Journal of Colloid and Interface Science 311 (2007) 276–284www.elsevier.com/locate/jcis

Formation of wormlike micelle in a mixed amino-acid based anionicsurfactant and cationic surfactant systems

Rekha Goswami Shrestha, Lok Kumar Shrestha, Kenji Aramaki ∗

Graduate School of Environment and Information Sciences, Yokohama National University, Tokiwadai 79-7, Hodogaya-ku, Yokohama 240-8501, Japan

Received 8 December 2006; accepted 8 February 2007

Available online 23 February 2007

Abstract

Formation of wormlike micelles in mixed anionic/cationic system without the addition of any salt has been studied. Amino-acid based an-ionic surfactant N -dodecylglutamic acid (LAD), which is practically immiscible with water at 25 ◦C upon neutralization by 2,2′,2′′-nitrilo-triethanol (TEA) forms small micellar aggregates and the solution behaves like a Newtonian fluid. The rheological behavior of LAD/water/hexa-decyltrimethylammonium bromide (CTAB) and LAD/water/dodecyltrimethylammonium bromide (DTAB) systems were investigated at differentdegrees of neutralization of the LAD depending on the concentration of the cationic surfactants and on temperature. Addition of CTAB to thedilute aqueous solution of the LAD–TEA-x (the neutralized product, where x represents the mole ratio of TEA) causes one dimensional micel-lar growth. After certain concentration the elongated micelles entangle forming a rigid network of viscoelastic wormlike micelles. Thus formedviscoelastic solutions follow Maxwellian behavior over a wide range of frequency and thus are considered to consist of transient network ofwormlike micelles. By varying the degree of neutralization from 1:1 via 1:1.5 to 1:2 (molar ratio) phase and rheological behavior were modi-fied in that the highly viscous region of viscoelastic wormlike micelles shifted to higher CTAB concentrations and no maxima in the zero-shearviscosity could be observed for the higher degree of neutralization of the LAD (1:1.5 and 1:2). However, the obtained rheological parametersshowed scaling relationships that were consistent with the living polymer model. The zero-shear viscosity decays exponentially with temperaturefollowing Arrhenius behavior. The flow activation energy calculated from the Arrhenius plot is very close to the value reported for the typicalwormlike micellar solution. In contrast to CTAB no formation of viscoelastic wormlike micelles could be observed with DTAB, although, thesolution viscosity increases. The elongated micelles could not entangle to form a rigid network of wormlike micelles in this system. For the firsttime viscoelastic wormlike micelles could be obtained in salt-free mixed anionic/cationic surfactant systems.© 2007 Elsevier Inc. All rights reserved.

Keywords: Amino-acid surfactants; Phase behavior; Wormlike micelles; Rheology; Viscoelastic solutions

1. Introduction

Surfactant molecules can give a range of self-assembled mi-crostructures such as micelles, vesicles, and different liquidcrystalline phases in both polar [1–10] and nonpolar medium[11–13]. Above critical micelle concentration (CMC), hy-drophilic surfactants forms small spherical or rodlike micellaraggregates and the solutions show Newtonian flow behavior.However, under certain conditions such as temperature or thepresence of counterions, one dimensional micellar growth takesplace and long flexible aggregates called wormlike micelles are

* Corresponding author. Fax: +81 45 339 4300.E-mail address: aramakik@ynu.ac.jp (K. Aramaki).

0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.jcis.2007.02.050

formed [14–20]. Above some critical concentration, called theoverlapping concentration, degree of entanglement of wormlikemicelles increases and the solution exhibit viscoelastic prop-erties [21] similar to those observed in solutions of flexiblepolymers. Unlike polymers, self-organized wormlike micellesare in dynamic equilibrium with their monomers and the micel-lar chains can break and recombine [22].

Several reports on the formation of viscoelastic wormlikemicelles in mixed cationic/anionic, or ionic/nonionic surfactantsystems with the addition of strongly binding counter ions orvarious salts are available in the literature [23–30]. The saltscreens the electrostatic repulsion between the charged headgroups, and induces the one dimensional micellar growth bydecreasing the effective cross-sectional area per head group ofthe surfactant molecule [31]. Compared to the cationic counter-

R.G. Shrestha et al. / Journal of Colloid and Interface Science 311 (2007) 276–284 277

parts, the formation of wormlike micelles in anionic surfactantsis less studied so far. Among the few reports available, sodiumdodecyl sulfate is the center of focus as the choice of surfactant[32,33]. Recently Kalur and Raghavan [34] observed wormlikemicelles in anionic surfactant sodiumoleate by the addition ofvarious salts. To our knowledge, there are no reports on forma-tion of wormlike micelles in the mixed anionic/cationic surfac-tant system without the addition of any salt.

