem. Phys. 87 (1987) 3195. [19] J. Sagalowicz, M. Michel, M. Adrian, P. Frossard, M. Rouvet, H.J. Watzke, A. Yaghmur, L. De Campo, L.O. Glatter, M.E. Leser, J. Microscopy 221 (2006) 110. [20] M. Kahlweit, R. Strey, D. Haase, H. Kunieda, T. Schmeling, B. Faulhaber, M. Borkovec, H.-F. Eicke, G. Busse, F. Eggers, T.H. Funck, H. Richmann, L. Magid, O. Söderman, P. Stilbs, J. Winkler, A. Dittrich, W. Jahn, J. Colloid Interface Sci. 118 (1987) 436. [21] P.D.T. Huibers, D.O. Shah, Langmuir 13 (1997) 5762. [22] R.J. Hunter, Foundations of Colloid Science; Oxford Science Publications, Clarendon Press, Oxford, 1989, Vol. 2, pp. 948–991. [23] G. Li, X. Kong, R. Guo, X. Wang, J. Disp. Sci. Technol. 5 (1989) 29. [24] N. Garti, V. Clement, M. Fanun, M.E. Leser, J. Agri. Food Chem. 48 (2000) 3945. [25] M.J. Hou, D.O. Shah, Langmuir 3 (1987) 1086. [26] H.F. Eicke, W. Meier, H. Hammerich, Langmuir 10 (1994) 2223. [27] J.A. Lake, Acta Crystallogr. 23 (1967) 191. [28] S.A. Teukolsky, W.T. Vetterling, B.P. Flannery, Numerical Recipes in C: The Art of Scientific computing, Cambridge University Press, New York, 1992. [29] J. Brunner-Popela, R. Mittelbach, R. Strey, K.-V. Schubert, E.W. Kaler, O. Glatter, J. Chem. Phys. 21 (1999) 10623. [30] M. Fanun, E. Wachtel, B. Antalek, A. Aserin, N. Garti, Colloids and Surf. A: 180 (2001) 173. [31] S. Ezrahi, E. Wachtel, A. Aserin, N. Garti, J. Colloid Interface Sci. 191 (1997) 277. [32] K.-V. Schubert, R. Strey, S.R. Kline, E.W. Kaler, J. Chem. Phys. 101 (1994) 5343. [33] K.-V. Schubert, R. Strey, J. Phys. Chem. 95 (1991) 8532. [34] J.F. Billman, E.W. Kaler, Langmuir 7 (1991) 1609. [35] D. Wu, A. Chen, C.S. Johnson, J. Magn. Reson. 115 (1995) 260–264. [36] E.O. Stejskal, in: E.O. Stejskal (Ed.), Encyclopedia of Nuclear Magnetic Resonance, Jhon Wiley & Sons, New York, 1995, pp. 657–658. [37] K. Aramaki, T. Hayashi, T. Katsuragi, M. Ishitobi, H. Kunieda, J. Colloid Interface Sci. 236 (2001) 14. [38] L.D. Ryan, E.W. Kaler, Langmuir, 13 (1997) 1510. [39] C. Rodriguez, D.P. Acharya, S. Hinata, M. Ishitobi, H. Kunieda, J. Colloid Interface Sci. 262 (2003) 500. [40] M. Fanun, W. Salah Al-Dyin, J. Disp. Sci. Technol. 27 (2006) 1119. [41] H. Kunieda, N. Ushio, A. Nakano, M. Miura, J. Colloid Interface Sci. 159 (1993) 37. [42] H. Kunieda, M. Yamagata, Langmuir 9 (1993) 3345. [43] H. Kunieda, A. Nakano, M. Akimura, J. Colloid Interface Sci. 170 (1995) 78. [44] H. Kunieda, A. Nakano, Ma.A. Pes, Langmuir 11 (1995) 3302. [45] Ma.A. Pes, K. Aramaki, N. Nakamura, H. Kunieda, J. Colloid Interface Sci. 178 (1996) 666. [46] M. Fanun, W. Salah Al-Diyn, J. Disp. Sci. Technol. 28 (2007) 164. [47] A. Yaghmur, A. Aserin, B. Antalek, N. Garti, Langmuir 19 (2003) 1063. [48] S.K. Mehta, R.K. Dewan, Phys. Rev. E 50 (1994) 4759. [49] S.K. Mehta, K. Bala, Fluid phase equilib. 172 (2000) 197. [50] S.A. Safran, G.S. Grest, A.L.R. Bug, Microemulsion system, in: H. Rosano, M. Clausse (Eds.), Surfactant Science Series, Marcel Dekker, Inc, New York, 1987. [51] S. Ezrahi, A. Aserin, N. Garti, J. Colloid Interface Sci. 202 (1998) 222. [52] L. Djordjevic, M. Primorac, M. Stupar, D. Krajisnik, Int. J. Pharm. 271 (2004) 11. [53] P. Ekwall, L. Mandell, K. Frontell, J. Colloid Interf. Sci. 33 (1970) 215. [54] S. Ajith, A.K. Rakshit, J. Surf. Technol. 8 (1992) 365. [55] S. Ajith, A.K. Rakshit, Langmuir 11 (1995) 1122. [56] S. Ajith, A.K. Rakshit, J. Phys. Chem. 99 (1995) 14778. [57] A. Maitra, C. Mathew, M. Varshney, J. Phys. Chem. 94 (1990) 5290. [58] C. Boned, J. Peyrelasse, J. Surf. Sci. Technol. 7 (1991) 1. [59] J. Peyrelasse, C. Boned, Phys. Rev. A 41 (1990) 938. [60] Z. Saidi, C. Matthew, J. Peyrelasse, C. Boned, Phys. Rev. A 42 (1990) 872. [61] J. Peyrelasse, C. Boned, in: S.H. Chen, et al., (Eds.), Structure and Dynamics of Strongly Interacting Colloids and Supramolecular Aggregates in Solution, Kluwer Academic Publisher, The Netherlands, 1992, pp. 801–806. [62] D.I. Iampietro, L.L. Brasher, E.W. Kaler, A. Strandner, O. Glatter, J. Phys. Chem., B. 102 (1998) 3105. [63] O. Regev, S. Ezrahi, A. Aserin, N. Garti, E. Wachtel, E.K. Kaler, A. Khan, Y. Talmon, Langmuir 12 (1996) 668. [64] S. Ezrahi, A. Aserin, N. Garti, Aggregation behavior in one-phase (Winsor IV) microemulsion systems, in: P. Kumar, K.L. Mittal (Eds.), Handbook of Microemulsion Science and Technology, Marcel Dekker, Inc., New York, 1999, pp. 185–246. [65] H. Dave, F. Gao, M. Schultz, C.C. Co, Colloids surf., A Physicochem. eng. asp. 296 (2007) 45. [66] H. Dave, F. Gao, J.-H. Lee, M. Liberatore, C.-C. Ho, C.C. Co, Nat. Materials 6 (2007) 287. [67] B. Lindman, U. Olsson, O. Soderman, in: P. Kumar, K.L. Mittal (Eds.), Handbook of Microemulsion Science and Technology, Marcel Dekker, New York, 1999, pp. 309–356.
