Unified Concept of Solubilization in Water by Hydrotropes and

Jun 15, 2005 - In the present work hydrophobic dyes, i.e. disperse red 13 (DR-13; (2-[4-(2-chloro-4-nitrophenylazo)-. N-ethylphenylamino]ethanol) and ...
0 downloads 0 Views 215KB Size
Download PDF
Langmuir 2005, 21, 6769-6775

6769

Unified Concept of Solubilization in Water by Hydrotropes and Cosolvents P. Bauduin, A. Renoncourt, A. Kopf, D. Touraud, and W. Kunz* Institute of Physical and Theoretical Chemistry, University of Regensburg, D-93040 Regensburg, Germany Received March 1, 2005. In Final Form: May 8, 2005 In the present work hydrophobic dyes, i.e. disperse red 13 (DR-13; (2-[4-(2-chloro-4-nitrophenylazo)N-ethylphenylamino]ethanol) and Jaune au gras W1201 (1H-indene-1,3(2H)-dione,2-(2-quinolinyl)), are solubilized in water with the help of different additives: acetone and 1-propanol as typical cosolvents, sodium xylene sulfonate (SXS) as a representative of a classical hydrotrope, sodium dodecyl sulfate (SDS) as a typical surfactant, and finally some “solvosurfactants” [ propylene glycol monoalkyl ether derivatives (CiPOj: i ) 1, j ) 1 and 3; i ) 3, j ) 1 and 2; i ) 4 and tertio-butyl, j ) 1) and 1-propoxy-2-ethanol (C3EO1)]. These solvosurfactants are short amphiphiles that do not form well-defined structures in water such as micelles. For all additives an exponential increase in the solubilizations of the two studied hydrophobic dyes was observed when their concentrations in water were increased. Except for the SDS solution, no difference in the overall shapes of the solubilization curves (dye solubility against additive concentration) was found. All the studied molecules were classified according to their hydrotropic efficiencies, i.e., their abilities to solubilize a hydrophobic, sparingly soluble compound in water. The volume of the hydrophobic parts of the studied additives, roughly evaluated by simple calculations, was found to influence strongly the hydrotropic efficiency; i.e. the larger the hydrophobic part of the additive, the better the hydrotropic efficiency. By contrast, the hydrophilic part carrying a charge or not is of minor importance. Taking the hydrophobic part of the molecules as the key parameter, the water solubilization efficiency of cosolvents, hydrotropes, and solvosurfactants can be described in a coherent way.

1. Introduction At the beginning of the past century, the term hydrotropy was defined as the large increase in the water solubility of a variety of hydrophobic compounds brought about by the addition of certain water-soluble organic compounds, named hydrotropes.1 Hydrotropes comprise hydrophilic and hydrophobic moieties, with the hydrophobic moiety being typically too small to induce micelle formation.2 Hydrotropes are usually anionic compounds and are composed of an aromatic ring substituted by a sulfate, sulfonate, or carboxylate group, typical examples of hydrotropes being sodium xylene sulfonate (SXS) or sodium benzoate. This definition was later extended to cationic and neutral aromatic compounds.3 Hydrotrope molecules are assumed to aggregate by a stacking mechanism of the planar aromatic ring present in their chemical structures. This type of aggregation is believed to be at the origin of the solubilization process of sparingly soluble hydrophobic compounds in water, in analogy to a micellization process. Nevertheless some aliphatic compounds such as short sodium alkanoates4,5 or alkyl sulfates6 show also a hydrotropic behavior as it was observed for aromatic derivatives. For such aliphatic compounds the assumed stacking mechanism does not make sense. It was observed that the hydrotrope concentration, at which the increase in the hydrophobic compound solubility becomes significant, sometimes called minimum hydro* To whom correspondence should be addressed. E-mail: [email protected]. (1) Neuberg, C. Biochem. Z. 1916, 76, 107. (2) Matero, A. In Handbook of Applied Surface and Colloid Chemistry, Vol. 1; Holmberg, K., Ed.; Wiley-VCH: New York, 2002; p 407. (3) Saleh, A. M.; El-Khordagui, L. K. Int. J. Pharm. 1985, 24, 231. (4) Danielsson, I.; Stenius, P. J. Colloid Interface Sci. 1971, 37, 264. (5) Lumb, E. C. Trans. Faraday Soc. 1951, 47, 1049. (6) Firman, P.; Haase, D.; Jen, J.; Kahlweit, M.; Strey, R. Langmuir 1985, 1, 718.

