Molecular Variations in Aromatic Cosolutes: Critical Role in theRheology of Cationic Wormlike MicellesThiago H. Ito,† Paulo C. M. L. Miranda,‡ Nelson H. Morgon,† Gabriel Heerdt,† Cecile A. Dreiss,§

and Edvaldo Sabadini*,†

†Department of Physical-Chemistry, Institute of Chemistry and ‡Department of Organic-Chemistry, Institute of Chemistry,University of CampinasUNICAMP P.O. Box 6154, 13084-862, Campinas, SP Brazil§Institute of Pharmaceutical Science, King’s College London, 150 Stamford Street, SE1 9NH London, U.K.

*S Supporting Information

ABSTRACT: Wormlike micelles formed by the addition to cetyl-trimethylammonium bromide (CTAB) of a range of aromatic cosoluteswith small molecular variations in their structure were systematicallystudied. Phenol and derivatives of benzoate and cinnamate were used,and the resulting mixtures were studied by oscillatory, steady-shearrheology, and the microstructure was probed by small-angle neutronscattering. The lengthening of the micelles and their entanglementresult in remarkable viscoelastic properties, making rheology a usefultool to assess the effect of structural variations of the cosolutes onwormlike micelle formation. For a fixed concentration of CTAB andcosolute (200 mmol L−1), the relaxation time decreases in the followingorder: phenol > cinnamate> o-hydroxycinnamate > salicylate > o-methoxycinnamate > benzoate > o-methoxybenzoate. The variations inviscoelastic response are rationalized by using Mulliken population analysis to map out the electronic density of the cosolutes andquantify the barrier to rotation of specific groups on the aromatics. We find that the ability of the group attached to the aromaticring to rotate is crucial in determining the packing of the cosolute at the micellar interface and thus critically impacts the micellargrowth and, in turn, the rheological response. These results enable us for the first time to propose design rules for the self-assembly of the surfactants and cosolutes resulting in the formation of wormlike micelles with the cationic surfactant CTAB.

■ INTRODUCTION

Wormlike micelles (WLMs) are one of several possible self-assembled morphologies adopted by surfactants; their formationcan be rationalized by the critical packing parameter (cpp),1

which considers geometrical features of the surfactants todetermine the optimal shape of the aggregate into which they canpack. The elongation of aggregates into flexible cylindricalmicelles can reach contour lengths on the order of micro-meters2,3 Their remarkable, tunable viscoelastic characteristicshas led to their exploitation from personal care products to oilexploration fields.3−7At low concentration, these giant structurescan be used as a mechanical and nondegradable hydrodynamicdrag-reducing agent.5,8−10 One of the most widely studied WLMsystem is based on the combination of the cationic surfactantcetyltrimethylammonium bromide (CTAB) with the aromaticcosolute o-sodium salicylate.3,11 The unidimensional elongationof the micelles can be obtained at high ionic strength withcommon ions, but at low concentration, the process is stronglyfavored by using aromatic ions such as salicylate, which isclassified as a strongly binding counterion. This is because,besides the charge neutralization of the headgroups, which bringsthem closer together, the insertion of the aromatic anions intothe palisade layer, driven by hydrophobicity, reduces the cpp andconsequently increases micellar curvature.12 However, this

sphere-to-rod transition is highly sensitive to the substitutionpattern of the aromatic anion. For example, the self-assemblystructures formed by cationic surfactants are strongly affectedwhen different isomers of salicylate13 and chlorobenzoate14 areused. This has been explained by considering the microenviron-ment of the counterions at the interface: in the case of salicylate,the hydroxyl group at the para position leads to an unfavorableenvironment, while this is improved with the meta substitution,the counterion being tilted; therefore, the most favorableorientation at the aqueous interface is obtained with the o-salicylate.15

Another interesting packing scenario to mention is that of theresponsive CTAB micelles formed by the addition of o-methoxycinnamate (OMCA).7,11,16 The transition from trans-to cis-OMCA induced byUV light markedly affects the packing ofthe counterions at the micellar interface, leading to a shorteningof the micelles and remarkable variations in viscosity.11,16 Thissystem is an interesting example demonstrating the possibility ofswitching macroscopic properties via a simple trigger thatcontrols the molecular detail of the constitutive units.

