Chloroform Reverse Micelles: A Closed or Open

Oct 16, 2012 - Three scenarios of the reverse micelle formation, the closed, open and Eicke's association models, were considered in the interpretatio...
6 downloads 0 Views 1005KB Size
Download PDF
Article pubs.acs.org/Langmuir

CTAB/Water/Chloroform Reverse Micelles: A Closed or Open Association Model? L'ubica Klíčová,†,‡ Peter Šebej,†,‡ Peter Štacko,† Sergey K. Filippov,§ Anna Bogomolova,§ Marc Padilla,† and Petr Klán*,†,‡ †

Department of Chemistry, Faculty of Science, Masaryk University, Kamenice 5, 625 00 Brno, Czech Republic Research Centre for Toxic Compounds in the Environment, Faculty of Science, Masaryk University, Kamenice 3, 625 00 Brno, Czech Republic § Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, v. v. i., Heyrovsky Sq. 2, 162 06 Prague 6, Czech Republic ‡

S Supporting Information *

ABSTRACT: The micellization of cetyltrimethylammonium bromide (CTAB) in chloroform in the presence of water was examined. Three scenarios of the reverse micelle formation, the closed, open and Eicke’s association models, were considered in the interpretation of the experimental data. The growth of the aggregates was observed through the changes of NMR signals of associated water, probing the microenvironment of the premicellar aggregates and the interior of reverse micelles. This technique if combined with isothermal titration calorimetry (ITC) revealed that hydrated surfactant premicellar aggregates are already present at ∼6 mM CTAB. NMR, ITC and conductometry were used to determine the critical micelle concentration (cmc) to be ∼40 mM CTAB. It is suggested that the variation of the cmc values reflects the fact that the NMR analysis indicated the beginning of the reverse micelle formation, whereas conductometry and ITC measurements provided the upper limit and an average value of a so-called apparent cmc, respectively. The cmc values were found to be unaffected by the water content. The presence of reverse micelles, the existence of multiple equilibria, and high polydispersity of the samples were evidenced by DOSY NMR spectroscopy. As a result, we validated Eicke’s association model, according to which cyclic inverse micelles are formed by a structural reorganization of linear associates within a narrow concentration range, called the apparent cmc. New experimental results have also been gained for micellization of cetyltrimethylammonium chloride (CTAC) in chloroform in the presence of water; a similar mechanism of reverse micelle formation has been suggested.

1. INTRODUCTION

shown to disrupt hydrogen bonding in the reverse micellar interior by changing the pH of the water pool.20 The probe molecules can also unexpectedly reside at the hydrophobic part of the reverse micelles.21 The addition of new species to the reverse micellar system can change micellar size and surfactant aggregation number, or cause electrical percolation.14 Additives have been shown to influence the cmc in water-containing systems,16 although their effect was excluded in some cases.4 Direct methods to determine the cmc, such as isothermal titration calorimetry (ITC), conductometry, or NMR, are available. While ITC characterizes the energetics of the aggregation,6 1H NMR chemical shifts of water can be used to monitor the extent of hydrogen bonding, thereby providing information about the water-pool size in the core.15 Various models of the micellization process have been proposed. The mass action (closed association) model of

Micelles, self-organized molecular assemblies of amphiphilic molecules in water, which possess hydrophobic cores and hydrophilic shells, have been the subject of interest for decades. Reverse (inverse) micelles can be formed in nonpolar solvents in the absence of water where the hydrophilic groups of the amphiphiles form the core while the hydrophobic chains the outer shell.1−7 The presence of water influences the process of micellization.8−10 Sodium 2-ethylhexylsulfosuccinate (AOT), an anionic surfactant, has been the most frequently studied surfactant forming reverse micelles; cetyltrimethylammonium bromide (CTAB) and chloride (CTAC) reverse micelles have also been examined.8,11−15 UV and fluorescent probes, for example Safranine T,7,16 tetrasodium 1,3,6,8-pyrenetetrasulfonate,17 7,7,8,8-tetracyanoquinodimethane,18 or other systems, have frequently been used to study reverse micellar systems and to determine the critical micelle concentrations (cmc). However, their presence can considerably disturb the process of micellization and influence the cmc value.19 For example, the polyoxometalate probe was © 2012 American Chemical Society

Received: August 10, 2012 Revised: September 26, 2012 Published: October 16, 2012 15185

dx.doi.org/10.1021/la303245e | Langmuir 2012, 28, 15185−15192

Langmuir

Article

micellization22 assumes a dynamic equilibrium between the monomers and the molecules associated in an aggregate. On the other hand, the multiple equilibrium (open association) model of micellization assumes a continuous growth of the micelles,23,24 in which the equilibrium constants can differ in the consecutive steps of the formation. The former model was used for the determination of the cmc using NMR in the case of both normal and reverse micelles.25−29 General agreement exists about micellization in water where the cmc is usually well-defined.22 In nonaqueous solvents, a multiple equilibrium model seems to be preferable due to much smaller aggregation numbers of the initially formed aggregates.26,29−31 The features characteristic for both models can coexist in some systems.23 A model of the surfactant aggregation in nonpolar media proposed by Eicke describes the formation of linear aggregates in a stepwise process, in which cyclic inverse micelles are formed by a structural reorganization of linear aggregates within a narrow concentration range referred to as the apparent cmc.31 Several studies have validated this model for the formation of reverse micelles formed from anionic as well as cationic surfactants.10,19,32−34 The aim of this study was to investigate the mechanism of reverse micelle formation in a water/CTAB/chloroform-d system without using any probe molecules and to determine the cmc for different water concentrations in this ternary system. A water/cetyl trimethylammonium chloride (CTAC)/ chloroform-d system was also examined. Various techniques, such as NMR, conductometry, and ITC, were used to characterize the reverse micelle formation.