tropic concentration (MHC), coincides with a change of the slope of the surface tension as a function of hydrotrope concentration.7 And as pointed out by da Silva et al.,8 cooperative aggregation such as micellization is accompanied by such a phenomenon. However, in hydrotrope-water mixtures such a micellization process does not occur. Further, typical critical micelle concentrations (CMC) are of the order of 10-2-10-3 M, and so far below the MHC of usual hydrotropes (≈1 M). Independently it was shown that for short amphiphiles, which present a high apparent CMC according to the plot of surface tension against molar concentration, the chemical activity has to be used instead of molar concentration.9 If this is done, the break in the surface tension curve, which is supposed to indicate the sudden onset of aggregation, can completely disappear.9 Usually this point is not considered in the studies concerning hydrotrope aggregation,7 but this must be taken into account to infer correct MHC values. Note that the MHC is not necessarily linked to a change in the slope of the surface tension. As Horvath-Szabo et al.10 showed recently, in the plot of surface tension as a function of the logarithm of hydrotrope activity no break occurred at the activity at which the sudden increase in solubilization was found. In their study a pronounced association between lecithin, which was to be solubilized, and SXS, as the hydrotrope, was proposed to explain the sudden increase of lecithin solubility in water. No cooperative aggregation of the SXS molecules was considered. It was also observed that the increase in solubilization induced by hydrotropes after a certain threshold concen(7) Balasubramanian, D.; Srinivas, V.; Gaikar V. G.; Sharma, M. M. J. Phys. Chem. 1989, 93, 3865. (8) da Silva, R. C.; Spitzer, M.; da Silva, L. H. M.; Loh, W. Thermochim. Acta 1999, 328, 161. (9) Strey, R.; Viisanen, Y.; Aratono, M.; Kratohvil, J. P.; Yin, Q.; Friberg, S. E. J. Phys. Chem. 1999, 103, 9112. (10) Horvath-Szabo, G.; Yin, Q.; Friberg, S. E. J. Colloid Interface Sci. 2001, 236, 52.

10.1021/la050554l CCC: $30.25 © 2005 American Chemical Society Published on Web 06/15/2005