Contents lists available at ScienceDirect
Journal of Molecular Liquids j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / m o l l i q
A study of the properties of mixed nonionic surfactants microemulsions by NMR, SAXS, viscosity and conductivity Monzer Fanun ⁎ Faculty of Science and Technology, Al-Quds University, P.O. Box 51000 East-Jerusalem, Palestine
A R T I C L E
I N F O
Article history: Received 4 February 2008 Received in revised form 9 May 2008 Accepted 16 May 2008 Available online 25 May 2008 Keywords: Surfactants mixture One-phase microemulsion Percolation threshold Periodicity of domain size Amphiphilicity factor Correlation length Microstructure of microemulsions Relative diffusion coefficients
A B S T R A C T The pseudoternary phase behavior of the water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)limonene systems was investigated for different surfactants mixing ratios (w/w) at 25 °C. The microemulsion boundaries were determined and the surfactants content at the interface of water- R (+)-limonene was estimated. For surfactants mixing ratio (w/w) equals unity, the area of the one phase microemulsion region reaches its maximum. The system with the maximum microemulsion area was investigated using electrical conductivity, dynamic viscosity, small angle X-ray scattering, and nuclear magnetic resonance. Electrical conductivity increases as the water volume fraction increases and a percolation threshold was observed. Dynamic Viscosity varies as function of the water volume fraction in a non-monotonic way giving twopeaked plot. The characteristics of the domain size of the microemulsions called periodicity measured by small angle X-ray scattering increases with the increase in the water volume fraction. The correlation length of the domain size reaches a maximum when plotted against the water volume fraction in the microemulsions. Relative diffusion coefficients of water increase and those of oil decrease with increasing the water volume fractions in the microemulsions indicating structural transitions. © 2008 Elsevier B.V. All rights reserved.
1. Introduction The formulation of nonionic, safe, biodegradable microemulsions with specific properties like high solubilization power and temperature insensitivity was focused in many researches [1,2]. The extensively reported [3,4] microemulsions physicochemical properties, which include thermodynamic stability, high solubilization power, low interfacial tensions, transparency, and low viscosity, are of great importance. Microemulsions are broadly used in food, cosmetic and pharmaceutical applications [5–8]. The surfactants used in the majority of the industrial applications almost consist of a mixture of surfactants. The tendency to form aggregates in systems containing mixed surfactants would be considerably different from that in systems based on single surfactant. The properties of a mixture of homologous surfactants belonging to the same class are simply predictable from the properties of the individual components. On the other hand, mixtures of surfactants formed from different classes usually show some deviation from ideal behavior. The incorporation of another surfactant into the mixture of water, oil and surfactant provides an additional degree of freedom, which enables the adjustment of phase behavior. Mixtures of nonionic surfactants are more effective than the other class of surfactants for the reason that they are more effective at lowering the monomer concentration and
⁎ Tel.: +972 522 40 60 61; fax: +972 22 79 69 60. E-mail addresses: [email protected], [email protected]. 0167-7322/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.molliq.2008.05.006
they are not susceptible to precipitation themselves [9–11]. Synergistic mixing properties of surfactants can only be achieved by understanding how surfactants interact with each other and how these interactions influence performance. Understanding the mixed surfactants behavior in the presence of water and oil is an important issue for the development and the optimal design of mixed surfactants based microemulsions for industrial applications [9]. Techniques used to study how changes in the relative amounts of the microemulsion components affect the microstructure of the solution in singlesurfactant systems also apply to mixed surfactant systems. These methods include electrical conductivity [12,13], dynamic viscosity [3,14], nuclear magnetic resonance [13,15,16], scattering methods [17,18], and imaging techniques [19,20]. In the present work, we evaluate the effect of adding ethoxylated mono-di-glyceride (EMDG) on the phase behavior and properties of the water/sucrose laurate/R (+)-limonene (W/L1695/LIM) system by exploring the following subjects: 1. Phase behavior of the single and mixed nonionic surfactants, in the presence of water and the biologically compatible R (+)-limonene oil, which permits the determination of the microemulsions phase regions. 2. Inter droplet interactions of the formulated microemulsions using electrical conductivity. 3. Microstructure transitions in the formulated microemulsions using dynamic viscosity, small angle X-ray scattering, and nuclear magnetic resonance.
104
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
2. Experimental 2.1. Materials The sucrose laurate (L1695) was obtained from Mitsubishi–Kasei Food Corp., (Mie, Japan) (see chemical structure in Fig. 1a). The purity of combined Lauric acid equals 95 %, the esters compositions are 80% monoester and 20% di, tri and polyester, its HLB equals 16. Ethoxylated mono-di-glyceride (EMDG) (MAZOL 80 MG KOSHER), its HLB equals 13.5. Ethoxylated mono-di-glyceride was obtained from BASF Corporation (Gurnee, Illinois, USA) (see chemical structure in Fig. 1b). R (+)-limonene (LIM), (98%) (see chemical structure in Fig. 1c) was purchased from Sigma Chemicals Co. (St. Louis, USA). Sodium chloride (NaCl) of analytical grade was purchased from J.T. Baker Inc. (Phillipsburg, USA). All of the components were used as supplied without further purification. Water was double distilled. 2.2. Methods 2.2.1. Pseudoternary phase diagrams at constant temperature The phase behavior of a system consisting of water, oil, surfactant (or mixture of surfactants) may be described on a phase tetrahedron whose apexes respectively represent the pure components. However, it is more convenient to describe the phase behavior on a pseudoternary phase triangles. Obviously, a fixed (weight, volume or mole) ratio must be chosen for any two of the components and one of the triangle vertices represents 100% of this binary mixture. Phase diagrams of such multi-component systems are generally based on constant ratios of surfactant-to-water or cosurfactant-to-surfactant [21–23]. Such a presentation enables us to follow directly and conveniently, the variation of the surfactant amount needed to solubilize a given amount of water. Mixtures at fixed weight ratios of oil phase and the surfactant or mixed surfactants were prepared in culture tubes sealed with Viton lined screw caps. Water was then added dropwise until its solubilization limit was reached. After this
Fig. 2. Schematic phase diagram presenting the solubilization parameters of mixed nonionic surfactants in water and oil at 25 °C. AT (%) is the area of the one phase microemulsions region. 1φ is the W/O, bicontinuous, and O/W microemulsions phase region; II(Wm + O) is the area of the two phase region that signifies water continuous micellar system with excess oil.
