Colloids and Surfaces A: Physicochemical and Engineering Aspeas, 70 (1993) 69-76Elsevier Science Publishers B. V., Amsterdam

69

Micellization and mixed micellization ofalkylxylenesulfonates - a calorimetric study

A. Sivakuma~, P. Somasundarana and S. Thachba Langmuir Center for Colloids and Interfaces, Henry Krumb School of Mines. Columbia University. New

York. NY 10027. USAb ARCO Exploration and Technology Company. PIano. TX 75075. USA

(Received 16 June 1992; accepted 29 September 1992)

Abstract

The effects or the position or sulfonate and or methyl groups on the aromatic ring or isomericatly pure alkylxylenesulro-nates on micelliution and mixed micellization have been investigated using microcalorimetry and surrace tensionmeasurements. It was observed that the position or the sulfonate with respect to the alkyl chain is more critical than thator the methyl groups in determining the surfactant behavior in solution. Steric constraints to the packing or the moleculesare proposed to be the main mechanism ror the differences in the micelliution or the sunactants. Calorimetric studiessuggested that entropy was the main driving rorce ror micellization. The mixed micelliution or two structurally differentalkylxylenesulronates was round to be non-ideal. Regular solution theory adequately fits the data and microcalorimetrywas used to veriry the validity or the thermodynamic assumptions or the theory.

Keywords: Alkylxylenesulfonates; calorimetry; micellization.

micellization has been studied previously for anumber of surfactants. An important structuralvariation is the position of the functional groupson the arom,atic ring of surfactants containing abenzene ring. Studies on the effect of the positionof functional groups in the aromatic ring on theproperties of surfactants are limited [15,16]. Bols-man and Daane [15] have studied the stability ofmicroemulsions using alkylxylenesulfonates, andhave monitored resulting performance characteris-tics such as optimal salinity, solubilization parame-ter and interfacial tension. They observed that thesurfactant with the sulfonate group in the orthoposition with respect to the alkyl chain had loweroptimal salinity, higher interfacial tension andlower solubility in water than surfactants with thesulfonate group in the para or meta positions.However, they used commercial grade surfactantswith isomeric purity of only about 75%. Thus, even

Introduction

Surfactants are increasingly used in processessuch as flotation, detergency, enhanced oil recov-ery, paint formulation, lubrication and microelec-tronics [1-5]. Also, surfactant assemblies such asmicelles have potential applications in novel sepa-ration and reaction schemes such as magneticisotope separation and polymer synthesis [6,7]. Anunderstanding of the solution behavior of surfac-tants, and in particular micellization, is importantfor improving the efficiency of the above processes.

Among the different factors that influence micel-lization, the surfactant structure plays an importantrole [8-14]. The effect of different structural varia-tions in alkyl chain length, branching of the alkylchain and the addition of functional groups on

Correspondence 10: A. Sivakumar, Nalco Chemical Company,NapervilIe, IL 60563, USA.

0927.7757{93{$06.00 @ 1993 - Elsevier Science Publishers B. V. All rights reserved.

70 A. SilJakllmar ft 1Il./Coiloids Surfaces A: Physicoc~. ErIg. Aspects 70 (1993) 69-76

s~-

Ort~.r8r82

Experimental conditions

All the experiments were carried out at 43°Cand at a constant ionic strength of 0.03 kmol m - 3

NaCI.

Methods

though they observed differences in properties withchanges in the position of the sulfonate group, theeffect of other isomers present cannot be excluded.In fact. in most of the studies on the effect ofsurfactant structure, the purity of the surfactant interms of its isomers is not high. This could lead toerroneous conclusions since it is known that impu-rities even in small amounts can mask the effectof structural variations [8]. Using isomericallypure alkylxylenesulfonates, we have shown that thesurfactant with the sulfonate group in the paraposition is more surface active, has a lower CMCand adsorbs onto alumina to a greater extent thanthe surfactant with the sulfonate group in the metaposition [17]. Recently, Varadaraj et al. [18] havealso observed tbe same behavior with alkylxylene-sulfonates with the sulfonate group in the paraand ortho position.

