- Email: [email protected]
The influence of surfactant mixing ratio on nano-emulsion formation by the pit method
Journal of Colloid and Interface Science 285 (2005) 388–394 www.elsevier.com/locate/jcis
The influence of surfactant mixing ratio on nano-emulsion formation by the pit method Paqui Izquierdo a , Jin Feng a , Jordi Esquena a , Tharward F. Tadros b , Joseph C. Dederen c , Maria José Garcia a,d , Núria Azemar a , Conxita Solans a,∗ a Dept. Tecnologia de Tensioactius, Institut d’Investigacions Químiques i Ambientals de Barcelona, CSIC, Jordi Girona 18-26, 08034 Barcelona, Spain b 89 Nash Grove Lane, Wokingham, Berkshire RG40 4HE, UK c ICI Belgium, Uniqema Health and Personal Care, Dorpsstraat 144A, B-3078 Meerbeek, Belgium d Dept. de Farmacia i Tecnologia Farmacèutica, Universitat de Barcelona, Avda. Joan XXIII s/n., 08028 Barcelona, Spain
Received 9 September 2004; accepted 27 October 2004 Available online 23 December 2004
Abstract The formation of O/W nano-emulsions by the PIT emulsification method in water/mixed nonionic surfactant/oil systems has been studied. The hydrophilic–lipophilic properties of the surfactant were varied by mixing polyoxyethylene 4-lauryl ether (C12 E4 ) and polyoxyethylene 6-lauryl ether (C12 E6 ). Emulsification was performed in samples with constant oil concentration (20 wt%) by fast cooling from the corresponding HLB temperature to 25 ◦ C. Nano-emulsions with droplet radius 60–70 nm and 25–30 nm were obtained at total surfactant concentrations of 4 and 8 wt%, respectively. Moreover, droplet size remained practically unchanged, independent of the surfactant mixing ratio, X C12 E6 . At 4 wt% surfactant concentration, the polydispersity and instability of nano-emulsions increased with the increase in XC12 E6 . However, at 8 wt% surfactant concentration, nano-emulsions with low polydispersity and high stability were obtained in a wide range of surfactant mixing ratios. Phase behavior studies showed that at 4 wt% surfactant concentration, three-liquid phases (W + D + O) coexist at the starting emulsification temperature. Furthermore, the excess oil phase with respect to the microemulsion D-phase increases with the increase in X C12 E6 , which could explain the increase in instability. At 8 wt% surfactant concentration, a microemulsion D-phase is present when emulsification starts. The low droplet size and polydispersity and higher stability of these nano-emulsions have been attributed, in addition to the increase in the surface or interfacial activity, to the spontaneous emulsification produced in the microemulsion D-phase. 2004 Elsevier Inc. All rights reserved. Keywords: Nano-emulsion; Mixed surfactant system; HLB temperature; Emulsion stability; Phase behavior; Ostwald ripening; Phase inversion temperature (PIT) method
1. Introduction The formation of emulsions with droplet size in the nanometer range, so called nano-emulsions [1,2], can be achieved either by high-energy emulsification methods (e.g., high-pressure homogenization) or by low-energy emulsification methods (e.g., the phase inversion temperature (PIT) method [3]). Although high-energy emulsification methods allow great control of the droplet size and a large choice of * Corresponding author. Fax: +34-93-2045904.
E-mail address: [email protected] (C. Solans). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.10.047
compositions, low-energy emulsification methods are interesting because they take advantage of the energy stored in the system to promote the formation of small droplets. In contrast to microemulsions (thermodynamically stable systems that form spontaneously), nano-emulsions are only kinetically stable. It is generally believed that to obtain stable emulsions, an optimum hydrophilic–lipophilic balance (HLB) number is required [4,5]. However, the HLB number concept, defined at 25 ◦ C, only takes into account the surfactant molecule alone and not the interactions of the surfactant with water and oil in the overall emulsion system. It has been shown that the HLB temperature or PIT [3,6,7], which takes
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
into account these interactions and other effects such as oil solubility and presence of additives [8,9], is a more suitable parameter for this purpose [10], although it only applies to ethoxylated nonionic surfactant systems. As is well known, there is a correlation between the HLB temperature of the system and the surfactant HLB number [9–11], as shown by the equation developed by Kunieda and Shinoda for mixtures of pure surfactants [11], mix mix THLB = Koil (NHLB − Noil ),
(1)
mix , H mix are the HLB temperature and the HLB where THLB HLB number of the surfactant mixture, respectively. In systems with pure surfactants, K oil is approximately 17 ◦ C/HLB unit for most alkanes and N oil is constant for a given oil [11] (it is considered as an “oil number,” which decreases as the hydrocarbon molecular weight increases). Based on the HLB temperature concept, Shinoda and Saito introduced the so-called PIT emulsification method [3], which is widely used in industry [12]. This method takes advantage of the extremely low interfacial tensions [13– 15] achieved at the HLB temperature or PIT to promote emulsification. It consists of a rapid cooling (or heating) of emulsions prepared at the HLB temperature by about 25–30 ◦ C to obtain kinetically stable O/W (or W/O) emulsions. However, it has been reported that the occurrence of phase inversion is no guarantee of the production of nanoemulsions [16,17]. Indeed, only the transition through a bicontinuous microemulsion (D) and/or a lamellar liquid crystalline phase during the cooling process produced fine O/W emulsions in a water/nonionic surfactant/oil system of cosmetic interest [16]. In this context, the results of studies on emulsification by the PIT method suggested that the main reason for the formation of small and uniform droplets is the spontaneous emulsification produced within the bicontinuous (D) microemulsion phase [13,17–19]. Similar results have been obtained in studies of nano-emulsion formation by low-energy emulsification methods at constant temperature [20,21]. We have recently reported the formation of O/W nanoemulsions with controlled droplet diameters in the range 50–160 nm, by the PIT method, in water/polyoxyethylene 4-lauryl ether/aliphatic hydrocarbon systems [19,22]. The lowest droplet sizes were obtained when at the starting emulsification temperature either a microemulsion (D) or a microemulsion and a lamellar liquid crystalline (D + Lα ) phase were the equilibrium phases [19]. However, the kinetic stability of these nano-emulsions was low. It was shown that the main mechanism of their destabilisation was Ostwald ripening [23–25]. The main objectives of the present work were to gain a better understanding of the relation between phase behavior, nano-emulsion formation, and stability. It has been reported that emulsion stability can be increased by mixing ethoxylated nonionic surfactants with similar polyoxyethylene chain lengths [26]. Accordingly, in the present study, the
389
technical grade surfactant C12 E6 was added to the previously studied [19] water/C12 E4 /isohexadecane system. The surfactant mixing ratio was varied, keeping the oil and the total surfactant concentrations constant. The PIT of samples of the water/mixed surfactants/oil system was first determined. Then nano-emulsions were prepared by the PIT method. Droplet size was determined by dynamic light scattering and stability was assessed from the changes of droplet size as a function of time. The phases involved in the emulsification process were determined by studying the phase behavior of the mixed surfactant system as a function of temperature and surfactant mixing ratio.