point, larger increments of aqueous phase were added. All of the aqueous phase additions were followed by vigorous stirring on a vortex mixer. The time for equilibration between additions of successive aliquots was typically, from a few minutes up to 24 h. Phase transitions detected visually by the appearance of cloudiness or sharply defined separated phases. The completion of this process was hastened by centrifuging the samples. The phase diagrams were determined at 25 ± 0.5°C. 2.2.2. Water solubilization parameter The water solubilization capacity of different amphiphilic systems should be compared at optimal conditions [24,25]. Li et al. [23] have employed as a solubilization parameter the area of the one-phase microemulsion region (AT) that is area limited by the microemulsification failure boundaries. In this study we used the one phase microemulsion region to compare the water solubilization capacity in the studied systems. The relative error in determining the AT (%) was estimated to be ±3% for all of the systems studied. Fig. 2 present a schematic phase diagram where we indicated the one phase region, the microemulsification failure and the multiple phase regions. 2.2.3. Electrical conductivity measurements Conductivity measurements were performed at 25 ± 0.5 °C on samples the compositions of which lie along the one phase channel, using conductivity meter, the conductivity cell used is Tetra Con® 325, the electrode material is graphite and the cell constant is 0.475 cm− 1 ± 1.5%. The range of application is between 1 µS/cm to 2 S/cm with an accuracy of ±0.5%, and the temperature range is from −5 to 100 °C. In the case of nonionic microemulsions, a small amount of an aqueous electrolyte must be added for electrical conduction [26]. Thus, a 0.01 M sodium chloride aqueous solution was used in the preparation of the microemulsion samples in place of pure water. The electrode was dipped in the microemulsion sample until equilibrium was reached and reading becomes stable. Reproducibility was checked for certain samples and no significant differences where observed. The constant of the conductivity cell was calibrated using standard KCl solutions and checked a minimum of three times during the course of the working shift.
Fig. 1. The chemical structures of [a] sucrose laurate [b] ethoxylated mono-di-glyceride (average structure) and [c] R (+)-limonene.
2.2.4. Viscosity measurements Viscosity was measured using a rotational viscometer, model DV1PL spindle from Anton Paar GmbH (Graz, Austria). “Double cylinder” geometry was used. Viscosities at 200 s− 1 shear rate were obtained at
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
25 ± 0.5 °C. Reproducibility (triplicate) was checked for the samples and no significant differences (±SD) where observed. 2.2.5. Small angle X-ray scattering (saxs) Scattering experiments were performed using Ni-filtered Cu Kα radiation (0.154 nm) from an Eliott GX6 rotating X-ray generator that operated at a power rating up to 1.2 kW. X-radiation was further monochromated and collimated by a single Franks mirror and a series of slits and height limits and measured by a linear position-sensitive detector. The sample was inserted into 1–1.5 mm quartz or lithium glass capillaries, which were then flame-sealed. Each sample was checked before and after the experiment to verify that, no fluid had been lost during the time of exposure, approximately 3 h. The temperature was maintained at 25 ± 1 °C. The sample-to-detector distance was 0.46 m, and the scattering patterns were measured using the Lake procedure [27]. 2.2.6. X-ray data analysis In this case, the scattering patterns after background subtraction were fit to the expression due to Teubner and Strey [18]: IðqÞ ¼ 1=a2 þ c1 q2 þ c2 q4 þ b
ð1Þ
With the constants a2, c1, c2 and b obtained by using the Levenburg–Marquardt procedure [28]. Such a functional form is simple and convenient for the fitting of spectra. Eq. (2) corresponds to a real space correlation function of the form
γ ðrÞ ¼ ðsin kr=kr Þe−r=
ð2Þ
The correlation function describes a structure with periodicity d = (2π/k) damped as a function of correlation length ξ. This formalism also predicts the surface to volume ratio, but because this ratio is inversely related to the correlation length and therefore must go to zero for a perfectly ordered system, calculated values are frequently found to be too low [29]. d and ξ are related to the constants in Eq. (1) by [18]: h i−1=2 d ¼ ð1=2Þðða2 =c2 ÞÞ1=2 −ðc1 =4c2 Þ
ð3Þ
h i−1=2 ¼ ð1=2Þðða2 =c2 ÞÞ1=2 þðc1 =4c2 Þ
ð4Þ
A third parameter, which can also be defined, is f a, the amphiphilicity factor [18,30–34], which relates to the behavior of the correlation function γ(r) and reflects the ability of the surfactant to impose order on the microemulsion: fa ¼ c1 =½4a2c2 1=2
105
gradient pulse length, Δ is the time between the two gradients in the pulse sequence (and hence defines the diffusion time). Typically, we use Δ = 100 ms, δ = 8 ms, and vary G from 1.7 to 32.3 G cm− 1 in 32 steps. 3. Results and discussion 3.1. Phase behavior A number of studies [37–39] reported on the investigation of sugar based surfactants microemulsions indicate the need of adding a cosurfactant or another surfactant to the sucrose esters in order to obtain better mutual solubilization of oil and water. In our previously published work [40] we studied the phase behavior of the water/ sucrose laurate/ethoxylated mono-diglyceride/R(+)-limonene systems as function of temperature where we specified the phase regions with labels. Two phase regions were observed designated by II (Wm + O) or (W + Om) and the three phase regions designated by III (W + D + O). The II phase (Wm + O) signifies water continuous micellar system with excess oil and (W + Om) means excess water with oil continuous micellar system. III phase (W + D + O) means a water-rich (W, lower), amphiphile-rich (D, middle), and an oil-rich (O, upper) phases. In this study, the pseudoternary phase diagrams at constant temperature equal to 25 °C of the water/sucrose laurate/ethoxylated-mono-diglyceride/R (+)-limonene system was explored. It indicates the presence of an isotropic and low-viscosity area which is a microemulsion one phase region (1φ), the remainder of the phase diagram represents a two phase region composed of water continuous micellar solution with excess oil designated by (Wm + O). A small microemulsion phase region was observed (i.e. the onephase microemulsion area (AT) equals 5%) in the ternary system water/sucrose laurate/R (+)-limonene. Adding ethoxylated-mono-diglyceride to the ternary system enhances the formation of the one phase microemulsion. For ethoxylated-mono-di-glyceride content in the mixture of (ethoxylated-mono-di-glyceride + sucrose laurate) equals 25 wt.%, the total one phase microemulsion area (AT) rises to 22%. At equal amounts of ethoxylated-mono-di-glyceride and sucrose laurate in the surfactants mixture (i.e. surfactants mixing ratio (w/w) equals unity), the one phase microemulsion region appears from the first drop of water added. This one phase region extends overall the water contents range along the dilution line N60 and the its area (AT) equals 62% (see Fig. 3). The one-phase microemulsion area (AT) decreases to the value of 47% by increasing the ethoxylated-mono-di-glyceride weight ratio in
ð5Þ
2.2.7. Pulsed gradient spin echo-nuclear magnetic resonance (PGSE-NMR) NMR measurements was performed on Bruker DRX-400 spectrometer with a BGU II [35,36] gradient amplifier unit and a 5-mm BBI probe equipped with a z-gradient coil, providing a z-gradient strength (G) of up to 55 G cm− 1. The self-diffusion coefficients were determined using bipolar-pulsed field gradient stimulated spin-echo (BPFG-SSE). In this work, we used bipolar gradient pulses as described by Wu et al. [35] to reduce the eddy-current effects. Experiments were carried out by varying the gradient strength and keeping all other timing parameters constant. The self-diffusion coefficient (D) is given by δ I ¼ Io exp γ2 G2 δ2 Δ− D 3
ð6Þ
Where I is the measured signal intensity, Io is the signal intensity for G = 0, γ is the gyro magnetic ratio for the 1H nucleus, δ is the
Fig. 3. Pseudoternary phase behavior of the system water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity. The one phase region is designated by 1ϕ, and the two phase region consisted of water continuous micellar solution with excess oil is designated by (Wm + O). N60 is the water dilution line where the weight ratios of sucrose laurate/ethoxylated mono-di-glyceride/R(+)-limonene equal to 3/3/4.
