Optical Spectroscopic and TEM Studies of CatanionicMicelles of CTAB/SDS and Their Interaction with a NSAID

Hirak Chakraborty and Munna Sarkar*

Chemical Sciences Division, Saha Institute of Nuclear Physics, 1/AF, Bidhannagar,Calcutta-700 064, India

Received November 14, 2003. In Final Form: February 11, 2004

If a vesicle is a better model of a membrane in the context of the hydrophobic effect, then from the chargedistribution point of view, a catanionic micelle is a closer model to a biomembrane. We have prepared andcharacterized two different types of catanionic micelles of sodium dodecyl sulfate (SDS) and cetyl N,N,N-trimethylammonium bromide (CTAB) having different surface charge ratios using optical spectroscopyand transmission electron microscopy. The average size of both types of mixed micelles was found to bemuch larger than that of micelles containing uniformly charged headgroups. Catanionic micelles containinghigher concentrations of positively charged headgroups (CTAB) are larger in size, less compact, and morepolar compared to the micelles containing higher concentrations of negatively charged headgroups (SDS).We have used these catanionic micelles as membrane mimetic systems to understand the interaction ofpiroxicam, a nonsteroidal anti-inflammatory drug (NSAID) of the oxicam group, with biomembranes. Incontinuation of our work on membrane mimetic systems, we have used spectral properties of the drug itselfto understand the effect of the presence of mixed charges on the micellar surface in guiding the interactionof catanionic micelles with piroxicam. Our earlier studies of the interaction of piroxicam with micelleshaving uniform surface charges have shown that the charge on the micellar surface not only dictates whichprototropic form of the drug will be incorporated in the micelles but also induces a switch-over betweendifferent prototropic forms of piroxicam. The equilibrium of this switch-over is extremely sensitive to theenvironment. In this study, we demonstrate how even small changes in the electrostatic forces obtainedby doping the uniformly charged surface of the micelles with oppositely charged headgroups (as in catanionicmicelles) are capable of fine-tuning this equilibrium. This implies that the surface charge of biomembranes,which are quite diverse in vivo, might play a significant role in selecting a particular form of the drug tobe presented to its targets.

Introduction

Micelles, dynamic nanostructures of surfactant mol-ecules, have the capability to solubilize a wide variety oforganic molecules with different polarities and hydro-phobicities. Several reactions, such as polymerization andcatalysis, are carried out in micelles to get a better yield.In these applications, the size and the stability of themicelle play an important role. The tailoring of micellarproperties may be achieved by adding salts, organicsolvents, or a second surfactant forming the so-calledmixed micellar system. The larger size and better ther-modynamic stability of the mixed micelles would enhancethe incorporation capability of solutes in the micellarphase. In this work, we have prepared and characterizedmixed micelles of cationic [cetyl N,N,N-trimethylammo-niumbromide (CTAB)]andanionic [sodiumdodecyl sulfate(SDS)] surfactants (called “catanionic micelles”), in whichtwo oppositely charged headgroups are distributed. Wehave taken two sets of catanionic micelles, one in whichpositive charge is more and another with more negativecharge compared to the other. Competition betweenvarious molecular interactions (van der Waals, hydro-phobic, electrostatic, hydration forces, etc.) may result ina variety of microstructures, viz., catanionic salts, mixedmicelles, and catanionic vesicles.1-3 So the phase behaviorof cationic/anionic surfactant mixtures strongly depends

on the molar ratio, actual concentration of individualsurfactant relative alkyl chain lengths, number of alkylchains per surfactant, and temperature, resulting in arich array of aggregates.2-9 In a particular region ofcomposition, twooppositely chargedsurfactants could formcatanionic micelles.4,6,10 So to prepare catanionic micellesone has to be very cautious about the composition of twooppositely charged surfactants. Generally, mixtures withan excess of either cationic or anionic surfactants havebeen shown to form catanionic micelles. As a result, theratio of two oppositely charged headgroups in thesemicelles is far from 1:1 and polydispersity in the micellarsize and charge distribution occurs.

