Phase Behavior and Micellar Properties of Carboxylic Acid EndGroup Modified Pluronic Surfactants

Johan P. A. Custers, Leo J. P. van den Broeke,* and Jos T. F. Keurentjes

Process DeVelopment Group, EindhoVen UniVersity of Technology, P.O. Box 513,5600 MB EindhoVen, The Netherlands

ReceiVed June 8, 2007. In Final Form: September 28, 2007

The micellar behavior of three different carboxylic acid end standing (CAE) surfactants has been characterized usingconductometry, differential scanning calorimetry, isothermal titration calorimetry, and dynamic light scattering. TheCAE surfactants are modified high molecular weight Pluronic (PEO-PPO-PEO triblock copolymer) surfactants. Theinfluence of pH and salt additives on the critical micellization temperature (CMT) and the cloud point of the CAEsurfactants have been studied. Both the CMT and the cloud points of the CAE surfactants increase as a function ofpH and decrease as a function of ionic strength. For the CAE surfactants, the CMT varies by about 5°C, and thecloud point shows a variation in the order of 20-30 °C, as compared to the unmodified Pluronics. From the differentexperimental techniques, it follows that at low pH values (pH< 3.5), the CAE surfactants show the same micellarbehavior as the unmodified Pluronic, while at high pH values (pH> 6), the micellar properties of the CAE surfactantsare considerably different from those observed for the corresponding Pluronic. It has been demonstrated that the CAEmicelles are capable of removing simultaneously divalent ions and phenanthrane. The CAE surfactants are the firstknown anionic surfactants that show cloud point behavior with the addition of low concentrations of simple salts,such as, for example, NaCl.

Introduction

Simultaneous removal of ions and organics is often desiredin water purification applications because organics often ac-cumulate and lead to a reduced performance (e.g., in ionexchangers or membrane systems). Typically, industrial sitesare contaminated with heavy metals as well as organics,which could cause a threat to the environment and publichealth.1-3

Recently, a new class of surfactants has been developed thatcan remove both cations and small organics (i.e., carboxylic acidend standing (CAE) surfactants).4 These surfactants are syn-thesized by a reaction of Pluronic surfactants (poly(ethyleneoxide)-poly(propylene oxide)-poly(ethylene oxide) blockcopolymers)5-7 with succinic anhydride to obtain high molecularweight anionic surfactants. The main feature is that these anionicCAE surfactants have a temperature-dependent micellizationtransition. Aggregates of these CAE surfactants are capable ofbinding multivalent ions above the critical micellization tem-perature (CMT) as well as organics in the micellar core, whilebelow the CMT, no binding occurs to the unimers. This featuremakes the CAE surfactant interesting in many (separation)processes (e.g., as a temperature-dependent ion exchanger or inthe removal of ions and organics simultaneously). In a previousstudy, the binding of cations to CAE-85 micelles has alreadybeen reported.4

Several papersareavailable thataddressendgroupmodificationof Pluronics,8-12 but most of them are related to the extensionof the Pluronics with a (charged) polymeric block. Only a fewpapers mention the modification with a small functional group.Although Pluronic surfactants have a high molecular weight, itis to be expected that modification of these polymers by attachingrelatively small ionic extensions to the end groups will result ina considerable change of the phase behavior. In general, thereis a large influence of the solution properties, such as ionicstrength, pH, and type of electrolyte, on the phase behavior ofsurfactants.13-16Different techniques have been described in theliterature to study the phase behavior of (un)modified Pluronicsurfactants.17-21

In this work, the phase behavior of different CAE surfactants,CAE-85, CAE-64, and CAE-81 (i.e., modified Pluronic P85,L64, and L81), has been studied. The CMT was investigatedusing conductometry. Differential scanning calorimetry (DSC)was used to study the dependence of the CMT on different saltadditives and on the pH. Cloud point (CP) measurements havebeen performed for two of the CAE surfactants, using different

* To whom correspondence should be addressed. E-mail: L.J.P.van.den.broeke@tue.nl.

(1) Dunn, R. O., Jr.; Scamehorn, J. F.; Christian, S. D.Colloids Surf.1989,35, 49-56.

(2) Alpatova, A.; Verbych, S.; Bryk, M.; Nigmatullin, R.; Hilal, N.Sep. Purif.Technol.2004, 40, 155-162.

(3) Maturi, K.; Reddy, K. R.Chemosphere2006, 63, 1022-1031.(4) Custers, J. P. A.; Kelemen, P.; Van den Broeke, L. J. P.; Cohen Stuart, M.

A.; Keurentjes, J. T. F.J. Am. Chem. Soc.2005, 127, 1594-1595.(5) Alexandridis, P.; Hatton, T. A.Colloids Surf., A1995, 96, 1-46.(6) Alexandridis, P.; Holzwarth, J. F.; Hatton, T. A.Macromolecules1994,

27, 2414-2425.(7) Pandya, K.; Bahadur, P.; Nagar, T. N.; Bahadur, A.Colloids Surf., A1993,

70, 219-227.

