Competition of Hydrophobic Steroids with Sodium Dodecyl Sulfate

Nov 14, 2012 - Dodecyltrimethylammonium Bromide, or Dodecyl β‑D‑maltoside for ... dodecyltrimethylammonium bromide, and dodecyl maltoside...
0 downloads 0 Views 331KB Size
Article pubs.acs.org/Langmuir

Competition of Hydrophobic Steroids with Sodium Dodecyl Sulfate, Dodecyltrimethylammonium Bromide, or Dodecyl β‑D‑maltoside for the Dodecane/Water Interface Shaoxin Feng†,‡ and Paul M. Bummer*,† †

College of Pharmacy, University of Kentucky, Lexington, Kentucky 40536, United States S Supporting Information *

ABSTRACT: The surface tension lowering abilities of insoluble steroids, progesterone and testosterone, were examined at the dodecane/water interface in the presence and absence of surfactants, sodium dodecyl sulfate, dodecyltrimethylammonium bromide, and dodecyl maltoside. In the absence of these surfactants, the steroids significantly lowered the interfacial tension while exhibiting no activity at the air/water and air/dodecane surfaces. Further, in mixtures of surfactants and steroids, significant enhancement of interfacial tension lowering was observed. At a sufficiently high concentration of surfactant, no further lowering of tension was observed in the presence of the steroids. The synergistic effects on interfacial tension of steroids and surfactants were characterized by the free energy of transfer to the interface of each solute based on a twodimensional solution equation of state. Assuming no significant interaction between the steroids and the surfactants in the interface, predictions of interfacial tensions were made based on the calculated free energies of transfer and interfacial area occupied. Good agreement was found between the predicted values and experimental values for interfacial tension. The results of these studies show that progesterone and testosterone, molecules not normally thought of as surface active, exhibit significant interfacial activity and can successfully compete with surfactants for the dodecane/water interface.



INTRODUCTION Mixtures of long-chain surfactants have many important applications in a wide range of fields from industrial processes to consumer and pharmaceutical products. In many of these applications, the behavior of the mixed surfactant systems at the oil/water interface is critical for the proper function of the product. As an example in the case of pharmaceutical products, the oil/water interfacial tension will influence the particle size and physical stability of oil droplets in parenteral emulsions and microemulsions.1−3 There have been a great number of studies of mixed long-chain surfactant systems employing a variety of methods, including experimental,4,5 thermodynamic modeling,6 and computational7,8 approaches. Most surface studies with drugs have been carried out with the intent of linking interfacial activity at a phospholipid monolayer to the biological effect.9−11 In these studies, drugs tending to show little-to-no activity at the air/water interface show strong tendencies to reduce the tension at a phospholipid monolayer-covered interface.11,12 Many of the same drug molecules are known to exhibit properties in solution mimicking long-chain surfactants, such as self-association with apparent critical micelle concentrations.13,14 It has long been suggested that small solutes solubilized in micelles often colocalize with the long-chain surfactant in the interfacial region.15−17 In our own studies of pharmaceutical emulsions, it became apparent that surface activities of steroids do influence critical properties of colloidbased drug delivery systems. Taken together, these studies © 2012 American Chemical Society

suggest that small solutes such as drug molecules may be more active at the oil/water interface than expected. The goal of the present work is to determine the extent to which model hydrophobic drug substances, progesterone and testosterone, may influence the interfacial tension of the dodecane/water interface in the absence and presence of model long-chain surfactants, sodium dodecyl sulfate, dodecyltrimethylammonium bromide, and β-D-dodecyl maltoside. Further, we test the applicability of a two-dimensional (2D) solution thermodynamic model, originally designed for mixtures of classical surfactants, to calculate the free energy of transfer in mixed drug/surfactant systems. The steroids are chemically stable, poorly soluble in both water and dodecane, and are nonionic. The surfactants too are chemically stable under the conditions employed, contain identical hydrocarbon chains, are poorly soluble in dodecane, and represent anionic, cationic, and neutral head groups.