6770

Langmuir, Vol. 21, No. 15, 2005

tration (MHC) is sometimes followed by a level-off of the solubilities at higher hydrotrope concentrations.7,11-13 In such cases the solubilization curve as a function of hydrotrope concentration, plotted in a linear scale, has a sigmoidal shape. Balasubramanian et al.7 attributed this sigmoidal curve to a cooperative process such as micellization. Aggregation would probably be the better word. Moreover, the fact that the MHC is rather insensitive to the nature of the hydrophobic compounds to be dissolved7 supports the hypothesis of hydrotrope cooperative selfaggregation. Thus the mechanism of self-association of hydrotrope molecules is still under debate with both noncooperative stepwise self-aggregation, i.e. in a pairwise manner by formation of dimers, trimers, and so on,14-17 and cooperative self-aggregation7,8 being suggested. Balasubramanian et al.7 differentiated clearly hydrotropy, showing typical sigmoidal solubilization profiles, from solubilization observed by cosolvency and by saltingin effects showing a monotonic increase in the solubilizing ability with increasing cosolvent concentration. More generally, the solubilization of a hydrophobic compound in water by a cosolvent is known to increase slightly and monotonically at low and moderate cosolvent concentrations and to increase exponentially at very high concentrations.19 In the present work we try a general classification of cosolvents and hydrotropes and related compounds (the solvosurfactants). To this purpose, we study spectrophotometrically the solubilization of hydrophobic dyes, i.e. disperse red 13 (DR-13) and Jaune au gras W1201, in water with the help of a typical hydrotrope, the SXS, by classical water cosolvents (acetone and 1-propanol) and using some short amphiphiles derived from propylene glycol (and ethylene glycol) monoalkyl ether CiPOj (CiEOj) with i < 5. These short amphiphiles do not form well-defined structures in water such as micelles and are usually considered as water cosolvents. Since they have also some features in common with surfactants, they are sometimes called “solvosurfactants”. The solubilization of the DR-13 by a classical anionic surfactant, sodium dodecyl sulfate (SDS), was also studied in order to ensure that the DR-13 behaves like a classical solubilizate in the presence of surfactant; i.e. the solubilization appears when micelles are formed at concentrations above the surfactant CMC. From these results all the studied compounds, whatever hydrotropes or cosolvents, were classified according to their “hydrotropic” efficiencies, which can be defined as their abilities to solubilize hydrophobic compounds at more or less low additive concentrations. Moreover, the volume of the hydrophobic parts of the studied solubilizing molecules, supposed to be the main factor influencing the hydrotropic efficiency, was roughly evaluated by simple calculations. According to these considerations the term of cosolvency and hydrotropy were discussed in a consistent way considering the usually admitted hydrotropy2,7,17 and cosolvency7,19 definitions that were mentioned above. (11) Booth, H. S.; Everson, H. E. J. Ind. Eng. Chem. 1949, 40, 1491. (12) Booth, H. S.; Everson, H. E. J. Ind. Eng. Chem. 1949, 41, 2627. (13) Booth, H. S.; Everson, H. E. J. Ind. Eng. Chem. 1949, 42, 1536. (14) Friberg, S. E. Curr. Opin. Colloid Interface Sci. 1997, 2, 490. (15) Schobert, B. Naturwissenschaften 1977, 64, 386. (16) Schobert B.; Tschesche, H. Biochem. Biophys. Acta 1978, 541, 270. (17) Muckerjee, P. J. Pharm. Sci. 1974, 63, 972. (18) Friberg, S. E.; Brancewicz, C. In Science Surfactant Series, Vol. 67; Lai, K. Y., Ed.; Dekker: New York, 1997; p 21. (19) Yalkowsky, S. H. In Solubility and Solubilization in Aqueous Media; Yalkowsky, S. H., Ed.; Oxford University Press: New York, 1999.

Bauduin et al.

2. Experimental Section 2.1. Materials. Sodium dodecyl sulfate (SDS, 99%), 1-propanol (99%), and acetone (p.a.) were supplied by Merck; sodium xylene sulfonate (SXS, 40 wt % in water), 1-propoxy-2-ethanol (C3EO1, 99.4%), disperse red 13 (2-[4-(2-chloro-4-nitrophenylazo)-Nethylphenylamino]ethanol (DR-13, 95%) by Aldrich; and the yellow hydrophobic dye Jaune au gras W1201 1H-indene-1,3(2H)-dione,2-(2-quinolinyl) from Lcw (France). Except for the 1-tert-butoxy-2-propanol (t-C4PO1, >99%) which was supplied by Lyondell all the propylene glycol monoalkyl ether derivatives were supplied by Dow Chemical with the following purity grades: 1-methoxy-2-propanol (C1PO1), >99.5%; tripropyleneglycol monomethyl ether (C1PO3), >97.5%; 1-propoxy-2propanol (C3PO1), 99%; dipropyleneglycol monopropyl ether (C3PO2), >98.5%; and 1-butoxy-2-propanol (C4PO1), >99%. 2.2. Solubilization Experiments. All experiments concerning the solubilization process and the optical density measurements were done in a thermostated room at 25 ( 0.2 °C. All solutions to be measured, i.e. containing water and the studied compounds at an appropriate concentration, were saturated with a sufficient amount of DR-13 or Jaune au gras W1201 and left equilibrated 24 h. The solutions were then filtered in order to separate the nonsolubilized excess of dye from the solutions. The optical density (OD) of the filtered solutions were measured in 1 cm path length quartz cells with a UV-visible spectrophotometer Cary-3E at a wavelength of 525 nm or of 430 nm corresponding to the wavelength λmax of, respectively, the DR-13 and the Jaune au gras W1201 where the dyes have their absorption maxima. Before each measure a zero absorbance was done with the corresponding solution without dye. For the studied hydrotropes and cosolvents λmax values were found to be not so sensitive to the chemical nature of these compounds ((10 nm). When the measured OD was above a critical value, suitable dilutions were done by using the same solution but without dye. The resulting ODs reflect then, through the Beer-Lambert law, the concentration of the hydrophobic dye solubilized in the corresponding aqueous hydrotrope solutions or water cosolvent mixtures.