2. Experimental 2.1. Materials Isohexadecane (commercial name Arlamol HD) was obtained from UNIQEMA (Belgium) and used as received. Technical grade polyoxyethylene lauryl ethers with an average of 4 (NHLB = 9.7) and 6 (NHLB = 11.7) moles of ethylene oxide (EO) per surfactant molecule were supplied by Sigma and UNIQEMA, respectively. In this work they are abbreviated as C12 E4 and C12 E6 , respectively (the bars indicate that this is an average value). The surfactant mixing ratio (X C12 E6 ) was defined as the weight fraction of C12 E6 in the surfactant mixture, calculated as follows: XC12 E6 = [C12 E6 ]/([C12 E6 ] + [C12 E4 ]). The HLB number of the surfactant mixture was considered to be the algebraic average of the HLB of the individual surfactants. NaCl (purity > 99.5%) was obtained from Merck. Water was deionized and Milli-Q filtered. 2.2. Methods 2.2.1. HLB temperature determination Emulsions with an isohexadecane concentration of 20 wt% and a total surfactant concentration of either 4 or 8 wt% and different surfactant mixing ratios (X C12 E6 ) were prepared at room temperature (∼25 ◦ C) by simple shaking of the oil, nonionic surfactants, and water (10−2 mol dm−3 NaCl was added for the conductivity determinations). The conductivity of the resulting emulsions was measured as a function of temperature using a Crison 525 conductivity meter and a dipping cell (with Pt/platinized electrode). 2.2.2. Preparation of nano-emulsions Emulsions with an isohexadecane concentration of 20 wt%, a total surfactant concentration of either 4 or 8 wt%, and different surfactant mixing ratios (X C12 E6 ) were prepared, according to the PIT method [3], at a temperature near the HLB temperature and then rapidly cooled to 25 ◦ C by immersing the emulsion in an ice/water bath. When the samples reached a temperature of 25 ◦ C, they were transferred to and kept in a water bath at 25 ◦ C.
390
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
2.2.3. Nano-emulsion droplet size The mean droplet size and polydispersity index of the nano-emulsions were determined by dynamic light scattering (DLS). A Malvern 4700 photon correlation spectrometer (Malvern Instruments, Malvern, UK) with an argon laser (λ = 488 nm) was used for this purpose. Measurements were always carried out at a scattering angle of 90◦ and at a constant temperature, 25 ◦ C. Samples were considered polydisperse when the polydispersity index was higher than 0.2. 2.2.4. Nano-emulsion stability Stability was assessed by measuring the droplet size as a function of time. Ostwald ripening rates, ω, were determined according to the LSW [23–25] theory. This theory predicts a linear relation between the cube of droplet radius, r 3 , and time, t, ω being the slope of the plots, ω=
dr 3 8 C(∞)γ Vm D = , dt 9 ρRT
(a)
(2)
where C(∞) is the bulk phase solubility (the solubility of an infinitely large droplet), γ is the interfacial tension, Vm is the molar volume of the oil, D is the diffusion coefficient of the dispersed in the continuous phases, ρ is the density of the oil, R is the gas constant, and T is the absolute temperature. (b)
2.2.5. Phase diagrams Oil, water, and surfactants were weighed in ampoules and they were sealed. The samples were first homogenized and then stored at −18 ◦ C for about 12 h before the phase behavior study started. The temperature of the samples was gradually increased from 25 ◦ C, homogenizing them at each change of temperature. To detect liquid crystalline phases, samples were observed under polarized light.
Fig. 1. Conductivity as a function of temperature in the aqueous 10−2 M NaCl/C12 E4 :C12 E6 /isohexadecane systems as a function of surfactant mixing ratio, X C12 E6 , at 20 wt% oil concentration and total surfactant concentrations of: (a) 4 wt% and (b) 8 wt%. Table 1 X C12 E6 , HLB number, HLB temperature, initial droplet size, and polydispersity index (in parentheses) at 25 ◦ C in the system water/C12 E4 : C12 E6 /isohexadecane at 20 wt% oil and 4.0 and 8.0 wt% total surfactant concentrations
3. Results and discussion
X C12 E6
N HLB
T HLB (◦ C) 4 wt%
8 wt%
4 wt%
8 wt%
3.1. HLB temperature (PIT)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
9.7 9.9 10.1 10.3 10.5 10.7 10.9 11.1 11.3 11.5 11.7
43.0 48.6 53.2 56.2 60.0 62.5 66.0 – – – –
32.7 36.4 41.9 45.5 51.0 52.9 57.5 61.1 62.5 68.1 70.9
63 (0.06) 60 (0.13) 61 (0.11) 64 (0.10) 67 (0.16) 64 (0.25) 62 (0.24) – – – 123 (0.76)
27 (0.12)a 26 (0.08) 26 (0.10) 26 (0.08) 26 (0.11) 25 (0.11) 27 (0.07) 27 (0.07) 27 (0.09) 29 (0.09) 32 (0.17)
Fig. 1 shows plots of conductivity versus temperature for samples with 20 wt% oil and 4 (Fig. 1a) and 8 wt% (Fig. 1b) total surfactant concentrations and different surfactant mixing ratios, X C12 E6 . Independent of the surfactant concentration, the conductivity of the samples initially increases with the increase in temperature, and at a certain temperature, the HLB temperature or PIT, conductivity suddenly decreases. In samples with 8 wt% total surfactant concentration (Fig. 1b), the decrease in conductivity is not sharp for X C12 E6 lower than 0.7. In these samples, the transition from O/W to W/O emulsions may take place through a lamellar liquid crystalline phase. The HLB temperature values shown in Table 1 were taken as the average between the maximum and the minimum values of conductivity (i.e., neglecting the second peak). The corresponding N HLB are indicated, as well, in Table 1.
r (nm) (25 ◦ C)
a This corresponds to a microemulsion as assessed by phase behavior studies.
As expected, the HLB temperature increases with the increase in N HLB (i.e., increase in X C12 E6 ) at both surfactant concentrations. The lower values of HLB temperature found with the increase in total surfactant concentration are attributed to the decrease of the ratio of higher hydrophilic homologues in the aqueous phase [11,15,27]. Although the
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
Fig. 2. HLB temperature as a function of N HLB in the aqueous 10−2 M NaCl/C12 E4 :C12 E6 /isohexadecane systems at 20 wt% oil concentration.
Table 2 Total surfactant concentration and the corresponding K oil and N oil in the system water/C12 E4 :C12 E6 /isohexadecane at 20 wt% oil concentration S (wt%)
K oil
4.0 8.0
18.5 19.0
(◦ C)
N oil 7.3 7.9
HLB temperature for pure ethoxylated nonionic surfactants is a system property, for technical grade surfactants it depends on composition. It should be noted that at 4 wt% surfactant concentration, the HLB temperature for XC12 E6 = 1 (system with only C12 E6 ) could not be determined by conductivity due to fast coalescence of the sample. It was estimated to be 80 ◦ C by phase behavior studies. Fig. 2 shows a linear variation of the HLB temperature with surfactant HLB number at 4 and 8 wt% total surfactant concentrations. According to Eq. (1), the slope and the intercept on the y-axis of the straight line give the values of K oil and N oil , shown in Table 2. The slight variation found in K oil and N oil values for isohexadecane at both surfactant concentrations can be attributed to the different partitioning of surfactant molecules at the oil/water interface. The K oil values differ from 17 ◦ C, a value reported for most alkanes in pure surfactant systems [11], probably due to the fact that both surfactants are of technical grade. N oil at 4 wt% total surfactant concentration is 7.3, lower than that of hexadecane (7.7). However, at 8 wt% total surfactant concentration, N oil is 7.9, as is that of tetradecane (7.9). These results indicate that isohexadecane behaves as a shorter-alkyl-chain-length hydrocarbon at the higher surfactant concentration. It has been reported that the optimum HLB number for the formation of stable O/W emulsions should be approximately Noil + 3 [11]. Consequently, emulsions with different properties are to be expected for a given surfactant HLB number (i.e., surfactant mixing ratio X C12 E6 ), depending on the total surfactant concentration.