In tbis work, the effects of the position of sulfonateand methyl groups on the benzene ring of isomer-icaUy pure alkylxylenesulfonates on micellizationand mixed micelle formation have been studiedusing surface tensiometry and microcalorimetry.

Surface tensiometrySurface tension was measured using a water-

jacketed du Nuoy ring tensiometer set to the test

temperature.

Experimental

Materials

Surfactants4CII 2,4-meta-xylenesulfonate (Meta), 4CII 3,5-

para-xylenesulfonate (Para I), 4C II 2,5-para-xylen-esulfonate (Para2) and 4Cll 4,S-ortho-xylenesulfo-nate (Ortho) obtained from ARCO Oil and Gas.Company were used in this work. The counterionfor aU the surfactants was Na. AU the surfactantswere specified to be at least 97% isomericaUy pureand were used as received. The structures of thesesurfactants are shown below:

MicrocalorimetryCalorimetric experiments were performed using

a LKB 2107 differential microcalorimetry system.The calorimeter consists of a rotating block placedinside an air bath maintained at a constant temper-ature. The block consists of a sample cell and areference cell. Each ceil has two compartments.Mixing of the reactants is achieved by rotating theblock. The temperature difference is then measuredusing thermopiles attached to each cell. The out-puts of the thermopiles associated with each vesselare connected in opposition to one another, sothat any electromotive force common to bothvessels is nullified. The resulting output voltage isthus a measure of the reaction heat alone. Thissignal is amplified and is passed on to a digitalread-out system and integrator/printer. The outputis then calibrated for the reaction heat using acalibration heater built into each vessel.

For the enthalpy of micellization, the heat ofdilution of a micellar solution to its premicellarconcentration was measured. The heat of dilution

s~-

Met. raraa

A. Sil}Qk.JllrlQ, It Dl.IC~ Surf«a A: Physicoc'-t. fill. AspectJ 70 (1993) 69-76

or the monomers was measured separately and wassubtracted rrom the total reaction heat to obtainthe heat or micellization.

The excess enthalpy or mixed micelle ronnationwas measured by mixing concentrated micellarsolutions or the two surractants. Solutions or equi-molar concentration were used ror each measure-ment. so that the mixture composition wasdetermined by the relative solution volumes.

Results and discussion

Surface tension

Surface tension data provide information on theadsorption of surfactants at the liquid/air interfaceand also on surfactant interactions in solution. TheCMC of Meta. Para I and Para2 were determinedfrom the break in the surface tension curves (Fig. I).The ortho-xylenesulfonate precipitated in the pres-ence of 0.03 kmol m - 3 NaCI and this result is

discussed in the next section. Absence of a mini-mum in the curves indicates that the surfactantsare pure. It can be seen from the figure that Para Iis more surface active than Meta and Para2. Also,Para I and Paral have a lower CMC and a largerlowering of surface tension at CMC than Meta.The difference in the surface activity and the CMCof the surfactants is attributed to:

(I) different hydrophobicities of the alkyl chainsdue to the inductive effect of the sulfonate group;

(2) different steric constraints arising from thepositions of the sulfonate and methyl groups tothe packing of the molecules at the interface andin micelles.

The relative hydrophobicities of the three surfac-tants were measured using reverse-phase high per-formance liquid chromatography [19]. Theretention times of the alkylxylenesulfonates alongwith a series of isomerically pure linear alkyl ben-zenesulfonates (LABS) were determined. Thedifferent retention times of the surfactants arisepurely from the hydrophobic interaction of thesurfactants with the column. A plot of the alkylchain length vs the logarithm of the retention timeof LABS (Fig. 2) is linear. From the retention timesof the alkylxylenesulfonates, their relative hydro-phobicities were determined in terms of effectivealkyl chain length. The results show that the twopara-xylenesulfonates have the same effective chainlength while the meta-xylenesulfonate is less hydro-phobic by 0.5 CHz group.