(8) Xiong, X. Y.; Tam, K. C.; Gan, L. H.J. Nano Technol.2006, 6, 2638-2650.

(9) Huang, K.; Lee, B. P.; Ingram, D. R.; Messersmith, P. B.Biomacromolecules2002, 3, 397-406.

(10) Su, Y.-L.; Li, C.React. Funct. Polym. 2007, 67, 233-240.(11) Bromberg, L. E.; Barr, D. P.Macromolecules1999, 32, 3649-3657.(12) Bromberg, L.Ind. Eng. Chem. Res.2001, 40, 2437-2444.(13) Pandya, K.; Lad, K.; Bahadur, P.Pure Appl. Chem.1993, 30, 1-18.(14) Bahadur, P.; Pandya, K.; Almgren, M.; Li, P.; Stilbs, P.Colloid Polym.

Sci.1993, 271, 657-667.(15) Desai, P. R.; Jain, N. J.; Sharma, R. K.; Bahadur, P.Colloids Surf., A

2001, 178, 57-69.(16) Anderson, B. C.; Cox, S. M.; Ambardekar, A. V.; Mallapragada, K.J.

Pharm. Sci.2002, 91, 180-188.(17) Alexandridis, P.; Holzwarth, J. F.Langmuir1997, 13, 6074-6082.(18) Armstrong, J. K.; Chowdhry, B. Z.; Snowden, M. J.; Leharne, S. A.

Langmuir1998, 14, 2004-2010.(19) Zhang, K.; Khan, A.Macromolecules1995, 28, 3807-3812.(20) Glatter, O.; Scherf, G.; Schillen, K.; Brown, W.Macromolecules1994,

27, 6046-6054.(21) Bordi, F.; Cametti, C.; Di Biasio, A.J. Phys. Chem.1988, 92, 4772-

4777.

12857Langmuir2007,23, 12857-12863

10.1021/la701697h CCC: $37.00 © 2007 American Chemical SocietyPublished on Web 11/20/2007

salt additives and different pH values. Dynamic light scattering(DLS) was used to investigate the effect of the surfactantmodification on the size of the micelles. Isothermal titrationcalorimetry (ITC) was used to study mixed (nonionic-ionic)micelles of Pluronic and CAE surfactants. Also, a study wasperformed on the simultaneous removal of phenanthrene andcalcium ions by CAE surfactants.

Experimental Procedures

CAE Surfactants. The carboxylic acid end standing (CAE-85)surfactant was synthesized by modifying the Pluronic P85 surfactant(EO39-PO52-EO39) with succinic anhydride to obtain a highmolecular weight anionic surfactant with temperature sensitivemicellization behavior1 (see Figure 1).

Three CAE surfactants were synthesized using Pluronic P85, L64,and L81 as the starting material. P85 is a paste at 25°C, while L81and L64 are liquids at room temperature. The difference in physicalstate induces minor changes in the synthesis procedure of the threedifferent CAE surfactants.

In Table 1, the number of ethylene oxide (EO) groups,p, thenumber of propylene oxide (PO) groups,q, and the molecular weightof the CAE surfactants are given. The molecular weights of thePluronic surfactants were taken from literature,5,6and the molecularweights of CAE surfactants were previously determined by GPC.The degree of polydispersity of the CAE surfactants was around 1.5.

Preparation of CAE Surfactant Solutions. The CAE-85solutions were prepared by dissolving an amount of CAE-85 intodemiwater, which always had a conductivity value below 3µS/cm.The solutions were stirred with a magnetic strirrer until CAE-85 haddissolved (ca. 30 min). In all the different experiments, the CAE-85solutions were freshly made. When necessary, the pH of the surfactantsolutions was adjusted to a higher value using NaOH and to a lowervalue using HCl.

Conductometric Analysis.All conductometric experiments wereperformed using a QiS M320 conductometer with a QiS QC203Tepoxy/graphite conductivity cell. The cell constant was periodicallychecked and calibrated by a 0.01 M KCl solution. The solutiontemperature was kept constant ((0.1 °C) with a Lauda C6 CPthermostat.

DSC.Experiments were performed using a Pyris Diamond DSC,type APP010, from PerkinElmer Instruments. The temperature wascalibrated between 0 and 200°C at a rate of 20°C/min using indiumas a calibration standard with a melting point of 157°C. Aluminumpans of 50µL with caps containing the sample or the referencematerial, usually water, were used. For each sample, multipletemperature scans were made from 5 to 60°C with a scanning rateof 20 °C/min. The determination of the onset temperature (CMT)was performed using software from PerkinElmer Series.