The anionic surfactants are biodegradable and less toxiccompared to their cationic counterparts. Therefore, the forma-tion of wormlike micelles in the anionic surfactant could of-fer many advantages in cosmetics, toiletry, and pharmaceuticalformulations. Especially, amino-acid based anionic surfactants(both natural and synthetic types) have been the subject of manystudies, due mostly to their huge potential applications in cos-metic and household products because of their mildness to theskin.

In the present contribution we report the formation of vis-coelastic wormlike micellar solution in the mixed anionic/ca-tionic surfactant without the addition of any salt. Anionicsurfactant N -dodecylglutamic acid (LAD) is neutralized by2,2′,2′′-nitrilotriethanol (TEA) at different degree of neutral-ization (1:1 via 1:1.5 to 1:2) and the phase behavior and rheo-logical behavior were studied depending on the concentrationand alkyl chain length of co-surfactants and on temperature.

2. Theoretical background

Wormlike micelles are very much similar to polymers withthe exception that the micelles are in equilibrium with theirmonomers. When the equilibrium network of the wormlike mi-celles is deformed, the relaxation occurs within a definite timeand the equilibrium condition is restored again. For a defor-mation with a time period shorter than the relaxation time, τR,the system exhibits an elastic property characteristic of a solidmaterial with a Hookean constant G0, called the shear modu-lus. For a slow deformation, however, the network has sufficienttime to dissipate the stress, and the viscoelastic system behavesas a viscous fluid with a zero-shear viscosity, η0.

Dynamics of wormlike micelles under oscillatory shear isdescribed by considering two different processes. When a smallstrain is applied, the stress relaxation occurs by reptation, i.e.,reptile-like motion of the micelle along its own contour. Be-sides, micelles may undergo reversible scission [22] takingplace at two time scales, namely, reptation time τrep and break-ing time τb. The τrep is the time taken by the wormlike micelleof length L̄ to pass through a hypothetical tube and the τb isthe average time taken by the chain of average length L̄ tobreak into two pieces. It is assumed that when a chain breaks,two daughter chains become uncorrelated, and recombine withthe micellar end in a random way. For a fast scission kinetics(τb � τrep), a single exponential stress decay is observed andthe viscoelastic behavior of such systems at low frequency fol-lows Maxwell model with a single relaxation time τR given by(τbτrep)

1/2.When the timescale of the reptation for an average micel-

lar contour length is too low in comparison to the timescale

of the scission, the viscoelastic micellar solutions behave as aMaxwell-fluid with single relaxation time, and variation of theelastic or storage modulus G′(ω) and the viscous or loss mod-ulus G′′(ω) as a function of oscillatory shear frequency ω isdescribed by following relations:

(1)G′ = (ωτR)2

1 + (ωτR)2G0,

(2)G′′ = ωτR

1 + (ωτR)2G0,

where, G0 is called shear (plateau) modulus. The relaxationtime, τR, can be estimated as (ωc)

−1, where ωc is the frequencyat which G′ and G′′ intersect and it is also called crossover fre-quency. At low ω, and ωτR � 1, G′ and G′′ show differentscaling behavior (G′ ∝ ω2 whereas G′′ ∝ ω), with G′′ > G′,which corresponds to a liquid-like behavior. For ωτR � 1, thatis, ω � ωc, G′ approaches a constant limiting plateau valueequal to G0, with G′ > G′′ and the system behaves as an elasticmaterial. At ω � ωc, the Maxwell equations predict a monoto-nous decrease in G′′. However, at ω � ωc, G′′ shows an upturn.This deviation is thought to have arisen from a transition of therelaxation mode from ‘slower’ reptation to other ‘breathing’ or‘Rouse’ modes [35] of cylindrical micelles, analogous to poly-mer chain. At a particular temperature, the G0 is the measure ofthe degree on entanglements, whereas relaxation time gives in-formation about the average micellar length. Eq. (3) allows oneto estimate the zero-shear viscosity, η0, from the oscillatory-shear measurement:

(3)η0 = G0 τR.

The detail of the theoretical background is given in the support-ing information.

3. Materials and methods

3.1. Materials

The anionic surfactant N -dodecylglutamic acid (designatedas LAD) was kindly received from Ajinomoto Co., Inc. Japan.The surfactant is 99.9% pure. The cationic surfactant hexadecyltrimethylammonium bromide (CTAB), dodecyltrimethylam-monium bromide (DTAB) and the neutralizing agent 2,2′,2′′-nitrilotriethanol (TEA) were purchased from Tokyo Kasei Ko-gyo Co. Inc. The purity of CTAB, DTAB, and TEA is more than99%. The LAD was neutralized by TEA at different molar ratio;1:1, 1:1.5, and 1:2. For the sake of clarity we have representedthe neutralized product as LAD–TEA-1, LAD–TEA-1.5, andLAD–TEA-2 respective to the molar ratio of TEA. The molec-ular structure of the LAD is presented in Scheme 1.