3. Modeling For SXS, acetone, 1-propanol, C3EO1, and C3PO1 isoelectrondensity molecular surfaces were computed and color-coded from electrostatic potentials. They were created with the software Molden based on ab initio calculations with Gaussian 03 (RHF/6-311++G** or B3LYP/6311++G**) using its cubegen utility. Red represents the most negative values, i.e. high electron density, increasing over green to blue, which indicates the most positive values, i.e. lower electron density. 4. Results and Discussion 4.1. Solubilization Curves: Comparison between Hydrotrope, Cosolvent, and Surfactant. The measured ODs obtained for the DR-13 as a function of molar concentration of surfactant, hydrotrope, cosolvent, or solvosurfactant in water are plotted in Figure 1 for SDS, SXS, 1-propanol, C3PO1, C3EO1, and acetone. In Figure 1a the OD is plotted in a logarithmic scale in order to compare all solubilization curves over a large range of concentrations. In the case of 1-propanol, C3PO1, and C3EO1 the OD could be measured over the whole range of concentrations from pure water to pure solvents. The OD was measured in more restricted ranges of concentrations: for SXS (solid at room temperature) solutions up to 2.248 mol/L (40 wt %) and for acetone up to 10.91 mol/L (80 wt %). In the case of acetone the DR-13 solubility becomes so high for acetone concentrations above 10.91 mol/L that quantitative results could not be obtained for these compositions. This is a consequence of the extremely high solubility of the DR-13 in acetone. In the case of the

Solubilization in Water by Hydrophobic Dyes

Langmuir, Vol. 21, No. 15, 2005 6771

Figure 2. Optical density (OD), proportional to the amount of dissolved DR-13 dye, versus the molar concentrations of SDS (4) in water. This curve is compared to the ones obtained with SXS (3) and C3EO1 (O).

Figure 1. Optical density (OD), proportional to the amount of dissolved DR-13 dye, versus the molar concentrations of hydrotropes (or cosolvents) in water, in log-linear (a) and linear-linear (b) plots: SXS (3), 1-propanol (4), C3PO1 (0), C3EO1 (O), and acetone (]).

SDS the OD was measured until 1 mol/L (around 25 wt %) because a hexagonal phase appeared at higher concentrations (around 27 wt %). The solubilities of the DR-13 in C3EO1, C3PO1, and 1-propanol are, in the order of decreasing solvent power, as follows: 2519, 2106, and 815 (in OD unit), respectively. The DR-13 solubilization by SXS seems to be less efficient, in terms of the maximum solubilization, than in the case of the other studied additives: the OD at 2.248 mol/ L (40 wt %) is only 347.4. The same is true for the dye solubilization by the anionic surfactant SDS: the maximum solubilization corresponds to an OD value of 331 at 1 mol/L (around 25 wt % SDS). The solubilization of a hydrophobic compound in water using a cosolvent is wellknown to be the most efficient solubilization method among the four common ones: micellar solubilization, complexation, hydrotropy, and cosolvency,19 and this is confirmed here. This point will not be further discussed here. For comparison to the SDS solubilization profile, a double logarithmic scale was used; see Figure 2. The DR13 solubilization curve obtained with SDS exhibits the classical evolution observed in the case of micellar solubilization; i.e. the DR-13 solubilization increases suddenly when micelles are formed for concentrations above the SDS CMC (8 × 10-3 M). Then, when the SDS concentration further increases, the DR-13 solubilization increases linearly.18,19 Thus the DR-13 behaves like a classical hydrophobic compound in the presence of a surfactant. The SXS and C3EO1 solubilization curves were added in Figure 2. As expected the dye solubilization starts at much lower SDS concentrations compared to the hydrotrope and solvosurfactant. The following discussions

