Aqueous Block Copolymer−Surfactant Mixtures and Their Ability in

Aqueous Block Copolymer−Surfactant Mixtures and Their Ability in Solubilizing Chlorinated Organic Compounds. A Thermodynamic and SANS Study. R. De L...
0 downloads 0 Views 274KB Size
J. Phys. Chem. B 2006, 110, 25883-25894

25883

Aqueous Block Copolymer-Surfactant Mixtures and Their Ability in Solubilizing Chlorinated Organic Compounds. A Thermodynamic and SANS Study R. De Lisi,† M. Gradzielski,‡ G. Lazzara,† S. Milioto,† N. Muratore,*,† and S. Prevost‡,§ Dipartimento di Chimica Fisica “F. Accascina”, UniVersita` degli Studi di Palermo, Viale delle Scienze, Parco D’Orleans II, 90128 Palermo, Italy, Stranski Laboratorium fu¨r Physikalische und Theoretische Chemie, Institut fu¨r Chemie, Technische UniVersita¨t Berlin, Strasse des 17, Juni 124, 10623 Berlin, Germany, and Hahn-Meitner-Institut Berlin, Glienicker Strasse 100, 14109 Berlin, Germany ReceiVed: August 4, 2006; In Final Form: October 6, 2006

Within the topic of surfactant enhanced solubilization of additives sparingly soluble in water, volumetric, solubility, conductivity, and small-angle neutron scattering (SANS) experiments on mixtures composed of R,ω-dichloroalkane, surfactant, copolymer, and water were carried out at 298 K. The triblock copolymers (ethylene oxide)132(propylene oxide)50(ethylene oxide)132 (F108) and (ethylene oxide)76(propylene oxide)29(ethylene oxide)76 (F68) were chosen to investigate the role of the molecular weight keeping constant the hydrophilic/hydrophobic ratio. The selected surfactants are sodium decanoate (NaDec) and decyltrimethylammonium bromide (DeTAB) with comparable hydrophobicity and different charged heads. The R,ωdichloroalkanes were chosen as contaminant prototypes. For the water + surfactant + copolymer mixtures, both the volume and the SANS results straightforwardly evidenced that (1) monomers of NaDec and copolymer unimers generate small mixed aggregates, (2) monomers of DeTAB combined with copolymer unimers do not form aggregates, and (3) unimeric copolymer is solubilized into NaDec and DeTAB micelles. The R,ωdichloroalkanes presence induces the F108 aggregation even at very low copolymer composition. The addition of surfactant disintegrates the F108 aggregates and, consequently, the additive is expelled into the aqueous phase. Once F108 is in the unimeric state, it forms copolymer-micelle aggregates which incorporate the oil. In the case of F68 both the volumetric and the SANS data reveal that the additive does not alter the copolymer unimeric state. Moreover, they show that for the aqueous DeTAB-F68 system the additive trapping in both the copolymer-micelle aggregate and the pure micelles takes place being enhanced in the former aggregate in agreement with solubility experiments. For the NaDec-F68 mixtures, an additional solubilization process in the premicellar copolymer-surfactant microstructures occurs. SANS and conductivity data show that the additive incorporation into the mixed and the pure micelles does not essentially influence the structural properties of the aggregates.

Introduction In the last years, scientific interest has been growing toward aqueous mixtures composed of conventional surfactants and poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) (PEO-PPO-PEO) copolymers (commercially available under the trade name of Pluronics or Synperonics). A number of studies1-7 have already been performed for a better comprehension of the complex mechanism of interaction between copolymer and surfactant. Sastry and Hoffman8 outlined and summarized the progress achieved on this topic. However, the issue of solubilization in such mixed systems has so far received only very little attention. Polymeric surfactants offer the great advantage to be properly tuned by modulating the molecular weight, the composition, and the hydrophobic/hydrophilic portions. Pluronics are more active7 at the water/air interface than conventional surfactants and can associate in aqueous solutions into micellar aggregates9,10 of various forms11 which have been exploited for solubilizing organic contaminants.12-16 Furthermore, they display the polymeric effect17,18 required for * Address correspondence to this author. E-mail: [email protected]. † Universita ` degli Studi di Palermo. ‡ Technische Universita ¨ t Berlin. § Hahn-Meitner-Institut Berlin.

stabilizing colloidal systems and remove organic materials adsorbed onto the solid surface.19 Developing a better understanding of aqueous copolymersurfactant mixtures is important because of their wide industrial applications (pharmaceuticals, enhanced oil solubilization, cosmetics, colloidal stabilization, etc.). For instance, the remediation technology for removing nonaqueous phase liquids from soil and water basins uses surfactant systems;20 one of the main roles played by the surfactant is to enhance the solubility of contaminants in water. At this stage, it is difficult to straightforwardly describe the effects of the surfactant and the copolymer nature on their interaction processes, but it can certainly be stated that, whatever are the characteristics of both components, mixed aggregates featured by hydrophobic domains are formed.1-7,21,22 Consequently, interesting solubilization effects of the Pluronicsurfactant mixtures toward additives sparingly soluble in water are expected. As far as we know, only a few solubilization studies within this topic are available.23-25 Therefore, our intention was to provide a contribution to the knowledge of the chemico-physical aspects within the solubilization of chlorinated additives in the copolymer-surfactant mixtures with respect to the additive hydrophobicity and the structure of both copolymer

10.1021/jp065035l CCC: $33.50 © 2006 American Chemical Society Published on Web 12/08/2006

25884 J. Phys. Chem. B, Vol. 110, No. 51, 2006 and surfactant. To this purpose, R,ω-dichloroalkanes (dichloromethane, 1,2-dichloroethane, and 1,3-dichloropropane) were chosen as prototype contaminants. Higher homologues were not studied because their low solubility in water26 prevented obtaining reliable volumetric data. As copolymers (ethylene oxide)132(propylene oxide)50(ethylene oxide)132 (EO132PO50EO132; F108) and (ethylene oxide)76(propylene oxide)29(ethylene oxide)76 (EO76PO29EO76; F68), which exhibit the same EO/PO ratio (ca. 5) and different molecular weight, were chosen. The selected surfactants are sodium decanoate and decyltrimethylammonium bromide, which have comparable hydrophobicity and different charged heads. Mixtures of cationic surfactants and nonionic polymers have been studied27 to a lesser extent because these systems possess weak interactions whereas the opposite has been observed for anionic surfactants. However, it is well-established that the presence of larger amounts of surfactant leads to a complete disintegration of Pluronic micelles and the copolymer molecules become solubilized unimerically in the surfactant micelles.28-30 The aqueous copolymer-surfactant mixtures were first characterized by keeping the copolymer composition below its critical micellar concentration,10 and by changing the surfactant concentration. Then, a comprehensive study of the R,ωdichloroalkanes behavior in the aqueous surfactant solutions and the aqueous surfactant-copolymer mixtures was performed. For a given surfactant system, the additive concentration was maintained low and constant. Volumetric, small-angle neutron scattering (SANS), solubility, and conductivity experiments were performed at 298 K. Experimental Section Materials. Sodium decanoate (NaDec, g99%) and decyltrimethylammonium bromide (DeTAB, g98%) were Sigma products. (Ethylene oxide)76(propylene oxide)29(ethylene oxide)76 (F68, average molecular weight Mw ) 8350 g mol-1) and (ethylene oxide)132(propylene oxide)50(ethylene oxide)132 (F108, average molecular weight Mw ) 14600 g mol-1) were obtained as gifts from BASF AG (Ludwigshafen). It should be noted that both copolymers are commercial products. For F68 and F108 polydispersity indices of 1.1 and 1.2 have been determined,31 which means that their behavior can substantially be affected by the fact that molecules of different hydrophobicity are present in one copolymer sample. D2O was obtained from Eurisotop in 99.9% isotopic purity. Dichloromethane (99.9%), 1,2-dichloroethane (99.8%), and 1,3-dichloropropane (99%) were Aldrich products. All compounds were used as received. The standard partial molar volumes of the surfactants, the copolymers, and the R,ω-dichloroalkanes in water, obtained from density experiments, agree with the values reported elsewhere.10,32-34 Care was taken during the experiments to avoid additive evaporation. To this aim, the proper amount of the R,ωdichloroalkane was added to an aliquot of the solvent mixture in a bottle furnished with a screw cap, where a Teflon tube, with a controlled opening, had been previously inserted. All solutions were prepared by mass ((0.01 mg), using water from reverse osmosis (Elga model Option 3) having a specific resistivity higher than 1 MΩ cm-1, and were equilibrated over 24 h. Equipment. Density. The densities of the various mixtures were determined at 298 K by using a vibrating tube flow densimeter (Model 03D, Sodev Inc.) sensitive to 3 ppm. The temperature was controlled within 0.001 K by using a closed