391
Fig. 3. Nano-emulsion r 3 as a function of time at 25 ◦ C in the water/ C12 E4 :C12 E6 /isohexadecane systems for various surfactant mixing ratios, X C12 E6 , at 20 wt% oil and 4 wt% total surfactant concentrations.
3.2. Formation of O/W nano-emulsions Nano-emulsions were prepared by rapid cooling of the samples from the HLB temperature to 25 ◦ C. Emulsion droplet radii, determined at 25 ◦ C immediately after emulsification, as well as the corresponding polydispersity indices, are shown in Table 1. At 4 wt% surfactant concentration, nano-emulsion droplet radii for the samples with XC12 E6 = 0–0.6 are in the range 60–67 nm. The polydispersity index values were lower than 0.2 except for samples with X C12 E6 higher than 0.4. The droplet size of nano-emulsions with XC12 E6 = 1 (only C12 E6 ) was about 123 nm and showed high polydispersity. At 8 wt% total surfactant concentration, the droplet size of nano-emulsions was considerably lower, in the range 25–32 nm in the whole range of X C12 E6 . By increasing surfactant concentration by a factor of 2, surface or interfacial activity is also increased, and consequently, droplet radii decrease to one-half, approximately. A minimum droplet radius should be expected at an optimum HLB number [28]. However, independent of the surfactant concentration, droplet radius remains practically unchanged in the range 9.7–11. Similarly, Sagitani et al. reported that the range of HLB numbers required to obtain fine emulsions with paraffin oil was wider than the theoretical with the so-called D-phase emulsification method [29], when a water/polyol mixture was used as aqueous phase. Droplet size of these nano-emulsions increased with time. As an example, Fig. 3 shows plots of r 3 versus time, for the samples with 4 wt% total surfactant concentration. The results for nano-emulsions with X C12 E6 higher than 0.4 are not shown in this figure because they experienced fast creaming. As already found for the O/W nano-emulsions of the C12 E4 systems [19,22], linear plots were obtained (Fig. 3). Therefore, the kinetics of nano-emulsion destabilization may be attributed to Ostwald ripening (i.e., oil diffusion from the small to the big droplets). The Ostwald ripening rates, ω, obtained, according to Eq. (2), from the slope of the straight lines are plotted as a function of X C12 E6 in Fig. 4. At 4 wt% surfactant concentration, the Ostwald ripening rate decreases with the increase in X C12 E6 up to 0.4.
392
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
(a) Fig. 4. Ostwald ripening rates, ω, as a function of surfactant mixing ratio, X C12 E6 , at 25 ◦ C in the water/C12 E4 :C12 E6 /isohexadecane systems at 20 wt% oil concentration.
However, the trend was different at 8 wt% surfactant concentration. Nano-emulsions with XC12 E6 = 0.1 are highly unstable, polydispersity increases very fast (probably due to coalescence), and creaming starts within 24 h. In contrast, the Ostwald ripening rate, ω, decreases at X C12 E6 values between 0.2 and 0.7 (increase in C12 E6 concentration), and at X C12 E6 higher than 0.7, ω is very low and remains practically unchanged, as shown in Fig. 4. The decrease in the destabilization rate, ω, with the increase in X C12 E6 , up to 0.4 for samples with 4 wt% and to 0.7 for samples with 8 wt% surfactant concentration, cannot be explained only by the LSW theory [23–25]. Although an increase in the oil molecular diffusion rate should be expected with an increase in the interfacial tension (Eq. (2)), the oil/water interfacial tension for these samples, at 25 ◦ C, is likely to be of the same order of magnitude. Therefore, this decrease in ω could be attributed to the decrease in oil diffusion due to other factors, such as micellar solubilization [30,31], partition of surfactant molecules between the oil and aqueous phase [19], and proximity to the HLB temperature [13,14]. It has been reported that the oil transference between emulsion droplets is controlled by interfacial phenomena (i.e., micellar solubilization) due to the presence of micelles. By increasing C12 E6 concentration (i.e., increasing X C12 E6 ), micellar properties and consequently micellar solubilization change. The increase in X C12 E6 also favors the increase in concentration of molecules with higher EO chains at the oil/water interface. It is likely that accumulation of high HLB molecules results in an increase of the Gibbs elasticity and this may result in a decrease of ω [32]. On the other hand, this increase in stability is what it would be expected considering that an increase in HLB temperature is produced. It is well known that although emulsification is favored at the HLB temperature due to the extremely low values of interfacial tension [13–15], emulsion stability is very low due to fast coalescence of the droplets [13,14]. Moreover, with a higher difference between the HLB temperature and the experimental temperature (i.e., 25 ◦ C), emulsion stability against coalescence is enhanced [3] because of the increase in interfacial tension as
(b) Fig. 5. Phase diagrams of the water/C12 E4 :C12 E6 /isohexadecane systems as a function of temperature and surfactant mixing ratio, X C12 E6 , at 20 wt% oil concentration and total surfactant concentrations of (a) 4 and (b) 8 wt%. W: isotropic liquid aqueous phase; D: isotropic bluish liquid phase (bicontinuous microemulsion); Lα : anisotropic phase (lamellar liquid crystalline phase); O: isotropic liquid colorless oil phase; Wm : isotropic bluish liquid phase (micellar solution or O/W microemulsion); Om : isotropic liquid colorless phase (inverse micellar solution or W/O microemulsion).
the system moves away from the minimum interfacial tension values [14]. The different stability of nano-emulsions with the increase in X C12 E6 and total surfactant concentration could also be related to different phase transitions during the emulsification process. The different type of phases involved in the emulsification, depending on the surfactant concentration, could play an important role in the process, as already observed for the water/C12 E4 /isohexadecane system [19]. Accordingly, the phase behavior of the mixing surfactant system as a function of temperature was determined. 3.3. Phase behavior The phase behavior of the water/mixed surfactant/oil system, as a function of temperature and surfactant mixing ratio, is shown in Fig. 5a (4 wt% surfactant) and Fig. 5b (8 wt% surfactant). Independent of the total surfactant concentration, two liquid phases (Wm + O), consisting of O/W microemulsion (Wm ) and oil (O) phases coexist at lower temperatures, while water (W) and W/O microemulsion (Om ) phases coexist at high temperatures. At intermediate temperatures, the phase
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
393
D-phase and during the cooling process to 25 ◦ C, transitions through (D + Lα )-(D)-(Wm + O) are produced. The lower droplet sizes found at this surfactant concentration could be attributed to the spontaneous emulsification at the D-phase in addition to the increase in surface activity as a consequence of the increase in surfactant concentration.