From Fig. I, the slopes of the lines on the lowconcentration side of the break for Meta, Para Iand Paral were calculated to be -17.1 mN m - 1,

-21.7 mN m-l and -15.3 mN m-', respectively.Using these values in the Gibbs adsorption equa-tion, the area per molecule for Meta, Para I and

_.Go CO..:..; ,:;'-Ce.1

...

XI-CD

:; 14.0... .Z

;: Il.u...... ~~ 10.0

i 8.0 octul

w 0.0l.O

, lelrodecyl&

'i

z0-CISZW0-

WUa:u.~~on

dodecy I

/".Bl~4/"PO"O

6 Porol

c Meto

D Paroldecyl,c'L

a.ol.. I

o.~ I.CI! IO.aI 100.CI! 2IXI.CI!

SURfACTANt CONC.. k80I/.3. 104

Fig. t. Surf~ tension or alkylxylenesulronate solutions.

~..RETENTION TIME. .inules

Fig. 2. Relative hydrophobicity of alkylxylenesulfonates

72 A. SivaOmar et aI./CoIloi4s Surfaces A: Physic«Mm. E",. Aspects 70 (1993) 69-76

Para2 was calculated to be 0.6 nmz, 0.46 nmz and0.66 nm2, respectively. The area per molecule ofMeta and Para2 is similar. The higher surfaceactivity of Para2 as compared to Meta is thereforedue to its higher hydrophobicity. However, thearea per molecule of Para I is smaller than that ofPara2. Thus, even though the hydrophobicities ofPara I and Para2 are similar, Para I is more surfaceactive due to lower steric constraints to the packingof the molecules at the interface and in the micelles.The reason for the precipitation of the ortho-xylenesulfonate is due to the inability of the mole-cules to pack in the micelles due to high stericconstraints. Since the surfactant monomers cannotpack into the micelles and also since it is unfavor-able for them to remain in the aqueous phase, thesurfactant precipitates out.

using the ideal solution and regular solution theo-ries are shown in the figure and it can be seen thatthe system shows a small negative deviation fromthe ideal behavior. This is interesting becausemixtures or similar surfactants normally show idealbehavior [20.21]. Regular solution theory fits thedata well with an interaction parameter (W) or-0.28. The deviations rrom ideality can be attrib-uted to the difficulty in packing or the moleculesin the mixed micelles [22]. Even though regularsolution theory adequately fits the data, the validityor the thermodynamic assumptions ror the systemhad to be verified [23]. Microcalorimetry was usedto veriry the assumptions and the results are dis-cussed below.

Enthalpy of micellization

Mixed micelleformation

The effect of the position of the sulfonate andmethyl groups on the benzene ring on the behaviorof these surfactants in mixtures was investigatedby measuring the mixed micelle formation of meta-xylenesulfonate and para-xylenesulfonate (Para 1).

Surface tension was used to measure the CMCof the mixtures of the two surfactants and theresults are given in Fig. 3. Also, the theoreticalCMC values for the different mixtures calculated

~

.0~

Calorimetry has been used extensively to studymicellization of surfactants and has proved to bea valuable tool for understanding the formationand structure of micelles [24-26]. For example,based on their study of heat capacity, Mazer andOlofsson [26] found that the major contributionto the heat capacity change for micelle formationof dodecyl sulfate arises (rom changes in the alkylchain and that the changes in the head groupcapacity are of secondary importance. In his workon the micellization of polyoxyethylene glycoldodecyl ethers. Olofssoo [27] observed that theenthalpy of micellization corresponds to the dehy-dration of between 2 and 2.5 ethylene oxide groupsimplying that the rest of the ethylene oxide groupsremain in full contact with wafer. Interestingly, thecontribution of the alkyl chain to dilution enthalpywas zero while it dominated the entropy of micelli-zation.