Isothermal Titration Calorimetry (ITC). The cmc values were determined by an isothermal titration microcalorimeter. The ITC experiments were performed three times using either 76 or 380 injections of a CTAB solution into chloroform (a constant titration volume of 0.5 or 0.1 μL; 180 s intervals). The thermograms were recorded and analyzed using Origin software.

3. RESULTS 1 H NMR Spectroscopic Investigations. Effects of the Surfactant Concentration. 1H NMR spectra of the CTAB/ water/chloroform-d and CTAC/water/chloroform-d systems for various water-to-surfactant molar ratios cwater/cCTAX (x; X = B or C, respectively) and cCTAX 1−200 mM were measured (Figure 1). The chemical shifts (δobs) of the surfactant

Figure 1. 1H NMR spectra of the CTAB/water/chloroform-d system (x = 3) obtained for different CTAB concentrations (cCTAB). The signals of water protons are marked by an asterisk.

2. EXPERIMENTAL SECTION

hydrogens’ signals did not exhibit any significant (≤0.1 ppm) dependence on the surfactant concentration. A distinct (single) signal was attributed to water in all equilibrated mixtures, and its shift moved downfield with the increasing surfactant concentration. Figure 2a represents a plot of δobs (H2O) versus (1/cCTAB) for the CTAB/water/chloroform-d system at x = 3 and two linear fits to the experimental data. The intersection point of the straight lines at cCTAB = (31 ± 4) mM defines the cmc. The dependence of water loading w, the ratio of cwater (core) (water in the core of reverse micelles), and cCTAB (micelle) (surfactant molecules of the assemblies) was evaluated as the ratio of the normalized peak areas associated with the water and the terminal CH3 group of CTAB proton signals at different cCTAB. The corresponding plot of two sequential linear dependencies for x = 3 provided the intersection point which corresponds to cCTAB = 5.6 mM (Figure 2b). Effects of the Water-to-Surfactant Molar Ratios (x). The influence of x (cwater/cCTAB) ranging from 1 to 5 on the point of intersection for two linear functions in three different plots ([δobs (H2O) vs (1/cCTAB)], [w vs cCTAB], and [κ vs cCTAB]; Figures S1 and S2) is shown in Table 1. The dependence of the water loading w on x for a CTAB solution (cCTAB = 100 mM) in chloroform-d was determined (Figure 3a). The water loading at x ≤ 4 exhibited a linear rise with a slope which equals to unity (w = x), and then stayed constant at x > 4. In a freshly prepared or vigorously agitated solution with x > 4, an 1H NMR signal corresponding to bulklike water (δ ∼ 4.7 ppm)35 was also observed, although it disappeared within a few minutes. When the samples were left to equilibrate overnight, a complete phase separation was observable by bare eyes as droplets of water on top of the chloroform solution.

Materials. Cetyltrimethylammonium bromide (CTAB), cetyltrimethylammonium chloride (CTAC), and chloroform-d of the highest purity available were used as purchased. Chloroform-d was stored in amber bottles over flame-dried molecular sieves (3 Å). Chloroform (99.9%) stabilized with max 55 ppm of amylene was used in the conductometry measurements as purchased. 1 H NMR and Diffusion Coefficient Measurements. The chloroform-d solutions were prepared by direct weighting of the amphiphile into an NMR tube. Water was subsequently added to the solution, and the mixture was vigorously agitated and then kept overnight in dark to equilibrate. 1H NMR spectra were measured at 303.15 K on a 300 MHz spectrometer. 2D-DOSY NMR experiments were carried out on selected samples employing a QNP probe-head equipped with the z-gradients set to 7 (Figures 3b and S4). Analogous δobs (H2O) versus w dependencies obtained at different CTAB concentrations (Figure 4a) provided a relation between the slopes of the linear fits from Figure 4a and cCTAB (Figure 4b) with the inflection point of the sigmoidal curve at cCTAB ∼ 38 mM. Diffusion Coefficients Measured by DOSY NMR Spectroscopy. The obtained diffusion coefficients and the corresponding hydrodynamic radii (Rh) are shown in Table 2. The observed 2D-DOSY peak maxima provided the diffusion coefficients attributed to both the water and CTAB signals (see Figures S5 and S6 in SI). The hydrodynamic radii corresponding to the DOSY peak maxima were smaller than fully stretched surfactant molecular chains (∼2.2 nm36,37) in the case of 8 mM CTAB solutions. At 40 mM CTAB concentration, the aggregate sizes varied from that corresponding to a single surfactant to about 110 nm.