EXPERIMENTAL METHODS

Materials. Two model hydrophobic drugs, progesterone (>99%) and testosterone (>98%), and three model surfactants, sodium dodecyl sulfate (>99%) (SDS), dodecyltrimethylammonium bromide (>99%) (DTAB), and dodecyl β-D-maltoside (>98%) (DM), were obtained from Sigma-Aldrich (St. Louis, MO). Progesterone, Received: September 18, 2012 Revised: November 13, 2012 Published: November 14, 2012 16927

dx.doi.org/10.1021/la303751e | Langmuir 2012, 28, 16927−16932

Langmuir

Article

testosterone, SDS, and DTAB were purified before use. To purify surfactants and drugs, the solid state adsorption method reported by Rosen18 was employed. Briefly, solutions of solutes were passed through Sep-Pak Plus C18 columns, collected, and freeze-dried. After the purification, the surface tensions of the surfactant concentrations showed no evidence of a local minimum near the critical micelle concentration (CMC). DM was used as received. All CMC values were in good agreement with the literature. In the cases of the model drugs, HPLC chromatograms showed only minor impurity peaks. The putative impurities were found to exhibit less than 0.1% of the area of the main peaks. Surface Tension Measurements. The DuNouy ring method was used to measure the liquid/air and liquid/liquid interfacial tension. The equilibrium time was between 30 min to 1 h for each of three replicate measurements at each concentration. A water-jacked beaker was employed to control the temperature of samples at 25 ± 0.1 °C. The surfactants were dissolved in an aqueous phase to make solutions with known concentrations (e.g., 0.1, 1.0, and 10 mg/mL SDS solutions). Because of negligible solubility in hydrocarbons, the surfactants were not prepared in the oil phase. When the drugs were present, both the aqueous and organic phases were saturated separately. Prior to interfacial tension measurements, aqueous solutions were filtered using 0.2 μm hydrophilic PTFE filter (Millipore Inc.) and the hydrocarbon solutions were filtered using 0.2 μm hydrophobic PTFE filter (Pall Corp.). Since testosterone can form different crystal forms, an anhydrous form in desiccated oil and a hydrate in water, preparing dodecane and aqueous solutions separately would not maintain the same thermodynamic activity of the drug in the two media. Therefore the oil/water interfacial tension measurements would have a starting point where the oil/water partitioning of the drug was far from equilibrium, and the kinetics of partitioning could affect the interfacial tension determinations. To avoid this complexity, a direct partitioning method to produce the lower-energy form hydrate was used: the watersaturated dodecane and aqueous solutions in the presence or absence of surfactants were premixed with an excess of testosterone (anhydrous). After gentle mixing for 48 h, the two liquid layers were filtered using 0.2 μm hydrophilic (for the aqueous layer) and hydrophobic (for the oil layer) PTFE filters separately and were combined for dodecane/water interfacial tension measurements. Previous experiments have shown that complete conversion of testosterone anhydrous to hydrate was accomplished within 48 h. No evidence for emulsification was observed. Drug concentrations were determined directly by a Waters HPLC system employing a Supelcosil ABZ+ C18 column and 240 nm detection. The mobile phase was 1:1 (v/v) acetonitrile/water.

Figure 1. Chemical structures of model steroids with “A” rings indicated.

water and dodecane phases. The results are summarized in Table 1. Table 1. Interfacial Tensions in the Absence and Presence of Model Drugs interfacial tension (mN/m)* no drug progesterone testosterone

water/air

dodecane/air

dodecane/water

71.2 ± 0.6 65.2 ± 1.8 66.8 ± 0.7

24.2 ± 0.5 23.8 ± 1.1 23.6 ± 0.9

51.2 ± 1.0 28.9 ± 0.8 33.5 ± 0.5

*

The interfacial tension average values have ±95% confidence limits, where n = 3.