will be focused on the comparison between cosolvents, the hydrotrope, and the solvosurfactants. For all these additives an exponential increase in the OD, i.e. in the DR-13 solubilization, is observed when the additive concentration increases; see Figure 1b. This is confirmed in Figure 1a, where the OD is plotted in a logarithmic scale; the increases in the OD with increasing cosolvent or hydrotrope concentrations appear to be linear up to around 2-3 M. The typical sigmoidal shape sometimes observed in the case of hydrotrope solubilization could not be clearly detected here, not even in the case of the SXS (see Figure 1b) in which the OD is plotted in a linear scale instead of a logarithmic one. Nearly identical results were obtained for the solubilization curves of the hydrophobic yellow dye Jaune au gras W1201 (curves not shown here). An exponential increase in solubility with increasing concentration is typical of cosolvent solubilization.19 Moreover according to Balasubramanian et al.7 the solubilization obtained in the case of cosolvents “shows a continuous and monotonic increase”. It is important to remark that Balasubramanian et al.7 only considered one cosolvent to draw up their conclusions and that this cosolvent was poly(ethylene glycol) 400 (PEG-400) which is not representative of a water cosolvent. Anyway, from our studies it can be inferred that the shape of the solubilization curves obtained in the cases of cosolvents, of the hydrotrope, and of solvosurfactants are similar. We conclude therefore that there is no need to distinguish between these different types, and in the following paragraphs we will use “cosurfactant”, “solvosurfactant”, and “hydrotropes” as synonyms. In all cases the additive concentration, at which the solubilization of a hydrophobic compound becomes significant and measurable could be called MHC, as for typical hydrotropes. These MHC values characterize the hydrotropic efficiencies of the studied compounds (hydrotrope or solvent); i.e. the lower the MHC the more efficient the additive. However, the term “MHC” refers to a critical phenomenon and suggests thus that the amphiphile or cosolvent molecules self-aggregate above a well defined concentration in water. But for short amphiphile and cosolvent molecules the MHC values cannot be precisely determined because the solubilization of a hydrophobic compound, here a hydrophobic dye, does not occur suddenly but exponentially with increasing concentration.

6772

Langmuir, Vol. 21, No. 15, 2005

Bauduin et al. Table 1. Selected Properties of the Cosolvents/ Hydrotropes Studieda

SXS C3PO1 C3EO1 acetone 1-propanol

aY

aDR-13

Vm - V O m

4.22 3.27 2.37 1.26 1.51

5.1 3.62 2.5 1.8 1.8

140.4 102.4 82.2 57.5 59.13

d(ln CMC)/ dYA 6.16 5.46 4.96

log P 0.42 0.613 0.265 -0.157 0.344

a

a is given in OD units per (mol/L); Vm, the molar volume, in cm3/mol; and CMC and YA (the amount of added hydrotrope or cosolvent) in mass fraction. Table 2. Intercorrelation among Co-solvents/ Hydrotropes Propertiesa parameter

aY

aDR-13

Vm - VO m

d(ln CMC)/ dYA

log P

aY aDR-13 Vm - VO m d(ln CMC)/dYA log P

1 0.9701 0.9804 0.9939 0.4836

0.9701 1 0.992 0.9987 1-propanol) as the ordering inferred from their interfacial behaviors.20-22 For (20) Bauduin, P.; Basse, A.; Touraud, D.; Kunz, W. Colloids Surf., A, in press. (21) Sokolowski, A.; Chlebicki, J. Tenside Deterg. 1982, 19, 282. (22) Sokolowski, A.; Burczyk, B. J. Colloid Interface Sci. 1983, 94, 369.