Figure 3. (a) Dependence of w on x (●) for equilibrated CTAB (cCTAB = 100 mM) solutions in chloroform-d. (b) The dependence of δobs (H2O) on w for 100 mM solutions of CTAB (●, black) and CTAC (▲, red) in chloroform-d. 15187

dx.doi.org/10.1021/la303245e | Langmuir 2012, 28, 15185−15192

Langmuir

Article

Figure 4. (a) Dependence of δobs (H2O) on w for equilibrated (full symbols) and agitated (empty symbols) solutions for different cCTAB in chloroform-d. (b) The sigmoidal dependence of the slopes obtained from the linear fits in part a on cCTAB.

Table 2. 2D-DOSY NMR Investigations a

x

δ/2/ms

b

cCTAB/mM = 40 5.1 5

gradient stepsc 32

8

128

6

64

cCTAB/mM = 8 10.4 1 0.9 1

64 64

2

−1d

× × × × × × × ×

10−11 10−11 10−11 10−10 10−11 10−11 10−11 10−11

D/m s 5.62 3.98 2.00 1.58 3.16 2.00 5.01 2.51

1.58 × 10−9 2.00 × 10−9 6.31 × 10−10

Rh/nm

Rh/nm (lower limite)

Rh/nm (upper limite)

7.74 10.94 21.82 2.75 13.77 21.82 8.69 17.33

4.89 3.46 6.90 1.54 6.90 8.69 2.18 2.75

12.27 34.58 69.00 4.89 27.47 54.81 34.58 109.4

0.27 0.22 0.69

0.07 0.09 0.35

1.09 0.55 1.38

e

a

Water-to-surfactant molar ratio. bGradient pulse length (δ/2). Number of steps in which the gradient changed during the experiments. dDiffusion coefficients. eHydrodynamic radii (the lower and upper limits are shown and corresponds to 80% confidence interval) were calculated using eq 1. The diffusion delay Δ was set to be constant (50 ms) in all experiments. c

Figure 6. Enthalpy change as a function of the CTAB concentration for titration of chloroform-d to the cCTAB = 300 mM solution in chloroform-d at x = 3 and 298 K. The data are fitted with a double Boltzmann function (red line).

Figure 5. Dependence of κ on cCTAB for (a) x = 0 and (b) x = 3. (c) The dependence of κ on x at cCTAB = 30 (▲, black) and 70 (●, red) mM. 15188

dx.doi.org/10.1021/la303245e | Langmuir 2012, 28, 15185−15192

Langmuir

Article

titration curves are commonly of a sigmoidal character.6,40 The cmc has already been obtained by employing the sigmoidal Boltzmann equation.39

slopes level off due to the formation of reverse micelles, which are able to solubilize larger amounts of water inside their cores compared with the solubility of water in chloroform. Table 1 shows that the evaluated cmc values (the data from a δobs (H2O) vs (1/cCTAB) plot) are rather independent of x in our system. This is in contrast to the study of Nakashima and Fujiwara,8 where the cmc decreased from 11 to ∼2 mM when x increased from 1 to 8. However, the (5,10,15,20-tetraphenylporphyrinato)-zinc(II) probe used in the Nakashima and Fujiwara study forms a complex with the surfactant counterion either in bulk organic or interfacial phases,8 and thus it does not reside in the water pool. It is possible that the concentration dependence of its absorbance would refer only to the formation of premicellar aggregates and not to reverse micelles. This may explain the fact that the cmc measured by us was about 8-fold higher than the cmc (5.12 mM) determined by Nakashima and Fujiwara. Verbeeck and co-workers, using a combination of UV absorption and fluorescence decay parameters of the ionic and neutral probes, showed that the cmc values of reverse micelle formation for ionic surfactants were not influenced by water content, whereas stimulated aggregation of monomers by water molecules was observed at low surfactant concentrations.10 This can be explained by solvation of surfactant counterions. In nonaqueous solvents, surfactant molecules can form much smaller assemblies than those observed in water. In some cases, their formation obeys a multiple equilibrium model of association with continuous growth of the aggregates. The multiple equilibria can also be observed in systems with pronounced cmc transition, and the preaggregation can precede the formation of spherical reverse micelles (Eicke’s association model31). We observed minor deviations from linearity in the dependence of δobs (H2O) on (1/cCTAB) close to the intersection point (Figure 2a). This may indicate that the mass action model is only an approximation of the real behavior of our system. We suggest that hydrated premicellar aggregates exist below the cmc in equilibrated solutions, and they further stepwise aggregate in the process of reverse micelle formation. As a result, the intersection point in the plot of w versus cCTAB at x = 3 (Figure 2b) can be attributed to the CTAB concentration (cCTAB = (5.6 ± 0.9) mM) (and (8.1 ± 2.3) mM for x = 1; Table 1), at which monomeric hydrated surfactants start to form higher aggregates due to mutual interactions of the trimethylammonium groups. Interestingly, this cCTAB value is close to the cmc reported by Nakashima and Fujiwara for the water/CTAB/(chloroform/cyclohexane) system described above.8 Overlap of the signals of water and the surfactant protons at higher CTAB concentrations leads to scatter of the data (Figure 2b). At x = 5, a continuous dependence of w on cCTAB without any apparent intersection point of two linear dependencies was observed (Figure S2b). This is probably related to the phase separation (maximum attainable w). A linear trend with steep rise up to a phase separation limit at x ∼ 4 is well demonstrated in a water loading w versus x plot (equilibrated 100 mM CTAB samples; Figure 3a and S3). The threshold of formation of a water-in-oil microemulsion is known to be x = 5.15 Since the phase separation starts at lower x values in our systems, any aggregates formed above the cmc must be reverse micelles. A microemulsion was observed in different solvent mixtures [CDCl3/n-heptane (7:3),15 and CHCl3/isooctane (2:1)]14 and in dichloromethane, where the phase separation starts at x = 8.12