Scheme 1. Molecular structure of N -dodecylglutamic acid (LAD).

278 R.G. Shrestha et al. / Journal of Colloid and Interface Science 311 (2007) 276–284

(a) (b)

Fig. 1. Partial ternary phase diagram of the LAD–TEA-x (x = 1, 1.5, and 2 are the molar ratio of TEA)/water/CTAB (a), and the LAD–TEA-1/water/DTAB (b) at25 ◦C. Wm is the micellar phase, H1, Lα , and I1 are the hexagonal, lamellar and discrete cubic phases respectively. The shaded area shows highly viscous regionwithin the Wm phase.

3.2. Methods

3.2.1. Phase behaviorRequired amount of chemicals were weighed in clean and

dry glass ampoules and immediately flame sealed. The sam-ples were mixed at higher temperature and homogenized byrepeated centrifugation. After mixing, the samples were keptin thermostated bath at 25 ◦C for several days to equilibrate andthe equilibrium phases were identified by visual observationsthrough normal and crossed polarizer.

3.2.2. Rheological measurementsSamples for all the rheological measurements were pre-

pared in clean and dry deciliter glass vessels by weighingrequired amount of reagents. The samples were mixed athigher temperature (50 ◦C) with constant stirring by mag-netic stirrer for several hours. After mixing, the samples werekept in a thermostated water bath at 25 ◦C for more thana week to ensure equilibration before measurements. Mea-surements were done in an ARES7 rheometer (RheometerScientific) at 25 ◦C (unless otherwise stated) using couette(cup diameter, 17.0 mm, bob diameter, 16.5 mm; bob length,13.3 mm) and cone plate (50 diameter, having cone angleof 0.04 rad) geometries depending on viscosity of the sam-ples.

4. Results and discussion

4.1. Phase behavior

The amino-acid based anionic surfactant LAD is practicallyimmiscible with water at all composition at 25 ◦C. However,upon neutralization by TEA, the Krafft temperature is reducedand it forms micellar solution. The phase behavior of LAD inthe mixed LAD/water/CTAB and DTAB systems has been stud-

ied at different degree of neutralization (1:1 via 1:1.5 to 1:2).The partial ternary phase diagrams are shown in Fig. 1. Theshaded areas in the phase diagrams indicate the highly viscousregion. Depending on the degree of neutralization the phaseand rheological behavior were modified. Namely, the micellardomain size increases and the highly viscous region of worm-like micelles shifted towards higher CTAB concentration. Inthe LAD–TEA-1/water/CTAB system, with increase in CTABconcentration, the solution viscosity increases up to certain con-centration afterwards decreases and eventually phase separationoccurs. However, no such viscosity maxima are observed athigher degree of neutralization (1:1.5 and 1:2). Independent tothe mole ratio of the LAD–TEA, successive addition of CTABthe aggregates loose the flexibility and transform into differentliquid crystal structures depending on the mixing fraction ofthe surfactant. For example in the dilute region, lamellar liquidcrystal is observed, whereas, hexagonal phase is found in thehigher surfactant concentration regions. In the surfactant/waterbinary system, discrete cubic phase is common at higher sur-factant concentration at all degree of neutralization (see Fig. 1a)and its boundary to the Wm phase is determined by the small-angle X-ray scattering measurements. The boarder between thehexagonal and lamellar liquid crystalline phases is not deter-mined.

The viscous samples (as shown by the shaded areas in thephase diagrams) show shear-birefringence, which is regardedas a typical behavior of wormlike micellar solution. In con-trast to CTAB no such viscous region could be observed withDTAB. The DTAB favors the micellar growth in the LAD–TEA system but the elongated micelles could not entangleto form a rigid network of wormlike micelles. The detail isdescribed in the rheology section. The micellar domain be-comes wider in the DTAB compared to the CTAB system (seeFig. 1b).

R.G. Shrestha et al. / Journal of Colloid and Interface Science 311 (2007) 276–284 279

Fig. 2. Viscosity versus shear rate curves for the LAD–TEA-1/water/CTAB sys-tem as a function of CTAB at fixed LAD–TEA-1/water = 15/85 at 25 ◦C.

4.2. Rheological measurements

4.2.1. Steady-state viscositySteady rate sweep tests were performed to achieve the

steady-state viscosity of all the systems shown in the phasediagram. The steady shear rate–viscosity curves for the LAD–TEA-1/water/CTAB system with LAD–TEA-1/water ratio of15/85 at different CTAB concentration at 25 ◦C is shown inFig. 2.