De Lisi et al. TABLE 1: Critical Micellar Concentration and Degree of Ionization of the Micelles in the Various Solvents Mixtures at 298 Ka solvent mixtures

cmc

β

water water + Cl2CH2 water + Cl2(CH2)3 water + F108 water + F108 + Cl2(CH2)3

DeTAB 67.3 ( 1.1 54.3 ( 1.0 64 ( 3 64.1 ( 1.2 60.0 (1.2

0.270 ( 0.003 0.259 ( 0.003 0.278 ( 0.006 0.271 ( 0.003 0.280 ( 0.003

water water + Cl2CH2 water + Cl2(CH2)3 water + F108 water + F108 + Cl2(CH2)3

NaDec 90 ( 3 94 ( 4 88 ( 3 95 ( 2 97 ( 3

0.483 ( 0.006 0.524 ( 0.006 0.510 ( 0.007 0.550 ( 0.005 0.517 ( 0.006

a cmc in mmol kg-1. The mA values for Cl2CH2 and Cl2(CH2)3 are 80 and 25 mmol kg-1, respectively. The mP value for F108 is 0.6 mmol kg-1.

loop temperature controller (Model CT-L, Sodev Inc.). The procedure used to calibrate the densimeter is described elsewhere.35 The apparent molar volume (VΦ,A) of the additive in a given solvent mixture was calculated by means of the following equation

VΦ,A )

3 M 10 (d - do) d mAddo

(1)

where mA and M are the molality and the molecular weight of the additive, respectively, and d and do are the densities (g cm-3) of the solution and the solvent mixture, respectively. In the case of the water + additive + surfactant system, the solvent is the aqueous surfactant solution whereas for the water + additive + surfactant + copolymer system the solvent is the aqueous copolymer-surfactant mixture. The apparent molar volume (VΦ,P) of the copolymer in the aqueous surfactant (or additive) solution was also calculated. Therefore, in eq 1 the concentration of the copolymer (mP) replaces mA, d represents the density of the water + surfactant + copolymer (or water + additive + copolymer) ternary system, and do is the density of the water + surfactant (or water + additive) binary system. The volumes of transfer of both the additive (∆Vt,A) and the copolymer (∆Vt,P) were calculated as the difference between the apparent molar volume (VΦ,A or VΦ,P) in a given solvent mixture and that in water. ConductiVity. The conductivity measurements were carried out at 298.0 ( 0.1 K by using a digital conductimeter Analytical Control 120. The specific conductivity corrected for the solvent (κ - κo) as a function of the surfactant concentration (mS) was calculated. The critical micellar concentration (cmc) was determined as the intersection point of the two straight lines, which may be discriminated in the graph of (κ - κo) vs mS, while the degree of ionization of the micelles (β) was obtained from the ratio of the slopes of these straight lines.36 The calculated values are collected in Table 1. This procedure was not applied to the systems containing F68 because the (κ - κo) dependence on mS did not evidence a clear break. The plots of (κ - κo) vs mS not shown in the paper are reported in the Supporting Information. Solubility. The solubility of R,ω-dichloroalkanes in the aqueous surfactant solutions and the aqueous copolymersurfactant mixtures at various surfactant concentrations was determined at 298.0 ( 0.1 K. To this aim, the solute was added

Surfactant Enhanced Solubilization of Additives in Water

J. Phys. Chem. B, Vol. 110, No. 51, 2006 25885

Figure 1. Volume of transfer of the copolymer from water to the aqueous sodium decanoate (b) and decyltrimethylammonium bromide (4) solutions as functions of the surfactant concentration at 298 K. Lines are the best fits according to eq 2.

Figure 2. Specific conductivity, corrected for the solvent, of NaDec (a) and DeTAB (b) as a function of the surfactant concentration in water (×), water + F108, mP ) 0.6 mmol kg-1 (4), and water + F68 at mP ) 15 (O) and 25 mmol kg-1 (b) at 298 K.

to the solvent mixture until turbidity was observed by eye. The systems were equilibrated over ca. 24 h. SANS. The small-angle neutron scattering (SANS) experiments were performed at the Hahn-Meitner Institute, Berlin (Germany), on the instrument V4.37 A wavelength of 6.0 Å was chosen, and sample-to-detector distances of 1 and 4 m, with a collimation distance of 4 m, were employed. The data were recorded on a 64 × 64 two-dimensional detector. They were afterward radially averaged and converted into absolute units, i.e., the differential cross-section, by comparison with the scattering of a 1 mm H2O sample and proper correction for detector background and the scattering of the empty cell.38,39 The data treatment was done by the standard software BerSANS provided by the HMI.40 It should be noted that the incoherent scattering (Iinc) was not subtracted from the intensity curves shown in the following. In the analysis of the scattering data Iinc was considered in the analysis procedure as a fitting parameter and it was always verified that the fitted Iinc was in agreement with the theoretically calculated value according to the concentration of the corresponding nuclei and their incoherent scattering length. The solutions (about 1 g per sample) were prepared by mass (to a precision of (0.1 mg) with D2O as solvent; they were equilibrated over 24 h.

values. Also, (κ - κo) varies linearly to ca. 0.04 mol kg-1 (Figure 2) where the onset of the volume sharp variation occurs. As concerns the cationic surfactant, the volume of F108 shows a smooth maximum at ca. 0.08 mol kg-1 (Figure 1). Moreover, F108 influences the DeTAB (κ - κo) values above 0.05 mol kg-1. The profile of the ∆Vt,P vs mS curve for F68 in DeTAB is equal to that for F68 in NaDec (Figure 1) and the (κ - κo) vs mS trend is monotonic and different from that in water. Such a dissimilarity is enhanced by the higher F68 amount, 25 mmol kg-1 (Figure 2). The ∆Vt,P data were quantitatively treated by using a recent model.41 This approach assumes that the aqueous mixtures of unimeric copolymer and surfactant are characterized by three equilibria which allow the formation of the following aggregates: (1) the surfactant-copolymer aggregation complexes (composed of z surfactant molecules and 1 copolymer molecule) which are formed at mS lower than the cmc in water; (2) the copolymer-micelle aggregates constituted by w copolymer molecules and 1 micelle having the aggregation number equal to the value in water (N); and (3) the pure micelles (at constant N values), the formation of which starts at mS near the cmc in water. Moreover, the reliability of this approach was proved.5,6,41,42 The following equation was used

Results and Discussion I. Aqueous Copolymer-Surfactant Mixtures. (i) Volume and ConductiVity Results. Figure 1 illustrates that the volume of transfer of the copolymer from water to the surfactant solution, ∆Vt,P, of F108 (mP ) 0.6 mmol kg-1) in NaDec sharply increases with mS to ca. 0.1 mol kg-1 and upon further addition of the surfactant it decreases. The variation of the specific conductivity (κ - κo) with mS is equal to that for NaDec in water (Figure 2), slightly differing above 0.4 mol kg-1. The cmc determined from this graph (Table 1) is coincident with the mS value at which the ∆Vt,P vs mS plot (Figure 1) exhibits the extremum. For F68 (mP ) 15 mmol kg-1) in NaDec, ∆Vt,P changes linearly to ca. 0.04 mol kg-1 thereafter it monotonically increases reaching a constant value. The addition of F68 to the NaDec + water mixture shifts the (κ - κo) vs mS trend to lower

∆Vt,P ) 2BV,PSxP[m] + xC∆VC + [m0] - [m] - zxCmP ∆Vm + xD∆VD (2) mP where BV,PS is the copolymer-surfactant interaction parameter in the aqueous phase and ∆Vm is the volume of micellization of the surfactant. ∆VC and ∆VD are the volume changes for the formation of the surfactant-copolymer aggregation complexes and the copolymer-micelle aggregates, respectively, the fractions of which are related to the equilibrium constants (KC and KD) as

xC ) KCxP[m]z;

xD )

(xPmP)wmMKD mP

(3)

where xP is the fraction of the free copolymer, and [m] and mM

25886 J. Phys. Chem. B, Vol. 110, No. 51, 2006

De Lisi et al.