4. DISCUSSION 1 H NMR Investigations. 1H NMR spectra provide interesting information on the character of water present in nonaqueous cationic micellar solutions (Figure 1). The chemical shift δobs (H2O) = 1.56 ppm observed for cCTAB = 8 mM is a typical value for water dissolved in chloroform-d35 (water solubility in chloroform is ∼80.0 mM at 298 K and increases with temperature41). A downfield shift of δobs (H2O) at higher CTAB concentrations is known to be caused by formation of the hydrogen bonds42 indicating growing water pools inside the assembly (Figures 2a, 3b, 4a). δobs (H2O) = ∼ 4.7 ppm at 24 ± 1 °C (HDO in D2O signal)35 is the upper limit and a typical value for bulk water. During our experiments, the signal corresponding to bulk water confirmed phase separation in freshly agitated samples. A single 1H NMR signal of water in samples equilibrated overnight corresponds to an average resonance indicating the proton exchange rate between water molecules located in the micellar cores be shorter than 10−4 s.43 Indeed, the exchange rate was shown to be on the order of nanoseconds in AOT reverse micelles.44 On the other hand, the chemical shifts of water protons associated with the surfactant monomers can also be obtained with a relatively good accuracy directly from the measurements at cCTAB ≪ cmc (Figure 2). At high dilution, they eventually approach δobs (H2O) = 1.56 ppm, which corresponds to water dissolved in chloroform.35 Indeed, the change of δobs (H2O) was negligible at low CTAB concentrations (cCTAB = 8 mM; Figure 4a). It means that aggregates do not change their character with increasing water loading. No apparent change of the 1H NMR signals of the surfactant molecules was observed in samples with cCTAB = 1−200 mM. It is presumably related to the fact that the nonpolar hydrocarbon chains from a micelle exterior have very similar nonpolar environment (chloroform molecules) in both monomeric and aggregated states. More pronounced concentration-dependent changes (Δδobs ≤ 0.1 ppm) were observed for the N(CH3)3 group, attributed to incomplete hydration of the group in monomers compared to the state when it becomes a part of the micellar water core. In contrast, changes of the water signal shift approached 1.7 ppm. A plot of the chemical shifts of water, δobs (H2O), as a function of the total surfactant concentration (cCTAB) was used to determine the cmc via the mass action (close association) model.25−29 According to this model, the observed chemical shift δobs (H2O) at the given cCTAB is represented by eq 1. δobs(H 2O) = δ(H 2Omicellar ) + (cmc/cCTAB)[δ(H 2Omonomer ) − δ(H 2Omicellar )]

(1)

A plot of δobs (H2O) versus 1/cCTAB shows that the data essentially follow two straight lines with different slopes; their intersection is related to 1/cmc. The cmc value determined from the dependence of δobs (H2O) on 1/cCTAB at x = 3 (Figure 2a) was (31 ± 4) mM. The δ values above the cmc (1/cCTAB → 0) correspond to micellar (core) water (H2Omicellar). A similar cmc value, i.e., 38 mM, was obtained from the dependence of the slopes on cCTAB (Figure 4b). The slopes of the dependences of δobs (H2O) on water loading w describe the extent of the growth of aggregates. At increased CTAB concentrations, the 15189

dx.doi.org/10.1021/la303245e | Langmuir 2012, 28, 15185−15192

Langmuir

Article

(above the cmc), the specific conductance decreases with x until the phase separation occurs. Micellization Energetics. To our knowledge, only a few microcalorimetric studies have been carried out to determine the cmc of reverse micelles formed in nonpolar solvents. AOT micellization in water-free nonpolar solvents showed a single transition assigned to the cmc which was below cAOT = 1 mM.6 In this work, ITC was performed on CTAB/water/ chloroform-d solutions at x = 3 to validate the cmc value obtained from the preceding analyses. Figure 6 shows two transitions on the ITC titration curve. The transition at cCTAB = (41.5 ± 0.1) mM, accompanied with a high enthalpy change, is in accord with that of NMR study and conductometry measurements (Table 3), and it is presumably closely related