CP Measurements.The clouding behavior of the CAE surfactantswas determined by slow heating ((0.2 °C/min) of 5 mL samples

using a Lauda C6 CP thermostat ((0.1 °C). At each temperature,the samples were equilibrated before further heating. The first steadysign of turbidity, by visual detection, was taken as the CP. A watersample was used as a reference. The standard deviation of the visualCP detection was determined to be less than 0.5°C for the systemsmeasured. In general, it can be said that the accuracy of the CPmeasurements is lower for the CAE surfactants, as compared toPluronic surfactants, because the CAE surfactants have a less sharpCP transition.

ITC. Experiments were performed using a MicroCal VP-ITCapparatus with a cell volume of 1.4431 mL. The experimentalprocedure consisted of adding 70 injections of 4µL of a surfactantsolution (P85 or CAE-85) to the sample cell. In all experiments, thereference cell was filled with demineralized water, having aconductivity value smaller than 3µS/cm. Before each experiment,the temperature was equilibrated at the desired value. In the mixedmicelle experiments, the surfactant concentration in the sample cellwas 2.16× 10-3 mol/L for CAE-85 and 2.14× 10-3 mol/L for P85.The molar concentrations were calculated from the molar mass forthe different surfactants, as given in Table 1. The syringe surfactantconcentration was 16.7× 10-3 mol/L for the CAE-85 solution and16.85 × 10-3 mol/L for the P85 solution. The concentration ofcalcium chloride in the sample cell was 2.25× 10-3 mol/L.

DLS. DLS experiments were performed with a Coulter N4 plusSubmicron Particle Sizer apparatus. The scattered light was correlatedwith a size distribution program (SDP) using the CONTIN routine.All measurements were performed at a 90° scattering angle. Solutionswith a concentration of 1 wt % surfactant were used, and allexperiments were performed in triplicate. In general, it can be saidthat the accuracy of the measurements was lower for the CAEsurfactants as compared to the Pluronic surfactants. The standarddeviation of the DLS measurements for the P85 solutions and CAE-85 solutions at low pH values was less than 3%; however, for highpH CAE-85 solutions, the standard deviation was less than 10%.

UV-vis Spectroscopy.Phenanthrene was chosen as a modelcompound to study the simultaneous removal of organics and cationsby the CAE micelles. Phenanthrene has good UV absorbanceproperties and dissolves very well into Pluronic surfactant solutions.The concentration of phenanthrene in the surfactant solutions wasmeasured using UV-vis (Uvikon XL from Bio-Tek Instruments,Inc.). In the UV spectrum of phenanthrene, four peaks can be observed(i.e., at wavelengths of 325, 332, 340, and 348 nm). The maximumUV absorption takes place at 332 nm, and therefore, this wavelengthwas chosen to determine the concentration of phenanthrene. Theinfluence of the surfactant concentration on the UV absorption ofa fixed amount of phenanthrene was investigated, and the influencewas negligible. The path length in the quartz cell was 10 mm, andthe temperature was 50°C.

For the solubilization experiments, different solutions wereprepared by adding phenanthrene to the CAE-85 surfactant or Pluronicsurfactant solution at temperatures of 25°C (below the CMT) and50 °C (above the CMT). The solutions were stirred at 50°C for 24h to reach equilibrium conditions. Solid particles that were notdissolved after 24 h were left to precipitate, and only the clear solutionwas used in the experiments. At 25°C, no detectable amount ofphenanthrene was dissolved in a CAE surfactant solution after 24h of stirring. It was observed that the UV absorbance of phenanthrenein a 10 wt % P85 solution, using an excess micelle concentration,showed a linear dependence on the phenanthrene concentration forthe range studied. The solubility of phenanthrene in water,Sw, isabout 1.0× 10-3 g/L (UV absorption, 0.01 au).37,38

Results and Discussion

Dependence of CMT on pH and Additives.In low molecularionic surfactant systems, such as sodium dodecyl sulfate (SDS),conductometry is a well-proven technique to determine the criticalmicellization concentration (CMC).22 For low molecular ionic

(22) Gurau, M. C.; Lpim, S.-M.; Castellana, E. T.; Albertorio, F.; Kataoka,S.; Cremer, P. S.J. Am. Chem. Soc.2004, 126, 10522-10523.

Figure 1. Chemical structure of CAE surfactant. The values ofpandq are given in Table 1.

Table 1. Molecular Compositions of Different Surfactants andMeasured CPs for Pluronics P85, L64, and L81 Solutions and

for CAE Surfactants CAE-85, CAE-64, and CAE-81 Solutions atDifferent Concentrations

composition CP (°C)

surfactant p q Mw (g/mol) 0.1 wt % 1 wt %

P85 26 39 4548 85CAE-85 26 39 4768 81 60L64 13 30 2918 58CAE-64 13 30 3102 25L81 3 43 2792 20CAE-81 3 43 3039 30 8

12858 Langmuir, Vol. 23, No. 26, 2007 Custers et al.

surfactant systems, the slope of the conductivity as a functionof the surfactant concentration above the CMC is lower than theslope below the CMC. As a result of counterion binding abovethe CMC, the conductivity per surfactant molecule is reduced.This method can also be used to determine the CMT for CAEsurfactants, but the conductivity should be plotted as functionof temperature instead of the surfactant concentration.