Received: July 4, 2014Revised: August 29, 2014Published: September 15, 2014

Article

pubs.acs.org/Langmuir

© 2014 American Chemical Society 11535 dx.doi.org/10.1021/la502649j | Langmuir 2014, 30, 11535−11542

Clearly, subtle variations in the architecture of the aromaticmolecule affect the interaction between the surfactant and thearomatic salt and, thus, the packing of the amphiphiles within theaggregates and, as a result, the rheological properties. Therefore,rheology is a powerful technique to investigate the influence ofsystematic structural changes in the aromatic cosolutes on thepropensity to form wormlike micelles and their properties. Whilethe rheology of CTAB micelles with a range of counterions hasbeen extensively studied,3 the precise role of the molecularstructure of the aromatic cosolutes in dictating the packing at theinterface is not well-understood. The objective of this work istherefore to rationalize the impact of the architecture of thearomatic cosolutes on the surfactant assemblage into wormlikemicelles or otherwise and therefore their rheological response.More specifically, the following parameters are examined inbenzoate and cinnamate derivatives and phenol: the distancebetween the aromatic ring and the carboxyl group, thesubstitution or otherwise of the hydroxyl group by a methoxygroup or hydrogen, and the absence of charge in the aromaticcosolute. Their impact on micellar self-assembly is studied bylinear rheology and small-angle neutron scattering (SANS)measurements, and the results are rationalized by mapping outthe electronic density of the cosolutes and analyzing the barrierto rotation of specific groups on the aromatics using computa-tional chemistry.

■ EXPERIMENTAL SECTIONMaterials. The surfactant CTAB was obtained from Sigma-Aldrich.

The aromatic cosolutes (Scheme 1) were obtained from Sigma-Aldrich[OMCA, cinnamic acid, o-methoxybenzoic acid (OMBA), and 3-phenylpropanoic acid], Merck (sodium salicylate and phenol), andSynth (sodium benzoate). Sodium o-hydroxycinnamic acid (OHCA)was synthesized (as described in the Supporting Information). The saltsof the cosolutes were prepared from their acid form by neutralizing therespective THF solutions with NaOH, followed by the lyophilization ofthe solvent. Phenol was purified by dissolving the samples in toluene andmaintaining contact with CaSO4 for 8 h. The process was repeated threetimes and the phenol was then recrystallized at low temperature.17

Solutions of WLMs were prepared using ultrapure water (18 MΩ cm)and then heated at 75 °C for 1 h before being cooled gradually to roomtemperature.Rheological Measurements. Oscillatory experiments were

conducted on a Haake RS1 rheometer equipped with a water bathand plate−plate sensor (35 mm diameter and 1 mm gap). The stressapplied was 3 Pa (which was chosen well within the linear viscoelasticrange). The temperature was maintained at 25 °C. In order to minimizeevaporation, the measurements were conducted using a solvent trap.Small-Angle Neutron Scattering Measurements. SANS experi-

ments were carried out on LOQ at ISIS (Rutherford AppletonLaboratory, Didcot, UK). The instrument uses incident wavelengthsfrom 2.2 to 10 Å, sorted by time-of-flight, with a fixed sample-to-detectordistance of 4.1 m, providing a q-range between ∼0.007 and 0.28 Å−1.Samples holders were 1 or 2 mm thick quartz cuvettes. All the systemscontained an equimolar concentration (100mmol L−1) of surfactant andcosolute in D2O. All scattering data were first normalized for sampletransmission and then background-corrected using the pure solvent(D2O). The data were then converted to the differential scattering crosssections using the standard procedures at ISIS.18

SANS data were fitted with the SasView software,19 using the cylindermodel. This model considers WLM as composed of a sequence of rigidcylinders (Figure 1), with cross-section radius r and length lp, which forlong structures whose contour length falls outside the detection limit ofneutrons can be assimilated to a persistence length (Figure 1). Thescattering length density of the wormlike micelles was calculated as avolume-weighted average, assuming an equimolar mixing of the CTABmolecules and cosolutes.

■ RESULTS AND DISCUSSIONThe vials in Figure 2 contain equimolar concentrations of CTABand aromatic cosolutes (200mmol L−1). The pale yellow color ofthe solutions formed with OHCA and OMCA is due to the weaklight absorption of these two organic cosolutes in the blue regionof the electromagnetic spectrum. The vials were inverted at thesame time to show the variations in retardation of flow under

Scheme 1. Molecular Structure of the Various AromaticCosolutes Studied, Together with Values of the RelaxationTime and Plateau Elastic Moduli Obtained from the MaxwellModel for 200 mmol L−1 Equimolar Mixtures of CTAB andCosolutea

aThe arrows highlight the comparative features under study: (1) Thepresence of an o-hydroxyl group on the aromatic ring; (2) thepresence of additional carbons between the ring and carboxylategroup; (3) the substitution of the o-hydroxyl group by a methoxygroup; and (4) the presence or otherwise of a double-bond in thecinnamate derivative.