In CTAC solutions, no phase separation was observed (Figures 3b and S4). Continuing solubilization of water at high w (>10) indicates formation of a water/oil microemulsion or gel phase, in which water exhibits more bulklike behavior. For CTAC microemulsion, percolation was observed with an increasing extent of the short-lived connections between water droplets upon increase of w in aromatic solvents.11 A slope change in the δobs (H2O) versus w dependencies was also observed for CTAC solutions with lower water loadings in the rising part of the sigmoidal fits below a percolation limit (Figure S4). As a result, the mechanism of CTAC reverse micelle formation is expected to be the similar to that of CTAB. Diffusion Coefficients. To evidence the existence of CTAB assemblies in the solution, DOSY NMR spectra for samples with cCTAB = 8 and 40 mM at different x were obtained (Figures S5 and S6). The particle sizes were then estimated from the diffusion coefficients derived from the hydrodynamic radii (Rh). The DOSY peak maxima for 8 mM CTAB solutions at various x had Rh < 0.7 nm, indicating that no micelles are present in the solution. The corresponding 1.4-nm diameter is smaller than the length of the CTAB molecule (∼ 2.2 nm for the fully stretched hydrocarbon chain), suggesting that the species are in the form of packed amphiphile molecules. In equilibrated 40 mM CTAB solutions at x = 5.1, several peaks were observed in all of the measured DOSY spectra. Their Rh values, varying from 1.54 to 109.4 nm, correspond to the sizes of anticipated premicellar as well as reverse micelle assemblies. The radii corresponding to the DOSY peak maxima were found in the range from 2.8 to 21.8 nm, and are comparable to the size of reverse micelles of 100 mM CTAB solutions in CHCl3/isooctane (2:1) at x = 5 (7 nm) determined by fluorescence quenching experiments using the monodispersed radii approximation of a water pool.14 The estimated parameters from the fluorescence decay data for quenching in polydisperse micelle systems are highly dependent on the micelle size distribution.45 Our samples exhibited a high degree of polydispersity (Figure S5). The distribution of aggregates of different sizes in the sample indicated that multiple equilibria are required to correctly describe our system. This observation is consistent with Eicke’s model. Conductometry. We utilized the specific conductivity as a function of the total surfactant concentration (cCTAB) at x = ∼0−3 to estimate the cmc according to the mass action law.30,38 Below the cmc, the concentration of dissociated ions (thus the electrical conductivity) is proportional to the square root of the concentration of undissociated surfactant.46 Above the cmc, the specific conductivity is induced by charge disproportionation of reverse micelles.47−49 The breaking point of the curves in the conductivity dependencies (Figures 5a, b and S7) is associated with the cmc transitions. The cmc values (∼54 mM, Table 1) were comparable for x equal to 0 (Figure 5a), 1 (Figure S7), as well as 3 (Figure 5b), which is in agreement with our 1H NMR experiments (Table 1). With more water in the system, the size of micelles increases, causing their lower diffusion rates. A smaller number of reverse micelles in relation with lower diffusion rates are then responsible for a substantial drop in the conductivity. The influence of water addition was evaluated at a constant surfactant concentration below and above the cmc value. Below the cmc, the specific conductivity does not change upon water addition, because the fraction of dissociated surfactant molecules (which can form premicellar aggregates) remains constant (Figure 5c, cCTAB = 30 mM). At cCTAB = 70 mM

Table 3. Comparison of the CTAB Initial Aggregation Concentrations with the Critical Micelle Concentration Values Measured by Various Techniques aggregation event

techniques

premicellar aggregation

1

cmc

1

H NMR

ITC H NMR

1

H NMR

ITC conductometry

experimental methods

cCTAB/mM

(w vs cCTAB) plot (Figure 2b) ITC titration (Figure 6) (δobs (H2O) vs (1/cCTAB)) plot (Figure 2a) (slope [(δobs (H2O) vs w] vs cCTAB) plot (Figure 4b) ITC titration (Figure 6) (κ vs cCTAB) plot (Figures 5a, b and S7)

7.9 ± 0.7 5.6 ± 0.9 31 ± 4 38 ± 3 41.5 ± 0.1 53 ± 10

to the formation of reverse micelles. In this process, TΔSmic° > ΔHmic° (where ΔHmic° = (−9608 ± 41) cal mol−1 and ΔSmic° = 48.0 cal mol−1 K−1), which suggests that the reverse micelle formation is an entropy-driven process, as was also found for AOT self-association in nonpolar solvents.6 We suggest that a positive entropy change is associated with the removal of chloroform molecules from the vicinity of the surfactant head groups during formation of reverse micelles. The best agreement was found between the cmc obtained by ITC and that determined from Figure 4b (cCTAB = (38 ± 3) mM). Interestingly, the onset of this transition concentration coincides with the cmc determined by NMR, whereas the concentration at its end corresponds to the cmc value obtained by conductometry. The second transition on the ITC curve at cCTAB = (7.9 ± 0.7) mM is accompanied by a low heat change and can be related to a formation of hydrated premicellar aggregates below the cmc as observed by NMR. This corresponds to Eicke’s model, according to which initially linear premicellar aggregates undergo association to form reverse micelles within a narrow amphiphile concentration range.31 A growing water pool in the aggregates with increasing CTAB concentration (Figure 4) indicates that the aggregation and subsequent micellization is a continuous process. The change of the slopes of the dependencies in Figure 4a obviously indicates that the premicellar aggregates and reverse micelles exist in the solution at variable molar ratios as described by Eicke’s model. The range of the cCTAB values at which reverse micelles are formed (the apparent cmc), indicated by the steepest change of the sigmoidal dependence of the slopes of the (δobs (H2O) vs w) dependencies in Figure 4b, was in the range 25−55 mM. At 15190