In Figure 2, the total conductivity for 0.5 wt % solutions ofCAE-85, CAE-64, and CAE-81 is plotted as a function oftemperature. For CAE-85 and CAE-64, a change in slope isobserved, which indicates the presence of a CMT. For the CAE-85 surfactant, this is at about 34°C, and for the CAE-64 surfactant,this is at about 35°C. In both cases, the CMT is about 1-2 °Chigher, as compared to the CMT of the corresponding Pluronicsurfactants P85 (32°C) and L64 (33°C).5,6 This increase canbe explained by the higher solubility of the modified (CAE)surfactants caused by the presence of the ionic charge.

The change in slope of the conductivity curve observed forCAE-85 and CAE-64 follows from a linear fit of the conductivitycurve below the CMT and above the CMT. Each of the linearfits of the two separate parts is significantly better than the linearfit of the total curve. On the other hand, for the CAE-81 surfactant,almost no change in the slope can be observed. The ratio of theslope of the upper part to the slope of the lower part for CAE-85and CAE-64 is about 1.4. For CAE-81, the ratio of the slope ofthe upper part to the slope of the lower part of the conductivitycurves is about 1.1. Furthermore, it is noted that the slope of theconductivity of CAE-81 is the same as the slope obtained forCAE-85 and CAE-64 below the CMT. It is therefore concludedthat CAE-81 has no CMT in the temperature range investigated.

As a second technique, DSC has been used to measure theCMT of a CAE surfactant solution. For Pluronic-based surfactants,the micellization enthalpy is relatively low, and therefore, a highaccuracy is required. For this reason, it was not possible to measurethe CMT of the CAE-64 surfactant. In Figure 3, the CMT ofCAE-85 is given as a function of pH. From titration curves, apKa value of 4.51 has been determined for CAE-85.4 At low pHvalues, the carboxylic acid groups of CAE-85 are protonated,and the CAE-85 surfactant can be considered as nonionic. Withan increase in pH, the carboxylic acid groups will deprotonate,and the CAE-85 surfactant head groups will become anionic.The anionic head groups are more hydrophilic because of thecharge, with the result that the solubility and the CMT of the

CAE-85 surfactant both increase. This increase of CMT continuesuntil the full deprotonation point at a pH of 7 is reached.

The influence of different electrolytes on the CMT of CAE-85at low pH values (pH 3.5) is shown in Figure 4a. It follows thatthe CMT decrease or the salting-out effect on CAE-85 followsthe order NaCl> BaCl2 > CaCl2. This is opposed to what is tobe expected from the Hofmeister series.23-28 It is noted that the

(23) Bostrom, M.; Williams, D. R. M.; Ninham, B. W.Langmuir2002, 18,6010-6014.

(24) Bostrom, M.; Williams, D. R. M.; Stewart, P. R.; Ninham, B. W.Phys.ReV. E 2003, 68, 41902.

Figure 2. Conductivity as a function of temperature for 0.5 wt %CAE-85, 0.5 wt % CAE-64, and 0.5 wt % CAE-81 solutions witha pH of∼6. The CMT is given by the intersection of the slopes atlow and high temperatures.

Figure 3. DSC results for a 1 wt %CAE-85 solution as functionof pH.

Figure 4. (a) DSC results for the CMT of P85 and CAE-85 at apH value of 3.5 as a function of ionic strength with CaCl2, BaCl2,and NaCl as the additional electrolytes. (b) DSC results for the CMTof P85 and CAE-85 at a pH value of 10.0 as a function of low ionicstrength with CaCl2 and NaCl as additional electrolytes.

Carboxylic Acid End Group Modified Surfactants Langmuir, Vol. 23, No. 26, 200712859

anion, which is chloride in all cases, has a substantially largerinfluence on the salting out mechanism of the surfactant than thecation.29-31 However, in Figure 4, the main difference is causedby the cations, which interact with the ether groups of the polymersurfactant,12,13 and where divalent cations have a strongerinteraction than monovalent cations.13 This means that thedecrease in ionic strength is larger for divalent ions, with theresult that in the case of Pluronic-based surfactants, the salting-out behavior of divalent cations will be weaker as compared tothe salting-out behavior of monovalent cations.