We have characterized the catanionic micelles by usingoptical spectroscopic techniques. Both steady-state andtime-resolved fluorescence studies were done with pyreneas the reporter chromophore. Transmission electronmicroscopy (TEM) has been used to confirm the formationof catanionic micelles, and the average diameter of thesemicelles was calculated from TEM photographs.

We have also studied the interactions of these mixedmicelles with a nonsteroidal anti-inflammatory drug(NSAID), piroxicam [4-hydroxy-2-methyl-N-(pyridin-2-yl)-

* Corresponding author. Fax: 91-33-23374637. E-mail: munna@nuc.saha.ernet.in or munna@hp1.saha.ernet.in.

(1) Kaler, E. W.; Murthy, A. K.; Rodriguez, B. E.; Zasadzinski, J. A.N. Science 1989, 245, 1371-1374.

(2) Kaler, E. W.; Herrington, K. L.; Murthy, A. K.; Zasadzinski, J.A. N. J. Phys. Chem. 1992, 96, 6698-6707.

(3) Brasher, L. L.; Herrington, K. L.; Kaler, E. W. Langmuir 1995,11, 4267-4277.

(4) Yatcilla, M. T.; Herrington, K. L.; Brasher, L. L.; Kaler, E. W. J.Phys. Chem. 1996, 100, 5874-5879.

(5) Soderman, O.; Herrington, K. L.; Kaler, E. W.; Meller, D. D.Langmuir 1997, 13, 5531-5538.

(6) Tomasic, V.; Stefanic, I.; Filipovic, N. Colloid Polym. Sci. 1999,277, 153-163.

(7) Marques, E. F. Langmuir 2000, 16, 4798-4807.(8) Jung, H. T.; Coldren, B.; Zasadzinski, J. A.; Iampietro, D. J.;

Kaler, E. W. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 1353-1357.(9) Edlund, H.; Sadaghiani, A.; Khan, A. Langmuir 1997, 13, 4953-

4963.(10) Karukstis, K. K.; McCormack, T. M.; McQueen, T. M.; Goto, K.

F. Langmuir 2004, 20, 64-72.

3551Langmuir 2004, 20, 3551-3558

10.1021/la0361417 CCC: $27.50 © 2004 American Chemical SocietyPublished on Web 03/17/2004

2H-1,2-benzothiazine-3-carboxamide1,1-dioxide] (Figure1). The physiological target of this drug is cyclooxygenase,which is a membrane active enzyme.11 Catanionic micellesserve as a better model of biomembranes in the contextof charge distribution, and hence the study of theinteraction of this drug with catanionic micelles would beof biological relevance. In our earlier work, we have studiedthe interactions of this drug with simple micelles havinguniform headgroup charges.12 A switch-over or changefrom one prototropic form of piroxicam to another in thepresence of differently charged micelles was observed,which was correlated to the change in the pKa value in thepresence of charged surfactants. This change in pKa valuesmay be due to either electrostatic interaction and/orhydrophobic interaction, which in turn depends on thelocation and orientation of the drug in the micellar phase.In this present study, we have made an attempt to see theeffect of changing electrostatic interaction on the switch-over between different prototropic forms. That is why wehave taken two sets of catanionic micelles, one in whicha predominantly positively charged micellar surface isdoped with a small amount of negative charges and theother in which the negatively charged micellar surface isdoped with a small amount of positive charges.

The interaction of piroxicam with the catanionic mixedmicelles was studied using the intrinsic absorption andfluorescence properties of the drug. Piroxicam can existin three different prototropic forms, viz., anionic, neutral,and zwitterionic forms.13 The neutral and zwitterionicforms are spectroscopically indistinguishable and arethereby termed together as the “global neutral” form. Theincorporation of different prototropic forms in differentcatanionicmicelleshasbeenshownbyoptical spectroscopictechniques. We have also measured the change in freeenergy (∆G) value of the switch-over in the presence ofdifferent catanionic micelles to probe how the change inelectrostatic interaction modulates this equilibrium.