TABLE 2: Thermodynamic Properties for the Formation of the Surfactant-Copolymer Aggregation Complexes and the Copolymer-Micelle Aggregates at 298 Ka DeTAB F68 1b

mP BV,PS KC z -∆GoC ∆VC KD -∆GoD ∆VD

NaDec F108

F68

F108

1c

15; 23 ( 15 na

0.6 413 ( 17 na

2450 ( 990; 2860 ( 140b 19 ( 1; 19.72 ( 0.12b 44.9 ( 0.9

2900 ( 600 19.8 ( 0.5 87.3 ( 0.8

0.25c

15; 280 ( 10 29000c 4.7c 25.5 ( 1.1c 116 ( 5 1800c 18.6 ( 0.3c 95.3 ( 1.4

0.6; 1500 ( 400 340 ( 150; 359 ( 9c 2.3 ( 0.2; 2.3c 14.4 ( 1.1; 14.58 ( 0.06c 170 ( 6 142 ( 10; 50 ( 12c 12.28 ( 0.17; 9.7 ( 0.6c 0.05 ( 0.04

a m in mmol kg-1; B 3 -2 kg; K in kg mol-1; K in kgz mol-z; volume in cm3 mol-1; standard free energy in kJ mol-1. na, no P V,PS in cm mol D C aggregates are formed. b From ref 41. c From ref 5.

Figure 3. Scattering function of D2O + F68 (O), D2O + F68 + Cl2(CH2)2 (3), D2O + F68 + surfactant (4), and D2O + F68 + surfactant + Cl2(CH2)2 (0). Left side: NaDec 66 mmol kg-1. Right side: DeTAB 40 mmol kg-1. mP ) 15 mmol kg-1 and mA ) 80 mmol kg-1. The curves showing maxima were fitted as a linear combination of eqs 4 and 6; the other curves were fitted by means of eq 4. Lines are best fits.

are the concentrations of the monomeric surfactant and the micelles in the presence of the copolymer. To calculate the third term at the right-hand side of eq 2, i.e., the shift of the micellization equilibrium induced by the copolymer, the monomer surfactant concentration in water [m0] and ∆Vm were calculated according to a mass action model43 by using proper data.44 The minimizing procedure was carried out by means of a nonlinear least-squares fitting method. For the F68/NaDec system, the obtained equilibrium constants were comparable to those determined5 from the enthalpies of transfer at mP ) 1 mmol kg-1 but affected by large errors. This is ascribable to the small xC and xD values due to the high mP (15 mmol kg-1). Therefore, we used the literature equilibrium constants5 as fixed parameters in the minimizing procedure. The best fits are shown in Figure 1. Whenever it is possible to make comparisons, the provided parameters (collected in Table 2) agree with those obtained from calorimetry.5,41 Moreover, for all the systems, it was found that one copolymer molecule binds to one surfactant micelle, i.e., w ) 1. (ii) SANS Results. SANS measurements were carried out on solutions of both NaDec and DeTAB in the absence and the presence of F68 kept in the unimeric state (mP ) 15 mmol kg-1). Pure F68 in D2O produces a quite low scattered intensity, which monotonically decreases with q (Figure 3). The data can be described well by the Debye equation45 for a noninteracting random coil

I(q) )

cg M w

2 (Fp - Fs)2 2(x - 1 + e-x) + Iinc x

NAvdp2

(4)

where x is (qRg)2 and Iinc is the incoherent background, cg is the concentration (g cm-3), Mw is the mass-average molecular

weight, NAv is the Avogadro constant, and dp is the density of the polymer (it is 1.173 g cm-3 for F68), whereas Fp and Fs are the scattering length densities of polymer and solvent, respectively (it is 6.1 × 109 and 63.6 × 109 cm-2 for F68 and D2O, respectively). The best fit provided a molecular weight of 780 g mol-1 and an Rg value of 12.2 ( 0.9 Å. This is much lower than the molecular weight of 8350 g mol-1 of the F68 molecule and as the Rg estimated to be 39 Å by a random coil model46 for the whole molecule

Rg )

x

C∞nl2 6

(5)

where l is the monomer length (3.6 Å), C∞ is the characteristic ratio in the limit of long chains (4.0 for PEO), and n is the number of monomer units () 181). The explanation is that here only the PPO core of the monomerically dissolved F68 molecules is visible, which is rather compact due to its hydrophobicity. For the PPO block only Rg from eq 5 would yield 15.8 Å, in reasonable agreement with the experimental observation, even showing that apparently in the SANS experiment not the whole part of the PPO is visible as also the apparent molecular weight is substantially lower than the theoretical value of 1682 g mol-1. In contrast to the more compact PPO, the PEO blocks are meandering through the aqueous solution. Such a picture is realistic as the mean distance between the polymer molecules at the given concentration can be estimated to be 49 Å. The addition of small amounts of surfactant to the D2O + F68 mixture generates a scattering behavior strongly dependent on the surfactant nature (Figure 3). DeTAB does not interact with F68. In fact, the I(q) vs q curve practically superimposes

Surfactant Enhanced Solubilization of Additives in Water

J. Phys. Chem. B, Vol. 110, No. 51, 2006 25887

Figure 4. Scattering function of micellar solutions of NaDec (0.6 mol kg-1) (left side) and DeTAB (0.4 mol kg-1) (right side). Top: D2O + surfactant (4) and D2O + surfactant + F68 (O). Bottom: D2O + surfactant + Cl2(CH2)2 (4) and D2O + surfactant + F68 + Cl2(CH2)2 (O). mP ) 15 mmol kg-1 and mA ) 80 mmol kg-1. Lines are best fits according to eq 6.

to that of D2O + F68, and applying eq 4 to the experimental data, the Rg value of 12 ( 2 Å is obtained. The anionic NaDec behaves very differently as even at a concentration much lower than its cmc it produces a substantial increase in the scattered intensity (Figure 3) and the appearance of a structure factor peak at q ) 0.070 Å-1. Evidently even under these conditions mixed micellar aggregates are formed, a phenomenon that has similarly been observed for F6847 and EO97PO69EO97 (F127)28 for the case of sodium dodecylsulfate (SDS) addition. As Figure 4 shows, for higher concentrations of both NaDec and DeTAB (mS ≈ 0.5 mol kg-1) one always observes a structure factor peak at q ) 0.139 and 0.101 Å-1, respectively, typical of strongly interacting micelles. This peak and the scattering curves in general are only little affected by the addition of F68 and/or oil (Figure 4). The addition of F68 slightly shifts the structure factor peak toward higher q values, explainable by a higher concentration of micellar aggregates. On this basis, one may deduce that F68 only slightly affects the size of the aggregates and the involved interactions. The q value of the maximum is directly related to the mean spacing of the aggregates as 2π/q and from the peak position together with the absolute intensity the aggregation number (N) was evaluated (Table 3). For all systems exhibiting a correlation peak the scattering intensity could always be described well by a model of polydisperse, homogeneous spheres that are interacting via a screened Coulomb potential. For such a model the scattering intensity is given by