4. Summary
Fig. 6. Phase equilibrium as a function of X C12 E6 for samples with 20 wt% oil and 4 wt% total surfactant concentrations in the systems water/C12 E4 :C12 E6 /isohexadecane at the corresponding HLB temperature. W: isotropic liquid aqueous phase; D: isotropic bluish liquid phase (bicontinuous microemulsion); O: isotropic liquid colorless oil phase.
behavior shows distinctive features depending on the surfactant concentration. At 4 wt% surfactant (Fig. 5a), a threeliquid-phase region (W + D + O) appears, consisting of water (W), shear-birrefringent microemulsion (D), and oil (O) phases. As the surfactant mixing ratio X C12 E6 increases, this region shifts to higher temperatures within the range 40– 75 ◦ C. At 8 wt% surfactant (Fig. 5b), two microemulsion (D) phase regions are separated by a two-phase region (D + Lα ), consisting of microemulsion (D) and lamellar liquid crystalline (Lα ) phases. These phases are also shifted to higher temperatures, within the range 25–72 ◦ C, as the X C12 E6 increases. It is noteworthy that the HLB temperature values determined by conductivity, shown in Table 1, are located in the upper part of the (W + D + O) region at 4 wt% surfactant and in the upper D-phase at 8 wt% surfactant. Therefore, at 4 wt% surfactant concentration, emulsification starts in the upper part of the three-phase region (W + D + O) and by decreasing the temperature to 25 ◦ C, the system evolves to an O/W structure. Fig. 6 shows the phase equilibria and the corresponding oil (O)-to-microemulsion (D) phase ratios (O/D) at the starting emulsification temperature as a function of X C12 E6 in the range 0.3–1. The excess oil phase increases with respect to the D-phase as the X C12 E6 increases. This increase in excess oil phase (i.e., decrease in microemulsion D-phase), which is more pronounced at X C12 E6 values higher than 0.5, may influence the emulsification process. It has been shown that emulsion droplet size and polydispersity are related to the presence of a microemulsion D-phase in the system at the starting temperature of emulsification [13]. Therefore, the increase in droplet size, polydispersity, and consequently instability produced at X C12 E6 values higher than 0.4 could be due to the decrease in microemulsion D-phase. At 8 wt% surfactant concentration, emulsification starts in the upper
O/W nano-emulsions with droplet radii 60–70 nm and 25–30 nm have been obtained by the PIT method in water/C12 E4 :C12 E6 /isohexadecane systems at 20 wt% oil and total surfactant concentrations of 4 and 8 wt%, respectively. Nano-emulsion droplet size remained practically unchanged independent of the surfactant mixing ratio, X C12 E6 , and it decreased by increasing surfactant concentration from 4 to 8 wt%. At 4 wt% surfactant concentration, nanoemulsion polydispersity and instability increased with the increase in C12 E6 concentration at X C12 E6 values higher than 0.4. This increase in nano-emulsion polydispersity and instability may be related to the increase in excess oil phase with respect to the microemulsion D-phase at the starting emulsification temperature, in which three liquid phases (W + D + O) coexist. At 8 wt% surfactant concentration, the range of surfactant mixing ratios for the formation of nano-emulsions with high stability is higher. It has been attributed to the spontaneous emulsification produced in the microemulsion D-phase at the starting emulsification temperature.
Acknowledgments The authors acknowledge financial support by UNIQEMA and partial support by “Generalitat de Catalunya,” DURSI (Grant 2001SGR-00357), and the Spanish Ministry of Science and Technology, DGI (Grant PPQ2002-04514CO3-03).
References [1] H. Nakajima, S. Tomomasa, M. Okabe, in: Proceedings of First World Emulsion Conference, vol. 1, EDS, Paris, 1993, pp. 1–11. [2] C. Solans, J. Esquena, A. Forgiarini, N. Usón, D. Morales, P. Izquierdo, N. Azemar, M.J. García, in: K.L. Mittal, D.O. Shah (Eds.), Adsorption and Aggregation of Surfactants in Solution, Dekker, New York, 2002, pp. 525–554. [3] K. Shinoda, H. Saito, J. Colloid Interface Sci. 26 (1968) 70. [4] W.C. Griffin, J. Soc. Cosmet. Chem. 1 (1949) 311. [5] J.T. Davies, Interfacial Phenomena, Academic Press, New York, 1963, pp. 371–383. [6] K. Shinoda, H. Arai, J. Phys. Chem. 68 (1964) 3485. [7] K. Shinoda, J. Colloid Interface Sci. 24 (1967) 4.
394
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
[8] L. Marszall, in: M.J. Shick (Ed.), Nonionic Surfactants, Surfactant Sciences Series, vol. 23, Dekker, New York, 1987, pp. 493–547. [9] K. Shinoda, H. Kunieda, in: P. Becher (Ed.), Encyclopedia of Emulsion Technology, vol. 1, Dekker, New York, 1983, pp. 337–367. [10] H. Kunieda, Y. Sato, in: S.E. Friberg, B. Lindman (Eds.), Organized Solutions: Surfactant in Science and Technology, Surfactant Science Series, vol. 44, Dekker, New York, 1992, pp. 67–88. [11] H. Kunieda, K. Shinoda, J. Colloid Interface Sci. 107 (1) (1985) 107– 121. [12] T. Förster, W. Von Rybinski, in: B.P. Binks (Ed.), Modern Aspects of Emulsion Science, Royal Soc. Chemistry, 1998, pp. 395–426. [13] S. Friberg, C. Solans, J. Colloid Interface Sci. 66 (1978) 66. [14] H. Kunieda, S. Friberg, Bull. Chem. Soc. Jpn. 54 (4) (1981) 1010– 1014. [15] M. Bourrel, R.S. Schechter, in: Microemulsions and Related Systems, Surfactant Science Series, vol. 30, Dekker, New York, 1988. [16] A. Wadle, T. Förster, W. Von Rybinski, Colloids Surf. A Physicochem. Eng. Aspects 76 (1993) 51–57. [17] D. Morales, J.M. Gutiérrez, M.J. García-Celma, C. Solans, Langmuir 19 (2003) 7196–7200. [18] C. Solans, J.J. García-Domínguez, S.E. Friberg, Proc. Jornadas CED 12 (1981) 509–525. [19] P. Izquierdo, J. Esquena, Th.F. Tadros, C. Dederen, J. Feng, M.J. García, N. Azemar, C. Solans, Langmuir 20 (2004) 6594–6598.
[20] H. Sagitani, in: S. Friberg, B. Lindman (Eds.), Organized Solutions: Surfactant in Science and Technology, Surfactant Science Series, vol. 44, Dekker, New York, 1992, pp. 259–271. [21] A. Forgiarini, J. Esquena, C. González, C. Solans, Langmuir 17 (7) (2001) 2076–2083. [22] P. Izquierdo, J. Esquena, Th.F. Tadros, C. Dederen, M.J. García, N. Azemar, C. Solans, Langmuir 18 (2002) 26–30. [23] I.M. Lifshitz, V.V. Slezov, J. Phys. Chem. Solids 19 (1961) 35. [24] C.Z. Wagner, Elektrochem. 65 (1961) 581. [25] A.S. Kabalnov, D.E. Shchuckin, Adv. Colloid Interface Sci. 38 (1992) 69. [26] K. Shinoda, H. Saito, H. Arai, J. Colloid Interface Sci. 35 (1971) 624. [27] H. Kunieda, N. Ishikawa, J. ColIoid Interface Sci. 107 (1985) 122– 128. [28] T.J. Lin, H. Kurihara, H. Ohta, J. Soc. Cosmet. Chem. 26 (1975) 121– 139. [29] H. Sagitani, Y. Hirai, K. Nabeta, M. Nagai, J. Jpn. Oil Chem. Soc. 35 (1986) 102–107. [30] J. Weiss, J.N. Coupland, D. Brathwaite, D.J. Clements, Colloids Surf. A Physicochem. Eng. Aspects 121 (1997) 53–60. [31] A.A. Peña, C.A. Miller, J. Colloid Interface Sci. 244 (2001) 154– 163. [32] P. Walstra, in: P. Becher (Ed.), Encyclopedia of Emulsion Technology, vol. 4, Dekker, New York, 1996, pp. 1–62.