106
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
the mixed surfactants to 75 wt.%. In the ternary system water/ ethoxylated mono-di-glyceride/R (+)-limonene, the one-phase microemulsion area (AT) decreases to the value of 30%. It is considered that surfactant monolayers at the interface of water and R (+)-limonene domains inside the microemulsions are directly related to the solubilization of water and R (+)-limonene. The monodisperse solubilities of sucrose laurate and ethoxylated mono-di-glyceride in R (+)-limonene are very small [37]. This means that the R (+)-limonene domains in the microemulsion phase are almost the same as bulk R (+)limonene phase when solubilization is large. It can be assumed that the monomeric solubilities of sucrose laurate (SL1695) and that of ethoxylated mono-di-glyceride (SEMDG) in the water phase forming the microemulsions are similar to their respective critical micelle concentrations (CMC) which equal 3.4 ⁎ 10− 4 and 1.1 ⁎ 10− 5 M for sucrose laurate and ethoxylated mono-di-glyceride, respectively. Since surfactant molecules at the water-R(+)-limonene interface inside microemulsions are directly related to the solubilization, it is important to estimate the mixing fraction of each surfactant. The surfactant content at interface could be obtained by simple mass balance equations as follows:
min CEMDG ¼ X min XEMDG −
1−X min SEMDG 2ð1−SL1695 −SEMDG Þ
ð7Þ
and 1−X min S L1695 min − CL1695 ¼ X min 1−XEMDG 2ð1−SL1695 −SEMDG Þ
ð8Þ
where CL1695 and CEMDG indicate the weight of sucrose laurate and ethoxylated mono-di-glyceride at the water-R (+)-limonene interface, Xmin is the minimum weight fraction of mixed surfactants capable of solubilizing equal amounts of water and R(+)-limonene in the microemulsions, Xmin EMDG is the minimum weight fraction of the lipophilic surfactant (in our case the ethoxylated mono-di-glyceride) corresponding to Xmin. CL1695 + CEMDG is the weight fraction of total surfactants in surfactants monolayer at the water-R(+)-limonene interface inside the mixed surfactants microemulsion system and is directly related to the net maximum solubilizing power of the mixed surfactants. The value of CL1695 + CEMDG is much smaller than the CL1695 and CEMDG in single surfactant based system, which indicates that the mutual solubilization of water and R (+)-limonene increases due to the mixing of surfactants. At equal amounts of water and R (+)-limonene the values of CEMDG/(CL1695 + CEMDG) depend on the surfactants mixing ratio. Table 1 shows the values of the mixing fractions of ethoxylated mono-di-glyceride surfactant at the water-R (+)-limonene interface. From Table 1 we can see that a minimum value of CEMDG/(CL1695 + CEMDG) is obtained at surfactants mixing ratio equals unity which corresponds to maximum water solubilization. Another way to determine the surfactants content at the interface of water-R (+)limonene in the microemulsion systems is by calculating the mixing weight fraction of the surfactant at the interface using the equation: XEMDG ¼ SSEMDG þ
SEMDG SSL1695 −SL1695 SSEMDG 1 −1 ROW X 1−SL1695 −SEMDG
ð9Þ
Where XEMDG represents the weight fraction of ethoxylated monodi-glyceride (the lipophilic surfactant) in the total mixed surfactants, SSL1695 and SSEMDG represents the mixing weight fraction at the waterTable 1 Mixing fractions of ethoxylated mono-di-glyceride at the water-R (+)-limonene interfaces Surfactants mixing ratio
CEMDG/(CL1695 + CEMDG)
SSEMDG
1/3 1/1 3/1
0.150 0.120 0.150
0.148 0.122 0.147
R (+)-limonene interface of sucrose laurate and ethoxylated mono-diglyceride respectively, Row is the weight fraction of R (+)-limonene in water + R (+)-limonene, and X is the weight fraction of mixed surfactants in the microemulsions. In order to do the calculations we estimate that the water and R (+)-limonene are pure and do not dissolve in each other. By plotting XEMDG versus 1x −1 a straight line is S obtained. SEMDG is the intercept and should be equal to the value of CEMDG/(CL1695 + CEMDG). The obtained values of SSEMDG are in good agreement with the values obtained for CEMDG/(CL1695 + CEMDG) at equal amounts of water and R (+)-limonene in the microemulsions as shown in Table 1. In other words, it is assumed that the sucrose laurate molecules are present only at the surfactant layers inside the microemulsion phase. The ethoxylated mono-di-glyceride molecules are distributed between the micro-water domains and the interface inside the microemulsion phase in a one phase microemulsions. Kuneida et al. [1,41–45] reported on similar results obtained with mixtures of sucrose monolaurate and polyethylene glycol alkyl ether systems in the presence of heptane, decane and hexadecane oils. The solubilization capability increases with mixing of surfactants in particular when surfactants with different hydrophilic–lipophilic balances are mixed [21]. The monomeric solubility of lipophilic surfactant (ethoxylated mono-di-glyceride) in R (+)-limonene is low as was reported in our previous study [40] and its mixing with sucrose laurate enables us to obtain large solubilization capacity of water and R (+)-limonene. 3.2. Transport properties 3.2.1. Electrical conductivity (σ) The electrical conductivity was measured on the (water + sodium chloride)/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene ((W + NaCl)/L1695/EMDG/LIM) where the mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity along the dilution line N60 (see phase diagram Fig. 3). The concentration of sodium chloride in water is 0.01 M. All samples appeared transparent and isotropic. Similar independence of phase behavior in the presence of a small amount of electrolyte is reported in the literature [13,26,46,47]. Fig. 4a displays the influence of water volume fraction on the electrical conductivity (σ). As the volume fraction of water increases the electrical conductivity increases exponentially. The increase in the electrical conductivity as function of water volume fraction is due to the increase in the fraction of sodium chloride ions that are not enclosed in the core of the microemulsions. The high values of electrical conductivity at high water volume fractions are explained by the fact that the sodium chloride ions are present in the external phase, which is the water. These results permit to distinguish between water-in-oil and oil-in-water microemulsions. Similar results and interpretations were provided by a number of authors [14,48,49] who studied electrical conductivity variations as function of water volume fraction as a method of characterization of microemulsions interdroplet interactions. Fig. 4b shows a plot of d log (σ)/d(ϕ) as function of water volume fraction for the system presented in Fig. 4a. The changes observed in the curve presented in Fig. 4b have been attributed to the occurrence of a percolation transition. Some authors [48,50] who studied the electrical conductivity of microemulsions as function of water volume fraction suggested a model for the explanation of the behavior of d log (σ)/d(ϕ) as function of water volume fraction. In this percolation model, the conductivity remains low up to a certain volume fraction (ϕc) of water. It must be emphasized that these conducting water-in-oil droplets, below ϕc are isolated from each other embedded in non-conducting continuum oil phase and hence contribute very little to the conductance. However, as the volume fraction of water reaches the percolation threshold ϕc = 0.17, some of these conductive droplets begin to contact each other and form clusters which are sufficiently close to each other. The number of such clusters increases very rapidly above the
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
107
the prefactors A, B by numerical analysis with adjustment by the least squares method using simultaneously the Eqs. (10) and (11). The computed s and t values are 0.16 and 1.0, respectively. The resulting ϕc = 0.17 obtained in this manner is close to the values obtained by the numerical estimate of the maximum of (d logσ/dϕ) versus ϕ (Fig. 4b). There is a reasonable agreement between calculated (by Eqs. (10) and (11)) and experimental values within prescribed range of composition with a mean deviation of 6%. The above equations are valid only near ϕc and cannot be extrapolated to infinite dilution and unit concentration. In addition, these are not applicable at the immediate vicinity of ϕc, where there is a continuous variation within a narrow interval around the percolation threshold.