I(q) ) 1NP(q,R) S(q,σ,ζ·e0) + Iinc

(6)

where 1N is the number density of aggregates, P(q,R) is the particle form factor, and S(q,σ,ζ·e0) is the structure factor. The particle form factor for polydisperse spheres can be written as

P(q,R) )

∫0



2

f(r,R)‚|F(q,r)| dr

TABLE 3: Structural Data from SANS Analysis for the D2O + F68 + 1,2-Dichloroethane + Surfactant Mixtures at 298 Ka N(S)b

mA

βd

N(P)c

mS

mP

66 66

15 15

604 604 531 530

0 0 15 15

NaDec micellar region 0 28 80 29 0 22 80 22

25 25

15 16

DeTAB pre micellar region 0 na 80 na

417 417 380 380

0 0 16 16

DeTAB micellar region 0 49 83 46 0 29 85 35

NaDec pre micellar region 0 5.1 0.7 81 7.1 1.0

0.6 0.8

0.52 0.56 0.62 0.60

1.8 2.0

0.25 0.24 0.41 0.39

a Concentration in mmol kg-1. b Number of surfactant molecules in one aggregate from a model of polydisperse, homogeneous spheres that are interacting via a screened Coulomb potential. c Number of copolymer molecules in one aggregate. na, no aggregates are formed. d Calculated as the ratio between the charge and the aggregation number parameters obtained from the fit according to eq 6.

3(sin(qr) - qr cos(qr))

F(q,r) ) Vp∆F

(8)

(qr)3

where Vp is the volume of the sphere and ∆F the difference of the scattering length densities of particle and solvent. f(r,R) is the distribution function of the radii for which we employed a Schulz distribution48

f(r,R) )

( ) Z+1 R

(

)

(Z + 1)r rz exp R Γ(Z + 1)

Z+1

(9)

(7)

where F(q,r) for a homogeneous sphere is simply given by

where R is the mean radius and the parameter Z is related to the polydispersity index p via

25888 J. Phys. Chem. B, Vol. 110, No. 51, 2006

p2 )

〈r2〉 〈r〉

) 2

1 Z+1

De Lisi et al.

(10)

Finally, for the structure factor S(q,σ,ζ‚e0), which is determined by an effective hard sphere diameter σ and an effective charge ζ·e0, we used a simple random phase approximation49 as it has been employed by us successfully before for the description of charged microemulsion droplets.50 This S(q,σ,ζ·e0) depends on the Debye screening length and thereby on the ionic strength, which we approximated by the cmc of the pure surfactants or for the mixed aggregates by correspondingly lower values. In all fits we fixed the effective hard sphere diameter σ to be 2(R + 3 Å), which should be a reasonable assumption for the hydration shell fixed to the micellar surface. In our modeling we assumed the dispersed aggregates to be composed of the PPO part of the Pluronic copolymer, the surfactant molecule (excluding the counterion), and the 1,2-dichloroethane when present, respectively. The EO part was assumed to be evenly distributed in the aqueous solution and therefore was accounted for within the solvent scattering length density. All experimental data in the postmicellar region could be fitted very well with the described model (Figure 4) and the deduced data are summarized in Table 3. For the concentrated surfactant solutions well above the cmc one always observes formation of quite monodisperse micellar aggregates (the polydispersity index is ca. 0.13). However, the presence of F68 leads to a small reduction of the aggregation number, i.e., the mixed aggregates are somewhat smaller than the pure surfactant micelles. It is apparent that for mixtures of NaDec at a concentration below the cmc with F68 a strong synergistic effect for micellization is observed, while this is not the case for DeTAB. For the mixed NaDec/F68 case, small structures of an aggregation number are observed, which has similarly been observed before for SDS/F6847 and SDS/F12728 mixtures. No such aggregates are formed for the cationic DeTAB. Apparently the interaction is much less favorable with the cationic surfactant compared to the anionic NaDec. A more careful look at the SANS data shows that these premicellar mixed aggregates cannot simply be composed of individual F68 molecules decorated with a corresponding amount of NaDec. The experimental scattering data are best described by a linear combination of eqs 4 and 6, i.e., with a contribution from individual copolymer molecules and one from spherical charged aggregates. From fitting the experimental data, the molecular weight of 7600 g mol-1 for the aggregate is obtained, which yields the aggregation number of 36 for NaDec assuming that only the PPO block of the copolymer is in the aggregate. However, that number has to be taken with substantial precaution as it refers to a situation where absolutely no change in the aggregational and conformational behavior of the copolymer molecules is assumed, i.e., only a Mw of 780 g mol-1 would be ascribed. But that is not necessarily a realistic assumption as one may expect the Pluronic molecules to be much more compact in the situation of interacting with the NaDec molecules (as it is well-known that anionic surfactants typically interact strongly with PEO28,29). Accordingly, more of the PPO block and possibly also of the PEO block should become visible in the scattering experiment thereby rendering the real aggregation number of the NaDec substantially lower. Already this effect would effectively reduce the real aggregation number by at least 10 NaDec molecules. Another aspect not accounted for is the possibility of not only having one F68 molecule per aggregate, which would lead to an even more substantial lowering of the effective aggregation number of the

NaDec molecules. Given the fact that here the F68 concentration is much higher than the NaDec concentration these effects are certainly expected to be substantial and are much more important than at higher surfactant concentration. It is noteworthy the consistency between thermodynamic and SANS results. In particular, the SANS data analysis performed through the linear combination of eqs 4 and 6 provided the amount of the copolymer molecules present in the mixed aggregates as a fitting parameter (the number density of the aggregates here is determined by the position of the correlation peak). Its value (ca. 7%, assuming one copolymer molecule per aggregate) agrees with the thermodynamic one (7.5%). Moreover, the value of the volume fraction of the mixed aggregates of 0.0084 is consistent with that evaluated from the volume fit (0.0088). As concerns the aggregation number, one has to be reminded that the thermodynamic approach provides the average aggregation number weighted by the number of particles. The latter was therefore calculated from the SANS data (N ) 11). Such a value, in spite of being an upper estimation according to the above arguments, is in good agreement with the thermodynamic value (Table 2) and with that derived from a simple primitive cubic cell model (Table 3). Finally, the polydispersity index of such premicellar aggregates is much larger (ca. 0.38) than that obtained for the micelles. The low polydispersity index of the latter makes the mass and number averages nearly equal. Note that the thermodynamic approach cannot take into account the polydispersity. (iii) Discussion. The independent analysis of both the volume and the SANS data clearly highlighted that the copolymers discriminate between the cationic and the anionic surfactants in agreement with our previous calorimetric studies.41 Namely, NaDec forms small aggregates in the region below the cmc whereas DeTAB does not. At higher concentrations above the cmc, both surfactants form mixed micelles. A sketch representation of these structures is given in Figure 5. The copolymer/ surfactant stoichiometry of the aggregation complexes obtained from the fit of our thermodynamic data is in fair agreement with that calculated from the analysis of SANS data (Tables 2 and 3). The formation of these small aggregates leads to the ∆Vt,P increase as a consequence of the positive ∆VC value, which reflects the hydrophobic desolvation of both the components. Such microstructures start to disappear at mS ≈ cmc because of the formation of the copolymer-micelle aggregates (Figure 5). These findings agree with the dialysis results51 that the calculated moles of adsorbed SDS per gram of nonassociating polymer as functions of the equilibrium SDS concentration reach a maximum at mS ≈ cmc. For each system, the copolymermicelle aggregates are present in the concentrated domain (above the cmc) and their amount is dependent on the KD value as well as the copolymer composition. Moreover, their contribution to the volume is system specific as observed for the enthalpy.5 The thermodynamic and the structural information enable the interpretation of the conductivity data. For the aqueous F68NaDec mixture, the monotonic (κ - κo) change reflects the presence of the aggregation complexes which, upon increasing mS, tend to disappear because the mixed and the pure micelles are forming (Figure 5). The monotonic change of (κ - κo) with mS for DeTAB in the aqueous F68 solution (Figure 2) is ascribable to the presence of two kinds of micellar aggregates. According to the thermodynamic results, in the region to 0.2 mol kg-1 the mixed aggregates are the most predominant species, which are replaced by the pure micelles in the more concentrated domain. On this basis, we calculated the degree of ionization β at mP ) 15 mmol kg-1 by analyzing the

Surfactant Enhanced Solubilization of Additives in Water

J. Phys. Chem. B, Vol. 110, No. 51, 2006 25889

Figure 5. Representation of the pure and mixed aggregates for the sodium decanoate/water/copolymer and decyltrimethylammonium bromide/ water/copolymer mixtures in both the pre- and the postmicellar regions.