In the absence of model drugs, the tension of the dodecane/ water interface was 51.2 mN/m in good agreement with the literature value.21 The presence of progesterone or testosterone showed only a weak effect on the surface tension at the air/ water interface and no effect at the air/dodecane interface. These results indicate that the model drugs are not significantly surface active at the water/air and air/dodecane interfaces. In contrast, there was a dramatic change in the dodecane/water interfacial tension upon the addition of the drugs. In the presence of saturated solutions of progesterone or testosterone the dodecane/water interfacial tensions were reduced to 28.9 and 33.5 mN/m, respectively. Thus, it seems that both steroids do possess an ability to decrease the dodecane/water interfacial tension. A similar trend was observed in an octane/water system (data not shown). Interfacial Studies in the Presence of Surfactants. Listed in Table 2 are the experimental interfacial tensions of the dodecane/water system in the presence of both SDS and steroids. In the absence of a drug, addition of 0.1 mg/mL SDS to the aqueous phase resulted in an interfacial tension of 40.1 mN/m. On the other hand, in the presence of either progesterone or testosterone and 0.1 mg/mL SDS, the interfacial tensions of the dodecane/water systems were reduced significantly. The ability of the drugs to lower the interfacial tension, even in the presence of SDS, suggests that both model drugs have some ability to either compete with or to coadsorb with the surfactant under these conditions. Similarly, the presence of either steroid along with 1.0 mg/mL SDS resulted in interfacial tensions less than that of the SDS/dodecane/water system. When the concentration of SDS was further increased to 10 mg/mL, a value higher than the CMC of SDS (approximately 2.3 mg/ mL), the steroids were still able to lower the interfacial tension from 7.86 (no drug) to 6.29 (progesterone) and 7.68 mN/m (testosterone). The lack of the effect of the drugs on the air/ water surface tension (Table 1) suggests that the phenomena



RESULTS The surface activities of model steroids were measured at the air/water, air/dodecane, and dodecane/water interface, in the presence and absence of surfactants, to observe the extent to which the hydrophobic solutes can modulate interfacial tension. A thermodynamic model was then utilized to quantitatively simulate the observed phenomena at the dodecane/water interface. The octanol/water partitioning coefficients and the aqueous and dodecane solubilities19,20 for progesterone and testosterone are shown in the Supporting Information. These physical values indicate that both of the compounds are quite hydrophobic and poorly soluble in water and dodecane. The chemical structures are shown in Figure 1 and indicate that both steroids are nonelectrolytes. The three surfactants are quite soluble in water but essentially insoluble in dodecane. Interfacial Studies in the Absence of Surfactants. The surface tensions of the air/water, air/dodecane, and dodecane/ water interfaces were measured in the presence and absence of the steroids. Saturated solutions of the drugs were employed for 16928

dx.doi.org/10.1021/la303751e | Langmuir 2012, 28, 16927−16932

Langmuir

Article

Table 2. Dodecane/Water Interfacial Tensions in the Absence and Presence of Saturated Solutions of Model Steroids (Progesterone and Testosterone) and Sodium Dodecyl Sulfate (SDS), Dodecyltrimethylammonium Bromide (DTAB), and Dodecyl-β-Maltoside (DM) Solutions progesterone no drug no surfactant 0.1 mg/mL SDS-dodecane 1.0 mg/mL SDS-dodecane 10 mg/mL SDS-dodecane 0.1 mg/mL DTAB-dodecane 1.0 mg/mL DTAB-dodecane 10 mg/mL DTAB-dodecane 0.003 mg/mL DM-dodecane 0.03 mg/mL DM-dodecane 0.30 mg/mL DM-dodecane

51.2 40.1 19.4 7.86 41.1 25.8 8.11 31.2 12.7 4.92

± ± ± ± ± ± ± ± ± ±

0.7 1.3 0.6 0.03 1.7 0.5 0.02 0.4 0.2 0.06

experimental 28.9 24.2 12.9 6.29 21.1 15.1 6.59 24.2 11.3 4.22

± ± ± ± ± ± ± ± ± ±

a

0.8 0.2 0.3 0.03 0.5 0.4 0.17 0.6 0.3 0.16

testosterone predicted − 26.5 16.8 7.13 26.8 21.2 6.61 24.3 11.9 4.75

b

experimentala

predictedb

± ± ± ± ± ± ± ± ± ±

− 32.4 18.5 7.51 32.9 24.1 6.93 27.8 12.3 4.87

33.5 30.4 16.3 7.68 22.7 19.0 7.35 26.9 11.8 4.59

0.5 0.1 0.1 0.13 0.2 0.9 0.07 1.0 0.4 0.11

Interfacial tension (mN/m); average value ± 95% confidence limits; n = 3. bPredicted interfacial tension (mN/m) from eq 4 (SDS, DTAB) or eq 5 (DM).