The calorimetric data for the three alkylxylene-sulfonates are given below.

uzu

Enthalpy of micellization of para-xylenesulfonate

{Paral}

The heat of dilution data of premicellar andmicellar Para} solutions are given in Tables I and

MOLE FRACTION Helo

rig. 3. CMC or mixlures oC ptra-xyleoesulfonate (Parat) andmeta-xylenesulConate (Meta).

A. Sillakumar et al./Colloids Surfaces A: Physicochem. Eng. Aspects 70 (1993) 69-16

TABLE 3TABLE 1

Heat of dilution of premicellar solutionsHeat of dilution of premicellar solutions

IWCr;,..,(kmol m-J)

1000Cinit;a1

(kmol m-3)JH~il(cal mol-I)

-33-39

6.226.22

4.964.97

"Average value is -36 cat mol'"Average value is -118 tal mol-I.

TABLE 4

2, respectively. In these tables Cinilial and Cfinal rep-resent the initial and final concentrations of thesolutions, respectively, L1Hdil and L1Hm representthe enthalpies of dilution and micellization, respec-tively and Qdil represents the heat of dilution.

The dilution of premicellar solutions results inan exothermic heat change of -118 cal mol- 1,and this heat results from the dilution of sulfonatemonomers. For the dilution of micellar solutions,demicellization accounts for the bulk of the heatchange, which for Paral is endothermic. The micel-lization enthalpy (L1Hm) values are given in thefourth column of Table 2 and show excellent con-sistency. with a mean value of -1091 cal mol-1.

9.96 S.67 4.949.96 S.68 4.74

12.0 6.61 9.31

'Average value is -774 caJ mol-

-745-720-ass

6.3.10-4 kmol m-3. the above data yield a meanARm of - 774 cal mol-i.

Enthalpy of micellization of para-xylene sulfonate

{Paral}

Enthalpy of micellization of meta-xylenesulfonate

( Meta)The heat of dilution data of premicellar and

micellar Para2 solutions are given in Tables 5 and

6, respectively.The mean heat of monomer dilution is - 354

tal mol-i. Applying this correction to the dataobtained with demicellized solutions and using aCMC of 5.3-10-4 kmol m-3 yields a mean L1Hmof -1671 cal mol-i.

The heat of dilution results of premicellar andmicellar Meta solutions are given in Tables 3 and4, respectively.

The mean heat of monomer dilution is - 36 cal

mol-I. Applying this correction to data obtainedwith demicelli2ed solutions and using a CMC of

TABLE 5TABLE 2

Heat of dilution or premicellar solutions--Heat of dilution of micellar solutions

1000C;nilla.

(kmol m-3)lQ4Cfi (kmol m -J)

IO"Ci8iu..(kmol m-J)

1000Cn..1

(kmol m-J)IO"'Q...,(caJ)

.iH~il(caJ mol-I)~w.

(cal mo)-1-363-348-365-344

2.862.863.983.98

1.431.421.982.00

5.778.921.13

2.802.802.58

3.935.444.87

-1041-1084-1148

"Average value is -1091 cal mol- I. "Average value is -354 ca1 mol-I.

74 A. Sivakumar et al./Colloids Surfaces A: Physicochem. Eng. Aspects 70 (1993) 69-76

TABLE 6 molecules structured around the hydrocarbonchain.

Excess enthalpy of mixed micelle formation

The free energy of micellization LtG accordingto the phase separation model [28] is given by

LtG=2RT In(CMC) (I)

Mixed micellization of the meta-xylenesulfonateand para-xylenesulfonate {Para I) showed negativedeviations from ideal solution theory and it wasfound that regular solution theory adequately fittedthe mixed CMC data. However, it has been shownthat in certain systems, thermodynamic assump-tions of the regular solution theory are not valid,even though the theory fits the surface tension datawell [22,23]. Therefore, it was considered impor-tant to verify the validity of the assumptions priorto using the theory.