In Figure 4b, results for the CMT at high pH (pH 10.0) areshown. From a comparison with Figure 4a, it follows that theresults obtained at a higher pH show the same behavior as obtainedfor a low pH. In both cases, the CMT decreases with an increasein ionic strength. The decrease of the CMT in the presence ofNaCl is larger than the decrease of the CMT in the presence ofCaCl2. However, when the behavior of the CMT is examined atlow ionic strength and with a pH equal to 10.0, a clear differencebetween CAE-85 and P85 can be seen. The CMT of P85 showsa linear dependence on the ionic strength. In the CAE-85 system,a clear transition is seen: initially the CMT shows a sharpdecreasing behavior, followed by a slope that resembles the slopeof the P85 system. The shielding of the carboxylic charges ofthe surfactant by the calcium ions makes the head group morehydrophobic, resulting in a larger decrease of the CMT ascompared to the P85 case. At the ionic strength where all chargesare screened (i.e., at 2.5× 10-3 mol/L), CAE-85 behaves in asimilar way as P85. The calculated value, 2.5× 10-3 mol/L, isbased on the amount of carboxylic acid groups determined frompotentiometric titration.4

Dependence of CP on pH and Additives.Only for a fewionic surfactants, in particular, for anionic surfactants with alarge hydrophobic cation, a CP has been reported.32,33 On theother hand, most nonionic surfactants with PEO blocks, includingthe Pluronic surfactants, display a CP. The behavior of the CAEsurfactants can be changed from nonionic to anionic by increasingthe pH, and in this respect, applications of CAE surfactants relatedto the CP seem promising. The presence of anionic carboxylategroups will have a clear influence on the clouding behavior ofthe Pluronic-based surfactants. In Table 1, results for CPmeasurements are given for both the CAE surfactants and theunmodified Pluronics at low pH (pH 3.5) (i.e. in the protonatedform). For each CAE surfactant, a substantial decrease in CP isobserved, which is also highly dependent on the actual surfactantconcentration. Contrary to the small difference between the CMTobserved for the CAE and the Pluronic surfactants, a considerabledifference is obtained between the CP for the CAE and thePluronic surfactants.

This difference in CMT and CP can be explained by consideringtheir mechanisms. The PPO block mainly determines themicellization behavior with the consequence that modificationof the P85 surfactant has almost no effect on the micellizationbehavior.5 The clouding behavior is largely determined by thetype of head group of the surfactant, and in turn, this determinesthe intermicellar interactions. Modification with succinic an-

hydride makes the head group more hydrophobic at low pH,resulting in a considerable decrease in CP.

The CP of the CAE-85 surfactant as a function of pH is shownin Figure 5. For all cases, a large increase in CP is observed whenthe pH is increased, which is a similar behavior as observed forthe CMT (see Figure 3). For the case with no additional electrolyte,the CP vanishes when the surfactant becomes anionic. However,a CP depression can be obtained by adding salts to the surfactantsolution.5 Screening of the charged head group will reduce therepulsive electrostatic forces between micelles, resulting in adecrease in CP. Contrary to the behavior observed for the CMTwith salt additives, the CP depression follows the Hofmeisterseries (i.e., CaCl2 > BaCl2 > NaCl). For the case of 0.5 M NaCl,a small deviation from this behavior is observed at a pH of 3.5.In the case of NaCl at a pH value of 3.5 (a 1.0 wt % CAE-85surfactant solution has a pH of 3.5; to obtain other pH values,the pH is adjusted by adding NaOH or HCl, which means extracations), a larger CP depression is observed as compared to otherpH values. The explanation here can be found in the competitionof cations. When it is assumed that cations will have someinteraction with the ether groups of the surfactant molecules,7,13

the extra cations will be located near the ether groups. Bariumand calcium ions experience some competition from the extraH+ and Na+ in the cases where the pH is adjusted, which makesit more difficult for barium and calcium to bind to the ethergroups. Therefore, the (free) concentration of barium and calciumis higher, and consequently, a larger CP depression is observedfor the solutions with a pH of 2.4 and a pH of 10.0. For thesituation with a pH value of 3.55, barium and calcium cationswill bind more successfully to the surfactant chain, and therefore,the decrease of the CP will be less pronounced as compared tothe presence of sodium chloride. As a result, a reversed trendas compared to the Hofmeister series is obtained at pH 3.5.

The hydrophobic Pluronic L81 has a low CP of about 20°C.In this case, the unimer surfactants will not form micelles butwill immediately separate into two phases at the CP.5 Aftermodification of L81 to CAE-81, no CMT was observed, whichwas confirmed by conductivity measurements (see Figure 2).From Table 1, it follows that for hydrophobic CAE-81, themodification has a considerable influence on the cloudingbehavior. It has been determined that at pH 3.5, the nonionicCAE-81 has a CP of 8°C, and at a pH 6.0, the CP has disappeared.The CP decreases significantly in the presence of the differentsalts, already at low ionic strength as shown in Figure 6. Again,these results follow the Hofmeister series, which explains thelarge decrease of the CP with the addition of MgSO4. In thiscase, both the cation and the anion are good kosmotropes.