Solubilization in Water by Hydrophobic Dyes

Figure 4. Optical density (OD), obtained with DR-13, as a function of (a) molar concentrations of hydrotropes (or cosolvents) in water and (b) molar concentrations of hydrotropes (or cosolvents) in water multiplied by the evaluated molar volume of the hydrophobic part of the respective hydrotrope (or cosolvent) molecules: C3PO2 (b), t-C4PO1 (2), C3PO1 (0), C1PO3 (9), C3EO1 (O), 1-propanol (4), and C1PO1 ([).

these three compounds, the intercorrelations obtained between aDR-13 and aY values and their ability to lower the CMC of an ionic surfactant, measured by their d(ln CMC)/dYA values, are remarkably good, namely 0.9987 and 0.9939, respectively (see Table 2). Note that the dielectric constant of the hydrotrope is sometimes used as a parameter to estimate its hydrophobicity. Although this strategy can be dangerous, because the dielectric constant is a macroscopic parameter and its dependence on microscopic interactions is not straightforward, it was successfully used to classify the hydrotropic efficiencies of several cosolvents in the case of the solubilization of indomethacin:23 ethanol > propylene glycol (PG) > glycerol. This order is in agreement with our postulation, namely that the hydrotropic efficiency correlates with the hydrotropic hydrophobicity. To have a broader data basis, we also determined the DR-13 solubilization by some other poly(propylene glycol) monoalkyl ether derivatives in mixture with water. The solubilization curves, as a function of “hydrotrope” molarity, obtained for C3PO2, t-C4PO1, C1PO3, and C1PO1 are shown in Figure 4a and compared to the ones already discussed for C3PO1, C3EO1, and 1-propanol. The solubilization or hydrotropic efficiency appears to be in the following order: C3PO2 > t-C4PO1 > C3PO1 ≈ C1PO3 > C3EO1 > 1-propanol > C1PO1. For the C3 derivatives, including 1-propanol, the order is again in agreement with the one obtained from recent results which classified the studied molecules according to their hydrophobic behaviors.20-22 The DR-13 solubilization by C4PO1, t-C4PO1, and C3PO2 water mixtures were measured up to the limits of solubility at 25 °C, which are respectively 5.3 wt % (23) Etman, M. A.; Nada, A. H. Acta Pharm. 1999, 49, 291. (24) Kaatze, U.; Menzel, K.; Pottel, R.; Schwerdtfeger, S. Z. Phys. Chem. 1994, 186, 141. (25) Kaatze, U.; Kettler, M.; Pottel, R. J. Phys. Chem. 1996, 100, 2360. (26) Menzel, K.; Rupprecht, A. U.; Kaatze, U. J. Phys. Chem. 1997, 101, 1255.