Figure 6. Volumes of transfer of Cl2CH2 (a), Cl2(CH2)2 (b), and Cl2(CH2)3 (c) from water to the aqueous decyltrimethylammonium bromide solution as functions of the surfactant concentration at 298 K. Lines are the best fits according to eq 11.

Figure 7. Volumes of transfer of Cl2CH2 (a), Cl2(CH2)2 (b), and Cl2(CH2)3 (c) from water to the aqueous sodium decanoate solution as functions of the surfactant concentration at 298 K. Lines are the best fits according to eq 11.

conductivity data to 0.2 mol kg-1. Its value (0.37) is larger than β in water (Table 1), presumably due to an additional shielding of the charges of the surfactant provided by the dipolar nature of PEO and PPO blocks. Furthermore, it is very close to β (0.39) calculated by taking data at mP ) 25 mmol kg-1 to mS ) 0.3 mol kg-1, where the mixed micellar aggregates are the main species. These β values agree with that obtained from the SANS analysis (Table 3). Finally, the shape of the (κ - κo) vs mS trends for both NaDec and DeTAB in water + F108 mixture (equal to those in water) indicate that the main aggregates are the pure micelles due to the quite low mP value. II. R,ω-Dichloroalkanes in Aqueous Surfactant Solutions. (i) Results. The profile of the transfer volume of the additive from water to the surfactant solution (∆Vt,A) vs mS curves for the R,ω-dichloroalkanes in the DeTAB solutions is independent of the additive nature (Figure 6). ∆Vt,A smoothly changes with mS to ca. 0.05 mol kg-1 thereafter it increases monotonically. As Figure 7 illustrates, the volume of Cl2CH2 in the NaDec solution is peculiar as it augments with mS to 0.1 mol kg-1 and thereafter it changes only slightly. According to the cmc in Table 1, one may deduce that Cl2CH2 experiences the presence of the monomeric surfactant whereas it does not the presence of the micelles. The other additives exhibit profiles of the ∆Vt,A vs mS trends similar to those in the DeTAB solution. The onset of

the steep increase of the volume takes place at mS values close to the cmc obtained from conductivities (Table 1). In agreement with the volume results, the solubility, normalized with respect to that in water (S/Sw), shows that the presence of the micellized NaDec and DeTAB essentially does not influence the solubility of Cl2CH2 in water whereas it enhances the solubility of the higher homologues (Figure 8). Finally, as Figure 4 shows, Cl2(CH2)2 in the NaDec and DeTAB micellar solutions has almost no effect on the scattering curves (a small increase in the absolute intensity is observed, due to the incorporation of the additive into the aggregate) and on the aggregation number (Table 3). (ii) Volume Modeling. The shape of the volume curves (Figures 6 and 7) is reminiscent of those of polar additives forming mixed micelles.33,52-54 Therefore, the volume data were quantitatively analyzed by means of the following equation, independently derived by De Lisi et al.54,55 and Desnoyers et al.52,53

∆Vt,A ) 2BV,AS[m]Nf + (1 - Nf)(Vb - VA,w) - ∆Vm{[m0] [m]}/mA (11) where BV,AS is the additive-surfactant interaction parameter, Vb and VA,w are the partial molar volumes of the additive in the micelles and water, respectively, and Nf is the fraction of the

25890 J. Phys. Chem. B, Vol. 110, No. 51, 2006

De Lisi et al. From the solubility data a Kb evaluation can be performed by means of the following equation54

Nf ) 1/{1 + Kb(mS - [m])}

(12)

Accordingly, Nf can be expressed as mA,aq/(mA,aq + mA,M), where mA,aq and mA,M are the moles of the additive in the aqueous and the micellar phases per kg of water, respectively. By stating that mA,aq ) Saq and mA,M ) (S - Saq), where Saq is the solubility of the additive in the aqueous phase, eq 12 assumes the following form

S/Saq ) 1 + Kb(mS - [m])

Figure 8. Solubility, normalized to that in water, of R,ω-dichloroalkanes in aqueous mixtures containing DeTAB (a) and NaDec (b) as a function of the surfactant concentration at 298 K: (b) Cl2CH2 in water + surfactant mixture; (O) Cl2(CH2)3 in water + surfactant; (0) Cl2(CH2)3 in water + F108 + surfactant at mP ) 0.6 mmol kg-1; (4) Cl2(CH2)3 in water + F68 + surfactant at mP ) 15 mmol kg-1; ()) Cl2(CH2)2 in water + surfactant; and (1) Cl2(CH2)2 in water + F68 + surfactant at mP ) 15 mmol kg-1.

additive in the aqueous phase. The other symbols have the same meaning as above. The De Lisi et al. approach employed the pseudo-phase transition model for micellization and a mass action model for the additive solubilization in the micellar phase.54,55 Desnoyers et al.52,53 used a mass action model for the micellization process and the pseudo-phase transition model for the distribution of the additive between the aqueous and the micellar phases. Both the methods assumed that the physico-chemical properties of the micelles are not affected by the presence of the additive. This is verified for the present systems as the data for conductivity (Table 1) and SANS (Table 3) indicate. We decided to use the Desnoyers et al. approach that is successful for short alkyl chain surfactants. The minimizing procedure was performed by means of a nonlinear least-squares fitting method, using the [m0] and ∆Vm values calculated according to the procedure aforementioned. The best fits shown in Figures 6 and 7 provided Vb, BV,AS, and the binding constant (Kb) for the additive distribution between the aqueous and the micellar phases (Table 4).

(13)

where Kb is the slope of the straight line obtained by plotting S/Saq against (mS - [m]). The mS value at which the S/Sw vs mS trend shows the break was stated to be [m] and the corresponding solubility was taken as Saq/Sw. The determined Kb (collected in Table 4) agree with the values provided by the volumes. This finding allows the supposition that the activity coefficient (γ) contribution is nearly unitary. As far as we know, no γ values on the present systems are known. However, the γ data of R,ωdichloroalkanes in ketones56 and aromatic hydrocarbons57 are very close to unity, consistent with our results. On the other hand, the increased solubility is not drastic; for the system presenting the largest Kb, i.e. Cl2(CH2)3 in the DeTAB solution, the highest measured S/Sw is ca. 10 (Figure 8). The standard free energy of transfer of the R,ω-dichloroalkanes from the aqueous to the micellar phases (∆Got,M), in the molarity scale, was calculated according to the literature procedure54

∆Got,M ) -RT ln(Kb,c/VM)

(14)

where VM (dm3 mol-1) is the partial molar volume of the surfactant in the micellar phase32,33 and Kb,c (dm3 mol-1) is the binding constant in the molarity scale. The latter, within the uncertainties, is equal to Kb (kg mol-1). (iii) Discussion. For a given surfactant, the BV,AS parameter exhibits a nonlinear change with the additive hydrophobicity. Also, its value is specific of the surfactant nature being smaller for DeTAB. The interactions between the surfactant polar head and the chlorine atoms of the additive are enhanced for DeTAB. As concerns ∆Got,M, it decreases with the number of carbon atoms in the additive alkyl chain (nC) evidencing that the micellar solubilization is enhanced upon increasing the additive hydrophobicity. Compared to DeTAB, the ∆Got,M values in NaDec are less negative (Figure 9) likely due to the less