Figure 1. Schematic representation of wormlike micelles formed bysmaller rigid cylinders with persistence length lp and radius r as seen bysmall-angle neutron scattering measurements.

Langmuir Article

dx.doi.org/10.1021/la502649j | Langmuir 2014, 30, 11535−1154211536

gravitational force for the range of cosolutes studied. They arepositioned in a sequence that follows their flow behavior, frommore solidlike to more liquidlike (left to right): phenol >cinnamate > OHCA > salicylate > OMCA > benzoate > OMBA.Viscosity measurements (Supporting Information) performed

on a range of equimolar solutions (30−200 mmol L−1) confirmwide variations in the rheological response, which follow thetrend observed visually (Figure 2). Interestingly, we note that inthe phenol/CTAB mixtures that show the strongest solidlikebehavior (Figure 2), the viscosity drops very rapidly withdilution: the samples lose their viscoelastic characteristics atconcentrations below 100 mmol L−1. In this respect, phenoldeparts from the behavior of the other cosolutes, a point which isdiscussed later on.These results clearly demonstrate that small variations in

cosolute architecture lead to very different macroscopicresponses in mixtures with CTAB. A notable observation isthat a liquid solution is formed with OMBA (Figure 2), whichdiffers from salicylate by a methoxy group substituting the o-hydroxyl group or from OMCA by a closer proximity of thecarboxylate group to the ring (Figure 2, Scheme 1). At theopposite extreme, the strong gel-like characteristics obtainedwith phenol are very surprising considering its lack of electriccharge (the negative charge on all other cosolutes provides adrive for the insertion of CTA+ cations in the micelles).On the basis of these observations, the formation of wormlike

micelles, and thus the viscoelastic properties resulting from theirentanglement, is affected by the following features of thearomatics:

(I) The presence or otherwise of a negative charge.(II) The presence of a hydroxyl group or its methoxyl variant.(III) The presence of two additional carbons between the

aromatic ring and the carboxylate group.

Understanding how these features impact the molecularpacking of the cosolutes, and thus impact the rheologicalresponse, is the objective of this study.For salicylate/CTAB2,20,21 and OMCA/CTAB11,16 systems, it

is well-documented that the high viscoelasticity is associated withthe entanglement of the WLM chains. Although WLMs can becompared with polymeric system, their rheology is uniquebecause, in contrast to a covalent polymer, the assemblies formedby the surfactant and aromatic additive are constantly breakingand re-forming. Two excellent reviews on the structure andrheology of wormlike micelle solutions have been published byCates and Fielding22 and Berret.23

The dynamics of the WLM chains is described by acombination of reptation and breaking processes, which arecharacterized by two relaxation times, τr and τb, respectively.Cates and Candau demonstrated that the overall relaxation time,τR, can be obtained by eq 2,24 if

≪t tr b (1)

τ τ τ=R r b (2)

In this situation, WLMs behave as a Maxwellian fluid, and theelastic modulusG′ and the viscous modulusG″ are given by eqs 3and 4, respectively

Figure 2.Gravitational effect on gels containing 200mmol L−1 of CTAB and 200mmol·L−1 of (1) phenol, (2) cinnamate, (3) OHCA, (4) salicylate, (5)OMCA, (6) benzoate, and (7) OMBA. The values (on the left) correspond to the time of observation after inversion of the vials.

Langmuir Article

dx.doi.org/10.1021/la502649j | Langmuir 2014, 30, 11535−1154211537

ωωτ

ωτ′ =

+G G( )

( )1 ( )0

R2

R2

(3)ω

ωτωτ

″ =+

G G( )( )

1 ( )0R

R2

(4)

Figure 3.Dynamic moduli as a function of oscillation frequency, at 25 °C, for equimolar mixtures of CTAB and the different aromatic cosolutes studied(200 mmol L−1): (A) derivatives of benzoate, (B) derivatives of cinnamate, and (C) phenol. The lines correspond to fits to the Maxwell model.