Regular solution theory is based on the assump-tion that the excess Gibbs free energy of mixing is

given by

where CMC is the critical micelle concentrationexpressed in mole fractions, R is the gas constantand T is the absolute temperature.

The entropy of micellization tJS is then calcu-lated from the following equation

~S = (A.H - A.G)/T (2)(3)LlGmi. = X lX2WRT

where X I and X 2 are the mole fractions of the twosurfactants in the mixed micelle and W is theinteraction parameter. Regular solution theory alsoassumes that the excess entropy and volume ofmixing are zero and hence the excess enthalpy ofmixing is given by

The values for the thermodynamic parametersof micellization for the three surfactants are shownin Table 7.

'There is no particular trend in the enthalpy andentropy of micellization with the CMC of thesurfactants. Both the enthalpy and entropy arefavorable for the micellization of the surfactants.The enthalpies of the more hydrophobic para-xylenesulfonates are more exothermic than thoseof the meta-xylenesulfonate. However, these differ-ences are very small when compared to the micelli-zation entropy of the surfactants which is the maindriving force for the micellization. The positiveentropy change arises from the removal of water

(4)A Hmi. = X lXzWRT

The excess enthalpy of mixed micelle formationof the two surfactants in this study at varioussurfactant compositions is shown in Fig. 4. Thefact that the excess enthalpy of mixing is non-zeroat all compositions suggests that the mixtures arenon-ideal and the exothermic value indicates thatthe deviations are negative with respect to idealmixing. An interaction parameter W = - 0.21 fits

the calorimetric data and it compares well withW = - 0.28 obtained from the surface tension data.

This indicates that the assumption of the regularsolution theory (excess entropy of mixing is zero)is valid for the system of meta-xylenesulfonate and

para-xylenesulfonate (Para I).

TABLE 7

Para!MetaPara2

-1f1J1-774

-1671

-15226-\4407-14626

454341

A. SivaklUllGr et al./Collo/d.t SurfGCu A: Physicocl-.. Ellg. Aspects 70 (1993) 69-76 75

and para-xylenesulfonate (Para I) is non-ideal. Reg-ular solution theory fits the data well with aninteraction parameter of -0.28. Microcalorimetricmeasurements of the excess enthalpy of mixedmicelle formation showed that the thermodynamicassumptions of the theory are valid for the surfac-tant system studied.

Acknowledgments

The authors acknowledge the National ScienceFoundation, Department of Energy, ARCO andBP America for support of this work.. MOL~ FRACTION OF Meta S IN MICELLE

Fig. 4. Excess enthalpy of mixing of para-xylenesulfonate(Para I) and meta-xylenesulfonate (Meta). References

Summary

I P. Somasundaran and B.M. Moudgil (Eds), RcaplS inMineral Technology, Marcel Dekker, New York, 1987.

2 GJ. Shilling and G.S. Bright, lubrication, 63 (1977) 13.3 D.B. Hough and H.M. Rendall, in G.D. Parfitt and C.H.

Rochester (Eds), Adsorption from Solution at the Solid-Liquid Interrace, Academic Press, New York, 1983.

4 MJ. Schwugr, in E.H. lucasscn-Reynden (Ed.), AnionicSurractants, Surfactant Sci. Ser., Vol. II, Marcel Dekker,New York, 1981.

S J.D. Swalen, D.L AlIara, J.D. Andrade, E.A. Chandross,S. Garolf, J. Israelachvili, TJ. McCarthy, R. Murray, R.F.Pease, J.F. Rabolt, KJ. Wynne and H. Yu, langmuir, 3(1987) 932. .

6 M.S. Wrighton (Ed.), InterraciaJ Process: Energy Conver-sion and Synthesis, Adv. Chem. Ser. 184, American OIemi-cat Society, Washington, DC, 1980.

7 NJ. Turro and B. Kraeutlcr, in E. Buncel and C.C. lee(Eds), Isotopes in Organic Chemistry, Vol. 6, Elsevier,Amsterdam, 1984.