TABLE 4: Thermodynamic Properties for r,ω-Dichloroalkanes Solubilization in Surfactant Micelles at 298 Ka

Cl2CH2

BV,AS

Kb

5.4 ( 0.9

3.5 ( 0.8 2.8 ( 0.5c 11.3 ( 0.9 25 ( 3 21 ( 3c

Cl2(CH2)2 Cl2(CH2)3

-24 ( 4 -18 ( 5

Cl2CH2

13.4 ( 0.8

Cl2(CH2)2

1.8 ( 0.8

Cl2(CH2)3

5.2 ( 1.0

0.0 ( 0.7 1.0 ( 0.5c 3.4 ( 0.4 4.4 ( 0.3c 7.2 ( 0.6 7.6 ( 0.4c

∆Got,M

∆Gos,w

∆Gos,M

Vb

V* b

VA,wb

-12.9 ( 0.3

61.1

64.6

60.3

-16.6 ( 0.2 -19.2 ( 0.3

80.1 95.9

79.0 95.7

75.7 90.2

-7.34 ( 0.3

-14.6 ( 0.3

79.6

79.0

75.7

-9.2 ( 0.2

-17.1 ( 0.2

96.2

95.7

90.2

DeTAB -7.0 ( 0.3 -5.88d -9.3 ( 0.2 -11.3 ( 0.3

-7.25d -7.93d

NaDec

3 -2 kg. Volume in cm3 mol-1. K in kg mol-1. Free energy in kJ mol-1. b From ref 34. c Obtained by applying eq 13 to solubility aB V,AS in cm mol b data. d From ref 59.

Surfactant Enhanced Solubilization of Additives in Water

Figure 9. Standard free energy and volume of transfer of R,ωdichloroalkanes from the aqueous phase to the micelles of DeTAB (b) and NaDec (O) as functions of the number of carbon atoms in the additive alkyl chain at 298 K.

favorable interactions between the chlorine atoms and the polar head. In particular, Cl2CH2 does not solubilize into the NaDec micellar phase in agreement with the solubility experiments, and the fraction of the more hydrophobic Cl2(CH2)3 in the micelle at the highest mS value analyzed is 0.6. To compare the solubilizing ability of the micelles toward each additive, the standard free energy of transfer (∆Gos,M) of the additive from the gaseous state to the micelles was calculated as ∆Got,M + ∆Gos,w, being ∆Gos,w the standard free energy of solvation of the additive in water58 (Table 4). The enhancement of the hydrophobic interactions is small whereas the affinity of the gaseous chlorine atoms to the aqueous micelles is strongly favored. The Vb values are very close to the volumes of pure additives (Table 4), which evidence that the site of solubilization of the additive into the micelle is the palisade layer. The Vb independence on the surfactant nature of a given additive is amazing. A reasonable explanation of this result is that the void space in the palisade layer is comparable for both micelles. III. R,ω-Dichloroalkanes in the Copolymer-Surfactant Mixtures. The discussion of the volumes, solubilities, conductivities, and SANS data successfully contributes to the comprehension of the additive behavior in the aqueous surfactantcopolymer systems. For the sake of clarity, the effect of the copolymer structure will be discussed in two paragraphs. (i) R,ω-Dichloroalkanes in the Aqueous F108-Surfactant Mixtures. As Figures 10 and 11 show, the addition of 0.6 mmol kg-1 of F108 to the water + surfactant mixtures influences the VΦ,A vs mS trends in the dilute region while above 0.06 and 0.09 mol kg-1 for DeTAB and NaDec, respectively, it does not affect VΦ,A, which assume values coincident with those determined in the absence of F108. The comparable slopes of S/Sw vs mS in the presence and the absence of F108 (Figure 8) for both surfactants reveal that the additive solubilization in the micelles is not affected by the presence of the copolymer. The negligible effect of F108 on the volumetric behavior of the additives in the micellar region might be ascribed to the copolymer concentration significantly lower than the surfactant composition. It may be verified that even relevant interactions between the additive and the copolymer are not detected (for instance, the data are not normalized for the copolymer concentration). Therefore, additional measurements were carried out by determining VΦ,A of Cl2(CH2)3 in the water + NaDec +

J. Phys. Chem. B, Vol. 110, No. 51, 2006 25891

Figure 10. Apparent molar volume of Cl2CH2 (a), Cl2(CH2)2 (b), and Cl2(CH2)3 (c) in the mixtures: water + DeTAB (O), water + F108 + DeTAB at mP ) 0.6 mmol kg-1 (4), and water + F68 + DeTAB at mP ) 15 mmol kg-1 (2). Lines are the best fits according to eq 4.

Figure 11. Apparent molar volume of Cl2CH2 (a), Cl2(CH2)2 (b), and Cl2(CH2)3 (c) in the mixtures: water + NaDec (O), water + F108 + NaDec at mP ) 0.6 mmol kg-1 (4), and water + F68 + NaDec at mP ) 15 mmol kg-1 (2). Lines are the best fits according to eq 4.

F108 mixtures at mP ) 2.5 and 5 mmol kg-1. The choice of these concentrations was dictated by the low F108 cmc (7.6 ( 0.2 mmol kg-1, at 298 K).10 As a general result, by increasing mP the VΦ,A vs mS curves are moved to larger values and the minima tend to disappear (Figure 12); moreover, the VΦ,A values at high mS are slightly larger than those in the water + NaDec mixtures. For a correct interpretation of these experimental findings, the effect of the additive on the F108 aggregation behavior was first highlighted. For all the additives studied, the apparent molar volumes of F108 in the water + additive mixture is equal to the partial molar volume of the aggregated copolymer in water10 (Table 5). The effectiveness of the additive in enhancing the copolymer aggregation even at very low composition is interesting. A similar finding was obtained from the dynamic light scattering and fluorescence studies16 of the water + F108 + Cl2(CH2)2 systems. The VΦ,A decrease with mS can reflect the expulsion of the additive from the F108 micelles which are undergoing a complete disintegration induced by the surfactant in agreement with the literature reports.28-30 The higher the mP is the larger is the surfactant amount required to

25892 J. Phys. Chem. B, Vol. 110, No. 51, 2006

De Lisi et al.

Figure 12. Apparent molar volume of Cl2(CH2)3 in the mixtures: water + NaDec (b); water + F108 + NaDec at mP ) 0.6 ()), 2.5 (∆), and 5 mmol kg-1 (O); and water + F68 + NaDec at mP ) 5 (4) and 15 mmol kg-1 (O).

TABLE 5: Volumes of Copolymers in Water and in the Water + r,ω-Dichloroalkanes Mixtures at 298 Ka mP

w+Cl2CH2 VΦ,P

w+Cl2(CH2)2 VΦ,P

w+Cl2(CH2)3 VΦ,P

w b VΦ,P

12492 12493 12490

12393 12393 12393

7113 7120

7119 7114

VMc

F108 12496 0.6 2.5 5

12464

12450

F68 7232 5 15

7104

7114

mP in mmol Volume in cm3 mol-1. bApparent molar volume of unimeric copolymer in water (ref 10). c Partial molar volume of the copolymer in the micellar state (ref 10). a

kg-1.