Fig. 4. [a] Variation of the electrical conductivity (σ) of (water + sodium chloride)/ sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene system as function of water volume fraction along the dilution line N60 at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity. Sodium chloride concentration in water is 0.01 M. The phase diagram is presented in Fig. 3. [b] Plot of d log (σ)/ d(ϕ) as function of water volume fraction for the system presented in a.
percolation threshold (between ϕc = 0.17 and 0.5), giving rise to the observed changes of properties, in particular to the increase of electrical conductivity. The increase in electrical conductivity for the volume fractions ϕ above ϕc has been attributed to the transfer of counter ions from one droplet to another through water channels opening between droplets during sticky collisions through transient merging of droplets [3]. The existence and position of this threshold depends on the interactions between droplets, which control the duration of the collision and the degree of the interface overlapping, hence the probability of merging. Building up of conductivity needs attractive interactions and ϕc decreases when the strength of these interdroplet interactions increases as predicted by recent theoretical calculations [48,49]. In the present study, a theoretical model of Safran et al. [50], which is based on the dynamical picture of percolation, has been utilized to analyze the conductivity results of the system. According to the theory σ ¼ Að/c −ϕÞ−s σ ¼ Bð/−ϕc Þt
if / b /c if / N /c
3.2.2. Dynamic viscosity (η) The formation of microemulsions is a dynamic self-organizing phenomenon where aggregating disaggregating processes operate in conjunction. In their process dynamics, exchange of matters between different phases continuously occur resulting in an overall equilibrium. The flow of microemulsions under stress and the transport of molecules through them are also of potential importance in the study of microemulsion dynamics. The study of the dynamic viscosity of microemulsions can provide information on the intrinsic and derived processes in the microemulsion system, as well as furnish knowledge on the overall geometry of the particles of the dispersed phase [3,48,51,52]. Viscosity measurements evidenced the dependence of the size and shape of microemulsions droplets on the amount of solubilized water [53–56]. The understanding of structural consistencies in microemulsions has also been attempted from viscosity measurements by others [57,58]. Fig. 5 shows the variation in the dynamic viscosity as function of water volume fraction in the water/sucrose laurate/ethoxylated monodi-glyceride/R (+)-limonene system along the dilution line N60, the mixing ratio of sucrose laurate/ethoxylated mono-di-glyceride equals unity (see phase diagram in Fig. 3). The existence of two peaks in viscosity values with increasing water volume fraction implying the existence of at least three microstructural regions: one before, one after and the other in between two peaks. The maxima are indicative of the transition from a mono- to a bicontinuous structure. The gradual increase of viscosity in the range of 0–0.2 water volume fraction suggests that the dispersed phase is present in the form of droplets in the microemulsion systems. These results are interpreted considering clustering of water droplets with the mixed surfactants by the formation of a mixed monomolecular layer of the surfactants couple. For water volume fraction above 0.20, the viscosity of the system decreased and continues to
ð10Þ ð11Þ
Where s and t are, scaling exponents express the dynamic aspect of the percolation phenomenon. We have determined ϕc, s and t and
Fig. 5. Variation of the dynamic viscosity (η) of the water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene system as function of water volume fraction along the dilution line N60 at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/ sucrose laurate equals unity. The phase diagram is presented in Fig. 3.
108
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
decrease as the water volume fraction increases. At 0.7 water volume fraction the viscosity increases again and reaches a second maximum at water volume fraction equals 0.75. The increase of dynamic viscosity for water volume fractions below 0.20 indicates attractive interaction and aggregation of droplets of water phase including molecular reorganization on the interface where the water-in-oil microemulsions are present. The slow decrease in dynamic viscosity for water volume fractions between 0.20 and 0.70 indicates a transition from water-in-oil microemulsions droplets to a bicontinuous structure. The increase in dynamic viscosity for water contents between 0.70 and 0.75 indicates a structural transition from bicontinuous structure to an oil-in-water microemulsions microstructure. The sharp decrease in dynamic viscosity for water volume fractions above 0.80 indicates that the water, which is the least viscous component of the microemulsion system, becomes the outer phase and oil-in-water microemulsions are formed. The percolation phenomenon in microemulsion essentially involves droplet association i.e. clustering and fusion. It must, therefore, have a direct influence on the internal structure and hence viscosity. The study of viscosity confirmed the presence of two percolation processes for this microemulsion system. The structural inversion of the water-in-oil microemulsion to the oil-in-water type without any phase separation takes place in two stages. With increasing water volume fraction, the oil-continuous microemulsion transforms into the bicontinuous form at roughly 0.2 water volume fraction (water percolation threshold), and then at roughly 0.75 water volume fraction the bicontinuous form transforms into the water continuous structure (oil percolation threshold). Other authors [48,49,51,52] reported about similar behavior of dynamic viscosity of microemulsions based on nonionic surfactant in the presence of cosurfactants, oil and water. Djordjevic et al. [52] demonstrated also similar effect of increasing water volume fraction on the dynamic viscosity based on nonionic surfactants in the presence of water and isopropylmyristate microemulsion system. Like electrical conductivity, viscosity may also follow scaling type equations in the water-in-oil microemulsion region [59–61]. η ¼ Að/−ϕc Þ−μ if / N /c
ð12Þ
η ¼ Bð/c −ϕÞ−s if / /c
ð13Þ
where ϕ is the volume fraction of water, ϕc, is the percolation threshold, A and B are parameters, and µ and s are scaling exponents. The slopes of the log η versus log (ϕc − ϕ) plot for ϕc N ϕ and the log η versus log (ϕ − ϕc) for ϕ N ϕc plot, yield s and µ parameters. The average values of μ and s in Eqs. (12) and (13) are 1.04 and 0.18, respectively. These values are fairly near to the t and s values 0.90 and 0.16, respectively, obtained for conductivity percolation. The s value corresponds to the dynamic value of 0.16 obtained for static conductance percolation. The estimated scaling parameters which are in good agreement with the experimental values obtained with both electrical conductivity and dynamic viscosity signifies that these microemulsion system show an interdependence of the viscosity– conductivity, especially at the stage of water percolation.