Langmuir Article

dx.doi.org/10.1021/la502649j | Langmuir 2014, 30, 11535−1154211538

where G0 is the plateau modulus (at high frequency) and τR theinverse of the frequency at which G′ and G″ crossover.The rheological behavior of the systems formed from

equimolar solutions of CTAB and the aromatic cosolutes isshown in Figure 3. For ease of comparison, the graphs have beengrouped according to their structure: derivatives of benzoate (A),derivatives of cinnamate (B), and phenol (C).The Maxwell model appropriately describes the behavior of

the systems. The values of τR and G0 obtained from fitting therheological curves at 25 °C are given in Scheme 1. These valuescorrelate well with the observations of the vials’ inversion (Figure3), with τR decreasing in the following order: phenol > cinnamate> OHCA > salicylate > OMCA > benzoate > OMBA.The value of τR for OMBA is too short to be measured. Under

gravitational stress, the WLM chains lose their mechanicalenergy, mainly by reptation. In this case, the gravitational force isapplied continuously and is equivalent to a frequency tending tozero. If the frequency of the stress, ω, is increased, as during arheological experiment, some elastic energy is stored, sincereptation becomes more restricted. Therefore, G′ progressivelyincreases, crossing the G″ curve, and the behavior of both curvesare described by the Maxwell model (eqs 3 and 4). Our resultsdemonstrate that τR is very sensitive to the molecular structure ofthe aromatic cosolute. According to the reptation-reactionkinetics model of Cates and Candau,24 the reptation andbreaking times depend on the average contour length of themicelle (L), and the corresponding relaxation times are given bythe following scaling laws (see, for example, ref 23):

τ ∼ L cr3 3/2

(5)

τ ∼

cLb (6)

By combining with eq 2, τR ∼ L. Therefore, for a fixedconcentration of CTAB, τR can be directly correlated to theaverage contour length of the micellar systems containing CTABand the aromatics. In other words, τR can be used as a criterion todiscuss the interaction between CTAB and the cosolutes andtheir incorporation in the palisade layer, leading to the formationof longer WLMs.On the basis of the evolution of τR (in decreasing order),

cinnamate > OHCA> salicylate > OMCA > benzoate, somegoverning rules based on the cosolute structure can beestablished (Scheme 1). The case of phenol is treated separately.First, the presence (or not) of a hydroxyl group at the ortho

position (comparison 1 in Scheme 1) is considered. On the basisof the stronger viscoelastic response obtained for salicylatecompared to benzoate (Figures 2 and 4, Scheme 1), it can beinferred that the presence of the hydroxyl group at the ortho-position promotes a better positioning of the cosolute at themicellar interface, favoring electrostatic interaction between thecarboxyl and the cationic heads of the surfactants and, thus,driving the elongation of the wormlike micelles. This isconsistent with the interpretation described previously for o-salicylate.12 In aromatics bearing a hydroxyl group, the distancebetween the carboxyl and hydroxyl groups may be a key torationalize these results, for instance, when comparing salicylatewith OHCA (comparison 2, Scheme 1). The geometry of o-salicylates favors the formation of a hydrogen bond betweenthese two groups, avoiding the free rotation of the carboxylgroup, thus halting the aromatic anion in the planar form andfavoring its packing within the micelle palisade.25,26

Next, the effect of additional carbons between the ring and thecarboxyl group is considered for all cosolutes (Scheme 1,comparative feature 2). This can be assessed by comparing thearomatic cosolutes above each other in rows 1 and 2 of Scheme 1.For instance, cinnamate has two additional carbons between thecarboxyl and the aromatic ring compared to benzoate, and sodoes OHCA compared to salicylate, and OMCA compared toOMBA. In all cases, the two additional carbon atoms induce ahigher relaxation time. Thus, in mixtures with CTAB (see alsoFigure 2), they must favor WLM growth by a betterincorporation of the aromatic cosolute at the interface. Finally,if we inspect the effect of substituting the hydroxyl group by amethoxy group at the ortho position (comparison 3 of Scheme 1,salicylate vs OMBA, and OHCA vs OMCA), in both cases, thisresults in a drastic drop of the rheological properties; we note inparticular that WLMs formed with OMBA are too short incomparison those formed with the other cosolutes (the samplesremain liquidlike).In order to investigate the correlation between the carboxyl/

hydroxyl group distance and the mechanisms of wormlikemicelle formation, we have calculated the Mulliken electrostaticpotential distributions of the aromatic cosolutes (details of themethod are given in the Supporting Information). This method,which is based on the electronic density division in individualatomic charges, gives a quantitative description of molecularproperties, such as chemical reactivity and molecular inter-actions.27

Figure 4A shows that a hydrogen bond is formed between thecarboxylate and hydroxyl group in salicylate, while in OHCA thedistance between those two groups prevents the formation ofsuch a bond. In salicylate, therefore, the presence of thisintramolecular interaction hinders the rotation of the carboxyl

Figure 4.Mulliken electrostatic potential distribution for (A) salicylateand (B) OHCA. The hydrogen bond formed between the carboxyl andhydroxyl groups can be clearly observed in salicylate. (C) Barriers for therotation (considering the dihedral angle) of the carboxyl groups ofbenzoate and salicylate.