Langmuir, Vol. 21, No. 15, 2005 6773

(0.42 mol/kg), 16.6 wt % (1.51 mol/kg), and 17.1 wt % (1.17 mol/kg).27 Since the aqueous solubility of C4PO1 is too low, no hydrotropic solubilization could be observed. By contrast, for t-C4PO1 and C3PO2 solutions, exponential increases in the solubilization can be noticed at concentrations below their limits of solubility. These two compounds have the highest solubilization efficiencies, due to their pronounced hydrophobic character, among all studied compounds. Nevertheless, the maximum amount of solubilized dye remains limited in these two systems, at least in the water-rich part of the phase diagram. The optimum solubility is given by a balance of a sufficiently high water solubility of the hydrotrope and a pronounced hydrophobicity. We propose that these are the two parameters to be optimized in parallel and that, compared to them, other properties of cosolvents or hydrotropes, such as the shape or the aromaticity, are of minor importance. A last point: it is interesting to compare C1PO3 and C1PO1. The solubility efficiency is much higher in the case of C1PO3. Obviously propylene glycol units are hydrophobic, and the more of such groups are present, the higher is the hydrotropic efficiency. 4.4. Extent of the Hydrophobic Part in Hydrotrope Molecules: A Determining Factor in the Hydrotropic Efficiency. It is our hypothesis that the “amount of hydrophobicity” is an important parameter for the hydrotropic efficiency of additives. To make a more quantitative estimation, some quantum calculations were done to estimate the electron densities in some selected hydrotropic molecules. The results can be seen in Figure 5: the blue and green areas represent the low electron density parts in the molecules, whereas the yellow and red ones display the high electron density parts. Roughly speaking, the yellow/red areas, i.e. the oxygen atoms in ether and hydroxyl groups, are responsible for the water solubility of the hydrotropic molecules, whereas the green/ blue areas are the hydrophobic areas, hydrocarbon parts of the molecules. According to the extent of the hydrophobic parts, the following ordering is found: SXS > C3PO1 > C3EO1 > 1-propanol ≈ acetone. This is in agreement with the order observed for the hydrotropic efficiency, confirming thus the importance of the “amount of hydrotropic hydrophobicity” for the solubilization process. To illustrate this fact in a more quantitative manner, we proceed as follows. The volume of the hydrophobic parts of each studied compound is estimated by doing a simple calculation. The molar volume (Vm) of each studied pure compound at 25 °C is calculated from density values. It was found to be 73.5, 75.125, 114.2, and 134.3 cm3/mol respectively for acetone, 1-propanol, C3EO1, and C3PO1. For SXS the molar partial volume was evaluated from the density of an aqueous solution at 40 wt % and was estimated to be about 188.4 cm3/mol. As a first approximation the “hydrotropic” molar volume was assumed to be equal to the partial molar volume of the “hydrotrope” in solution in water. This approximation was checked in the case of C3PO1, for which the partial molar volumes at different concentrations were determined at 25 °C. In the diluted region the C3PO1 partial molar volume was found to differ from the molar volume of pure C3PO1 by not more than 9%. In the next step the molar volume of the hydrophobic part of these molecules was evaluated by subtracting from Vm the molar volume VO m, i.e. the “molar volume” of the oxygen atom(s) contained in each molecule. The molar volume of one oxygen (27) Bauduin, P.; Wattebled, L.; Schroedle, S.; Touraud, D.; Kunz, W. J. Mol. Liq. 2004, 115, 23.

6774

Langmuir, Vol. 21, No. 15, 2005

Bauduin et al.

aromatic rings of SXS and dyes and despite the fact that SXS is a salt in contrast to the other hydrotropes or cosolvents. To summarize, the intercorrelations among cosolvent/ hydrotrope parameters reported in Table 1 are given in Table 2. These parameters evaluate the hydrophobicity of the additives and are all given at 25 °C. log P values represent the partition coefficient of the additives between water and 1-octanol. This parameter is commonly used in various fields such as in pharmacy to evaluate the hydrophobicity of drugs, which determines their partitions between blood and biological tissues or their abilities to pass through biological barriers. It can be seen in Table 2 that the intercorrelations between a, Vm - VO m, and d(ln CMC)/dYA values are good. By contrast there is no correlation of these parameters with the log P values. This is not surprising because the log P values are correlated with the aqueous solubility of an additive, and we already noticed that the aqueous solubility of additives was not correlated, neither with the hydrotropic efficiency (see above) nor with the d(ln CMC)/dYA values (see ref 20). It should be further noted that the slight λmax shifts, for the two studied dyes, measured between the different additives do not correlate with any of the additive parameters reported in Table 2. Concerning the mechanism of the hydrotrope solubilization, it is interesting to remark that the short amphiphiles CiEOj, especially C4EO1, have been recently studied by ultrasonic and dielectric spectroscopy, and the results show that such molecules form microheterogeneous structures when mixed with water.24-26,28 The related aggregates appear at relatively high concentrations (around 1 M), and these concentrations correlate well with the concentrations at which the solubilization becomes significant in the case of C3EO1 and C3PO1 (chemical isomer of C4EO1). We have to bear in mind that at 1 M the average distance between two molecules is only 18.8 Å, assuming as a rough guide that the molecules are arranged on a cubic lattice and that the “length” of an SXS molecule is about 7 Å, the length of C3PO1 about 10 Å. Small hydrophobic association effects are then sufficient to bring the hydrotropic molecules together so that they can form bigger aggregates, in which hydrophobic molecules can be dissolved. 5. Conclusion Figure 5. Representations of the electron density in hydrotrope and cosolvent molecules. Green/blue and yellow/red areas represent respectively the low and high electron density parts in the molecules: SXS (a), acetone (b), 1-propanol (c), C3EO1 (d), and C3PO1.