(25) Bostrom, M.; Craig, V. S. J.; Albion, R.; Williams, D. R. M.; Ninham,B. W. J. Phys. Chem. B2003, 107, 2875-2878.

(26) Bauduin, P.; Renoncourt, A.; Touraud, D.; Kunz, W.; Ninham, B. W.Curr. Opin. Colloid Interface Sci.2004, 9, 43-47.

(27) Bauduin, P.; Renoncourt, A.; Touraud, D.; Kunz, W.; Ninham, B. W.Curr. Opin. Colloid Interface Sci.2004, 9, 43-47.

(28) Omta, A. W.; Kropman, M. F.; Woutersen, S.; Bakker, H. J.Science(Washington, DC, U.S.)2003, 301, 347-349.

(29) Larsen, J. W.; Magid, L. J.J. Am. Chem. Soc.1974, 96, 5774-5782.(30) Chai, J.-L.; Mu, J.-H.Colloid J. 2002, 64, 550-555.(31) Jain, N. J.; George, A.; Bahadur, P.Colloids Surf., A1999, 157, 275-

283.

Figure 5. CP results for a 1 wt %CAE-85 solution as function ofpH with NaCl, CaCl2, and BaCl2 as salt additives.

12860 Langmuir, Vol. 23, No. 26, 2007 Custers et al.

The results of the CP measurements for CAE-85 and P85 aredepicted in Figure 7. CAE-85 seems to be less sensitive for CPdepression as compared to CAE-81. This is the result of the morehydrophilic nature of the CAE-85 surfactant. From this, it canbe concluded that hydrophilic Pluronic-based surfactants are lesssensitive than the hydrophobic ones for CP depression. Asexplained, the results for CAE-85 and P85 at pH 3.55 do notfollow the Hofmeister series, and it can be noted that not a largedifference is observed between calcium and barium cations. Itcan be concluded that the CP depression is larger for CAEsurfactants as compared to Pluronic surfacants and that the CPof the CAE surfactants is (very) sensitive to the addition of salts,similar to the behavior observed for the Pluronic surfactants. Ithas been shown for the first time that the CAE surfactants inanionic form have the possibility to show clouding behavior,even in the presence of salts such as NaCl.

Determination of Micelle Size with DLS.Micellar aggregatesare dynamic structures because surfactant monomers are ex-changed between the aqueous surrounding and the micelles.Therefore, the micellar aggregates are sensitive to a change insolution properties of the unimers. A considerable amount ofdata are available about the size and structure of Pluronicsurfactants, as a function of various parameters.5 In Figure 8, acomparison is made between the micellar size of P85 and themicellar size of CAE-85 aggregates obtained using DLS. Thediameter of P85 and CAE-85 aggregates is shown as a functionof the calcium chloride concentration. With no calcium chloride,

the diameter of the P85 micelle is about 16.2 nm, which is inagreement with values reported in the literature.5 When only asmall amount of calcium chloride is added to the solution, theP85 micelle shrinks to a shape with a diameter of 13.8 nm. Thedifference in polarity between the PEO head groups of the P85surfactants and the aqueous solution with the calcium cationsincreases. It is known that the size of Pluronic micelles willincrease when the CP is approached.5 The addition of calciumchloride stimulates the dehydration of the surfactant unimersand, therefore, depresses the CP. As a result, the size of the P85micelles increases with an increase in calcium concentration toabout 16.7 nm at a calcium chloride concentration of 0.4 mol/L.The results for the diameter of the CAE-85 micelles at 20°C,which is below the CMT, and at 50°C, which is above the CMT,are measured as well. At 20°C, no CAE-85 micelles could bedetected. At 50°C, the diameter of the CAE-85 micelle is about15 nm, independent of the pH. The result for the CAE-85 micelleat a pH of 3.5 is consistent with the results for the P85 micelles(see Figure 8).

Mixed Micelle Interactions Studied with ITC. ITC is oftenused to measure interactions between surfactants in mixedmicellar systems.34-36 Here, mixed (nonionic-ionic) micellesof P85 and CAE-85 were studied using ITC. In Figure 9a, theresults for the titration of a P85 surfactant solution to a solutionof P85 micelles and to a solution of CAE-85 micelles are shown.The molar ratioz is defined as the amount of added mol ofsurfactant per mol of surfactant in the sample cell before the startof the injections.