Experimental SectionCTAB and SDS were purchased from Merck and USB,

respectively. Piroxicam was purchased from Sigma Chemicals(U.S.) and was used without further purification. Water wasdistilled thrice before use. Stock solutions of piroxicam ofconcentration 0.5 mM were prepared in ethanol (Merck, Ger-many), and the exact concentration was adjusted by triple-distilled water. Each sample contains a maximum of 6% (v/v) ofethanol. The pH of the working solutions was adjusted by addingdilute HCl to them. The volume of acid (HCl) added to the working

solutions is exactly equal to the volume of acid that is needed toacidify a volume of water equal to the working solution to attainthat particular pH. Solution at pH 5.5 indicates that no acid oralkali was added to the aqueous solutions. Samples weredeoxygenated by passing argon gas for about 20 min beforescanning to avoid photochemical changes. The temperature waskept constant at 298 K throughout all experiments.

We have prepared the catanionic micelles of two differentcompositionsbymixingaliquotsof concentratedmicellar solutionsof SDS and CTAB followed by sonication for 15 min. The solutionswere then left for 24 h before making any measurements. In one,the concentration of CTAB is higher than the SDS concentration(the concentration of SDS was kept constant at 0.03 mM, and theconcentration of CTAB was varied from 0.4 to 12 mM), and theopposite is true for the other (the concentration of CTAB waskept constant at 0.1 mM, and the concentration of SDS was variedfrom 2 to 30 mM). We have measured the critical micellarconcentration (cmc) of the catanionic micelles of differentconcentrations of SDS and CTAB at different pHs using pyreneas the chromophore following the environmental effect on thevibronic band intensities (3/1) of pyrene.14 Below the cmc, thereare no micelles present and the pyrene fluorescence spectrumhas a 3/1 band ratio that is the same as in water. However, asthe detergent concentration increases above the cmc, pyrene issolubilized in the hydrophobic interior as illustrated by theincreased 3/1 ratios. The micelle formation was followed by thesharp increase in the 3/1 vibronic band ratio of the fluorescencespectrum of pyrene, which corresponds to the cmc value of themixed surfactants. Pyrene solution was prepared in dimethylformamide (DMF), and a very low concentration of pyrene (5 ×10-6 M) was used to avoid excimer formation. The maximumconcentration of DMF in the working solution was 0.1% (v/v).

The fluorescence lifetime of pyrene was also used to monitorthe formation of mixed micelles. Fluorescence lifetimes weredetermined from total emission intensity decay measurements,using a time-resolved fluorimeter assembled in our laboratorywith components from Edinburgh Analytical Instruments (EIA,U.K.) and EG&G ORTEC (U.S.) and operated in the time-correlated single-photon-counting mode. A pulsed high-pressure(1.5 atm) N2 lamp operating at 25 kHz repetition rate was usedas a source. The pulse profile had a full width at half-maximum(fwhm) of 1.2 ns. Pyrene was excited by the 337 nm N2 line, andits emission was monitored at 373 nm. Slits with 32 nm band-pass were used in both excitation and emission channels.Intensity decay curves could be fitted to a biexponential series:

where A1 and A2 are pre-exponential factors representing thefractional contribution of the time-resolved decay of the com-ponent with a lifetime τ1 and τ2, respectively. A0 is a constant.The decay parameters were recovered using a software packagesupplied by EIA, which used the Marquardt iterative nonlinearleast-squares fitting procedure. Statistically, the goodness of thefit was evaluated by the reduced ø2 value.

Absorption spectra were recorded with a Shimadzu UV-visiblespectrophotometer model UV2101PC. Baseline correction wasdone with water before recording each set of data. Fluorescencemeasurements were performed using a Hitachi spectrofluorim-eter model F 4010. All emission spectra were corrected forinstrument response at each wavelength. The concentration ofthe drug was measured from Lambert-Beer’s law, as theextinction coefficients are known at the characteristic wavelengthof the global neutral and anionic forms.15 A 2 × 10 mm2 pathlength quartz cell was used for all fluorescence measurementsto avoid any blue edge distortion of the spectrum due to innerfilter effect.16

TEM was done with a Hitachi electron microscope model 600operating at 75 kV with a resolution of 5 Å. The samples werespread over a copper grid coated with carbon. Phospho tungstic

(11) Hawkey, C. J. Lancet 1999, 353, 307-314.(12) Chakraborty, H.; Banerjee, R.; Sarkar, M. Biophys. Chem. 2003,

104, 315-325.(13) Tsai, R. S.; Carrupt, P. A.; Tayar, N. E.; Giroud, Y.; Testa, A.