a

competition between the two classes of molecules for the dodecane/water interface. In this section, we subject interfacial tension data to a thermodynamic analysis to probe the free energy of transfer of the solutes to the dodecane/water interface. We then attempt to predict dodecane/water interfacial tension of surfactant/steroid mixtures based on the free energy of transfer to the interface and partial molar area occupied of component members. There are two main theoretical approaches to define the equilibrium relationship between adsorption and surface tension:22 a 2D gas model proposed by Langmuir23 and a 2D solution model proposed by Butler.24 The 2D gas approach works best for insoluble surfactants, while the 2D solution approach has advantages for soluble surface-active compounds.22 Lucassen-Reynders5,22 has outlined the derivation of the Butler approach for relating bulk concentration, interfacial tension, and free energy of transfer to the interface. In our studies, the 2D solution approach was chosen because the surfactants employed are quite soluble in the aqueous phase. Application of Butler Model to Single Solute Systems. The tension of the dodecane/water interface was measured as a function of the concentration of a single solute. The experimental γdodecane/water − concentration curves for DM, progesterone, and testosterone are shown in Figures 2 and 3. Following the development of the Butler equation by Lucassen-Reynders5,22 and Fainerman and Lucassen-Reynders,4 the free energies of transfer to interface (ΔGtransfer; i) were calculated from the experimental interfacial tension data according to eq 1;

observed at the dodecane/water interface are not due to highly surface-active impurities in the systems. The interfacial tensions of the dodecane/water system in the presence of both DTAB and steroids are listed in Table 2. In the absence of steroids, addition of 0.1 mg/mL DTAB to the aqueous phase resulted in a dodecane/water interfacial tension of 41.1 mN/m. In the presence of either steroid along with 0.1 mg/mL DTAB, the interfacial tensions were reduced by approximately 20 mN/m. Similarly, the presence of the steroid and 1.0 mg/mL DTAB resulted in interfacial tensions of 7 mN/ m less than that of the DTAB/dodecane/water system. When the concentration of DTAB was further increased to 10 mg/mL in the aqueous phase, a value higher than the CMC of DTAB (approximately 4.9 mg/mL), the ability of model drugs to compete for the interface was further weakened: the interfacial tensions of the steroid-DTAB/dodecane/water system dropped to 6.59 (progesterone) and 7.35 mN/m (testosterone) from 8.11 mN/m in the absence of a drug. The experimental interfacial tensions of the dodecane/water system in the presence of both DM and steroids are also listed in Table 2. In the absence of steroids, addition of 0.003 mg/mL DM to the aqueous phase of the dodecane/water system resulted in an interfacial tension of 31.2 mN/m. On the other hand, in the presence of steroid along with 0.003 mg/mL DM reduced the interfacial tensions to 24.2 and 26.9 mN/m for progesterone and testosterone, respectively. At the 0.03 mg/mL concentration of DM in the aqueous phase, the interfacial tensions of the progesterone- (11.3 mN/m) and testosterone(11.8 mN/m) containing systems were less than that in the absence of the steroids (12.7 mN/m). When the concentration of DM was increased to 0.3 mg/mL in the aqueous phase, a value higher than the CMC of DM (approximately 0.08 mg/ mL), the ability of the steroids to further lower the interfacial tension was abolished. Overall, the results in this section confirm that the steroids tested are able to lower the interfacial tension of the water/ dodecane interface, even in the presence of moderate concentrations of surfactants. In a qualitative sense, reduction in interfacial tension was not dependent upon the electrical charge on the surfactant. Thermodynamic Model of Dodecane/Water Interfacial Tension in the Presence of Drug and Surfactant. The surfactant concentration-dependent manner by which the tension is modulated in the presence of steroids suggests a

⎛ ΔGtransfer, i ⎞ ⎛ πa ⎞ ⎛ πai ⎞ ⎟ = 1 exp⎜ − water ⎟ + xil exp⎜ − ⎟ exp⎜ − ⎝ ⎠ ⎝ RT ⎠ RT RT ⎠ ⎝ (1)

In this expression, xli is the aqueous bulk mole fraction of DM or steroid, and ai is the occupied interfacial area of the solute. The value employed for the occupied interfacial area of water (awater) was 7.62 Å2/molecule. This value was taken from Gumkowski25 and was based on average hard sphere diameter of water ranging from 2.50−2.93 Å.26 Surface pressure (π) is defined as the difference between the interfacial tensions in the absence and presence of the surfactant or solutes. The lines in Figures 2 and 3 represent a good fit of eq 1 to the data. The 16929

dx.doi.org/10.1021/la303751e | Langmuir 2012, 28, 16927−16932

Langmuir

Article

⎛ ΔGtransfer,ionicsurfact ⎞ ⎛ πa ⎞ l l f surfact exp⎜ − water ⎟ + 2xsurfact exp⎜ − ⎟ ⎝ RT ⎠ RT ⎠ ⎝ ⎛ πasurfact ⎞ ⎟ = 1 exp⎜ − ⎝ 2RT ⎠ (2)