For the P85 solution titrated with P85, a small exothermic heateffect is observed. This is caused by the dilution of a concentratedP85 solution, from the syringe, in the sample cell, which containsa solution with a lower concentration P85 solution. This heateffect becomes smaller during the titration because the concen-trations of P85 in the sample cell solution and in the syringesolution converge to the same value. For the case that aconcentrated P85 solution is titrated to a CAE-85 surfactant

(32) Moroi, Y.; Yoshida, N.Langmuir1997, 13, 3909-3912.(33) Kumar, S.; Sharma, D.; Khan, Z. A.; Kabir-ud-DinLangmuir2002, 18,

4205-4209.(34) Zana, R.; Benrraou, M.; Bales, B. L.J. Phys. Chem. B2004, 108, 18195-

18203.(35) Couderc, S.; Li, Y.; Bloor, D. M.; Holzwarth, J. F.; Wyn-Jones, E.Langmuir

2001, 17, 4818-4824.(36) Li, X.; Wettig, S. D.; Verrall, R. E.Langmuir2004, 20, 579-586.(37) Fetterolf, G. J. Characterization of a Creosote-Contaminated Tie Yard

Site and the Effects of Phytoremediation. Ph.D. Thesis, Virginia PolytechnicInstitute and State University, Blacksburg, VA, 1998.

(38) Shin, K.-H.; Kim, K.-W.; Seagren, E. A.Appl. Microbiol. Biotechnol.2004, 65, 336-343.

Figure 6. CP results for CAE-81 at a pH value of∼6 as a functionof ionic strength with NaCl, CaCl2, and MgSO4 as additionalelectrolytes.

Figure 7. CP results for P85 and CAE-85 at a pH value of 3.55as a function of ionic strength with CaCl2, BaCl2, and NaCl asadditional electrolytes.

Figure 8. Diameter of the P85 micelle as a function of the addedcalcium chloride concentration obtained with DLS.

Carboxylic Acid End Group Modified Surfactants Langmuir, Vol. 23, No. 26, 200712861

solution at pH 3.5, almost the same heat change curve is observedas for the case where the concentrated P85 solution has beentitrated to a P85 solution. It can be concluded that the behaviorof the modified CAE-85 surfactant at low pH resembles thebehavior of the P85 surfactant. A larger difference is seen whenthe P85 solution is titrated to a CAE-85 solution at pH 11.0. Athigh pH, the end groups of CAE-85 are deprotonated, and aconsiderable difference exists between CAE-85 and P85 sur-factant. As a result of this difference, the formation of mixedmicelles is less favorable (i.e., endothermic).

Initially, the P85 surfactant unimers will be incorporated inthe CAE-85 micelle without affecting the micellar structure.This results in a constant heat change, until a molar ratio,z, of0.4. At this point, micelles have been formed that containpredominantly P85, and the heat change begins to decrease untilno net heat effect will be observed. For the final situation, thesolution in the sample cell and the solution in the syringe willconverge to the same composition. From the results shown inFigure 9a, it can be concluded that the intermicellar interactionsincrease considerably when the end groups of the Pluronic-basedsurfactant are charged.

In Figure 9b, mixed micelles of P85 and CAE-85 are studiedin the presence of calcium. For the two experiments, a referenceexperiment was performed without calcium chloride, and thesedata were subtracted from the heat change data for the experimentswith calcium chloride. The heat effects shown in Figure 9b canbe attributed solely to the presence of calcium chloride. An

exothermic heat effect is seen when a solution of P85 is titratedto a solution of CAE-85 with calcium. This exothermic effectis a result of the release of the calcium from the CAE-85 micellesbecause the hydration enthalpy of calcium is exothermic. At acertain point, all calcium cations are released while the P85unimers begin to dominate the composition of the mixed micelles.For the situation where CAE-85 is titrated to a solution of P85with calcium chloride, initially no calcium will be bound. Withthe increase in CAE-85 surfactant concentration, calcium willhave an increasing electrostatic interaction with the mixedmicelles. This is an endothermic effect because the dehydrationof the calcium cation is the dominant contribution. At a certainpoint, calcium will no longer contribute to the measured heateffect because all calcium ions are bound to CAE-85. Theformation of the CAE-85 micelles will become the dominantfactor for the measured heat effect, which again converges tozero.

Comparing Figure 9a,b it follows that below a molar ratio,z,of about 0.4, the observed behavior is predominantly determinedby mixed micelles, while at higher values ofz, the contributionof the homogeneous micelles becomes the dominate behavior.

Solubilization of Phenanthrene.To study the simultaneousremoval of metal ions and organic compounds and the effect ofthe surfactant modification on the solubilization behavior,phenanthrene was used as the solute. In Figure 10, the resultsare shown for the maximum solubility of phenanthrene in 1.0wt % P85 and CAE-85 surfactant solutions.