The influence of surfactant mixing ratio on nano-emulsion formation by the pit method Paqui Izquierdo a , Jin Feng a , Jordi Esquena a , Tharward F. Tadros b , Joseph C. Dederen c , Maria José Garcia a,d , Núria Azemar a , Conxita Solans a,∗ a Dept. Tecnologia de Tensioactius, Institut d’Investigacions Químiques i Ambientals de Barcelona, CSIC, Jordi Girona 18-26, 08034 Barcelona, Spain b 89 Nash Grove Lane, Wokingham, Berkshire RG40 4HE, UK c ICI Belgium, Uniqema Health and Personal Care, Dorpsstraat 144A, B-3078 Meerbeek, Belgium d Dept. de Farmacia i Tecnologia Farmacèutica, Universitat de Barcelona, Avda. Joan XXIII s/n., 08028 Barcelona, Spain
Received 9 September 2004; accepted 27 October 2004 Available online 23 December 2004
Abstract The formation of O/W nano-emulsions by the PIT emulsification method in water/mixed nonionic surfactant/oil systems has been studied. The hydrophilic–lipophilic properties of the surfactant were varied by mixing polyoxyethylene 4-lauryl ether (C12 E4 ) and polyoxyethylene 6-lauryl ether (C12 E6 ). Emulsification was performed in samples with constant oil concentration (20 wt%) by fast cooling from the corresponding HLB temperature to 25 ◦ C. Nano-emulsions with droplet radius 60–70 nm and 25–30 nm were obtained at total surfactant concentrations of 4 and 8 wt%, respectively. Moreover, droplet size remained practically unchanged, independent of the surfactant mixing ratio, X C12 E6 . At 4 wt% surfactant concentration, the polydispersity and instability of nano-emulsions increased with the increase in XC12 E6 . However, at 8 wt% surfactant concentration, nano-emulsions with low polydispersity and high stability were obtained in a wide range of surfactant mixing ratios. Phase behavior studies showed that at 4 wt% surfactant concentration, three-liquid phases (W + D + O) coexist at the starting emulsification temperature. Furthermore, the excess oil phase with respect to the microemulsion D-phase increases with the increase in X C12 E6 , which could explain the increase in instability. At 8 wt% surfactant concentration, a microemulsion D-phase is present when emulsification starts. The low droplet size and polydispersity and higher stability of these nano-emulsions have been attributed, in addition to the increase in the surface or interfacial activity, to the spontaneous emulsification produced in the microemulsion D-phase. 2004 Elsevier Inc. All rights reserved. Keywords: Nano-emulsion; Mixed surfactant system; HLB temperature; Emulsion stability; Phase behavior; Ostwald ripening; Phase inversion temperature (PIT) method
1. Introduction The formation of emulsions with droplet size in the nanometer range, so called nano-emulsions [1,2], can be achieved either by high-energy emulsification methods (e.g., high-pressure homogenization) or by low-energy emulsification methods (e.g., the phase inversion temperature (PIT) method [3]). Although high-energy emulsification methods allow great control of the droplet size and a large choice of * Corresponding author. Fax: +34-93-2045904.
E-mail address: [email protected] (C. Solans). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.10.047
compositions, low-energy emulsification methods are interesting because they take advantage of the energy stored in the system to promote the formation of small droplets. In contrast to microemulsions (thermodynamically stable systems that form spontaneously), nano-emulsions are only kinetically stable. It is generally believed that to obtain stable emulsions, an optimum hydrophilic–lipophilic balance (HLB) number is required [4,5]. However, the HLB number concept, defined at 25 ◦ C, only takes into account the surfactant molecule alone and not the interactions of the surfactant with water and oil in the overall emulsion system. It has been shown that the HLB temperature or PIT [3,6,7], which takes
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
into account these interactions and other effects such as oil solubility and presence of additives [8,9], is a more suitable parameter for this purpose [10], although it only applies to ethoxylated nonionic surfactant systems. As is well known, there is a correlation between the HLB temperature of the system and the surfactant HLB number [9–11], as shown by the equation developed by Kunieda and Shinoda for mixtures of pure surfactants [11], mix mix THLB = Koil (NHLB − Noil ),
(1)
mix , H mix are the HLB temperature and the HLB where THLB HLB number of the surfactant mixture, respectively. In systems with pure surfactants, K oil is approximately 17 ◦ C/HLB unit for most alkanes and N oil is constant for a given oil [11] (it is considered as an “oil number,” which decreases as the hydrocarbon molecular weight increases). Based on the HLB temperature concept, Shinoda and Saito introduced the so-called PIT emulsification method [3], which is widely used in industry [12]. This method takes advantage of the extremely low interfacial tensions [13– 15] achieved at the HLB temperature or PIT to promote emulsification. It consists of a rapid cooling (or heating) of emulsions prepared at the HLB temperature by about 25–30 ◦ C to obtain kinetically stable O/W (or W/O) emulsions. However, it has been reported that the occurrence of phase inversion is no guarantee of the production of nanoemulsions [16,17]. Indeed, only the transition through a bicontinuous microemulsion (D) and/or a lamellar liquid crystalline phase during the cooling process produced fine O/W emulsions in a water/nonionic surfactant/oil system of cosmetic interest [16]. In this context, the results of studies on emulsification by the PIT method suggested that the main reason for the formation of small and uniform droplets is the spontaneous emulsification produced within the bicontinuous (D) microemulsion phase [13,17–19]. Similar results have been obtained in studies of nano-emulsion formation by low-energy emulsification methods at constant temperature [20,21]. We have recently reported the formation of O/W nanoemulsions with controlled droplet diameters in the range 50–160 nm, by the PIT method, in water/polyoxyethylene 4-lauryl ether/aliphatic hydrocarbon systems [19,22]. The lowest droplet sizes were obtained when at the starting emulsification temperature either a microemulsion (D) or a microemulsion and a lamellar liquid crystalline (D + Lα ) phase were the equilibrium phases [19]. However, the kinetic stability of these nano-emulsions was low. It was shown that the main mechanism of their destabilisation was Ostwald ripening [23–25]. The main objectives of the present work were to gain a better understanding of the relation between phase behavior, nano-emulsion formation, and stability. It has been reported that emulsion stability can be increased by mixing ethoxylated nonionic surfactants with similar polyoxyethylene chain lengths [26]. Accordingly, in the present study, the
389
technical grade surfactant C12 E6 was added to the previously studied [19] water/C12 E4 /isohexadecane system. The surfactant mixing ratio was varied, keeping the oil and the total surfactant concentrations constant. The PIT of samples of the water/mixed surfactants/oil system was first determined. Then nano-emulsions were prepared by the PIT method. Droplet size was determined by dynamic light scattering and stability was assessed from the changes of droplet size as a function of time. The phases involved in the emulsification process were determined by studying the phase behavior of the mixed surfactant system as a function of temperature and surfactant mixing ratio.