Here, xlsurfact is the aqueous bulk mole fraction of the ionic surfactant, and flsurfact is the activity coefficient of the ionic surfactant in the bulk. The activity coefficient, flsurfact, for ionized surfactants (SDS and DTAB) in eq 2 was calculated in terms of the ionic strength, I, using the modified Debye−Huckel equation:27 l −log f surfact =

⎛ ΔGtransfer,surfact ⎞ ⎛ πa ⎞ l l exp⎜ − water ⎟ + 2xsurfact exp⎜ − f surfact ⎟ ⎝ RT ⎠ RT ⎝ ⎠

Figure 3. Dodecane/water interfacial tension as a function of the progesterone (▲) and testosterone (●) concentration in water. The points represent experimental data, and the solid lines are the fitted curves according to eq 1. Error bars represent standard deviation, n = 3.

⎛ ΔGtransfer,drug ⎞ ⎛ πa ⎞ l exp⎜ − surfact ⎟ + xdrug exp⎜ − ⎟ ⎝ 2RT ⎠ RT ⎝ ⎠ ⎛ πadrug ⎞ ⎟ = 1 exp⎜ − ⎝ RT ⎠

fitted values for the free energy of transfer of solutes to the dodecane/water interface were found to be −42.1 kJ/mol, −42.1 kJ/mol, and −36.2 kJ/mol for DM, progesterone, and testosterone, respectively (Table 3). To determine the free energies of transfer to the interface (ΔGtransfer;ionic surfact) for ionic surfactants SDS and DTAB, the experimental dodecane/water interfacial tension data were fitted by eq 2.

SDS DTAB DM progesterone testosterone

32.0 37.1 41.4 61.3 59.6

± ± ± ± ±

⎛ ΔGtransferDM ⎞ ⎛ πa ⎞ l ⎟ exp⎜ − water ⎟ + xsurfact exp⎜ − ⎝ ⎠ ⎝ ⎠ RT RT ⎛ ΔGtransferdrug ⎞ ⎛ πa ⎞ l exp⎜ − DM ⎟ + xdrug exp⎜ − ⎟ ⎝ RT ⎠ RT ⎝ ⎠ ⎛ πadrug ⎞ ⎟ = 1 exp⎜ − ⎝ RT ⎠

−23.5 −22.8 −42.1 −42.1 −36.2

± ± ± ± ±

b

0.8 1.2b 0.6c 0.5c 0.4c



DISCUSSION Adsorption of drugs to an oil/water interface has been shown to influence the physical stability of lipid emulsions30 and the

Values from ref 35. Values from fitting eq 2. Values from fitting Eq. 1. a

b

(5)

The comparisons between calculated and experimental surface tensions are shown in Table 2. Overall, the predictions are good with the exception of progesterone and testosterone in DTAB at 0.1 and 1 mg/mL, respectively. Possible reasons for the deviations will be discussed below.

ΔGtrans(kJ/mol)

0.6 0.6a 2.7c 4.8c 5.1c

(4)

For systems containing nonionic surfactant DM and steroids, eq 5 was applied.

Table 3. Area Per Molecule at the Interface and Free Energy of Transfer of Surfactants and Solutes a

(3)

At concentrations of surfactant greater than the CMC, the CMC was employed in the fitting of eq 2. The CMC values of SDS28 and DTAB29 in water saturated with dodecane (6.9 mM for SDS and 12.5 mM for DTAB) were employed. From eq 2, the fitted ΔGtransfer;ionic surfact values were −23.5 ± 0.8 kJ/mol for SDS and −22.8 ± 1.2 kJ/mol for DTAB (Table 3). Within error, these surfactants have the same free energy of transfer to the dodecane/water interface. Further, the free energy of transfer of the ionic surfactants was about half the value found for DM and for the steroids. Application of Butler Model to Surfactant/Drug Mixtures. The goal of this section is to predict the interfacial tensions of systems containing both surfactant and model drugs by employing the free energy of transfer to the interface and area occupied by each solute at the interface. The full derivation of this model by Butler24 and Lucassen-Reynders5,22 is supplied in the Supporting Information. For systems containing ionic surfactants SDS and DTAB and model drugs, eq 4 was applied.