We first describe the effects of CTAB concentration on thesteady-state viscosity in the LAD–TEA-1/water/CTAB systemwithin the Wm region. At low mixing fraction of CTAB, the vis-cosity is very low and is independent of shear applied i.e., thesample follows Newtonian fluid like behavior. Upon increasingthe CTAB, the shear thinning, which is a typical behavior ofwormlike micelles, is observed [14]. As is clear from Fig. 2the samples follow a Newtonian behavior at low shear ratesand exhibit shear thinning above the attainment of critical shearrates γ̇c. The decreasing viscosity with increasing shear rate canbe taken as an evidence for the formation of long wormlikemicelles. All the results obtained can be fitted to the Carreaumodel and the zero-shear viscosity (η0) is estimated.

Fig. 3 shows the η0, obtained through the fitting of thesteady-state curves to the Carreau model. Addition of CTAB in-creases the zero-shear viscosity of the solution till a maximumis reached. Afterwards, further addition of CTAB decreasesthe viscosity. The maximum of the viscosity curve shiftedtowards the higher CTAB concentration upon increasing theLAD–TEA/water mixing ratio. However, no such maxima areobserved at higher degree of neutralization that is in the LAD–TEA-1.5/water/CTAB and LAD–TEA-2/water/CTAB systems(see Fig. 3b). At higher mole ratio of TEA (1:1.5 and 1:2)the viscosity of the solution continuously increase with theCTAB and eventually phase separation to liquid crystallinephase is occurred. The steady-shear rheology for the LAD–TEA-2/water/CTAB system is presented in Fig. S1 in the sup-porting information.

The viscosity of the dilute aqueous solution of LAD–TEA isvery low indicating that the solution consists of small spherical

(a)

(b)

Fig. 3. Effect of CTAB and DTAB on the zero-shear viscosity dependingon the mixing fraction of the surfactant (LAD–TEA-1) for the LAD–TEA-1/water/CTAB and DTAB systems at 25 ◦C (a), and the zero-shear viscosity ver-sus CTAB depending on the degree of neutralization of LAD (b). Note, theLAD–TEA-1/water/DTAB system is represented by the black color open circlein panel (a).

or rodlike micelles. Addition of CTAB increases the viscos-ity of the solution, which can be taken as an evidence of theone-dimensional micellar growth. After certain concentration,these elongated micelles start to entangle with each other form-ing a rigid transient network of wormlike micelles [36,37]. Atthis condition, the viscosity of the solution increases extremely(∼4th order of magnitude compared to that of pure water) andthe solution exhibits viscoelastic behavior. The reduction inthe viscosity and poor viscoelastic behavior with the furtheraddition of CTAB, may be due to strong interaction betweencationic and anionic counterparts, which facilitate a transitionfrom linear micelle to a branched network or breaking of mi-celles [20], or to a change in the curvature of aggregates asso-ciated with the transition to the lamellar phase. Most probably,formation of micellar joints and its growth to a bilayer struc-ture is the possible mechanism of the decreasing viscosity withfurther addition of CTAB after maximum [38].

Next, we have studied the effect of DTAB on the rheologyof the LAD–TEA-1/water/DTAB system. The DTAB favors the

280 R.G. Shrestha et al. / Journal of Colloid and Interface Science 311 (2007) 276–284

(a)

(b)

Fig. 4. Effect of temperature on the steady shear rheology of the 5.5% CTABat fixed surfactant concentration of LAD–TEA-1/water = 15/85. The viscosityversus shear rate at different temperatures (a), and the zero-shear viscosity ver-sus 1/T in a semilog (Arrhenius) plot (b), where T is the absolute temperature.

micellar growth and shows an increase of viscosity upon in-creasing concentration. However, the system could not formrigid network of wormlike micelles. The difference effect be-tween the CTAB and DTAB on the change in viscosity is shownin Fig. 3a.

The effect of temperature on the steady-shear rheology of theLAD–TEA-1/water/CTAB system is shown in Fig. 4a. With in-creasing temperature, the critical shear rate γ̇c (the shear rateat which shear thinning appears) increases. The value of γ̇cincreases by ∼2nd order of magnitude upon increasing tem-perature from 20 to 40 ◦C. On the other hand, the zero-shearviscosity decreases exponentially with temperature as provenby the Arrhenius plot of logη0 versus 1/T in Fig. 4b. Thisexponential decay of the viscosity with temperature is in ac-cordance with the well-known equation for wormlike micelles[39,40].

(4)η0 = AeEa/(RT ),

where Ea is the flow activation energy, R is the universal gasconstant, and A is a constant. The value of flow activation en-

Fig. 5. Frequency sweep measurements as a function of CTAB in the LAD–TEA-1/water/CTAB system at fixed LAD–TEA-1/water = 15/85 at 25 ◦C.(—) stand for the best fit to the Maxwell model.

ergy calculated from the slope of the straight line fit in Fig. 4bis found to be about 150 kJ/mol, which is close to the valuereported for wormlike micelles [34,39,40]. The exponential de-cay of viscosity with temperature may be due to the formationof the micellar joints in the networked structure [40,41]. Thedetail of temperature effect on rheology is described in the sec-tion of oscillatory measurements.