B. Helv. Chim. Acta 1993, 76, 842-854.

(14) Kalyansundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1997,999, 2039-2044.

(15) Banerjee, R.; Chakraborty, H.; Sarkar, M. Spectrochim. Acta,Part A 2003, 59, 1213-1222.

(16) Lackowicz, J. R. Principles of Fluorescence Spectroscopy; KluwerAcademic/Plenum Publishers: New York, 1999.

Figure 1. Structure of different prototropic forms of piroxicam.

I(t) ) A0 + A1 exp(-t/τ1) + A2 exp(-t/τ2)

3552 Langmuir, Vol. 20, No. 9, 2004 Chakraborty and Sarkar

acid (PTA) was used as the stain for catanionic micelles containinga higher concentration of CTAB, and uranyl acetate was usedfor the catanionic micelles containing a higher concentration ofSDS. Samples were negatively stained with those heavy metalcompounds.

All measurements were done with solutions incubated at 298K.

Results and Discussion

(A) Characterization of Catanionic Micelles. Char-acterization by Pyrene Fluorescence 3/1 Band Ratio andLifetime. Interplay between various molecular interactionsmay result in a variety of microstructures, catanionicsurfactant salts, mixed micelles, catanionic vesicles, andso forth.6 Spontaneous vesicle formation has rarely beenobserved in single tailed chain systems without adequatechemical and mechanical treatment, but in cationic/anionic surfactant mixtures, spontaneous formation ofvesicles is frequently observed. The major reason behindthis phase transition is the electrostatic force. It was foundthat the phase transition proceeds in several stages withincreasing mole fraction of one surfactant. Figure 2a,bshows the plot of pyrene 3/1 band ratio versus concentra-tion of SDS at pH 5.5 and pH 3.8, respectively, withconstant CTAB concentration of 0.1 mM. Figure 3a,bdemonstrates the change of pyrene 3/1 band ratio versusconcentration of CTAB at pH 5.5 and pH 3.8, respectively,with constant SDS concentration of 0.03 mM. Themeasurement at pH 3.8 has been done for the interest ofthe second part of this work. From the above plots, thesaturation values of the peak ratio in the mixed micellesof higher concentration of SDS and CTAB at pH 5.5 are0.96 and 0.78, respectively. Below the cmc, there are nomicelles present and the pyrene fluorescence spectrumcorresponds to that in water with a 3/1 ratio of ∼0.64.14

The major contribution to the change in pyrene vibronicband intensities is from specific solute-solvent dipole-dipole coupling, although other effects due to π-orbitalinteractions between solute and solvent and the bulkdielectric constant of the solvent cannot be neglected.14

Despite the different contributions, qualitatively the 3/1band ratio serves as a measure of solvent polarity andincreases with decreasing dipole moment of the solvent.So from the peak ratio, we can compare different environ-ments in terms of dipole moment; at least we can saywhich one has the higher dipole moment and which onehas the lower dipole moment. Again, if two solvents havethe same dipole moment, then the peak ratio increaseswith decreasing dielectric constant.14 The higher 3/1 ratio(0.96) for catanionic micelles containing a higher SDSconcentration (Figure 2a,b) therefore reflects a less polarmicellar core compared to the catanionic micelles havinga larger concentration of CTAB. Generally the change ofthe pyrene 3/1 band ratio with surfactant concentrationis very sharp and then saturates as we have seen in Figure3a,b. But in Figure 2a,b between 4 and 7 mM concentrationof SDS, the value of the 3/1 band ratio overshoots and ishigher than the saturation value. This is not seen in single-surfactant micellar systems.