8 P. Somasundaran, R. Middleton and K.V. Viswanathan,in MJ. Rosen (Ed), Structure/Performance Relationshipin Surractants, ACS Symp. Ser. 253, American ChemicalSociety, Washington, DC, 1984, pp.270-290.

9 L. Matos, J. Ravey and G. Serratrice, J. Colloid Interracesa, 128 (1989) 341.

10 G. Zhao and D. Zhu, Colloids Surfaces, 33 (1988) 289.II H. Teo, s.G. Yeates. C. Price and C. Booth, J. OleIn. Soc.,

faradayTrans. 1,80(1984) 1787.12 R. Granet and S. Piekarsi, Colloids Surfaces, 33 (1988)

321.13 MJ. Rosen, z. Zhu, B. Gu and D.5. Murphy. langmuir,

4 (1988) 1273.14 N.M. van Os, G.J. Daane and T.A.B.M. Bolsman. J. Colloid

Interrace Sci., 115 (1987) 402.IS T.A.B.M. Bolsman and GJ.R. Daane, Soc. Pet. Ens. Res.

Ens.. I (1986) S3.16 D.C.H. Cheng and E. Gulari, J. Colloid Interface Sci, 90

(1982) 410.

The position of sulfonate and methyl groups onthe benzene ring of alkylxylenesulfonates is shownto play an important role in determining theproperties of these surfactants. The surfactant withthe sulfonate group in the para position is morehydrophobic and has a lower CMC than thesurfactant with the sulfonate group in the metaposition. The surfactant with the sulfonate grou.pin the ortho position precipitated in the presenceof 0.03 kmol m - 3 NaO. The two para-xylenesulfo-

nates with different positions of the methyl groups,though having the same hydrophobicity, havedifferent surface activity and CMC values. Thissuggested that the steric constraints to the packingof the molecules at the liquid/air interface and inthe micelles is the main mechanism for the observeddifferences in the behavior of the structurallydifferent alkylxylenesulfonates. The position of thesulfonate group is found to be more critical thanthose of the methyl groups on the benzene ring indetermining the behavior of the surfactants.

Microcalorimetric studies showed that enthalpyand entropy are favorable for micellization andentropy is the main driving force for the micelliza-tion of the surfactants.

The mixed micellization of meta-xylenesulfonate

76 A. Sivakumar el al./Colloids Surfaces A: Physicochem. Eng. Aspects 70 (1993) 69- 76

23 P.M. Holland, in J.F. Scamehom (Ed.), Phenomena inMixed Surfactant Systems, ACS Symp. Ser. 311, AmericanChemical Society, Washington, DC, 1986, p. 102

24 K.S. Birdi, Colloid Polym. Sci., 261 (1983) 45.25 G.C. Kresheck and W.A. Hargraves, J. Conoid Interface

Sci., 129 (1989) 175.26 N.A. Mazer and G. Olofsson, J. Phys. Chern., 86 (1982)

4584.27 G. 0lofsson, J. Phys. Chem~ 89 (1985) 1473.28 B.W. Barry and G.J.F. Russell, J. Colloid Interface Sci~ 40

(19r2) 174.

17 A. Siva kumar, P. Sornasundaran and S. Thach, in K.LMittal and D.O. Shah (Eds). Surfactants in Solution, Vol.II, Plenum Press, New York, 1991, pp. 293-302.

18 R. Varadaraj, P- Va1int, Jr, J- Bock, S. Zushma andN. Brons, Langmuir, 7 (I99L) 658.

19 K.V. Viswanathan and P. Somasundaran, J. Colloid Inter-face Sci., 126 (1988) 634.

20 J.H. Oint, J. Chern. Soc-, Faraday Trans. 1,71 (1975) 1327.21 KJ. Mysels and R.J. Otter, J. Colloid Interface Sci~ 16

(1961) 462.22 J.F. Rathman and J.F. Scamehorn, Langmuir, 4(1988)474.