Temperature has dual effect on the viscosity at higher degreeof neutralization that is in LAD–TEA-2/water/CTAB systemwhen temperature is increased from 15 to 25 ◦C, the η0 in-creases but the critical shear rate decreases indicating that thesystem gets more structured. However, opposite trend is ob-served upon heating the sample above 25 ◦C, see Fig. S2 in thesupporting information. The η0 at 15 and 35 ◦C are very closeto each other (see steady-shear curve in Fig. S2) but the criti-cal shear rate differs, which might be due to a difference in thestructure of the system at the respective temperatures. The in-crease of viscosity at higher temperature may possibly be dueto one-dimensional growth if the rodlike aggregates that are al-ready formed or by the increase of the entanglement density ofthe long cylindrical micelles.

4.2.2. Oscillatory measurementsViscoelastic properties of the wormlike micellar solutions

were investigated by oscillatory-shear (frequency sweep) mea-surements. Frequency sweep measurements were carried out asa function of the CTAB concentration at fixed mixing ratio ofsurfactant and the results are presented in Fig. 5. The dynamicrheology data shows variation of elastic modulus (G′), and vis-cous modulus (G′′) as a function of oscillation frequency (ω).At high ω, the sample exhibits elastic behavior (G′ shows aplateau value and dominates over the G′′), whereas at low ω

the sample shows a viscous behavior (G′′ exceeds G′). Thus thesample shows the viscoelastic behavior and its relaxation timecan be estimated by the reciprocal of the crossover frequency(the frequency at which G′ and G′′ cross). The data points canbe well fitted to a single Maxwell Model in the range of low

R.G. Shrestha et al. / Journal of Colloid and Interface Science 311 (2007) 276–284 281

Fig. 6. Normalized Cole–Cole plot as a function of CTAB in the LAD–TEA-1/H2O/CTAB system at LAD–TEA-1/H2O = 15/85 at 25 ◦C.

and medium frequencies. The solid lines in the Fig. 5 representthe best fit with a single relaxation time. Such a viscoelasticbehavior indicates the formation of rigid network of entangledwormlike micelles. The deviation of the G′′ from the model es-pecially at higher ω region is another characteristic features ofthe wormlike micelles. This deviation corresponds to the factthat wormlike micelles are in a dynamic equilibrium and thereis rapid breaking and recombination process.

Next, we have studied the oscillatory-shear rheological mea-surements on the LAD–TEA-1/water/DTAB system. In contrastto the CTAB no formation of viscoelastic wormlike micellescould be observed with DTAB. Rheological measurements haveshown that addition of the DTAB increases the viscosity of thesolution up to few Pa s indicating the micellar growth. How-ever, the elongated micelles could not entangle to form a rigidnetwork of wormlike micelles. Hence, the solution does not ex-hibit viscoelastic properties. The variation of elastic modulus,the viscous modulus, and the modulus of complex viscosityas a function of oscillation frequency for the LAD–TEA-1/water/DTAB system is presented in Fig. S3 of the supportinginformation.

As shown in Fig. 5, with increasing the mixing fraction ofCTAB, the crossing point between G′ and G′′ moves towardsthe higher frequency region indicating lower relaxation time,and thus poorer viscoelastic behavior.

Cole–Cole plot is another way of verifying the viscoelas-tic wormlike micellar behavior. From the data shown in Fig. 6it can be seen that the shape of the curve follows perfectlya semicircular behavior indicating the presence of viscoelas-tic wormlike micellar solution at 5.5% CTAB added system.At other CTAB concentrations, the shape of the curve deviatesfrom semicircle and hence the poor viscoelastic behavior.

The rheological parameters G0 and τR may be obtainedby fitting the data from oscillatory shear measurements to theMaxwell equations (1), (2), and (3). The values of G0 and τRobtained from Maxwell fit are plotted as a function of CTABconcentration and are shown in Fig. 7. The continuous increaseof the plateau modulus G0 with CTAB can be taken as anevidence of one dimensional micellar growth and hence an in-crease in the entanglement degree of wormlike micelles. How-

Fig. 7. Variation of the G0 (filled symbols) and the τR (open symbols) as a func-tion of the CTAB depending on the wt% of the surfactant. (!, ") 10, (1, 2) 15,and (P, Q) 20 wt%.

ever, the maxima observed in the τR curve imply some struc-tural changes occur that allows a faster stress relaxation [42].