2. Experimental 2.1. Materials Isohexadecane (commercial name Arlamol HD) was obtained from UNIQEMA (Belgium) and used as received. Technical grade polyoxyethylene lauryl ethers with an average of 4 (NHLB = 9.7) and 6 (NHLB = 11.7) moles of ethylene oxide (EO) per surfactant molecule were supplied by Sigma and UNIQEMA, respectively. In this work they are abbreviated as C12 E4 and C12 E6 , respectively (the bars indicate that this is an average value). The surfactant mixing ratio (X C12 E6 ) was defined as the weight fraction of C12 E6 in the surfactant mixture, calculated as follows: XC12 E6 = [C12 E6 ]/([C12 E6 ] + [C12 E4 ]). The HLB number of the surfactant mixture was considered to be the algebraic average of the HLB of the individual surfactants. NaCl (purity > 99.5%) was obtained from Merck. Water was deionized and Milli-Q filtered. 2.2. Methods 2.2.1. HLB temperature determination Emulsions with an isohexadecane concentration of 20 wt% and a total surfactant concentration of either 4 or 8 wt% and different surfactant mixing ratios (X C12 E6 ) were prepared at room temperature (∼25 ◦ C) by simple shaking of the oil, nonionic surfactants, and water (10−2 mol dm−3 NaCl was added for the conductivity determinations). The conductivity of the resulting emulsions was measured as a function of temperature using a Crison 525 conductivity meter and a dipping cell (with Pt/platinized electrode). 2.2.2. Preparation of nano-emulsions Emulsions with an isohexadecane concentration of 20 wt%, a total surfactant concentration of either 4 or 8 wt%, and different surfactant mixing ratios (X C12 E6 ) were prepared, according to the PIT method [3], at a temperature near the HLB temperature and then rapidly cooled to 25 ◦ C by immersing the emulsion in an ice/water bath. When the samples reached a temperature of 25 ◦ C, they were transferred to and kept in a water bath at 25 ◦ C.
390
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
2.2.3. Nano-emulsion droplet size The mean droplet size and polydispersity index of the nano-emulsions were determined by dynamic light scattering (DLS). A Malvern 4700 photon correlation spectrometer (Malvern Instruments, Malvern, UK) with an argon laser (λ = 488 nm) was used for this purpose. Measurements were always carried out at a scattering angle of 90◦ and at a constant temperature, 25 ◦ C. Samples were considered polydisperse when the polydispersity index was higher than 0.2. 2.2.4. Nano-emulsion stability Stability was assessed by measuring the droplet size as a function of time. Ostwald ripening rates, ω, were determined according to the LSW [23–25] theory. This theory predicts a linear relation between the cube of droplet radius, r 3 , and time, t, ω being the slope of the plots, ω=
dr 3 8 C(∞)γ Vm D = , dt 9 ρRT
(a)
(2)
where C(∞) is the bulk phase solubility (the solubility of an infinitely large droplet), γ is the interfacial tension, Vm is the molar volume of the oil, D is the diffusion coefficient of the dispersed in the continuous phases, ρ is the density of the oil, R is the gas constant, and T is the absolute temperature. (b)
2.2.5. Phase diagrams Oil, water, and surfactants were weighed in ampoules and they were sealed. The samples were first homogenized and then stored at −18 ◦ C for about 12 h before the phase behavior study started. The temperature of the samples was gradually increased from 25 ◦ C, homogenizing them at each change of temperature. To detect liquid crystalline phases, samples were observed under polarized light.
Fig. 1. Conductivity as a function of temperature in the aqueous 10−2 M NaCl/C12 E4 :C12 E6 /isohexadecane systems as a function of surfactant mixing ratio, X C12 E6 , at 20 wt% oil concentration and total surfactant concentrations of: (a) 4 wt% and (b) 8 wt%. Table 1 X C12 E6 , HLB number, HLB temperature, initial droplet size, and polydispersity index (in parentheses) at 25 ◦ C in the system water/C12 E4 : C12 E6 /isohexadecane at 20 wt% oil and 4.0 and 8.0 wt% total surfactant concentrations
3. Results and discussion
X C12 E6
N HLB
T HLB (◦ C) 4 wt%
8 wt%
4 wt%
8 wt%
3.1. HLB temperature (PIT)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
9.7 9.9 10.1 10.3 10.5 10.7 10.9 11.1 11.3 11.5 11.7
43.0 48.6 53.2 56.2 60.0 62.5 66.0 – – – –
32.7 36.4 41.9 45.5 51.0 52.9 57.5 61.1 62.5 68.1 70.9
63 (0.06) 60 (0.13) 61 (0.11) 64 (0.10) 67 (0.16) 64 (0.25) 62 (0.24) – – – 123 (0.76)
27 (0.12)a 26 (0.08) 26 (0.10) 26 (0.08) 26 (0.11) 25 (0.11) 27 (0.07) 27 (0.07) 27 (0.09) 29 (0.09) 32 (0.17)
Fig. 1 shows plots of conductivity versus temperature for samples with 20 wt% oil and 4 (Fig. 1a) and 8 wt% (Fig. 1b) total surfactant concentrations and different surfactant mixing ratios, X C12 E6 . Independent of the surfactant concentration, the conductivity of the samples initially increases with the increase in temperature, and at a certain temperature, the HLB temperature or PIT, conductivity suddenly decreases. In samples with 8 wt% total surfactant concentration (Fig. 1b), the decrease in conductivity is not sharp for X C12 E6 lower than 0.7. In these samples, the transition from O/W to W/O emulsions may take place through a lamellar liquid crystalline phase. The HLB temperature values shown in Table 1 were taken as the average between the maximum and the minimum values of conductivity (i.e., neglecting the second peak). The corresponding N HLB are indicated, as well, in Table 1.
r (nm) (25 ◦ C)
a This corresponds to a microemulsion as assessed by phase behavior studies.
As expected, the HLB temperature increases with the increase in N HLB (i.e., increase in X C12 E6 ) at both surfactant concentrations. The lower values of HLB temperature found with the increase in total surfactant concentration are attributed to the decrease of the ratio of higher hydrophilic homologues in the aqueous phase [11,15,27]. Although the
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
Fig. 2. HLB temperature as a function of N HLB in the aqueous 10−2 M NaCl/C12 E4 :C12 E6 /isohexadecane systems at 20 wt% oil concentration.
Table 2 Total surfactant concentration and the corresponding K oil and N oil in the system water/C12 E4 :C12 E6 /isohexadecane at 20 wt% oil concentration S (wt%)
K oil
4.0 8.0
18.5 19.0
(◦ C)
N oil 7.3 7.9
HLB temperature for pure ethoxylated nonionic surfactants is a system property, for technical grade surfactants it depends on composition. It should be noted that at 4 wt% surfactant concentration, the HLB temperature for XC12 E6 = 1 (system with only C12 E6 ) could not be determined by conductivity due to fast coalescence of the sample. It was estimated to be 80 ◦ C by phase behavior studies. Fig. 2 shows a linear variation of the HLB temperature with surfactant HLB number at 4 and 8 wt% total surfactant concentrations. According to Eq. (1), the slope and the intercept on the y-axis of the straight line give the values of K oil and N oil , shown in Table 2. The slight variation found in K oil and N oil values for isohexadecane at both surfactant concentrations can be attributed to the different partitioning of surfactant molecules at the oil/water interface. The K oil values differ from 17 ◦ C, a value reported for most alkanes in pure surfactant systems [11], probably due to the fact that both surfactants are of technical grade. N oil at 4 wt% total surfactant concentration is 7.3, lower than that of hexadecane (7.7). However, at 8 wt% total surfactant concentration, N oil is 7.9, as is that of tetradecane (7.9). These results indicate that isohexadecane behaves as a shorter-alkyl-chain-length hydrocarbon at the higher surfactant concentration. It has been reported that the optimum HLB number for the formation of stable O/W emulsions should be approximately Noil + 3 [11]. Consequently, emulsions with different properties are to be expected for a given surfactant HLB number (i.e., surfactant mixing ratio X C12 E6 ), depending on the total surfactant concentration.