Figure 2. Dodecane/water interfacial tension as a function of dodecyl β-D-maltoside concentration in water. The points represent experimental data, and the solid line is the fitted curve according to eq 1.

a (Å2)

0.509 I I +1

c

16930

dx.doi.org/10.1021/la303751e | Langmuir 2012, 28, 16927−16932

Langmuir

Article

distribution and kinetics of the degradation of drugs.31 In the present work, we have hypothesized that progesterone and testosterone, which show no activity at the air/water interface, exert a significant effect on the interfacial tension of the oil/ water interface. To test this hypothesis, we carried out experiments at the dodecane/water interface in the absence and presence of model surfactants. We first turn our attention to the surface activity of the model steroids in the absence of added surfactants. From the free energy of transfer to the interface, it is possible to define Kinterface/water, the interfacial partition coefficient. K int erf ace/water = exp( −ΔGtransfer /RT )

Information. In the absence of surfactants, the experimentally observed areas occupied were 59.8 and 58.0 Å2 for testosterone and progesterone, respectively, indicating that the orientation of the steroids are neither lying flat (maximal cross section) nor are on-end (minimal cross section) at the interface. Whether or not the interfacial orientations of model steroids present in the absence of added surfactant are maintained in the presence of SDS, DTAB, or DM is not known. Interestingly, there appears to be no evidence of strong repulsive interactions between the anionic SDS and the pi electrons of the model steroids. The model steroids are nonionic, but other drug molecules, such as those containing amine functionalities, might be expected to interact with anionic surfactants.

(6)



In the absence of other solutes, progesterone and DM have the greatest ability to partition to the dodecane/water interface, followed by testosterone, while SDS and DTAB have the lowest partition coefficient. The results are quite surprising because they show that these steroids are very active at the dodecane/ water interface and are even more surface active than conventional surfactants. The ability of the model steroids to lower interfacial tension is probably limited by low aqueous solubility, while the surfactants SDS and DTAB, have much higher solubilities and CMCs. From the negative sign of the transfer free energies of the model steroids, the dodecane/water interface is always a more energetically favorable location than the bulk aqueous phase. Negative free energies of transfer were also observed in octane/water and for a variety of other steroid drugs (data not shown). The results in Table 2 indicate that the interfacial tensions of steroid mixtures with SDS and DM can be successfully fit by a 2D-solution free energy of transfer model. On the other hand, the 2D-solution free energy of transfer model was less successful in fitting the interfacial tensions of steroid mixtures with DTAB. An implicit assumption of the approach is that the free energy of transfer of the individual components is independent of the composition of the mixed interfacial film. Equations 4 and 5 do not take into account any possible surfactant/steroid interactions. Several literature reports suggest that binary mixtures of cationic surfactants with other amphiphiles tend to exhibit nonzero interaction parameters.32,33 Nonzero interactions have also been assumed in computational models in binary systems.34 It might be expected that the quaternary ammonium headgroup of DTAB would interact by an ion-induced dipole mechanism with the pielectrons of the A-ring of the steroids. The molecular details of an attractive interaction between DTAB and the model steroids that may influence the interfacial tension are not clear. Lower interfacial tensions would be brought about by greater molecular packing in the interfacial region of the oil/water system. Molecular attractions would be expected to result in a greater packing density, possibly by modifying the orientation of solutes at the interface. As an example, it might be possible that the steroids could self-associate in the interface. Such aggregates might result in a more dense packing of molecules at the interface and thereby exert an added effect on the interfacial tension. Molecular simulation studies were carried out to calculate the maximum and minimum possible cross-sectional areas occupied per molecule at the interface. For testosterone, the maximal and minimal cross-sectional areas were 95.8 and 39.5 Å2, respectively. For progesterone, the maximal and minimal areas were 104.4 and 39.6 Å2, respectively. The method employed in these calculations is outlined in the Supporting

CONCLUSION Although exhibiting little or no surface activity at the air/ dodecane or air/water interfaces, the presence of model steroids, testosterone and progesterone, significantly decreases the dodecane/water interfacial tension. Further, the tested steroids appear able to compete with conventional surfactants SDS, DTAB, and DM for the dodecane/water interface. Using a 2D solution model for the free energy of transfer to the interface, the interfacial tension of mixed steroid/surfactant systems could be predicted.