We have also performed the oscillatory-shear (frequencysweep) rheological measurements for the LAD–TEA-2/water/CTAB system. Like in the LAD–TEA-1/water/CTAB system,the crossover of the G′ and G′′ shifts towards lower frequencyregion with increasing the CTAB concentration. This indicatesthat the relaxation time increases rapidly at higher CTAB con-centration and the samples show more viscoelastic character.The values of G0 and τR calculated from the Maxwell fit tothe experimental data increases with CTAB, see Fig. S3 in thesupporting information. This may be due to the increased en-tanglement degree of wormlike micelles.

The temperature effect on the oscillatory-shear rheology hasbeen carried out. Fig. 8 shows the frequency sweep test at fixedmixing ratio of LAD–TEA-1/water/CTAB system at differenttemperature. As can be seen from Fig. 8a with increasing tem-perature, the entire frequency spectrum shifts to higher frequen-cies and consequently the relaxation time decreases, see Fig. 8b.Beyond 35 ◦C, the relaxation time is very low and above 40 ◦Cviscoelastic properties is poor. As was mentioned in the steady-

282 R.G. Shrestha et al. / Journal of Colloid and Interface Science 311 (2007) 276–284

(a)

(b)

Fig. 8. Effect of temperature on the oscillatory shear rheology of the5.5% CTAB in the LAD–TEA-1/water/CTAB system at fixed LAD–TEA-1/water = 15/85 solution. Frequency-sweep measurements at different temper-atures (a), and the relaxation time, τR versus 1/T in a semilog (Arrhenius)plot (b). (—) in the panel (a) represent the best fit to the Maxwell model.

shear rheology section that the samples becomes less viscousupon heating (see Fig. 4a). The decreasing viscosity upon heat-ing the samples in the LAD–TEA-1 system could be attributedto the micellar branching [38]. The number of branching pointsmay grow with temperature and ultimately decreases the vis-cosity [38]. An Arrhenius plot of τR versus 1/T shows anexpected decreases in the τR with temperature, as given by thefollowing equation [40,43]

(5)τR = AeEa/(RT ),

where Ea is the flow activation energy. The Ea is calculated to151 kJ/mol, which is comparable to the value reported for otherwormlike micelles [40].

In the Cole–Cole plot the data follows Maxwell model asconfirmed by the semicircular nature of the curves at lower tem-peratures. However, above 40 ◦C, the solution does not followthe Maxwellian behavior and the curves significantly deviatesfrom the semicircular behavior, see Fig. 9.

Similar temperature effect is observed in the LAD–TEA-2/water/CTAB system. The oscillatory-shear measurement for the15% CTAB in the system at LAD–TEA-2/water = 15/85 is pre-sented in Fig. S4 of the supporting information.

The variation of G0 as a function of temperature for boththe LAD–TEA-1/water/CTAB and the LAD–TEA-2/water/CTAB systems are presented in Fig. 10. The increase of theG0, with temperature indicates the increase of degree of en-tanglements of the micellar network. However, the viscosityof the LAD–TEA-1/water/CTAB system is decreasing linearlywith the temperature (see Fig. 4). Thus, the viscosity decreaseis not because of the cylinder-to-sphere type of transition inthe micellar structure but due to formation of micellar joints inthe network structure. With increasing temperature, the energycost of the free ends of the micelles becomes higher. Therefore,the free ends tend to fuse at the cylindrical part of if own orof the other micelles, thus forming micellar joints or branch-ing [36,38]. Such joints reduce the viscosity because, when astress is applied, the micellar joints can slide along the cylinderbody thereby allowing a fast relaxation process, but the networkstructure is retained [18]. On the other hand the increases of

Fig. 9. Normalized Cole–Cole plot for the 5.5% CTAB in the LAD–TEA-1/water/CTAB system at fixed LAD–TEA-1/water = 15/85 at different temperatures.

R.G. Shrestha et al. / Journal of Colloid and Interface Science 311 (2007) 276–284 283

Fig. 10. Variation of the plateau modulus, G0, as a function of temperature forthe LAD–TEA/water/CTAB system depending on the neutralization degree ofLAD.

viscosity from 15 to 25 ◦C (see Fig. S2 of the supporting infor-mation) in the LAD–TEA-2/water/CTAB system correspondsto the increase of the entanglement network density of worm-like micelles as indicated by the increased G0. The decreasingtrend of the viscosity above 25 ◦C is attributed to the formationof micellar joints.

If we compare the value of G0, for the LAD–TEA-1/water/CTAB and the LAD–TEA-2/water/CTAB systems at 20 ◦C, itcan be seen that the value of the G0 is higher in the formersystem. Higher value of G0 corresponds to the higher net-work or entanglement of the micelles. Therefore, the viscos-ity of the LAD–TEA-1/water/CTAB system is higher at 20 ◦C.The higher value of the G0 at 40 ◦C for the LAD–TEA-2/water/CTAB system supports the higher viscosity data of thesystem at this temperature.