Measurement of the fluorescence lifetime of pyrene isalso a very good tool to study the micellization process.14,17

Generally the lifetime of pyrene is increased in bothrestricted and less polar environments provided there isno excimer formation. As mentioned before, the concen-tration of pyrene was kept low to avoid excimer formation.Figure 4a,b shows the plot of lifetime with increasing

concentration of SDS (at constant CTAB concentration)and CTAB (at constant SDS concentration) at pH 5.5,respectively. The lifetime of pyrene in the catanionicmicelles containing a higher concentration of SDS is 175ns (Figure 4a), and that in the other mixed micellescontaining higher CTAB concentrations is 156 ns (Figure4b). From the lifetime data of pyrene in two types ofcatanionic micelles, we can say that the catanionic micellecontaining a higher concentration of SDS is much morecompact and less polar compared to the one containing ahigher CTAB concentration. To summarize, the micellarcores of catanionic micelles, having a higher concentrationof negatively charged headgroups, are both more compactand hydrophobic than the ones containing a higherconcentration of positively charged headgroups.

Characterization by TEM. We have measured the sizeof the catanionic micelles of two different types from theTEM photographs. Figure 5a,b shows representative TEMphotographs of mixed micelles at a higher concentration

(17) Siemiarczuk, A.; Ware, W. R. Chem. Phys. Lett. 1990, 167, 263-268.

Figure 2. Plot of IIII/II of pyrene vs concentration of SDS at(a) pH 5.5 and (b) pH 3.8 at constant CTAB concentration of0.1 mM.

Catanionic Micelles of CTAB/SDS Langmuir, Vol. 20, No. 9, 2004 3553

of SDS (the concentration of CTAB is 0.1 mM, and theconcentration of SDS is 10.0 mM) and a higher concen-tration of CTAB (the concentration of SDS is 0.03 mM,and the concentration of CTAB is 6.0 mM), respectively,at pH 5.5 (without adding acid or alkali). Figure 5c is theTEM photograph of the overshooting region obtained inFigure 2a,b. The photograph has been taken at 0.1 mMCTAB and 5 mM SDS concentration. Figure 6a-c dem-onstrates the bar diagrams of frequency versus diametercorresponding to Figure 5a-c, respectively. Several TEMphotographs were taken, and in all cases all the systemsshow a high level of polydispersity. This is because of thenonuniform charge ratio of catanionic and anionic sur-factants in the catanionic micelles. The average diameterof the catanionic micelles with a higher concentration ofSDS is around 50 nm (the diameter of a pure SDS micelleis 3.68 nm18), and that with a higher concentration of CTABis around 80-90 nm (the diameter of a pure CTAB micelleis 4 nm19). There have been suggestions that water can

enter the micelles and can extend up to four carbons fromthe headgroup.20 In micelles with compact headgroupssuch as SDS, water penetration is smaller than in micelleswith larger headgroups such as CTAB, making their coresmore hydrophobic. The larger size of catanionic micellescontaining a higher CTAB (having a less compact head-group) concentration could allow more water penetration,making their cores more polar and less compact than thoseof the catanionic micelles containing a higher SDS (havinga more compact headgroup) concentration. The larger sizeof the catanionic micelles compared to single-surfactantmicelles can be explained as follows. Micelle formation isa compromise between the extremes of a complete phasetransition and a molecular disperse solution. The seques-tering of the nonpolar tails of the surfactant in the micellarinterior is driven by the solvophobic effect which isbalanced by the solubilization of polar headgroups in polarsolvent. Micellar aggregation is characterized both by the

(18) Duplatre, G.; Marques, M. F. F.; Miguel, M. D. G. J. Phys. Chem.1996, 100, 16608-16612.

(19) Singh, M.; Trivedi, M. K.; Bellare, J. J. Mater. Sci. 1999, 34,5315-5323.

(20) Fendler, J. H.; Fendler, E. J. Catalysis in Micellar andMacromolecular Systems; Academic Press: New York, 1975.

Figure 3. Plot of IIII/II of pyrene vs concentration of CTAB at(a) pH 5.5 and (b) pH 3.8 at constant CTAB concentration of0.03 mM.

Figure 4. Trace of the fluorescence lifetime of pyrene withincreasing concentration of (a) SDS at constant CTAB con-centration of 0.1 mM and (b) CTAB at pH 5.5 at constant SDSconcentration of 0.03 mM.