and so all data are presented in arbitrary units. Only the polar head groups and water regions are visible with SAXS experiments because their electron densities are higher than the electron density of the surrounding oil. In this section, we used the SAXS technique to investigate the microemulsion system water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene microemulsion system as a function of the water volume fraction (ϕ) along the dilution line N60. The mixing ratio of ethoxylated mono-di-glyceride/sucrose laurate equals unity (see phase diagram in Fig. 3). Characteristic profiles of the system are presented in Fig. 6 as an example of what happens in these systems. In Fig. 6, we observe that in each case, the scattering profile exhibits a single intensity maximum at q ≠ 0, followed by a high-angle tail. With increasing water volume fraction, the position of the maximum moves to a lower angle. By fitting all the scattering curves (Fig. 6) to the Teubner–Strey equation [18] (Eq. (1)) we were able to derive from the values of the periodicity, d, correlation length, ξ, and the amphiphilicity factor, fa, as described in the experimental section (Eqs. (3)–(5)). The dependence of the d and ξ parameters on the volume fraction of water, ϕ was plotted in Fig. 7. In Fig. 7a we find that the periodicity, d, increases from 58 to 217 Å. Plots of d versus ϕ can probe the dimentionality of swelling along the dilution lines. Eq. (3) was used to determine the values of the correlation length, ξ. Initially, the growth of ξ (Fig. 7b) parallels that of d, with the former being smaller than the latter. ξ reaches maximum at ϕ equals 0.3, while d increases in a monotonic fashion over the whole range of water dilution. The increase of the correlation length, ξ , as function of water volume fraction is explained as follows: when the water is the dispersed phase, increasing the water volume fraction increases the size of the scattering units and the correlation length, ξ, whereas when water is in the bulk, increasing the water volume fraction dilutes the scattering units and ξ decreases. Eq. (5) was used to calculate the amphiphilicity factor, fa, (see values of fa in Table 2). In all cases, fa is negative with values that lie between −0.94 and −0.72, consistent with oscillatory behavior of the correlation function γ(r) and the appearance of a well-defined scattering peak at q ≠ 0. Within this range, for more negative fa, the microemulsion is more ordered. Other authors reported about similar results on the behavior of the microstructural parameters of nonionic microemulsions based on ethoxylated nonionic surfactants [31,63,64]. Recently, other authors [65,66] studied the formation of glass microemulsions using anhydrous powders of sugars or aqueous concentrated sugar solutions and sucrose ester surfactants in R (+)-limonene. Small-angle neutron scattering experiments confirm the presence of well-structured microemulsions with domain sizes ranging from ~35 to 60 nm in the
3.3. Structural properties 3.3.1. Periodicity, correlation length and amphiphilicity factor by saxs Small-angle X-ray scattering (SAXS) probes the pertinent colloidal length scales of 1–100 nm and therefore is the method of choice for determining size of colloidal particles. The scattering intensity depends on the different scattering length densities of the particles and the solvent. For SAXS, the scattering length density is proportional to the electron density, which is a linear function of the number region of aggregates composed of surfactant molecules [62]. However, at the time of these experiments, SAXS data were not measured on absolute scale
Fig. 6. Small angle X-ray scattering curves for samples whose compositions lie along the N60 dilution line for the system water/sucrose laurate/ethoxylated mono-di-glyceride/ R (+)-limonene at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/ sucrose laurate equals unity. The phase diagram is presented in Fig. 3.
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
Fig. 7. [a] Microemulsion periodicity, d as a function of volume fraction of water, ϕ, for samples along the dilution line N60 of water/sucrose laurate/ethoxylated mono-diglyceride/R (+)-limonene system at 25 °C. The mixing ratio (w/w) of ethoxylated monodi-glyceride/sucrose laurate equals unity. The phase diagram is presented in Fig. 3. [b] Microemulsion correlation length, ξ, as a function of volume fraction of water for the system presented in b.
109
3.3.2. Diffusion coefficients by nuclear magnetic resonance We used the diffusion coefficient obtained by nuclear magnetic resonance analysis to characterize the microstructure changes in the microemulsions system water/sucrose laurate/ethoxylated mono-diglyceride along the dilution line N60 for surfactants mixing ratio (w/w) equals unity (see phase diagram in Fig. 3). To evaluate the selfdiffusion data in terms of microstructure, the calculation of the relative diffusion coefficient, D/D0, of the different components of the microemulsion especially oil and water is needed [67]. Relative diffusion coefficients were obtained by dividing water (DWater) and oil (DOil) diffusion coefficients in the microemulsion by the diffusion coefficient of water in the pure water phase (D0Water) and oil in the neat phase (DOil 0 ). It is well-documented [15,16,67] that if the D/D0 values of water and oil differ by more than 1 order of magnitude, discrete particles of the slowly diffusing solvent are implied, whereas if the D/D0 values of water and oil are of the same order of magnitude, a bicontinuous structure is suggested. Fig. 8 shows the relative diffusion coefficients of water and R (+)-limonene as a function of the water volume fraction. One can clearly see that at the two extremes of aqueous-phase concentrations (up to 0.2 and above 0.7 wt water volume fractions), the DWater/DWater values are easily interpreted, while the in between 0 regions are somewhat more difficult to explain since gradual changes take place. As Fig. 8 indicates that microemulsions containing up to 0.2 water volume fraction, have a confined water molecules microstructure, since the relative diffusion coefficients of water and R (+)limonene differ by more than 1 order of magnitude. Microemulsions with 0.20 to 0.70 water volume fraction have a microstructure where the nor water neither oil can be distinguished as confined phases, as the diffusion coefficients of water and R (+)-limonene are of the same order of magnitude. Relative diffusion coefficients results for water volume fractions above 0.70 indicates that the oil phase is the confined one since the relative diffusion coefficients of water and R (+)-limonene differ by more than 1 order of magnitude. Comparing these results to the NMR results presented in our pervious published work [46], one can see that the relative diffusion coefficients of both water and oil are lower compared to the diffusion coefficients obtained for both oil and water in the previous study. As we suggested in our previous work the mixed surfactants content in the microemulsions samples affects the diffusion coefficients of both oil and water. This can be one of the causes for the difference in diffusion coefficients for both oil and water. The second reason that can cause the difference in the diffusion coefficients values is the ratio of mixed
case of the aqueous concentrated sugar solutions/sucrose ester surfactants/R (+)-limonene systems. With few exceptions, the patterns of microemulsion with concentrated sugar solutions are very similar to that of aqueous systems. The mixing of ethoxylated mono-di-glyceride with sucrose laurate in our systems induces high water solubilization and larger periodicity values compared to the reported systems [65,66] that are composed of single surfactant the sucrose ester.