Langmuir Article

dx.doi.org/10.1021/la502649j | Langmuir 2014, 30, 11535−1154211539

group, thus ensuring a better packing of the salicylate in themicellar palisade. In benzoate, obviously, such a bond does notexist. Hence, if we refer back to comparative feature 1 in Scheme1, the higher viscoelasticity of salicylate compared to benzoatecan now be attributed to the existence of a hydrogen bond insalicylate between the carboxylate and hydroxyl groups. Inbenzoate, the steric hindrance caused by the rotation of thecarboxyl group at the compacted surface of the WLM is likely toreduce the stabilization of the micelle, while in salicylate theexistence of the hydrogen bond favors the incorporation of thesalicylate inside the micelles, compared to benzoate. As shown inFigure 4C, the barriers for the free rotation of the carboxylate(dihedral angle) are low for benzoate (≈25 kJ mol−1) comparedto salicylate (≈75 kJ mol−1). For OHCA, no hydrogen bond canbe formed due to the larger distance between the hydroxyl andcarboxylate groups (Figure 4B). However, the relaxation time forOHCA is longer in comparison with the system formed withsalicylate. This means that another relevant structural effect inthe aromatic explains the favorable incorporation of OHCA atthe micelle palisade (this is discussed in the next paragraph).Finally, if the hydroxyl is substituted by a methoxyl group(feature 3 in Scheme 1, OMBA vs salicylate), the absence of ahydrogen bond enables the free rotation of the carboxylategroup, thus justifying the absence of viscoelasticity in mixtures ofCTAB and OMBA (Figures 2 and 3). In that case, the larger sizeof the methoxyl group is also likely to affect negatively thepacking of this cosolute in the micellar palisade.The formation of a hydrogen bond explains the results in the

benzoate family of compounds (Scheme 1, comparison ofsalicylate with both benzoate and OMBA). However, theconsideration of feature 2the presence of extra carbonsbetween the carboxylate group and the aromatic ringshowsthat all compounds in the cinnamate family induce largerrelaxation times compared to their benzoate counterparts.Therefore, another structural aspect must be responsible forthe more favorable formation of WLMs with cinnamatederivatives. Could the double bond also restrict the rotation ofthe group, thus favoring the incorporation of these cosolutes inthe palisade layer? In order to investigate this aspect, weintroduced the compound sodium 3-phenylpropanate, which hasa structure quite similar to sodium cinnamate but bears asaturated carbon chain instead (comparison 4, Scheme 1). Anequimolar system containing 200 mmol L−1 of CTAB and 3-phenylpropanate was prepared and visually compared to theequivalent mixture with cinnamate (Figure 5). Clearly, therheological characteristics of both samples differ completely.While the system containing cinnamate behaves as a viscoelasticsolidlike sample (see also Figures 2 and 3), a purely liquidlikeresponse is obtained with 3-phenylpropanate (Figure 5),pointing to the absence of wormlike micelle formation. Thisdemonstrates the critical the role of the double bond in thestructure of the cosolute. In order to further analyze this aspect,density functional theory (DFT) calculations were carried outwithin the hybrid generalized gradient approximation (Support-ing Information). Optimized geometries for cinnamate (Figure6, left) and 3-phenylpropanate (Figure 6, right) were obtained atthe B3LYP/6-31G(2df,p) level of theory. The green gridsrepresent the isosurface of the ground-state electron density ascalculated with DFT. The molecular volume is appropriate toindicate the steric hindrance caused by the rotation of the groupsof the two molecules. The calculations were performed at theseoptimized geometries, and the values for cinnamate (0.167 nm3

molecule−1) are around 25% lower than for 3-phenylpropanate

(0.220 nm3 molecule−1). The more planar conformation and thehigher internal rotation barrier of cinnamate compared to 3-phenylpropanate (see Supporting Information) are due to thehyperconjugation of cinnamate and are relevant to explain thedifference between the volumes of the two structures. Thisdifference in volume affects the insertion of the aromaticmolecule in the micelle palisade, explaining the more efficientpacking of cinnamate and therefore the more favorable drive forWLM formation. OHCA and OMCA have a double bondbetween the carboxyl group and the aromatic ring, and for thisreason, the hyperconjugation effect is also responsible for thehigh level of incorporation of these two cosolutes at the micellarpalisade.In summary, while it is well-known that the incorporation of