3554 Langmuir, Vol. 20, No. 9, 2004 Chakraborty and Sarkar

cooperativity of the formation and also by the mechanismthat stops aggregate growth to a macroscopic size andfinally to phase separation. For typical surfactants, it isthe repulsive interaction between the headgroups thatemanates the stop process. The magnitude of the elec-trostatic repulsion guides the predominant size of theaggregates in solution.21 A standard way of achievingaggregate growth is to reduce the headgroup-headgrouprepulsion. The two types of catanionic micelles studiedhere can be viewed as having either a positively chargedmicellar surface doped with negative charges or vice versa.In both cases, this doping with surfactants havingoppositely charged headgroups reduces headgroup-head-group repulsion and thereby promotes aggregate growth.This results in a much larger average diameter of thecatanionic micelles compared to the micelles with uni-formly charged headgroups. The average diameter of thespherical structures in the overshooting region in Figure5c is much larger (diameter, 90-100 nm) than the averagediameter of the mixed micelles with a higher SDSconcentration (diameter, 50 nm). This could indicate eitheran actual increase in the micellar size or more likely thepresence of mixed components, which could includevesicles. As has been mentioned earlier, the phasetransition between monomer and mixed micelles is not asclear as in the case of pure micelle formation. The borderof the phase transition in the case of mixed micelles is

characterized by the presence of other microstructuressuch as vesicles. For constant CTAB and varying SDSconcentration, the overshooting region (4-7 mM SDSconcentration in Figure 2a,b) corresponds to a region closeto the phase boundary6 where other microstructures arepresent. This is reflected in our pyrene 3/1 band ratio andTEM photograph. As we move away from the phaseboundary, the predominant structures are mixed micelles.

Figure 7 represents the TEM photograph of a mixtureof 0.03 mM SDS and 6 mM CTAB at pH 5.5, 2 h after thepreparation of the solution. Here we have found thatdistinct micelles have not formed; rather they are fusedtogether. From this, it is evident that catanionic micelleformation takes a certain hydration time. Accordingly,all experiments have been done with solutions incubatedat 298 K for at least 24 h. This is consistent with theobservations that the nature and size of the microstruc-tures depend on formation path, sonication, and aging.7

(B) Interaction of Catanionic Micelles with aNSAID. NSAIDs of the oxicam group can exist in differentprototropic forms, that is, cationic, global neutral (neutraland/or zwitterionic), and anionic forms, in differentphysiological conditions.13 These forms are extremelysensitive to their microenvironment22 because of theirdynamic structural features.13,23-25 The absorption maxima

(21) Evans, D. F.; Wennerstrom, H. The Colloidal Domain: WherePhysics, Chemistry and Biology Meet; Wiley-VCH: New York, 1999.

(22) Banerjee, R.; Sarkar, M. J. Lumin. 2002, 99, 255-263.(23) Yoon, M.; Chol, H. N.; Kwon, H. W.; Park, K. H. Bull. Korean

Chem. Soc. 1998, 9, 171-175.(24) Bordner, J.; Hammen, P. D.; Whipple, E. B. J. Am. Chem. Soc.

1989, 111, 6572-6578.

Figure 5. TEM photographs of mixed micelles (a) at a higher concentration of SDS (concentration of CTAB ) 0.1 mM andconcentration of SDS ) 10 mM), (b) at a higher concentration of CTAB (concentration of SDS ) 0.03 mM and concentration of CTAB) 10 mM), and (c) at the overshooting region obtained in Figure 2a,b (concentration of CTAB ) 0.1 mM and concentration of SDS) 5 mM) at pH 5.5.

Catanionic Micelles of CTAB/SDS Langmuir, Vol. 20, No. 9, 2004 3555

in water of the neutral and anionic forms are 330 and 353nm, respectively.22 With increasing concentration of SDSat a particular concentration of CTAB, the absorptionmaximum does not show any shift at pH 5.5. This isbecause the predominant population at pH 5.5 is theanionic form, which is expected to be repelled by thenegatively charged headgroup of the anionic surfactant.So the experiments with a higher concentration of SDShave been done at pH 3.8 to get an adequate amount ofthe global neutral form even though the major populationis the anionic one as indicated by the 353 nm absorptionmaximum. In the presence of a higher concentration ofSDS at pH 3.8, the absorption maximum shifts from 353to 343 nm (Figure 8a). On the other hand, increasing theconcentration of CTAB while keeping SDS concentrationconstant shifts the absorption maximum from 353 to 363nm (Figure 8b). The ground-state electronic transition ofthe anionic form of piroxicam is n, π*.15 For a moleculeshowing n, π* transition, a 10 nm red shift of the absorptionmaximum indicates that it is being incorporated in the