Table 2 Values of the Amphiphilicity factor (fa) for the microemulsion system water/sucrose laurate/ethoxylated mono-di-glyceride/R (+)-limonene as a function of the water volume fraction (ϕ) along the dilution line N60 Water volume fraction (ϕ)
Amphiphilicity factor (fa)
0 0.11 0.21 0.32 0.42 0.52 0.62
−0.79 −0.94 −0.94 −0.84 −0.72 −0.72 −0.72
The mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity. Values of fa calculated from Eq. (5) from data obtained at 25 °C.
Fig. 8. Relative diffusion coefficients of water (Δ) and R (+)-limonene (O) as function of water volume fraction for samples along the dilution line N60 of water/sucrose laurate/ ethoxylated mono-di-glyceride/R (+)-limonene system at 25 °C. The mixing ratio (w/w) of ethoxylated mono-di-glyceride/sucrose laurate equals unity. The phase diagram is presented in Fig. 3.
110
M. Fanun / Journal of Molecular Liquids 142 (2008) 103–110
surfactants to oil, which was variable in our previous study, and in this study the ratio is constant for all of the samples investigated. The diffusion coefficients of sucrose laurate and ethoxylated mono-diglyceride at low water contents are very low. The values of their diffusion coefficients at 0.10 water volume fraction are 0.013 × 10− 5 and 0.011 × 10− 5 cm2/s for sucrose laurate and ethoxylated mono-diglyceride, respectively. For high water volume fractions the diffusion coefficient of sucrose laurate increases and equals 0.063 × 10− 5 cm2/s at water volume fraction equals 0.90. On the other hand, the diffusion coefficient of ethoxylated mono-di-glyceride at water volume fraction equals 0.90 equals the same value as for 0.10 water volume fraction (i.e. 0.011 × 10− 5 cm2/s). These results evidenced the interpretations presented in the phase behavior section that ethoxylated mono-diglyceride is there in the center of a hexagonal arrangement bordered by the more mobile molecules of sucrose laurate. Nuclear magnetic resonance results provide a clear picture of structural transitions along the N60 dilution line studied. As in the dynamic viscosity results three different regions are observed. The first region indicates the presence of water-in-oil droplets microstructure. The second region indicates the transition from water-in-oil microstructure to a bicontinuous microemulsion. In the third region a transition from a bicontinuous microemulsion to an oil-in-water microstructure occurs and a discrete oil-in-water microstructure can be present. 4. Conclusions • The mixed surfactants used in this study enable the formation of alcohol free safe U-type microemulsions. • The values of the weight fraction of total surfactants (CL1695 + CEMDG) in surfactants monolayer at the water-R (+)-limonene interface inside the mixed surfactant microemulsion system is much smaller than the CL1695 and CEMDG in single surfactant based system which indicates that the mutual solubilization of water and R (+)-limonene increases due to the mixing of surfactants. • Conductivity results indicate that these systems are percolated. The estimated scaling parameters indicate that the percolation is static. • Viscosity variation as function of water volume fraction demonstrates the presence of two percolation thresholds (water and oil). The placement of these thresholds supports the occurrence of structural transitions from water-in-oil to bicontinuous to oil-inwater microemulsions. • Upon increasing the volume fraction of water in the microemulsions there is an increase in the periodicity of the microemulsions that indicates an increase in the domain size in one dimension with increasing water content. • Structural transitions from water-in-oil to bicontinuous to oil-inwater microemulsions were also evidenced by the use of NMR technique. References [1] H. Kunieda, C. Solans, How to prepare microemulsions: temperature insensitive microemulsions, in: H. Kunieda, C. Solans (Eds.), Industrial Applications of Microemulsions, Marcel Dekker, Inc., New York, 1996, pp. 21–45. [2] M. Kahlweit, G. Busse, B. Faulhaber, H. Ebil, Langmuir 11 (1995) 4185–4187. [3] S.P. Moulik, B.K. Paul, Adv. Colloid Interface Sci. 78 (1998) 99. [4] T. Sottmann, R. Strey, in: J. Lyklema (Ed.), Soft Colloids V — Fundamentals in Interface and Colloid Science, Elsevier, Amsterdam, 2005, chap. 5. [5] N. Patel, U. Schmid, M.J. Lawrence, J. Agric. Food Chem. 54 (2006) 7817. [6] J. Flanagan, H. Singh, Crit. rev. food sci. nutr. 46 (2006) 221. [7] M.J. Lawrence, G.D. Rees, Adv. drug deliv. rev. 45 (2000) 89. [8] M. Kreilgaard, Adv. drug deliv. rev. 54 (2002) S77–S98 (SUPPL.). [9] K. Ogino, M. Abe (Eds.), Mixed Surfactant Systems, Surfactant Science Series, 46, Marcel Dekker, Inc., New York, 1992. [10] P.M. Holland, D.N. Rubingh, Mixed Surfactant Systems, American Chemical Society, Washington, 1992.