aromatic cosolutes is mainly driven by the hydrophobic effectand charge neutralization, our results clearly show that the freerotation of the anionic group creates a steric hindrance. This is inagreement with the rupture of the long WLMs formed by CTABand trans-OMCA when irradiated by UV light.7,11,16 It is veryplausible that cis-OMCA has a huge steric effect reducing itsincorporation into the CTAB micelles. It is now quite clear thatthe meta- and para- isomers of salicylate are not able to producewormlike micelles, since the larger distance between the hydroxyland the carboxyl groups prevents the formation of a hydrogen

Figure 5. Gravitational effect on systems containing 200 mmol L−1 ofCTAB and 200 mmol L−1 of cinnamate (left) and 3-phenylpropanate(right). The photo was obtained just after inversion of the vials.

Figure 6. Fully optimized equilibrium structures of cinnamate (A) and3-phenylpropanate (B). The planarity of the cinnamate molecule isvisualized by comparing the length of the side of the box (x, y, and z) thatcontains the two molecules. Clearly, x is narrower for cinnamate incomparison with 3-phenylpropanate.

Langmuir Article

dx.doi.org/10.1021/la502649j | Langmuir 2014, 30, 11535−1154211540

bond, thus allowing the free rotation of the carboxyl group at thecrowded surface of the micelle. The importance of steric effectswas clearly demonstrated by comparing the system containingcinnamate and 3-phenylpropanate.Finally, we examine the surprising gel-like samples formed by

the only noncharged cosolute studied: phenol. WLMs formed byCTAB and phenol have already been characterized by cryo-TEM.28 The authors proposed that the repulsions between theCTAB headgroup are reduced by the introduction of phenol atthe micelle palisade, leading to the formation of WLMs.Compared to the other cosolutes, phenol does not have a stronginteraction with the cationic headgroup of CTA+, since it is notcharged. Therefore, WLM formation is essentially driven by thehydrophobic effect, resulting from the incorporation of thearomatic ring in the palisade layer of the micelles. The absence ofcharge on the cosolute results in high micellar electrostaticdensity, creating (similarly to polyelectrolytes) strong chargerepulsions between the WLM chains. Therefore, the chains mustbe more stretched, which could explain the longer relaxationtime.In order to investigate the structure of the WLM, SANS

measurements on CTAB (100 mmol L−1) and the variouscosolutes (100 mmol L−1) were carried out. The results areshown in Figure 7. As can be seen, except for OMBA and phenol,all cosolutes exhibit a similar scattering pattern with the intensitysmoothly increasing toward low q values, indicating the presenceof elongated structures.3 For OMBA, the presence of a plateau atlow q suggests a finite size of the aggregates compared to theother systems, i.e., shorter micelles. Fitting the curves with a rodmodel confirms the presence of elongated cylindrical structuresfor all samples, with shorter length for OMBA (ca. 49 Å). In allsystems studied, the cross-section r remains fairly constant at ca.21± 1 Å and in good agreement with the length of the alkyl chainof CTA inWLMs formed with this surfactant (r = 20.3 Å).29 Thescattering pattern obtained inmixtures of CTAB and phenol witha strong structural peak markedly departs from those of theothers and confirms the presence of very strong interactions. Theresult obtained for this system is compatible with high chargeinteractions and stiffer chains. By taking the value of 2π/qmax, acharacteristic correlation distance can be estimated at ≈18 nm.