less polar environment within the micelles from the bulkaqueous phase. Again, with increasing concentration ofSDS, keeping CTAB concentration constant, the absorp-tion maximum shifts from 353 to 343 nm, indicating thatthe population of the neutral form has increased insolution. The absorption maximum has not shifted to 335nm, which was observed in the case of pure SDS micelles.12

This will be explained in a later part of this paper.Another way of monitoring the incorporation of the drug

is by following the increment of fluorescence intensitywith increasing concentration of surfactant. Figure 9arepresents the plot of fluorescence intensity versusconcentration of SDS at pH 3.8 at constant CTAB

(25) Geckle, J. M.; Rescek, D.; Whipple, E. B. Magn. Reson. Chem.1989, 27, 150-154.

Figure 6. (a-c) Size distribution plots of mixed micellescorresponding to Figure 5a-c.

Figure 7. TEM photograph of a mixture of 0.03 mM SDS and6 mM CTAB at pH 5.5, 2 h after the preparation of the solution.

Figure 8. Change in the absorption maximum with change inthe concentration of (a) SDS at pH 3.8 at constant CTABconcentration of 0.1 mM and (b) CTAB at pH 5.5 at constantSDS concentration of 0.03 mM.

3556 Langmuir, Vol. 20, No. 9, 2004 Chakraborty and Sarkar

concentration monitored at the excitation wavelength ofthe neutral form, that is, at 330 nm. Figure 9b shows thechange of fluorescence intensity with concentration ofCTAB at a fixed concentration of SDS (pH 5.5) when theexcitation wavelength was kept at that of the anionic format 363 nm. These two plots make it clear that the neutralform is incorporated in catanionic micelles with a higherconcentration of anionic surfactant SDS and the anionicform is incorporated in the catanionic micelles with ahigher concentration of cationic surfactant CTAB.

Figure 10a,b shows the plot of the optical density ratioat 363 and 330 nm (the absorption maxima of the anionicand global neutral forms) with increasing SDS concentra-tion (at constant CTAB concentration of 0.1 mM at pH3.8) and with increasing CTAB concentration (at constantSDS concentration of 0.03 mM at pH 5.5), respectively.Interestingly, it is found that the ratio decreases with theconcentration of SDS at pH 3.8 and increases with theCTAB concentration at pH 5.5 though the extinction

coefficient of the global neutral form is higher than thatof the anionic form.15 This indicates that in the presenceof different catanionic micelles there is a switch-over orchange between two prototropic forms of piroxicam, viz.,the anion and the global neutral form. The amount ofneutral form is increased in the presence of negativelycharged SDS, and the anionic form is increased in thepresence of positively charged CTAB. In our earlier work,we have shown that this switch-over between two pro-totropic forms, that is, change of population between globalneutral and anionic forms, also occurred in single-surfactant systems.12 Consider the following equilibriumin the presence of catanionic micelles:

where N represents the global neutral and A the anionicform of piroxicam.

Since the micellar pseudophase is spectroscopicallysilent, the effect of micellar equilibrium is indirectlyreflected in the changes in spectral properties of the drugmolecule. The equilibrium constant is given by

Figure 9. Plot of the relative fluorescence intensity withincrease in the concentration of (a) SDS at pH 3.8 at constantCTAB concentration of 0.1 mM (λexc ) 330 nm) and (b) CTABat pH 5.5 at constant SDS concentration of 0.03 mM (λexc ) 363nm).

Figure 10. Plot of O.D363nm/O.D330nm with the change inconcentration of (a) SDS at pH 3.8 at constant CTAB concen-tration of 0.1 mM and (b) CTAB at pH 5.5 at constant SDSconcentration of 0.03 mM.