391
Fig. 3. Nano-emulsion r 3 as a function of time at 25 ◦ C in the water/ C12 E4 :C12 E6 /isohexadecane systems for various surfactant mixing ratios, X C12 E6 , at 20 wt% oil and 4 wt% total surfactant concentrations.
3.2. Formation of O/W nano-emulsions Nano-emulsions were prepared by rapid cooling of the samples from the HLB temperature to 25 ◦ C. Emulsion droplet radii, determined at 25 ◦ C immediately after emulsification, as well as the corresponding polydispersity indices, are shown in Table 1. At 4 wt% surfactant concentration, nano-emulsion droplet radii for the samples with XC12 E6 = 0–0.6 are in the range 60–67 nm. The polydispersity index values were lower than 0.2 except for samples with X C12 E6 higher than 0.4. The droplet size of nano-emulsions with XC12 E6 = 1 (only C12 E6 ) was about 123 nm and showed high polydispersity. At 8 wt% total surfactant concentration, the droplet size of nano-emulsions was considerably lower, in the range 25–32 nm in the whole range of X C12 E6 . By increasing surfactant concentration by a factor of 2, surface or interfacial activity is also increased, and consequently, droplet radii decrease to one-half, approximately. A minimum droplet radius should be expected at an optimum HLB number [28]. However, independent of the surfactant concentration, droplet radius remains practically unchanged in the range 9.7–11. Similarly, Sagitani et al. reported that the range of HLB numbers required to obtain fine emulsions with paraffin oil was wider than the theoretical with the so-called D-phase emulsification method [29], when a water/polyol mixture was used as aqueous phase. Droplet size of these nano-emulsions increased with time. As an example, Fig. 3 shows plots of r 3 versus time, for the samples with 4 wt% total surfactant concentration. The results for nano-emulsions with X C12 E6 higher than 0.4 are not shown in this figure because they experienced fast creaming. As already found for the O/W nano-emulsions of the C12 E4 systems [19,22], linear plots were obtained (Fig. 3). Therefore, the kinetics of nano-emulsion destabilization may be attributed to Ostwald ripening (i.e., oil diffusion from the small to the big droplets). The Ostwald ripening rates, ω, obtained, according to Eq. (2), from the slope of the straight lines are plotted as a function of X C12 E6 in Fig. 4. At 4 wt% surfactant concentration, the Ostwald ripening rate decreases with the increase in X C12 E6 up to 0.4.
392
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
(a) Fig. 4. Ostwald ripening rates, ω, as a function of surfactant mixing ratio, X C12 E6 , at 25 ◦ C in the water/C12 E4 :C12 E6 /isohexadecane systems at 20 wt% oil concentration.
However, the trend was different at 8 wt% surfactant concentration. Nano-emulsions with XC12 E6 = 0.1 are highly unstable, polydispersity increases very fast (probably due to coalescence), and creaming starts within 24 h. In contrast, the Ostwald ripening rate, ω, decreases at X C12 E6 values between 0.2 and 0.7 (increase in C12 E6 concentration), and at X C12 E6 higher than 0.7, ω is very low and remains practically unchanged, as shown in Fig. 4. The decrease in the destabilization rate, ω, with the increase in X C12 E6 , up to 0.4 for samples with 4 wt% and to 0.7 for samples with 8 wt% surfactant concentration, cannot be explained only by the LSW theory [23–25]. Although an increase in the oil molecular diffusion rate should be expected with an increase in the interfacial tension (Eq. (2)), the oil/water interfacial tension for these samples, at 25 ◦ C, is likely to be of the same order of magnitude. Therefore, this decrease in ω could be attributed to the decrease in oil diffusion due to other factors, such as micellar solubilization [30,31], partition of surfactant molecules between the oil and aqueous phase [19], and proximity to the HLB temperature [13,14]. It has been reported that the oil transference between emulsion droplets is controlled by interfacial phenomena (i.e., micellar solubilization) due to the presence of micelles. By increasing C12 E6 concentration (i.e., increasing X C12 E6 ), micellar properties and consequently micellar solubilization change. The increase in X C12 E6 also favors the increase in concentration of molecules with higher EO chains at the oil/water interface. It is likely that accumulation of high HLB molecules results in an increase of the Gibbs elasticity and this may result in a decrease of ω [32]. On the other hand, this increase in stability is what it would be expected considering that an increase in HLB temperature is produced. It is well known that although emulsification is favored at the HLB temperature due to the extremely low values of interfacial tension [13–15], emulsion stability is very low due to fast coalescence of the droplets [13,14]. Moreover, with a higher difference between the HLB temperature and the experimental temperature (i.e., 25 ◦ C), emulsion stability against coalescence is enhanced [3] because of the increase in interfacial tension as
(b) Fig. 5. Phase diagrams of the water/C12 E4 :C12 E6 /isohexadecane systems as a function of temperature and surfactant mixing ratio, X C12 E6 , at 20 wt% oil concentration and total surfactant concentrations of (a) 4 and (b) 8 wt%. W: isotropic liquid aqueous phase; D: isotropic bluish liquid phase (bicontinuous microemulsion); Lα : anisotropic phase (lamellar liquid crystalline phase); O: isotropic liquid colorless oil phase; Wm : isotropic bluish liquid phase (micellar solution or O/W microemulsion); Om : isotropic liquid colorless phase (inverse micellar solution or W/O microemulsion).
the system moves away from the minimum interfacial tension values [14]. The different stability of nano-emulsions with the increase in X C12 E6 and total surfactant concentration could also be related to different phase transitions during the emulsification process. The different type of phases involved in the emulsification, depending on the surfactant concentration, could play an important role in the process, as already observed for the water/C12 E4 /isohexadecane system [19]. Accordingly, the phase behavior of the mixing surfactant system as a function of temperature was determined. 3.3. Phase behavior The phase behavior of the water/mixed surfactant/oil system, as a function of temperature and surfactant mixing ratio, is shown in Fig. 5a (4 wt% surfactant) and Fig. 5b (8 wt% surfactant). Independent of the total surfactant concentration, two liquid phases (Wm + O), consisting of O/W microemulsion (Wm ) and oil (O) phases coexist at lower temperatures, while water (W) and W/O microemulsion (Om ) phases coexist at high temperatures. At intermediate temperatures, the phase
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
393
D-phase and during the cooling process to 25 ◦ C, transitions through (D + Lα )-(D)-(Wm + O) are produced. The lower droplet sizes found at this surfactant concentration could be attributed to the spontaneous emulsification at the D-phase in addition to the increase in surface activity as a consequence of the increase in surfactant concentration.