■ CONCLUSION

Small variations in the structure of aromatic cosolutes have beenshown to significantly affect the rheology of WLM formed withthe cationic surfactant CTAB. The formation of WLM can beexamined by considering the mass action law, in which the lengthof the micelles depends on the extent of incorporation of thearomatic cosolute into the micellar palisade. The length of theWLMs and, as a result, their entanglement directly affect theviscoelastic response of the system. Therefore, rheologicalmeasurements can be used to evaluate the effect of molecularvariations in aromatic cosolutes on wormlike micelle formation.For a fixed concentration of CTAB and cosolute, the followingorder of decreasing relaxation times was observed: phenol >cinnamate > OHCA > salicylate > OMCA > benzoate > OMBA.In the case of salicylate, a cosolute well-known to induce theelongation of CTABmicelles, our study reveals that the favorableincorporation of the aromatic molecule results from theformation of a hydrogen bond between the hydroxyl group (atthe ortho position) and the carboxylate group, which preventsthe free rotation of the single bond, thus reducing sterichindrance and allowing a better insertion of the aromaticmolecule into the micelle. In comparison, the combination ofCTAB with sodium benzoate induces a weaker viscoelasticresponse, due to the free rotation of the carboxylate group,resulting in poorer packing, while the addition OMBAwith amethoxy replacing the hydroxyl group in the ortho positionprevents the formation of wormlike micelles (as inferred fromthe liquidlike response); this was attributed again to the freerotation of the carboxylate group and bulky methyl group, whichhinders the favorable packing of the cosolute in the palisade layer.The addition of two carbons between the ring and thecarboxylate in cinnamate derivatives substantially enhances theviscoelastic properties, reflecting a favorable incorporation of thecosolutes. The presence of the double bond was found to becritical. The double bond in cinnamate derivatives was shown toreduce the free rotation of the group, and additionally, thehyperconjugation ensures a better packing of the aromatic insidethe micelle. This was observed by comparing the saturatedcounterpart of cinnamate, namely, 3-phenylpropanate, which didnot produce wormlike micelles. Surprisingly, the addition of a

Figure 7. SANS patterns for wormlike micelles formed by CTAB and different cosolutes. The concentration of each component is 100mmol L−1 in D2Oat 25 °C. Solid lines represent fits to the cylinder model.

Langmuir Article

dx.doi.org/10.1021/la502649j | Langmuir 2014, 30, 11535−1154211541

noncharged cosolute, phenol, leads to the formation of WLMswith the longest relaxation time, which was attributed to theelectrostatic repulsions between the highly charged cationicmicelles, leading to more rigid structures. Morphologicalinvestigations with SANS confirmed the formation of elongatedstructures in all cases and the occurrence of strong interactions inthe CTAB/phenol systems, which corroborates the rheologicalresults. Overall this work brings new fundamental insight into themolecular mechanisms that control wormlike micelle formation,showing, in addition to the hydrophobic effect and electrostaticinteractions, the importance of intermolecular hydrogen bondsand the steric effect.

■ ASSOCIATED CONTENT*S Supporting InformationPreparation of OHCA, flow curves and the determination of zeroshear viscosity in a range of CTAB and aromatics concentrations,computational analysis and electrostatic potential maps of thearomatic cosolutes, and calculated rotational barrier forcinnamate and 3-phenylpropanate. This material is availablefree of charge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: sabadini@iqm.unicamp.br.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThe authors thank the Brazilian agencies CAPES, CNPq,FAPESP, and PETROBRAS for financial support and fellowshipsand the Center for Scientific Computing (NCC/GridUNESP) ofthe Sao Paulo State University (UNESP) for computing time.ISIS (Rutherford Appleton Laboratory, Didcot, UK) is acknowl-edged for the provision of beam time and Ann Terry is thankedfor her help with the measurements.

■ REFERENCES(1) J. N. Israelachvili Intermolecular and Surface Forces, 2nd ed.;Academic Press: London, 1991.(2) Rehage, H.; Hoffmann, H. Rheological properties of viscoelasticsurfactant systems. J. Phys. Chem. 1988, 92, 4712−4719.(3) Dreiss, C. A.; C, D.Wormlike micelles: Where do we stand? Recentdevelopments, linear rheology and scattering techniques. Soft Matter2007, 2, 956−970.(4) Yang, J. Viscoelastic wormlike micelles and their applications. Curr.Opin. Colloid Interface Sci. 2002, 7, 276−281.(5) Ezrahi, S.; Tuval, E.; Aserin, A. Properties, main applications andperspectives of wormmicelles. Adv. Colloid Interface Sci. 2006, 128−130,77−102.(6) Magid, L. J. The surfactant−polyelectrolyte analogy. J. Phys. Chem.B 1998, 102, 4064−4074.(7) Chu, Z.; Dreiss, C. A.; Feng, Y. Smart wormlike micelles.Chem. Soc.Rev. 2013, 42, 7174−7203.(8) Rodrigues, R. K.; da Silva, M. A.; Sabadini, E. Worm-like micelles ofCTAB and sodium salicylate under turbulent flow. Langmuir 2008, 24,13875−13879.(9) Gyr, A.; Bewersdorff, H. W. Drag Reduction of Turbulent Flows byAdditives (Fluid Mechanics and Its Applications); Springer: Berlin, 1995.(10) Samuel, M. M.; Dismuke, K. I.; Card, R. J.; Brown, J. E.; England,K. W. Methods of fracturing subterranean formations. US Patent6306800, 2001.(11) Baglioni, P.; Braccalenti, E.; Carretti, E.; Germani, R.; Goracci, L.;Savelli, G.; Tiecco, M. Surfactant-based photorheological fluids: Effectof the surfactant structure. Langmuir 2009, 25, 5467−5475.