ASSOCIATED CONTENT

S Supporting Information *

Physical properties of the steroids, the derivation of eqs 2−5, and the calculation of the maximum and minimum interfacial surface areas occupied by the steroids are listed in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (859) 218-6522. Present Address ‡

Allegran Comany, Irvine, CA.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors wish to thank B. D. Anderson (University of Kentucky), Dr. Michael Gumkowski, and Dr. Pasupati Mukerjee (University of Wisconsin−Madison) for valuable discussions and suggestions and Boehringer-Ingelheim, USA, for financial support.



ABBREVIATIONS CMC, critical micelle concentration; SDS, sodium dodecyl sulfate; DTAB, dodecyltrimethyl-ammonium bromide; DM, dodecyl-β-maltoside



REFERENCES

(1) Malcolmson, C.; Lawrence, M. J. A comparison of the incorporation of model steroids into nonionic micellar and microemulsion systems. J. Pharm. Pharmacol. 1993, 45, 141−143. (2) Stalidis, G.; Svranas, A.; Jannakoudakis, D. Interfacial properties and stability of oil-in-water emulsion stabilized with binary mixtures of surfactants. J. Colloid Interface Sci. 1990, 135, 313−324. (3) Craig, D. Q. M.; Lievens, H. S. R.; Pitt, K. G.; Storey, D. E. An investigation into the physicochemical properties of self-emulsifying systems using low frequencies dielectric spectroscopy, surface tension

16931

dx.doi.org/10.1021/la303751e | Langmuir 2012, 28, 16927−16932

Langmuir

Article

(27) Robinson, R. A.; Stokes, R. H. Electrolyte Solutions; Butterworth: London; 1959. (28) Bonfillon, A.; Sicoli, F.; Langevin, D. Dynamic surface tension of ionic surfactant solutions. J. Colloid Interface Sci. 1994, 168, 497−504. (29) Medrzycka, K.; Zwierzykowski, W. Adsorption of alkyl trimethylammonium bromides at the various interfaces. J. Colloid Interface Sci. 2000, 230, 67−72. (30) Washington, C. Stability of lipid emulsions for drug delivery. Adv. Drug Delivery Rev. 1996, 20, 131−145. (31) Teagarden, D. K.; Anderson, B. D.; Petre, W. J. Distribution and chemical kinetics of PGE1 in lipid emulsions. Adv. Drug Delivery Rev. 1996, 20, 155−162. (32) Zhang, L.; Luo, L.; Zhao, S.; Yu, J. Studies of synergism/ antagonism for lowering dynamic interfacial tensions in surfactant/ alkali/acidic oil systems, Part 2: Synergism/antagonism in binary surfactant mixtures. J. Colloid Interface Sci. 2002, 251, 166−171. (33) Din, K.; Khan, A. B.; Naqvi, A. Z. Mixed micellization of antidepressant drug amitriptyline hydrochloride with cationic surfactants. Colloids Surf., B 2010, 80, 206−212. (34) Dominguez, H. Computer simulation studies of surfactant monolayer mixtures at the water/oil interface: Charge distribution effects. J. Colloid Interface Sci. 2004, 274, 665−672. (35) Haydon, D. A.; Taylor, F. H. Adsorption at the oil−water interface and the calculation of electrical potentials in the aqueous surface phase. I. Neutral molecules and a simplified treatment for ions. Philos. Trans. R. Soc. London, Ser. A 1960, 252, 225−248.