[11] J.F. Scamehorn, in: J.F. Scamehorn (Ed.), Phenomena in Mixed Surfactants Systems, ACS Symposium Series, vol. 311, American Chemical Society, Washington, D.C., 1986 [12] A. Bumajdad, J. Eastoe, J. Colloid Interf. Sci. 274 (2004) 268. [13] M. Fanun, W.S. Al-Diyn, Colloids surf., A Physicochem. eng. asp. 277 (2006) 83. [14] R.K. Mitra, B.K. Paul, J. Colloid Interface Sci. 283 (2005) 565. [15] O. El-Seoud, J. Mol. Liquids 72 (1997) 85. [16] O. Soderman, M. Nyden, Colloids Surfaces. A 158 (1999) 273. [17] O. Glatter, R. Strey, K.-V. Schubert, E.W. Kaler, Ber. Bunsenges. Phys. Chem. 100 (1996) 323. [18] M. Teubner, R. Strey, J. Chem. Phys. 87 (1987) 3195. [19] J. Sagalowicz, M. Michel, M. Adrian, P. Frossard, M. Rouvet, H.J. Watzke, A. Yaghmur, L. De Campo, L.O. Glatter, M.E. Leser, J. Microscopy 221 (2006) 110. [20] M. Kahlweit, R. Strey, D. Haase, H. Kunieda, T. Schmeling, B. Faulhaber, M. Borkovec, H.-F. Eicke, G. Busse, F. Eggers, T.H. Funck, H. Richmann, L. Magid, O. Söderman, P. Stilbs, J. Winkler, A. Dittrich, W. Jahn, J. Colloid Interface Sci. 118 (1987) 436. [21] P.D.T. Huibers, D.O. Shah, Langmuir 13 (1997) 5762. [22] R.J. Hunter, Foundations of Colloid Science; Oxford Science Publications, Clarendon Press, Oxford, 1989, Vol. 2, pp. 948–991. [23] G. Li, X. Kong, R. Guo, X. Wang, J. Disp. Sci. Technol. 5 (1989) 29. [24] N. Garti, V. Clement, M. Fanun, M.E. Leser, J. Agri. Food Chem. 48 (2000) 3945. [25] M.J. Hou, D.O. Shah, Langmuir 3 (1987) 1086. [26] H.F. Eicke, W. Meier, H. Hammerich, Langmuir 10 (1994) 2223. [27] J.A. Lake, Acta Crystallogr. 23 (1967) 191. [28] S.A. Teukolsky, W.T. Vetterling, B.P. Flannery, Numerical Recipes in C: The Art of Scientific computing, Cambridge University Press, New York, 1992. [29] J. Brunner-Popela, R. Mittelbach, R. Strey, K.-V. Schubert, E.W. Kaler, O. Glatter, J. Chem. Phys. 21 (1999) 10623. [30] M. Fanun, E. Wachtel, B. Antalek, A. Aserin, N. Garti, Colloids and Surf. A: 180 (2001) 173. [31] S. Ezrahi, E. Wachtel, A. Aserin, N. Garti, J. Colloid Interface Sci. 191 (1997) 277. [32] K.-V. Schubert, R. Strey, S.R. Kline, E.W. Kaler, J. Chem. Phys. 101 (1994) 5343. [33] K.-V. Schubert, R. Strey, J. Phys. Chem. 95 (1991) 8532. [34] J.F. Billman, E.W. Kaler, Langmuir 7 (1991) 1609. [35] D. Wu, A. Chen, C.S. Johnson, J. Magn. Reson. 115 (1995) 260–264. [36] E.O. Stejskal, in: E.O. Stejskal (Ed.), Encyclopedia of Nuclear Magnetic Resonance, Jhon Wiley & Sons, New York, 1995, pp. 657–658. [37] K. Aramaki, T. Hayashi, T. Katsuragi, M. Ishitobi, H. Kunieda, J. Colloid Interface Sci. 236 (2001) 14. [38] L.D. Ryan, E.W. Kaler, Langmuir, 13 (1997) 1510. [39] C. Rodriguez, D.P. Acharya, S. Hinata, M. Ishitobi, H. Kunieda, J. Colloid Interface Sci. 262 (2003) 500. [40] M. Fanun, W. Salah Al-Dyin, J. Disp. Sci. Technol. 27 (2006) 1119. [41] H. Kunieda, N. Ushio, A. Nakano, M. Miura, J. Colloid Interface Sci. 159 (1993) 37. [42] H. Kunieda, M. Yamagata, Langmuir 9 (1993) 3345. [43] H. Kunieda, A. Nakano, M. Akimura, J. Colloid Interface Sci. 170 (1995) 78. [44] H. Kunieda, A. Nakano, Ma.A. Pes, Langmuir 11 (1995) 3302. [45] Ma.A. Pes, K. Aramaki, N. Nakamura, H. Kunieda, J. Colloid Interface Sci. 178 (1996) 666. [46] M. Fanun, W. Salah Al-Diyn, J. Disp. Sci. Technol. 28 (2007) 164. [47] A. Yaghmur, A. Aserin, B. Antalek, N. Garti, Langmuir 19 (2003) 1063. [48] S.K. Mehta, R.K. Dewan, Phys. Rev. E 50 (1994) 4759. [49] S.K. Mehta, K. Bala, Fluid phase equilib. 172 (2000) 197. [50] S.A. Safran, G.S. Grest, A.L.R. Bug, Microemulsion system, in: H. Rosano, M. Clausse (Eds.), Surfactant Science Series, Marcel Dekker, Inc, New York, 1987. [51] S. Ezrahi, A. Aserin, N. Garti, J. Colloid Interface Sci. 202 (1998) 222. [52] L. Djordjevic, M. Primorac, M. Stupar, D. Krajisnik, Int. J. Pharm. 271 (2004) 11. [53] P. Ekwall, L. Mandell, K. Frontell, J. Colloid Interf. Sci. 33 (1970) 215. [54] S. Ajith, A.K. Rakshit, J. Surf. Technol. 8 (1992) 365. [55] S. Ajith, A.K. Rakshit, Langmuir 11 (1995) 1122. [56] S. Ajith, A.K. Rakshit, J. Phys. Chem. 99 (1995) 14778. [57] A. Maitra, C. Mathew, M. Varshney, J. Phys. Chem. 94 (1990) 5290. [58] C. Boned, J. Peyrelasse, J. Surf. Sci. Technol. 7 (1991) 1. [59] J. Peyrelasse, C. Boned, Phys. Rev. A 41 (1990) 938. [60] Z. Saidi, C. Matthew, J. Peyrelasse, C. Boned, Phys. Rev. A 42 (1990) 872. [61] J. Peyrelasse, C. Boned, in: S.H. Chen, et al., (Eds.), Structure and Dynamics of Strongly Interacting Colloids and Supramolecular Aggregates in Solution, Kluwer Academic Publisher, The Netherlands, 1992, pp. 801–806. [62] D.I. Iampietro, L.L. Brasher, E.W. Kaler, A. Strandner, O. Glatter, J. Phys. Chem., B. 102 (1998) 3105. [63] O. Regev, S. Ezrahi, A. Aserin, N. Garti, E. Wachtel, E.K. Kaler, A. Khan, Y. Talmon, Langmuir 12 (1996) 668. [64] S. Ezrahi, A. Aserin, N. Garti, Aggregation behavior in one-phase (Winsor IV) microemulsion systems, in: P. Kumar, K.L. Mittal (Eds.), Handbook of Microemulsion Science and Technology, Marcel Dekker, Inc., New York, 1999, pp. 185–246. [65] H. Dave, F. Gao, M. Schultz, C.C. Co, Colloids surf., A Physicochem. eng. asp. 296 (2007) 45. [66] H. Dave, F. Gao, J.-H. Lee, M. Liberatore, C.-C. Ho, C.C. Co, Nat. Materials 6 (2007) 287. [67] B. Lindman, U. Olsson, O. Soderman, in: P. Kumar, K.L. Mittal (Eds.), Handbook of Microemulsion Science and Technology, Marcel Dekker, New York, 1999, pp. 309–356.