N a A (1)

K ) [A]/[N]

Catanionic Micelles of CTAB/SDS Langmuir, Vol. 20, No. 9, 2004 3557

where [A] is the concentration of the anionic form and [N]is the concentration of the neutral form of piroxicam. Thechange in the free energy (∆G) can be determined fromthe equation

The ∆G value becomes more negative in the presence ofcatanionic micelles with a higher concentration of posi-tively charged CTAB (Figure 11b) and becomes lessnegative in the presence of a higher concentration ofnegatively charged SDS (Figure 11a). Negative ∆Gindicates the spontaneity of the equilibrium to the right-hand side, that is, formation of more anionic species,whereas positive or less negative ∆G indicates theformation of the more neutral form of the drug. Thedifference between the ∆G values in the presence ofmaximum concentration of surfactant and in water istermed as change in ∆G. The maximum change in ∆Gvalue in the presence of catanionic micelles with a higherconcentration of CTAB is -0.12 kJ mol-1, and that in the

presence of a higher concentration of SDS is 1.43 kJ mol-1.The change in ∆G in pure CTAB micelles was -0.3 kJmol-1, and that in the presence of pure SDS was 1.7 kJmol-1.12

In the presence of both types of catanionic micelles, thechanges in ∆G value are less compared to those of theirpure micellar counterpart. This may be due to the chargescreening effect. In pure SDS micelles, drug moleculesface a uniform negatively charged micellar surface, butin the catanionic micelles with a higher concentration ofSDS they face a micellar surface predominantly negativelycharged with a small amount of positive charge doped init. This is reflected in a smaller change in ∆G value. Thesame argument is also true for the other catanionicmicelles where CTAB concentration is higher. In thepresence of pure SDS micelles, the maximum value of ∆Gwas positive, which indicated that the forward reaction(eq 1) was thermodynamically forbidden thereby allowingthe formation of the neutral form only, resulting in a shiftat the absorption maximum to 335 nm. In the presenceof catanionic micelles with a higher concentration of SDS,the ∆G value becomes less negative but never reaches apositive value. So in the presence of this type of catanionicmicelles, the reaction predominantly proceeds to the left-hand side, but the forward reaction is not thermodynami-cally forbidden, which allows the coexistence of bothneutral and anionic forms of piroxicam. This results in ashift of the absorption maximum to 343 nm, characteristicof the mixed population, instead of 335 nm for the neutralform only. Such a strong effect of the micellar surfacechargeondrug incorporationandmodulationofprototropicforms could indicate that the probable locus of solubili-zation of the drug is nearer to the interface rather thandeep inside the micellar core. However, at this stage wecan only speculate on the exact location of the drug in themicelles.

Conclusion

Our study demonstrates that catanionic micelles con-taining a higher concentration of CTAB are larger in size,less compact, and more polar compared to the onescontaining a higher concentration of SDS. The larger sizeand less compact nature of the catanionic micellar corecontaining a higher concentration of CTAB might allowsome water penetration which could be the reason for itbeing more polar than the corresponding micelles con-taining a higher SDS concentration. The doping of somepositively charged headgroups in a predominantly nega-tively charged micellar surface and vice versa modulatethe equilibrium of the switch-over between prototropicforms of piroxicam in a different way compared to micelleshaving uniformly charged headgroups. This fine-tuningof the equilibrium even by such small changes in theelectrostatic properties of the environment is extremelyimportant in the context of biomembranes where chargesvary depending on the nature of the membrane.

Acknowledgment. We are extremely thankful to Ms.Rona Banerjee for her help and cooperation. We alsoacknowledge Mr. Pulak Ray, Mr. Tapan Kumar Ray, andMr. Ajay Chakrabarti of the Biophysics Division of theSaha Institute of Nuclear Physics for their help intransmission electron microscopy.

LA0361417

Figure 11. Dependence of ∆G (J mol-1) with concentration of(a) SDS at pH 3.8 at constant CTAB concentration of 0.1 mMand (b) CTAB at pH 5.5 at constant SDS concentration of 0.03mM.

∆G ) -RT ln K

3558 Langmuir, Vol. 20, No. 9, 2004 Chakraborty and Sarkar