destroy the F108 micelles. The VΦ,A increase can be due to the additive solubilization in the pure and mixed surfactant aggregates. Thus, it is likely that at mP ) 0.6 mmol kg-1 and for mS > cmc of the surfactant, the copolymer is in the unimeric state and the coincidence of the VΦ,A points with those in the water + surfactant mixtures reveals the additive incorporation into the pure micelles which are the dominant species. This assessment is corroborated by the measurements of conductivity and solubility carried out on the aqueous surfactant-F108Cl2(CH2)3 system. In fact, the cmc and the β values derived from conductivities are very close to those in the aqueous surfactant + F108 mixtures, which, in turn, are nearly equal to those in the aqueous surfactant solutions. Moreover, the copolymer does not influence the S/Sw data (Figure 8). (ii) R,ω-Dichloroalkanes in the Aqueous F68-Surfactant Mixtures. In the first step, it was verified whether the R,ωdichloroalkane induces the F68 aggregation as it does for F108. For all the investigated systems, the VΦ,P values in the water + additive mixtures correspond to those of the unimeric copolymer in water10 (Table 5). Similar conclusions were drawn by the analysis of SANS data of the D2O + Cl2(CH2)2 + F68 mixture (mA ) 80 mmol kg-1 and mP ) 15 mmol kg-1), which evidenced that the added oil does not modify the unimeric state of the copolymer. On this basis, the SANS data, fitted by using

the randomly distributed coil form factor (eq 4), allowed the Rg value of 13 ( 1 Å, which is equal to that in water. The addition of Cl2(CH2)2 (mA ) 80 mmol kg-1) to the water + F68 + DeTAB (mS ) 40 mmol kg-1 and mP ) 15 mmol kg-1) does not alter the I(q) vs q trend suggesting that the additive does not cause aggregation processes (Figure 3). Accordingly, the Rg value of 13 ( 2 Å is obtained. The slight rise of the intensity of the maximum induced by Cl2(CH2)2 in the F68 + DeTAB mixture at mS ) 0.4 mol kg-1 can be explained considering that small amounts of incorporated oil somewhat affect the micelles formation. The VΦ,A vs mS curves of all the additives in the DeTAB + water + F68 mixture (mP ) 15 mmol kg-1) show a maximum in the dilute region. Compared to the VΦ,A in the water + DeTAB mixtures at high mS values, the corresponding volumes exhibit a difference that becomes smaller upon increasing the additive hydrophobicity. VΦ,A incorporates contributions for the additive solubilization in the copolymer-micelle aggregate and the pure micelles as well as the shift of the aggregation equilibria induced by the additive. The presence of the extremum in the VΦ,A vs mS curves at ca. 0.06 mol kg-1 may be due to the shift of the copolymer-micelle formation equilibrium generated by their interactions with the additive also evidenced by SANS data (Figure 3) as well as solubility experiments extended to Cl2(CH2)3. Accordingly, above 0.05 mol kg-1, S/Sw increases with mS with a slope that is 2-fold that observed for the same additives in the water + DeTAB mixture (Figure 8). At mS ) 0.1 mol kg-1, the displacement of the pure micelles formation equilibrium is expected to produce a small positive contribution (based on ∆Vm and Kb values). From the volume and the SANS data, one may infer that in the dilute domain (mS < 0.05 mol kg-1), the additive does not generate aggregated structures and in the concentrated one, the additive is incorporated in the mixed and pure micelles. Conductivity and SANS results indicate that such a solubilization does not change the structural properties of the aggregates. Significant differences are observed when DeTAB is replaced by NaDec (Figure 11). The maximum in the VΦ,A vs mS profile appears at ca. 0.05 mol kg-1, which is lower than the cmc (Table 1), whereas that for DeTAB is located at mS ≈ cmc. Interestingly, in the concentrated region, the additive hydrophobicity enhances the difference between VΦ,A in the water + F68 + NaDec and water + NaDec mixtures whereas the opposite takes place for DeTAB. This suggests that the solubilization phenomena are specific of surfactants in agreement with the different kinds of formed aggregates. The maximum in the VΦ,A vs mS plot reflects the shift of the copolymer-surfactant aggregation complexes formation due to the oil incorporation. The SANS data support this conclusion and show a substantially higher scattering intensity I(q) compared to the situation without added Cl2(CH2)2. From a quantitative analysis of the increase in intensity (by fitting the data in Figure 3 with the combination of eqs 4 and 6) we can deduce that about 85% of the added Cl2(CH2)2 should be incorporated into the aggregated species, in which an average number of 7 surfactant molecules is bound to 1 copolymer molecule. In addition, the peak position is slightly shifted to lower values, from which it can be concluded that on average the aggregation number of surfactant and copolymer increases somewhat (Table 3), i.e., an increased tendency of aggregation is induced by the presence of the hydrophobic solubilizate. From this analysis, one may argue that all the additives show attractive forces for these small aggregates, even Cl2CH2, which exhibits no affinity to NaDec

Surfactant Enhanced Solubilization of Additives in Water micelles. Such a peculiarity can reflect the low hydrophobic character of such structures. VΦ,A at high mS reveals the additive solubilization into the pure and mixed micelles. The VΦ,A points indicate that Cl2CH2 does not experience the presence of the micellar aggregates. By increasing the additive hydrophobicity, the VΦ,A values are larger than those in the water + NaDec mixtures suggesting that the oils are also solubilized in the mixed aggregates. Same conclusions can be drawn by inspecting the solubility data. The SANS results (Figure 4) show that the presence of the oil induces a small increase of I(q) and their analysis allows the determination of the aggregation numbers not influenced by the additive (Table 3). A similar result was observed in the absence of F68. The above arguments may explain the VΦ,A data of Cl2(CH2)3 in the water + NaDec + F68 mixtures at mP ) 5 mmol kg-1. The lack of the maximum can be due to the negligible effect of the shift for the copolymer-surfactant aggregation complexes formation due to the low amount of these aggregates. The VΦ,A merging points for mS > 0.2 mol kg-1 reveal the Cl2(CH2)3 solubilization into pure micelles which are the dominant species. Conclusions The ability of aqueous block copolymer-surfactant mixtures in solubilizing chlorinated organic contaminants was demonstrated. A combination of techniques (volume, conductivity, solubility, and SANS) was used to obtain specific thermodynamic and structural insights. Materials were carefully chosen to elucidate particular aspects. The copolymers (F68 and F108) were selected to investigate the role of the molecular weight keeping constant the hydrophilic/hydrophobic ratio. The surfactants of comparable hydrophobicity, i.e., NaDec and DeTAB, were studied to show the effect of the charged head groups. The R,ω-dichloroalkanes were chosen as a model for chlorinated contaminants. Volume findings in conjunction with the SANS results straightforwardly evidenced that there are significant differences between NaDec and DeTAB in both the pre- and the post-micellar regions in terms of the formed structures and their ability to dissolve the organic compounds. Whatever is the copolymer nature, it occurs that monomers of DeTAB combined with copolymer unimers do not form aggregates whereas monomers of NaDec and copolymer unimers generate small mixed aggregates which are tools for entrapping the oil. As concerns the post-micellar region, the additive incorporation takes place in both the copolymer-micelle aggregate and the pure micelles being enhanced in the former structures. Moreover, the aggregates composed of the cationic surfactant exhibit more effectiveness in their solubilizing ability. The copolymer structure also plays a role showing a different sensitivity to the presence of the additive. All the solubilization processes do not influence significantly the structural properties of the aggregates. From these studies one may infer that the aqueous block copolymer-surfactant systems are better candidates than the surfactant aqueous solutions in removing organic contaminants from soil and aquifers. Acknowledgment. This work was financially supported by Vigoni project 2006, CORI project 2005, and MIUR (Italy). We thank the BASF for kindly providing the copolymer. The experiments at BENSC in Berlin were supported by the European Commission under the 6th Framework Programme through the Key Action: Strengthening the European Research Area, Research Infrastructures. Contract No. RII3-CT-2003505925 (NMI3).