The scaling relationship of different rheological parameterswith surfactant volume fraction is mainly based on the un-charged polymer solution. Nevertheless, the validity of the scal-ing relationships to wormlike micellar system has been testedto ionic surfactants in presence of counterions. In the presentstudy we have tested the validity of the scaling relationships.Fig. 11 shows the variation of η0, G0, and τR with the weightfraction of CTAB in the total amphiphile at 25 ◦C. Since theweight fraction is almost proportional to the volume fraction,the exponents of the scaling relationship can be evaluated fromsuch plots. The exponent of the G0 evaluated from the experi-mental data is 2.17, which is very much close to the predictedone [44]. Similarly, the exponent of η0 is calculated to 5.86,which is almost similar to the reported value of 5.8 to the ionicsurfactants [44]. The value of η0 5.86 indicates that the break-ing and recombination of the micelle is negligibly small on thetime scale of reptation [44].

5. Summary

We have studied the phase and rheological behavior of theN -dodecylglutamic acid (LAD)/water/CTAB and DTAB sys-tems at different degree of neutralization of LAD depending

Fig. 11. Variation of η0, G0, and τR as a function of weight fraction of CTABin the total amphiphile versus weight fraction of CTAB at 25 ◦C.

on the concentration of cationic surfactant and on temperature.Both the CTAB and the DTAB favor the micellar growth inthe LAD–TEA/water system. However, the elongated micellescould not entangle to form a rigid network of wormlike micellesin the DTAB system. Depending on the neutralization degree ofLAD (from 1:1 via 1:5 to 1:2), the phase behavior and rheolog-ical behavior were modified. Namely, the highly viscous regionof viscoelastic wormlike micelles shifted to higher CTAB con-centration and no maxima in the zero-shear viscosity could beobserved for the totally neutralized LAD system. Most proba-bly, formation of micellar joints and growth of the joints to abilayer structure is the mechanism for the decrease of viscosityafter maximum in the LAD–TEA-1/water/CTAB system.

With increase in temperature the rheological parameters η0and the τR decrease following the Arrhenius behavior in theLAD–TEA-1/water/CTAB system. On the other hand, a max-imum in the η0–T curve is observed in the higher degree ofneutralization (1:2 mole ratio). However, in both mole ratio (1:1and 1:2), the G0 increases with temperature indicating that theviscosity decrease with temperature is caused by the micellarjoints formation in the network structure. The observed rheo-logical parameters η0, τR, and G0 showed scaling relationshipsthat were consistent with the living polymer model. The in-vestigation of systems forming viscoelastic wormlike micellarsolutions is of great importance for both basic and applied re-search. For the first time viscoelastic wormlike micelles couldbe obtained in salt-free mixed anionic/cationic surfactant sys-tems. These systems offer many advantages for cosmetics, toi-letry and pharmaceutical formulations.

Acknowledgments

We are thankful to the Ajinomoto Co. Ltd. Japan for sup-plying the amino-acid surfactant. This work is partly sup-ported by Core Research for Evaluation Science and Tech-nology (CREST) of JST Corporation. L.K. Shrestha is thank-ful to the Ministry of Education, Culture, Sports, Science andTechnology (MEXT) of Japan for giving the MonbukagakushoScholarship.

284 R.G. Shrestha et al. / Journal of Colloid and Interface Science 311 (2007) 276–284

Supporting information

The online version of this article contains additional support-ing information.

Please visit DOI: 10.1016/j.jcis.2007.02.050.

References

[1] H. Kunieda, K. Shigeta, K. Ozawa, M. Suzuki, J. Phys. Chem. B 101(1997) 7952.

[2] K. Aramaki, U. Olsson, Y. Yamaguchi, H. Kunieda, Langmuir 15 (1999)6226.

[3] H. Kunieda, K. Shigeta, M. Suzuki, Langmuir 15 (1999) 3118.[4] H. Kunieda, H. Kabir, K. Aramaki, K. Shigeta, J. Mol. Liq. 90 (2001) 157.[5] K. Shigeta, U. Olsson, H. Kunieda, Langmuir 17 (2001) 4717.[6] H. Kunieda, M. Kaneko, M.A. López-Quintela, M. Tsukahara, Lang-

muir 20 (2004) 2164.[7] R. Strey, R. Schomäcker, D. Roux, F. Nallet, U. Olsson, J. Chem. Soc.

Faraday Trans. 86 (1990) 2253.[8] K. Aramaki, H. Kunieda, Colloid Polym. Sci. 277 (1999) 34.[9] H. Kunieda, N. Kanei, I. Tobita, K. Kitahara, A. Yuki, Colloid Polym.