4. Summary
Fig. 6. Phase equilibrium as a function of X C12 E6 for samples with 20 wt% oil and 4 wt% total surfactant concentrations in the systems water/C12 E4 :C12 E6 /isohexadecane at the corresponding HLB temperature. W: isotropic liquid aqueous phase; D: isotropic bluish liquid phase (bicontinuous microemulsion); O: isotropic liquid colorless oil phase.
behavior shows distinctive features depending on the surfactant concentration. At 4 wt% surfactant (Fig. 5a), a threeliquid-phase region (W + D + O) appears, consisting of water (W), shear-birrefringent microemulsion (D), and oil (O) phases. As the surfactant mixing ratio X C12 E6 increases, this region shifts to higher temperatures within the range 40– 75 ◦ C. At 8 wt% surfactant (Fig. 5b), two microemulsion (D) phase regions are separated by a two-phase region (D + Lα ), consisting of microemulsion (D) and lamellar liquid crystalline (Lα ) phases. These phases are also shifted to higher temperatures, within the range 25–72 ◦ C, as the X C12 E6 increases. It is noteworthy that the HLB temperature values determined by conductivity, shown in Table 1, are located in the upper part of the (W + D + O) region at 4 wt% surfactant and in the upper D-phase at 8 wt% surfactant. Therefore, at 4 wt% surfactant concentration, emulsification starts in the upper part of the three-phase region (W + D + O) and by decreasing the temperature to 25 ◦ C, the system evolves to an O/W structure. Fig. 6 shows the phase equilibria and the corresponding oil (O)-to-microemulsion (D) phase ratios (O/D) at the starting emulsification temperature as a function of X C12 E6 in the range 0.3–1. The excess oil phase increases with respect to the D-phase as the X C12 E6 increases. This increase in excess oil phase (i.e., decrease in microemulsion D-phase), which is more pronounced at X C12 E6 values higher than 0.5, may influence the emulsification process. It has been shown that emulsion droplet size and polydispersity are related to the presence of a microemulsion D-phase in the system at the starting temperature of emulsification [13]. Therefore, the increase in droplet size, polydispersity, and consequently instability produced at X C12 E6 values higher than 0.4 could be due to the decrease in microemulsion D-phase. At 8 wt% surfactant concentration, emulsification starts in the upper
O/W nano-emulsions with droplet radii 60–70 nm and 25–30 nm have been obtained by the PIT method in water/C12 E4 :C12 E6 /isohexadecane systems at 20 wt% oil and total surfactant concentrations of 4 and 8 wt%, respectively. Nano-emulsion droplet size remained practically unchanged independent of the surfactant mixing ratio, X C12 E6 , and it decreased by increasing surfactant concentration from 4 to 8 wt%. At 4 wt% surfactant concentration, nanoemulsion polydispersity and instability increased with the increase in C12 E6 concentration at X C12 E6 values higher than 0.4. This increase in nano-emulsion polydispersity and instability may be related to the increase in excess oil phase with respect to the microemulsion D-phase at the starting emulsification temperature, in which three liquid phases (W + D + O) coexist. At 8 wt% surfactant concentration, the range of surfactant mixing ratios for the formation of nano-emulsions with high stability is higher. It has been attributed to the spontaneous emulsification produced in the microemulsion D-phase at the starting emulsification temperature.
Acknowledgments The authors acknowledge financial support by UNIQEMA and partial support by “Generalitat de Catalunya,” DURSI (Grant 2001SGR-00357), and the Spanish Ministry of Science and Technology, DGI (Grant PPQ2002-04514CO3-03).
References [1] H. Nakajima, S. Tomomasa, M. Okabe, in: Proceedings of First World Emulsion Conference, vol. 1, EDS, Paris, 1993, pp. 1–11. [2] C. Solans, J. Esquena, A. Forgiarini, N. Usón, D. Morales, P. Izquierdo, N. Azemar, M.J. García, in: K.L. Mittal, D.O. Shah (Eds.), Adsorption and Aggregation of Surfactants in Solution, Dekker, New York, 2002, pp. 525–554. [3] K. Shinoda, H. Saito, J. Colloid Interface Sci. 26 (1968) 70. [4] W.C. Griffin, J. Soc. Cosmet. Chem. 1 (1949) 311. [5] J.T. Davies, Interfacial Phenomena, Academic Press, New York, 1963, pp. 371–383. [6] K. Shinoda, H. Arai, J. Phys. Chem. 68 (1964) 3485. [7] K. Shinoda, J. Colloid Interface Sci. 24 (1967) 4.
394
P. Izquierdo et al. / Journal of Colloid and Interface Science 285 (2005) 388–394
[8] L. Marszall, in: M.J. Shick (Ed.), Nonionic Surfactants, Surfactant Sciences Series, vol. 23, Dekker, New York, 1987, pp. 493–547. [9] K. Shinoda, H. Kunieda, in: P. Becher (Ed.), Encyclopedia of Emulsion Technology, vol. 1, Dekker, New York, 1983, pp. 337–367. [10] H. Kunieda, Y. Sato, in: S.E. Friberg, B. Lindman (Eds.), Organized Solutions: Surfactant in Science and Technology, Surfactant Science Series, vol. 44, Dekker, New York, 1992, pp. 67–88. [11] H. Kunieda, K. Shinoda, J. Colloid Interface Sci. 107 (1) (1985) 107– 121. [12] T. Förster, W. Von Rybinski, in: B.P. Binks (Ed.), Modern Aspects of Emulsion Science, Royal Soc. Chemistry, 1998, pp. 395–426. [13] S. Friberg, C. Solans, J. Colloid Interface Sci. 66 (1978) 66. [14] H. Kunieda, S. Friberg, Bull. Chem. Soc. Jpn. 54 (4) (1981) 1010– 1014. [15] M. Bourrel, R.S. Schechter, in: Microemulsions and Related Systems, Surfactant Science Series, vol. 30, Dekker, New York, 1988. [16] A. Wadle, T. Förster, W. Von Rybinski, Colloids Surf. A Physicochem. Eng. Aspects 76 (1993) 51–57. [17] D. Morales, J.M. Gutiérrez, M.J. García-Celma, C. Solans, Langmuir 19 (2003) 7196–7200. [18] C. Solans, J.J. García-Domínguez, S.E. Friberg, Proc. Jornadas CED 12 (1981) 509–525. [19] P. Izquierdo, J. Esquena, Th.F. Tadros, C. Dederen, J. Feng, M.J. García, N. Azemar, C. Solans, Langmuir 20 (2004) 6594–6598.
[20] H. Sagitani, in: S. Friberg, B. Lindman (Eds.), Organized Solutions: Surfactant in Science and Technology, Surfactant Science Series, vol. 44, Dekker, New York, 1992, pp. 259–271. [21] A. Forgiarini, J. Esquena, C. González, C. Solans, Langmuir 17 (7) (2001) 2076–2083. [22] P. Izquierdo, J. Esquena, Th.F. Tadros, C. Dederen, M.J. García, N. Azemar, C. Solans, Langmuir 18 (2002) 26–30. [23] I.M. Lifshitz, V.V. Slezov, J. Phys. Chem. Solids 19 (1961) 35. [24] C.Z. Wagner, Elektrochem. 65 (1961) 581. [25] A.S. Kabalnov, D.E. Shchuckin, Adv. Colloid Interface Sci. 38 (1992) 69. [26] K. Shinoda, H. Saito, H. Arai, J. Colloid Interface Sci. 35 (1971) 624. [27] H. Kunieda, N. Ishikawa, J. ColIoid Interface Sci. 107 (1985) 122– 128. [28] T.J. Lin, H. Kurihara, H. Ohta, J. Soc. Cosmet. Chem. 26 (1975) 121– 139. [29] H. Sagitani, Y. Hirai, K. Nabeta, M. Nagai, J. Jpn. Oil Chem. Soc. 35 (1986) 102–107. [30] J. Weiss, J.N. Coupland, D. Brathwaite, D.J. Clements, Colloids Surf. A Physicochem. Eng. Aspects 121 (1997) 53–60. [31] A.A. Peña, C.A. Miller, J. Colloid Interface Sci. 244 (2001) 154– 163. [32] P. Walstra, in: P. Becher (Ed.), Encyclopedia of Emulsion Technology, vol. 4, Dekker, New York, 1996, pp. 1–62.