dx.doi.org/10.1021/la303245e | Langmuir 2012, 28, 15185−15192

Langmuir

■

cCTAB > 60 mM, reverse micelles become the only species present in the solution, and thus the slope values leveled off. The range of the apparent cmc covers the cmc values obtained by the other techniques. 1H NMR determination of the cmc (a concentration dependence of water chemical shifts at constant x) reflects the lower limit of Eicke’s apparent cmc, i.e., the beginning of the reverse micelle formation. In contrast, the value obtained by conductometry measurements corresponds to the concentration, at which reverse micelles prevail, indicating the upper apparent cmc limit. The ITC transition provided an average value of Eicke’s apparent cmc (half of the aggregates is in the form of reverse micelles). Table 3 summarizes the CTAB aggregation concentrations at which monomers start to form premicellar aggregates and the cmc values measured by different experimental techniques. Micellar Sizes. Polydisperse species studied by DOSY NMR exhibited a broad distribution of the hydrodynamic radii in 40 mM CTAB solutions at x = 5.1 (5−100 nm; Table 2). In contrast, sizes smaller than those of the stretched surfactant molecules (∼2 nm) were estimated from the diffusion coefficients measured for 8 mM CTAB solutions. This obviously reflects the relative insensitivity of NMR to the onset of the premicellar aggregation, when small aggregates containing only few surfactant molecules are formed (Figure S12). On the other hand, the plot of w versus cCTAB (Figure 2b) obtained from 1H NMR measurements provides information on CTAB concentrations, where the surfactant molecules start to aggregate into clusters. Relative amounts of water molecules associated with the monomers and premicellar aggregates (Figures S11 and S12) are obviously dependent on experimental conditions.

AUTHOR INFORMATION

Corresponding Author

*Phone: +420-54949-4856. Fax: +420-54949-2443. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The project was supported by the Grant Agency of the Czech Republic (P503/10/0947) and the project CETOCOEN (CZ.1.05/2.1.00/01.0001) granted by the European Regional Development Fund. The authors express their thanks to Professor Hans-Friedrich Eicke and Professor Jakob Wirz for fruitful discussions.

■

REFERENCES

(1) Dhar, S.; Rana, D. K.; Sarkar, A.; Mandal, T. K.; Ghosh, S.; Bhattacharya, S. C. Physicochemical characterization of reverse micelles of Triton X using 1-anthracene sulphonate as fluorescent probe: A spectroscopic study. Colloid Surf., A 2009, 349, 117−124. (2) Desando, M. A.; Lahajnar, G.; Sepe, A. Proton magnetic relaxation and the aggregation of n-octylammonium n-octadecanoate surfactant in deuterochloroform solution. J. Colloid Interface Sci. 2010, 345, 338−345. (3) Tran, C. D.; Yu, S. F. Near-infrared spectroscopic method for the sensitive and direct determination of aggregations of surfactants in various media. J. Colloid Interface Sci. 2005, 283, 613−618. (4) Anand, U.; Jash, C.; Mukherjee, S. Spectroscopic determination of critical micelle concentration in aqueous and non-aqueous media using a non-invasive method. J. Colloid Interface Sci. 2011, 364, 400− 406. (5) Huang, C. C.; Hohn, K. L. Tetrakis(dimethylamino)ethylene chemiluminescence (TDE CL) characterization of the CMC and the viscosity of reversed microemulsions. J. Chem. Phys. B 2010, 114, 2685−2694. (6) Majhi, P. R.; Moulik, S. P. Microcalorimetric investigation of AOT self-association in oil and the state of pool water in water/oil microemulsions. J. Chem. Phys. B 1999, 103, 5977−5983. (7) Ghosh, S. K.; Khatua, P. K.; Bhattacharya, S. C. Physicochemical characteristics of reverse micelles of polyoxyethylene nonyl phenol in different organic solvents. J. Colloid Interface Sci. 2004, 279, 523−532. (8) Nakashima, T.; Fujiwara, T. Effects of surfactant counter-ions and added salts on reverse micelle formation of cetyltrimethylammonium surfactant studied by using (5,10,15,20-tetraphenylporphyrinato)zinc(II) as a probe. Anal. Sci. 2001, 17, 1241−1244. (9) Eicke, H.-F. Surfactants in nonpolar solvents. Aggregation and micellization. Top. Curr. Chem. 1980, 87, 85−145. (10) Verbeeck, A.; Geladé, E.; De Schryver, F. C. Aggregation behavior in inverse micellar systemsSpectroscopic evidence for a unified model. Langmuir 1986, 2, 448−456. (11) Zana, R.; Lang, J.; Canet, D. Ternary water-in-oil microemulsions made of cationic surfactants, water, and aromatic solvents. 3. Self-diffusion studies in relation to exchange of material between droplets and percolation. J. Phys. Chem. 1991, 95, 3364−3367. (12) Dokter, A. M.; Woutersen, S.; Bakker, H. J. Ultrafast dynamics of water in cationic micelles. J. Chem. Phys. 2007, 126, 124507. (13) Mukherjee, K.; Mukherjee, D. C.; Moulik, S. P. Thermodynamics of microemulsion formation. 3. Enthalpies of solution of water in chloroform as well as chloroform in water aided by cationic, anionic, and nonionic surfactants. J. Colloid Interface Sci. 1997, 187, 327−333. (14) Lang, J.; Mascolo, G.; Zana, R.; Luisi, P. L. Structure and dynamics of cetyltrimethylammonium bromide water-in-oil microemulsions. J. Phys. Chem. 1990, 94, 3069−3074. (15) El Seoud, O. A.; Pires, P. A. R. FTIR and 1H NMR studies on the structure of water solubilized by reverse aggregates of dodecyltrimethylammonium bromide; didodecyldimethylammonium