Sci. 273 (1995) 584.[10] Md.H. Uddin, Y. Yamasita, H. Furukawa, A. Harashima, H. Kunieda,

Prog. Colloid Polym. Sci. 123 (2004) 269.[11] L.K. Shrestha, K. Masaya, T. Sato, D.P. Acharya, T. Iwanaga, H. Kunieda,

Langmuir 22 (2006) 1449.[12] L.K. Shrestha, T. Sato, D.P. Acharya, T. Iwanaga, K. Aramaki, H. Ku-

nieda, J. Phys. Chem. B 110 (2006) 12266.[13] C. Rodriguez, Md.H. Uddin, K. Watanabe, H. Furukawa, A. Harashima,

H. Kunieda, J. Phys. Chem. B 106 (2002) 22.[14] H. Rehage, H. Hoffmann, J. Phys. Chem. 92 (1988) 4712.[15] F. Kern, P. Lemarechal, S.J. Candau, M.E. Cates, Langmuir 8 (1992) 437.[16] T.M. Clausen, P.K. Vinson, J.R. Minter, H.T. Davis, Y. Talmon, W.G.

Miller, J. Phys. Chem. 96 (1992) 474.[17] T. Sato, D.P. Acharya, M. Kaneko, K. Aramaki, Y. Singh, M. Ishitobi,

H. Kunieda, J. Dispers. Sci. Technol. 27 (2006) 611.

[18] D.P. Acharya, S.C. Sharma, C. Rodriguez, K. Aramaki, J. Phys. Chem.B 110 (2006) 20224.

[19] H. Hoffman, A. Rauscher, M. Gradzielski, S.F. Schulz, Langmuir 8 (1992)2140.

[20] C. Moitzi, N. Freiberger, O. Glatter, J. Phys. Chem. B 109 (2005) 16161.[21] Z. Lin, J.J. Cai, L.E. Scriven, H.T. Davis, J. Phys. Chem. 98 (1994) 5984.[22] R. Granek, M.E. Cates, J. Chem. Phys. 96 (1992) 4758.[23] R. Oda, J. Narayanan, P.A. Hassan, C. Manohar, R.A. Salkar, F. Kern, S.J.

Candau, Langmuir 14 (1998) 4364.[24] W.J. Kim, S.M. Yang, M. Kim, J. Colloid Interface Sci. 194 (1997) 108.[25] J.H. Mu, G.Z. Li, X.L. Jia, H.X. Wang, G.Y. Zhang, J. Phys. Chem. B 106

(2002) 11685.[26] T. Sikata, H. Hirata, Langmuir 5 (1989) 398.[27] T. Imae, J. Phys. Chem. 94 (1990) 5953.[28] H. Kunieda, C. Rodriguez, Y. Tanaka, Md.H. Kabir, M. Ishitobi, Colloids

Surf. B 38 (2004) 127.[29] B.A. Schubert, E.W. Kaler, N.J. Wanger, Langmuir 19 (2003) 4079.[30] S.R. Raghavan, G. Fritz, E.W. Kaler, Langmuir 18 (2002) 3797.[31] D.F. Evans, H. Wennerstrom, The Colloidal Domain: Where Physics,

Chemistry, and Biology Meet, Wiley–VCH, New York, 1999.[32] L.J. Magid, Z. Li, P.D. Butler, Langmuir 16 (2000) 10028.[33] P.A. Hassan, S.R. Raghavan, E.W. Kaler, Langmuir 18 (2002) 2543.[34] G.C. Kalur, S.R. Raghavan, J. Phys. Chem. B 109 (2005) 8599.[35] A. Maestro, D.P. Acharya, H. Furukawa, J.M. Gutierrez, M.A. Lopez-

Quintela, M. Ishitobi, H. Kunieda, J. Phys. Chem. B 108 (2004) 14009.[36] A. Khatory, F. Kern, F. Lequeux, J. Appell, G. Porte, N. Morie, A. Ott,

W. Urbach, Langmuir 9 (1993) 933.[37] A. Khatory, F. Lequeux, F. Kern, S.J. Candau, Langmuir 9 (1993) 1456.[38] Z. Lin, Langmuir 12 (1996) 1729.[39] M.E. Cates, S.J. Candau, J. Phys. Condens. Matter 2 (1990) 6869.[40] S.R. Raghavan, E.W. Kaler, Langmuir 17 (2001) 300.[41] G.C. Kalur, B.D. Frounfelker, B.H. Cipriano, A.I. Norman, S.R. Ragha-

van, Langmuir 21 (2005) 10998.[42] S. Engelskirkchen, D.P. Acharya, M. Garcia-Roman, H. Kunieda, Colloids

Surf. A 279 (2006) 113.[43] S.R. Raghavan, H. Edlund, E.W. Kaler, Langmuir 18 (2002) 1056.[44] M.E. Cates, J. Phys. Fr. 49 (1988) 1593.