atom was roughly approximated to be 16 cm3/mol. The Vm - VO m values are reported in Table 2 for the five hydrotropic molecules studied. With these approximations the optical densities given in Figures 3a and 4a were plotted again, but this time as a function of the “hydrotrope” concentration multiplied by the molar volume of the hydrophobic part. The results are given in Figures 3b and 4b. As expected the curves are much closer now, although a universal scaling law could not be detected. Partly, this is the consequence of the chemical specificity and different geometry of each hydrotrope; partly it may come from the limited number of hydrotropic molecules considered in this study. Nevertheless, such calculations may help to give a first estimation of the solubility efficiency of a hydrotropic or cosolvent molecule. Note that the SXS curve is now much more in agreement with the other curves, despite the postulated “specific” association between the

From the present study it is concluded that the solubility curves of hydrophobic dyes are all similar no matter if a cosolvent, a hydrotrope, or a solvosurfactant is used. The hydrophobic part of the hydrotrope or cosolvent molecules seems to be the major factor influencing the hydrotropic efficiency, and this is independent of the hydrophilic part, at least to first order. Other (second-order) effects may enhance the hydrotropic efficiency, for example special attractions between aromatic rings. But these effects neither are at the origin of hydrotropy nor are they indispensable to the hydrotropic effect. High hydrotropic efficiency does not mean that a large amount of hydrotropic molecules can be dissolved in water. If this is required, a hydrotrope must be used that is sufficiently soluble in water despite its necessarily high hydrophobicity. In this respect C3PO1 and C1PO3 offer a good compromise between a large hydrophobic part and a sufficient water comiscibility. To overcome the problem of the poor solubility of some potentially promising (28) Schroedle, S. Ph.D. Thesis, University of Regensburg, Germany, 2005.

Solubilization in Water by Hydrophobic Dyes

hydrotropes, it is preferable to mix them with other hydrotropes that are comiscible with water. By doing so an optimum in solubilization is often achieved. This principle of solublization is known as “facilitated hydrotropy”.29 The hydrophobicity of hydrophobic additives obtained by the evaluation of the volumes of their hydrophobic part correlates well not only with the efficiency to dissolve hydrophobic particles in water but also with surface tension measurements and the lowering of CMCs of classical surfactants in water. By contrast none of these (29) Simamora, P.; Alvarez, J. M.; Yalkowsky, S. H. Int. J. Pharm. 2001, 213, 25.

Langmuir, Vol. 21, No. 15, 2005 6775

parameters correlate neither with the aqueous solubilities nor with the log P values of the additives. Moreover the hydrotropic efficiencies of additives do not correlate at all with the solubility of hydrophobic molecules in the pure hydrotropes. For example, the DR13 dye used here is best soluble in pure acetone. However, the highest amount of dissolved dye in dilute water solutions is found with C3PO1 and C1PO3. A similar conclusion was already drawn in ref 20: It can be misleading to extrapolate the behaviors of pure hydrotropic compounds or mixtures of them with only a small amount of water to the properties of dilute aqueous solutions. LA050554L