J. Phys. Chem. B, Vol. 110, No. 51, 2006 25893 Supporting Information Available: Conductivity curves for DeTAB and NaDec in some solvent mixtures at 298 K and table of the scattering length densities used in SANS analysis. This material is available free of charge via the Internet at http:// pubs.acs.org. References and Notes (1) Senkow, S.; Roux, A. H.; Roux-Desgranges, G. Phys. Chem. Chem. Phys. 2004, 6, 822. (2) Roux, A. H.; Douhe´ret, G.; Roux-Desgranges, G. Colloids Surf., A 2005, 252, 43. (3) Li, X.; Wettig, S. D.; Verrall, R. E. J. Colloid Interface Sci. 2005, 282, 466. (4) De Lisi, R.; Milioto, S.; Munafo`, M.; Muratore, N. J. Phys. Chem. B 2003, 107, 819. (5) De Lisi, R.; Lazzara, G.; Milioto, S.; Muratore, N. J. Phys. Chem. B 2004, 108, 18214. (6) De Lisi, R.; Lazzara, G.; Milioto, S.; Muratore, N. Macromolecules 2004, 37, 5423. (7) Vieira, J. B.; Thomas, R. K.; Li, Z. X.; Penfold, J. Langmuir 2005, 21, 4441. (8) Sastry, N. V.; Hoffmann, H. Colloids Surf. A2004, 250, 247. (9) Kelarakis, A.; Yang, Z.; Pousia, E.; Nixon, S. K.; Price, C.; Booth, C.; Hamley, I. W.; Castelletto, V.; Funding, J. Langmuir 2001, 17, 8085. (10) De Lisi, R.; Lazzara, G.; Lombardo, R.; Milioto, S.; Muratore, N.; Turco Liveri, M. L. J. Solution Chem. 2006, 35, 659. (11) Loh, W. In Encyclopedia of Surface and Colloid Science; Hubbard, A. T., Ed.; Marcel Dekker: New York, 2002; p 802. (12) Zhou, M. F.; Rhue, R. D. EnViron. Sci. Technol. 2000, 34, 1985. (13) Paterson, I. F.; Chowdhry, B. Z.; Leharne, S. A. Langmuir 1999, 15, 6187. (14) Hurter, P. N.; Alexandridis, P.; Hatton T. A. In Solubilization in Surfactant Aggregates; Christian, S. D., Scamehorn, J. F., Eds.; Marcel Dekker: New York, 1995; p 191. (15) Melik-Nubarov, N. S.; Kozlov M. Yu Colloid Polym. Sci. 1998, 276, 381. (16) Lazzara, G.; Milioto S.; Gradzielski, M. Phys. Chem. Chem. Phys. 2006, 8, 2299. (17) Nelson, A.; Cosgrove, T. Langmuir 2005, 21, 9176. (18) De, Lisi, R.; Lazzara, G.; Lombardo, R.; Milioto, S.; Muratore, N.; Turco Liveri, M. L. Phys. Chem. Chem. Phys. 2005, 7, 3994. (19) De Lisi, R.; Lazzara, G.; Milioto S.; Muratore, N. Langmuir 2006, 22, 8056. (20) Chlorinated SolVent and DNAPL Remediation. InnoVatiVe Strategies for Subsurface Cleanup; Henry, S. M., Warner, S. D., Eds.; ACS Symp. Ser. No. 837; American Chemical Society: Washington, DC, 2003. (21) Jansson, J.; Schillen, K.; Nilsson, M.; So¨derman, O.; Fritz, G.; Bergmann, A.; Glatter, O. J. Phys. Chem. B. 2005, 109, 7073. (22) Ganguly, R.; Aswal, V. K.; Hassan, P. A.; Gopalakrishnan, I. K.; Kulshreshtha, S. K. J. Phys. Chem. B 2006, 110, 9843. (23) Lad, K.; Bahadur, A.; Panda, K; Bahadur, P. Indian J. Chem., Sect A: Inorg. Phys. Theor. Anal. 1995, 34, 938. (24) Contractor, K.; Bahadur, P. Eur. Polym. J. 1998, 34, 225. (25) Gandhi, H.; Mody, S.; Gorge, A.; Jain, N.; Bahadur, P. Tenside, Surfactants, Deterg. 2002, 39, 8. (26) Leighton, D. T.; Calo, J. M. J. Chem. Eng. Data 1981, 26, 382. (27) Polymer-Surfactant Systems; Kwak, J. C. T., Ed.; Marcel Dekker, Inc.: New York, 1998. (28) Hecht, E.; Hoffmann, H. Langmuir 1994, 10, 86. (29) Hecht, E.; Mortensen, K.; Gradzielski, M.; Hoffmann, H. J. Phys. Chem. 1995. 99, 4866. (30) Thurn, T.; Couderc, S.; Sidhu, J.; Bloor, D. M.; Penfold, J.; Holzwarth, J. F.; Wyn-Jones, E. Langmuir 2002, 18, 9267. (31) Shar, J. A.; Obey, T. M.; Cosgrove T. Colloids Surf. A 1998, 136, 21. (32) De, Lisi, R.; Milioto, S.; Triolo, R. J. Solution Chem. 1988, 17, 673. (33) Yamashita, F.; Perron, G.; Desnoyers, J. E.; Kwak, J. C. T. In Phenomena in Mixed Surfactant Systems; Scamehorn J. F., Ed.; ACS Symp. Ser. No. 311; American Chemical Society: Washington, DC, 1986; p 79. (34) De Lisi, R.; Lazzara, G.; Milioto, S.; Muratore, N. J. Colloid Interface Sci. 2006, 300, 368. (35) De Lisi, R.; Lazzara, G.; Milioto, S.; Muratore, N. J. Phys. Chem. B 2003, 107, 13150. (36) Zana, R. J. Colloid Interface Sci. 1980, 78, 330. (37) Keiderling, U.; Wiedenmann, A. Phys. B 1995, 213 and 214, 895. (38) Chen, S. H.; Lin, T. L. Methods Exp. Phys. 1987, 23B, 489. (39) Keiderling, U. Phys. B 1997, 234-236, 1111. (40) Keiderling, U. Appl. Phys. A 2002, 74, S1455. (41) De Lisi, R.; Lazzara, G.; Milioto S.; Muratore, N. J. Phys. Chem. B 2004, 108, 1189.

25894 J. Phys. Chem. B, Vol. 110, No. 51, 2006 (42) De Lisi, R.; Lazzara, G.; Milioto S.; Muratore, N. Langmuir 2004, 20, 9938. (43) Desnoyers, J. E.; Caron, G.; De Lisi, R.; Roberts, D.; Roux, A.; Perron, G. J. Phys. Chem. 1983, 87, 1397. (44) The data used to calculate [m0] and ∆Vm were taken from refs 32 and 41. Moreover, details on this procedure are reported in ref 41. (45) Debye, P. J. Phys. Colloid Chem. 1947, 51, 18. (46) Flory, P. J. Statistical Mechanics of Chain Molecules; Interscience: New York, 1969. (47) Almgren, M.; van Stam, J.; Lindblad, C.; Li, P.; Stilbs, P.; Bahadur, P. J. Phys. Chem. 1991, 95, 5677. (48) Schulz, G. V. Z. Phys. Chem. B 1939, 43, 25. (49) Baba-Ahmed, L.; Benmouna, M.; Grimson, M. J. Phys. Chem. Liq. 1987, 16, 235. (50) Farago, B.; Gradzielski, M. J. Chem. Phys. 2001, 114, 10105.

De Lisi et al. (51) Holmberg, C.; Nilsson, S.; Sundelo¨f, L. Langmuir 1997, 13, 1392. (52) He`tu, D.; Roux, A. H.; Desnoyers, J. E. J. Solution Chem. 1987, 16, 529. (53) He`tu, D.; Camire´, C.; Perron, G.; Desnoyers, J. E. Langmuir 1994, 10, 4039. (54) De Lisi, R.; Turco Liveri, V.; Castagnolo, M.; Inglese, A. J. Solution Chem. 1986, 15, 23. (55) De Lisi, R.; Turco, Liveri, V. Gazz. Chim. Ital. 1983, 113, 371. (56) Barhala, A.; Dragoescu, D. ReV. Roum. Chim. 2002, 47, 831. (57) Barhala, A.; Dragoescu, D. ReV. Roum. Chim. 2004, 49, 97. (58) Lafuente, C.; Pardo, J.; Rodrı´guez, F. M.; Royo, F. M.; Urı´eta, J. S. J. Chem. Eng. Data 1993, 38, 554. (59) Pardo, J.; Rodrı´guez, V.; Lo´pez, M. C.; Royo, F. M.; Urı´eta, J. S. J. Chem. Thermodyn. 1992, 24, 113.