measurements and particle size analysis. Int. J. Pharm. 1993, 96, 147− 155. (4) Fainerman, V. B.; Lucassen-Reynders, E. H. Adsorption of single and mixed ionic surfactants at fluid interfaces. Adv. Colloid Interface Sci. 2002, 96, 295−323. (5) Lucassen-Reynders, E. H. Surface Interactions in Mixed Surfactant Systems. J. Colloid Interface Sci. 1982, 85, 178−186. (6) Goldsipe, A.; Blankschtein, D. Molecular-thermodynamic theory of micellization of multicomponent surfactant mixtures: 1. Conventional (pH-insensitive) surfactants. Langmuir 2007, 23, 5942−52. (7) Ginzburg, V. V.; Chang, K.; Jog, P. K.; Argenton, A. B.; Rakesh, L. Modeling the interfacial tension in oil-water-nonionic surfactant mixtures using dissipative particle dynamics and self-consistent field theory. J. Phys. Chem. B 2011, 115, 4654−4661. (8) Dominquez, H. Computer simulations of surfactant mixtures at the liquid/liquid interface. J. Phys. Chem. B 2002, 106, 5915−24. (9) Felmeister, A. Relationships between surface activity and biological activity of drugs. J. Pharm. Sci. 1972, 61, 151−64. (10) Schreier, A.; Malheiros, S. V. P.; dePaula, E. Surface active drugs: Self-association and interaction with membranes and surfactants. Physicochemical and biological aspects. Biochim. Biophys. Acta 2000, 1508, 210−234. (11) Lucio, M.; Lima, J. L. F. C.; Reis, S. Drug-membrane interactions: Significance for medicinal chemistry. Curr. Med. Chem. 2010, 17, 1795−1809. (12) Zografi, G.; Auslander, D. E.; Lytell, P. L. Interfacial properties of phenothiazine derivatives. J. Pharm. Sci. 1964, 53, 573−4. (13) Taboada, P.; Attwood, D.; Ruso, J. M.; Garcia, M.; Sarmiento, F.; Mosquera, V. Influence of molecular structure on the ideality of mixing in micelles formed in binary mixtures of surface-active drugs. J. Colloid Interface Sci. 1999, 216, 270−275. (14) Attwood, A. T. and Florence, A. Surfactant Systems: Their chemistry, pharmacy and biology. Chapman Hall: London, 1983. (15) Mukerjee, P.; Ramachandran, C.; Pyter, R. A. Solvent effects on the visible spectra of nitroxides and relation to nitrogen hyperfine splitting constants. Nonemperical polarity scales for aprotic and hydroxylic solvents. J. Phys. Chem. 1982, 86, 3189−3197. (16) Croy, S. R.; Kwon, G. S. Polysorbate 80 and Cremophor EL micelles deaggregate and solubilize nystatin at the core-corona interface. J. Pharm. Sci. 2005, 94, 2345−2354. (17) Mukerjee, P.; Ko, J. S. Solubilization of ethyl o-, m-, and paminobenzoates in micelles of different charge types: Interfacial adsorption and orientation effects. J. Phys. Chem. 1992, 96, 6090− 6094. (18) Rosen, M. J. Purification of surfactants for studies of their fundamental surface properties. J. Colloid Interface Sci. 1981, 79, 587− 8. (19) Kamlet, M. J.; Doherty, R. M.; Carr, P. W.; Mackay, D.; Abraham, M. H.; Taft, R. W. Linear solvation energy relationships. 44. Parameter estimation rules that allow accurate prediction of octanol/ water partition coefficients and other solubility and toxicity properties of polychlorinated biphenyls and polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 1988, 22, 503−9. (20) Hansch, C.; Leo, A.; Hoekman, D. Exploring QSAR; American Chemical Society: Washington, DC, 1995; Vol.2. (21) Zeppieri, S.; Rodriguez, J.; Lopez de Ramos, A. L. Interfacial tension of alkanyl + water systems. J. Chem. Eng. Data 2001, 46, 1086−1088. (22) Lucassen-Reynders, E. H. Anionic Surfactants; LucassenReynders, E. H., Ed.; Marcel Dekker: New York, 1981; Chapter 1. (23) Langmuir, I. Constitution and fundamental properties of solids and liquids. II. Liquids. J. Am. Chem. Soc. 1917, 39, 1848−1906. (24) Butler, J. A. V. Thermodynamics of the surfaces of solutions. Proc. R. Soc. London, Ser. A 1932, 135, 348−375. (25) Gumkowski, M. J., Surfactant Micelles: Solubilization and Related Phenomena. Ph.D. Dissertation, University of Wisconsin− Madison, 1986. (26) Pierotti, R. A. Aqueous solutions of nonpolar gases. J. Phys. Chem. 1965, 69, 281−288. 16932

dx.doi.org/10.1021/la303751e | Langmuir 2012, 28, 16927−16932