(12) Nettesheim, F.; Kaler, E. W. Phase behavior of systems withwormlike micelles. In Giant Micelles: Properties and Applications; Zana,R., Kaler, E. W., Eds.; CRC Press, Boca Raton, FL, 2007; SurfactantScience Series Vol. 140, pp 223−247.(13) Sarac, B.; Meriguet, G.; Ancian, B.; BeSter-Rogac, M. Salicylateisomer-specific effect on the micellization of dodecyltrimethylammo-nium chloride: Large effects from small changes. Langmuir 2013, 29,4460−4469.(14) Smith, B. C.; Chou, L.-C.; Lu, B.; Zakin, J. L. In Structure and Flowin Surfactant Solutions; ACS Symposium Series; Herb, C. A.,Prud’homme, R. K., Eds.; American Chemical Society: Washington,DC, 1994; p 578.(15) Bijma, K.; Engber, J. B. F. N. Effect of counterions on properties ofmicelles formed by alkylpyridinium surfactants. 1. Conductometry and1H NMR chemical shifts. Langmuir 1997, 13, 4843−4849.(16) Ketner, A. M.; Kumar, R.; Davies, T. S.; Elder, P. W.; Raghavan, S.R. A Simple class of photorheological fluids: Surfactant solutions withviscosity tunable by light. J. Am. Chem. Soc. 2007, 129 (6), 1553−1559.(17) Armarego, W. L. F.; Chai, C. L. L. Purification of LaboratoryChemicals, 6th ed.; Butterworth-Heinemann: Oxford, UK, 2009.(18) Heenan, R. K.; King, S. M. Osborn, R.; Stanley, H. B. RAL Intl.Rep., 1989 (RAL-89-128).(19) SasView project, originally developed by the DANSE projectunder NSF award DMR-0520547; http://www.sasview.org/.(20) Shikata, T.; Sakaiguchi, Y.; Urakami, H.; Tamura, A.; Hirata, H.Enormously elongated cationic surfactant micelle formed in CTAB−aromatic additive systems. J. Colloid Interface Sci. 1987, 119, 291−293.(21) Shikata, T.; Hirata, H.; Kotaka, T. Micelle formation of detergentmolecules in aqueous-mediaViscoelastic properties of aqueouscetyltrimethylammonium bromide Solutions. Langmuir 1987, 3,1081−1086.(22) Cates, M. E.; Fielding, S. M. Rheology of giant micelles. Adv. Phys.2006, 55, 799−879.(23) Berret, J. F. In Molecular Gels: Materials with Self-AssembledFibrillar Networks; Weiss, R. G., Terech, P., Eds.; Springer: Dordrecht,The Netherlands, 2005.(24) Cates, M. E.; Candau, S. J. Statics and dynamics of worm-likesurfactant micelles. J. Phys.: Condens. Mater. 1990, 2, 6869.(25) Rakitin, A. R.; Pack, G. R. Necessity of aromatic carboxylateanions to be planar to induce growth of cationic micelles. Langmuir2005, 21, 837−840.(26) Liu, J.; Dong, B.; Sun, D.; Wei, X.; Wang, S.; Zheng, L. Enthalpymeasurements for the formation of salt-induced wormlike micelles usingisothermal titration microcalorimetry. Colloids Surf., A 2011, 380, 308−313.(27) Jensen, F. Introduction to Computational Chemistry; John Wiley &Sons: Chichester, UK, 2003.(28) Agarwal, V.; Singh, M.; McPherson, G.; John, V.; Bose, A.Microstructure evolution in aqueous solutions of cetyltrimethylammo-nium bromide (CTAB) and phenol derivatives. Colloids Surf., A 2006,281, 246−253.(29) Helgeson, M. E.; Hodgdon, T. K.; Kaler, E. W.; Wagner, N. J. Asystematic study of equilibrium structure, thermodynamics, andrheology of aqueous CTAB/NaNO3 wormlike micelles. J. ColloidInterface Sci. 2010, 349, 1−12.

Langmuir Article

dx.doi.org/10.1021/la502649j | Langmuir 2014, 30, 11535−1154211542