For the P85 solution, it can be observed that the UV absorptionlevels off at a phenanthrene concentration of about 0.95 g/L. Theformation of precipitates is a second indication that the uppersolubility limit was reached. This means that the maximumsolubility of phenanthrene in a 1 wt % P85solutuion,Smic, isequal to 0.95 g/L, which corresponds to a solubility ratio,â(Smic/Sw), of 950. In the presence of calcium chloride, nosignificant difference in UV absorption for P85 is noticed (seeFigure 10). For the 1.0 wt % CAE-85 solution, the UV absorptionlevels off at a phenanthrene concentration of about 0.90 g/L,which corresponds to a solubility ratio,â, equal to 900, a valueslightly lower than for the P85 solution. There can be twoexplanations for the slightly lowerâ observed for the CAE-85surfactant solution. First, in a 1.0 wt % P85 solution, there aremore surfactant molecules present than in a 1.0 wt % CAE-85solution, caused by the slightly lower molecular weight for P85(4568 g/mol) as compared to CAE-85 (4768 g/mol). Second, theformation of micelles is energetically less favored when thesurfactant has charged end groups. Overall, better defined micelles

Figure 9. (a) ITC heat change data for solution of P85 and CAE-85at pH 3.5 titrated with P85 at 50°C and for CAE-85 micelles at pH11.0 titrated with P85 at 50°C, as a function of the molar ratio,z.(b) ITC heat change data for solutions of P85 and CAE-85 at pH11.0 and in the presence of calcium chloride (using a ratio of molarconcentration calcium to molar concentration carboxylic acid groupsof 0.5) at 50°C, as a function of the molar ratio,z.

Figure 10. Maximum solubility UV results for phenanthrene in a1.0 wt % P85 solution and in a 1.0 wt % CAE-85 solution at 50°Cwith and without addition of CaCl2.

12862 Langmuir, Vol. 23, No. 26, 2007 Custers et al.

with a higher solubilization capacity will be formed in P85surfactant solutions.

In Figure 11, the effect of the pH of a 1.0 wt % CAE-85solution on the maximum phenanthrene concentration is shown.It can be observed that there is a quite strong decrease in maximumsolubility when the pH of the solution is increased.38 Again, itcan be concluded that the charged end groups of the CAE-85surfactant form less structured micellar aggregates than the P85surfactants. However, when the charged end groups are shieldedor covered by cations, such as calcium, the micelles becomemore structured, and the solubilization capacity of the micellesincreases. This is also shown in Figure 11, where a highersolubility for phenanthrene is observed with an increasing calciumchloride concentration.

It can be concluded that the charge density of the micelle isan important factor in the solubilization capacity for phenanthrene.Highly charged micelles form less structured aggregates with ahigher energy because of the electrostatic head group repulsion.The shielding of the charged end groups or the binding of cationswill decrease this electrostatic repulsion and increase the capacityto solubilize phenanthrene.

ConclusionThe phase behavior of three different CAE surfactants, CAE-

85, CAE-64, and CAE-81, was studied using various experimentaltechniques, and a comparison was made with the unmodifiedPluronics surfactants. Conductivity and DSC were used todetermine the CMT of CAE-85 and CAE-64. For CAE-81, noCMT could be obtained because of its hydrophobic nature. TheCMT of CAE-85 seems to be slightly affected by pH changes,where the addition of salts has about the same effect on the CMTof CAE-85 and P85. DSC was used to study the effect of pH andadditives on the CMT. A small pH effect was observed for theCMT of CAE surfactants. Contrary to the CMT, the CP changesconsiderably for the CAE surfactants as compared to the Pluronicsurfactants. The CP of the CAE surfactants is very sensitive tochanges in pH because of the carboxylate end group. It can beconcluded that the effect of salts on the CP will increase as thehydrophobicity of the Pluronic-based surfactant increases. DLShas been used to determine the micellar size of the Pluronic-based surfactants. The modification of the Pluronic surfactantshas almost no effect on the micellar diameter.

The solubilization of phenanthrene in CAE-85 micelles isslightly lower for CAE-85 as compared to P85, which can beexplained by the charged end groups of CAE-85. Additionally,the solubilization capacity of CAE-85 is favored by lowering thepH or increasing the calcium chloride concentration becausethen the charge of the end groups will be counterbalanced by theadded cations and electrostatic repulsion between the head groupsof the charged surfactants will be reduced.

By choosing a proper hydrophobic-hydrophilic balance ofthe starting Pluronic, CAE surfactants can be synthesized withdifferent properties, which can be tuned by the solution properties,such as pH and ionic strength. Understanding the phase behaviorof these CAE surfactants, such as the pH dependency, will opennew possibilities for temperature responsive applications. Fur-thermore, the CAE surfactants are the first ionic surfactants thatshow clouding behavior upon the addition of simple acids andsalts such as, for example, HCl or NaCl.

LA701697H

Figure 11. UV absorption results (332 nm) for phenanthrene in a1.0 wt % CAE-85 solution at 50°C with the addition of CaCl2 attwo different pH values.

Carboxylic Acid End Group Modified Surfactants Langmuir, Vol. 23, No. 26, 200712863