5. CONCLUSIONS We have investigated the micellization processes in a water/ CTAB/chloroform-d system with different water to surfactant molar ratios using NMR, conductometry, and ITC techniques. The NMR experiments provided evidence that aggregation/ micellization is a continuous process. Premicellar aggregates were observed and found to be in equilibrium with reverse micelles. Above the cmc, the incidence of CTAB reverse micelles is proposed. The ITC experiments revealed two pronounced enthalpy changes. The first transition at cCTAB ∼ 8 mM is attributed to the formation of hydrated premicellar aggregates; the second transition at cCTAB ∼ 40 mM is attributed to the apparent cmc. The NMR and conductometry methods provided apparent cmc values equal to 31 and 53 mM, respectively. We hypothesize that NMR analysis indicates the beginning of the reverse micelle formation, whereas conductometry measurements provide the upper limit of the cmc. As a result, Eicke’s association model is suggested to be useful for the description of CTAB micellization in chloroform in the presence of water. Finally, the size of the nanoparticles was estimated using DOSY NMR spectroscopy. Their polydispersity in size and shape is also in agreement with the model proposed by Eicke. A similar mechanism is suggested for the micellization in water/CTAC/chloroform-d system.

■

Article

ASSOCIATED CONTENT

* Supporting Information S

δobs (H2O) and w dependencies; DOSY NMR spectra; conductometry data; schematic illustrations of the micellization process. This material is available free of charge via the Internet at http://pubs.acs.org. 15191

dx.doi.org/10.1021/la303245e | Langmuir 2012, 28, 15185−15192

Langmuir

Article

bromide, and their mixtures in organic solvents. Prog. Colloid Polym. Sci. 2008, 134, 101−110. (16) Nandi, S.; Bhattacharya, S. C.; Moulik, S. P. Effects of additives (NaCl, urea, glucose, guanidine hydrochloride) on the physicochemical properties of reverse micelles of Tweens in chloroform. Indian J. Chem., Sect. A 2000, 39, 589−597. (17) Hasegawa, M.; Yamasaki, Y.; Sonta, N.; Shindo, Y.; Sugimura, T.; Kitahara, A. Clustering of AOT reversed micelles as studied by nonradiative energy transfer of solubilized probes. J. Phys. Chem. 1996, 100, 15575−15580. (18) Olesik, S. V.; Miller, C. J. Critical micelle concentration of AOT in supercritical alkanes. Langmuir 1990, 6, 183−187. (19) Jean, Y. C.; Ache, H. J. Study of micelle formation and effect of additives on this process in reversed micellar systems by positronannihilation techniques. J. Am. Chem. Soc. 1978, 100, 6320−6327. (20) Baruah, B.; Swafford, L. A.; Crans, D. C.; Levinger, N. E. Do probe molecules influence water in confinement? J. Chem. Phys. B 2008, 112, 10158−10164. (21) Crans, D. C.; Rithner, C. D.; Baruah, B.; Gourley, B. L.; Levinger, N. E. Molecular probe location in reverse micelles determined by NMR dipolar interactions. J. Am. Chem. Soc. 2006, 128, 4437−4445. (22) Blandamer, M. J.; Cullis, P. M.; Soldi, L. G.; Engberts, J.; Kacperska, A.; Vanos, N. M.; Subha, M. C. S. Thermodynamics of micellar systems: Comparison of mass action and phase equilibrium models for the calculation of standard Gibbs energies of micelle formation. Adv. Colloid Interface Sci. 1995, 58, 171−209. (23) Nyrkova, I. A.; Semenov, A. N. Multimerization: Closed or open association scenario? Eur. Phys. J. E: Soft Matter Biol. Phys. 2005, 17, 327−337. (24) Eicke, H. F.; Christen, H. Nucleation process of micelle formation in apolar solvents. J. Colloid Interface Sci. 1974, 48, 281−290. (25) Muller, N. Multiple-equilibrium model for the micellization of ionic surfactants in nonaqueous solvents. J. Phys. Chem. 1975, 79, 287−291. (26) Emin, S. M.; Denkova, P. S.; Papazova, K. I.; Dushkin, C. D.; Adachi, E. Study of reverse micelles of di-isobutylphenoxyethoxyethyldimethylbenzylammonium methacrylate in benzene by nuclear magnetic resonance spectroscopy. J. Colloid Interface Sci. 2007, 305, 133−141. (27) Faure, A.; Tistchenko, A. M.; Zemb, T.; Chachaty, C. Aggregation and dynamical behavior in sodium diethylhexyl phosphate/water/benzene inverted micelles. J. Phys. Chem. 1985, 89, 3373−3378. (28) Luchetti, L.; Mancini, G. NMR investigation on the various aggregates formed by a gemini chiral surfactant. Langmuir 2000, 16, 161−165. (29) Yan, J. F.; Palmer, M. B. A nuclear magnetic resonance method for determination of critical micelle concentration. J. Colloid Interface Sci. 1969, 30, 177−182. (30) Guo, Q.; Singh, V.; Behrens, S. H. Electric charging in nonpolar liquids because of nonionizable surfactants. Langmuir 2010, 26, 3203− 3207. (31) Eicke, H. F.; Hopmann, R. F. W.; Christen, H. Kinetics of conformational change during micelle formation in apolar media. Ber. Bunsen-Ges. 1975, 79, 667−673. (32) Verbeeck, A.; Voortmans, G.; Jackers, C.; De Schryver, F. C. Characterization and stabilization of inverse micelles. Langmuir 1989, 5, 766−776. (33) Djermouni, B.; Ache, H. J. Effect of temperature and counterion on the micelle formation process studied by positron-annihilation techniques. J. Phys. Chem. 1979, 83, 2476−2479. (34) Fucugauchi, L. A.; Djermouni, B.; Handel, E. D.; Ache, H. J. Study of micelle formation in solutions of alkylammonium carboxylates in apolar solvents by positron-annihilation techniques. J. Am. Chem. Soc. 1979, 101, 2841−2844. (35) Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. NMR chemical shifts of common laboratory solvents as trace impurities. J. Org. Chem. 1997, 62, 7512−7515.

(36) Warr, G. G.; Sen, R.; Evans, D. F.; Trend, J. E. Microemulsion formation and phase behavior of dialkydimethylammonium bromide surfactants. J. Phys. Chem. 1988, 92, 774−783. (37) Weidemaier, K.; Tavernier, H. L.; Fayer, M. D. Photoinduced electron transfer on the surfaces of micelles. J. Chem. Phys. B 1997, 101, 9352−9361. (38) Khan, A. M.; Shah, S. S. Determination of critical micelle concentration (Cmc) of sodium dodecyl sulfate (SDS) and the effect of low concentration of pyrene on its Cmc using Origin software. J. Chem. Soc. Pak. 2008, 30, 186−191. (39) Hait, S. K.; Moulik, S. P.; Palepu, R. Refined method of assessment of parameters of micellization of surfactants and percolation of W/O microemulsions. Langmuir 2002, 18, 2471−2476. (40) Tariq, M.; Podgorsek, A.; Ferguson, J. L.; Lopes, A.; Gomes, M. F. C.; Padua, A. A. H.; Rebelo, L. P. N.; Lopes, J. N. C. Characteristics of aggregation in aqueous solutions of dialkylpyrrolidinium bromides. J. Colloid Interface Sci. 2011, 360, 606−616. (41) Hu, H. S. Determination of vapour-liquid and vapour-liquidliquid equilibrium of the chloroform-water and trichloroethylene-water binary mixtures. Fluid Phase Equilib. 2010, 289, 80−89. (42) Becker, E. D. High Resolution NMR; Academic Press, Inc.: New York, 1972. (43) Eisenberg, D.; Kaufman, W. The Structure and Properties of Water; Oxford University Press: London, 1969. (44) Zhang, J.; Bright, F. V. Nanosecond reorganization of water within the interior of reversed micelles revealed by frequency-domain fluorescence spectroscopy. J. Phys. Chem. 1991, 95, 7900−7907. (45) Almgren, M.; Lofroth, J.-E. Effects of polydispersity on fluorescence quenching in micelles. J. Chem. Phys. 1982, 76, 2734− 2743. (46) Sainis, S. K.; Merrill, J. W.; Dufresne, E. R. Electrostatic interactions of colloidal particles at vanishing ionic strength. Langmuir 2008, 24, 13334−13337. (47) Hsu, M. F.; Dufresne, E. R.; Weitz, D. A. Charge stabilization in nonpolar solvents. Langmuir 2005, 21, 4881−4887. (48) Roberts, G. S.; Sanchez, R.; Kemp, R.; Wood, T.; Bartlett, P. Electrostatic charging of nonpolar colloids by reverse micelles. Langmuir 2008, 24, 6530−6541. (49) Eicke, H. F.; Borkovec, M.; Dasgupta, B. Conductivity of waterin-oil microemulsions: A quantitative charge fluctuation model. J. Phys. Chem. 1989, 93, 314−317.

15192

dx.doi.org/10.1021/la303245e | Langmuir 2012, 28, 15185−15192