Nonionic micelles in mixed water-glycerol solvent

Nonionic micelles in mixed water-glycerol solvent

Nonionic Micelles in Mixed Water-Glycerol Solvent LAURA CANTO,* MARIO CORTI, t VITTORIO DEGIORGIO,¢ HEINZ HOFFMANN,§ AND WERNER ULBRICHT§ *Dipartiment...

425KB Sizes 0 Downloads 37 Views

Nonionic Micelles in Mixed Water-Glycerol Solvent LAURA CANTO,* MARIO CORTI, t VITTORIO DEGIORGIO,¢ HEINZ HOFFMANN,§ AND WERNER ULBRICHT§ *Dipartimento di Chimica e Biochimica Medica, Universitd di Milano, 20133 Milano, Italy; tCISE S.p.A., Segrate (Milano), Italy; 5;Dipartimento di Elettronica-Sezione di Fisica Applicata, Universitd di Pavia, 27100 Pavia, Italy; and §Physicalische Chemie L Universitdt Bayreuth, 8580 Bayreuth, West Germany Received February 4, 1986; accepted April 21, 1986 The aggregation properties of the nonionic amphiphile C~2Es in water-glycerol mixtures are investigated at various glycerol volume fractions X b y surface tension, light scattering, refractometry, and viscosimetry measurements. In the range 0 < X < 0.4, the critical micelle concentration increases slowly with X, and the aggregation number m decreases with X. For X > 0.4, the CMC grows considerably, and the Krafft temperature becomes larger than room temperature. The minimum of the cloud point curve decreases from 74°C a t X = 0 to 43°C a t X = 0.6. © 1987AcademicPress,Inc. 1. INTRODUCTION

It is important to know for both fundamental and practical reasons the aggregation properties of amphiphiles in nonaqueous or mixed solvents. Concerning in particular nonionic amphiphiles, very few data are available (1, 2) on the critical miceUe concentration (CMC), whereas no information exists about the aggregation number and the size of the micelle, and about the phase diagram of the system. It is a well known phenomenon that nonionic amphiphiles with polyoxyethylene groups as hydrophilic moieties show clouding in aqueous solution if the solution is heated above a characteristic temperature which is called the cloud point. The cloud point depends on the nature of the amphiphile, on the length of the hydrophobic and the hydrophilic chains, on the amphiphile concentration, and on the concentration of additives (3). Approaching the cloud point the turbidity of the solution increases considerably with increasing temperature. When the cloud point is reached, the solution separates into two isotropic micellar phases (4) having different concentrations.

The increase of the turbidity of the micellar solution was first interpreted in terms o f a micellar growth. It was concluded from light scattering measurements that very large micelles are formed near the cloud point (5). More recent investigations have shown that the increasing turbidity is mainly due to the increasing range of concentration fluctuations (4, 6). The solutions behave like binary liquid mixtures on approaching a demixing line. If the minimum of the cloud curve (critical point) is approached following a constant concentration path, both the osmotic compressibility and the correlation range of concentration fluctuations, expressed as functions of the temperature distance from the critical point, show a power-law divergence. In the critical region, the solution is highly nonideal, that is, its properties are determined by collective effects (intermicellar interactions) more than by individual micelle properties. This implies that the aggregation number can be easily derived from a light scattering experiment only if the measurement is performed at a temperature which is sufficiently far from the critical temperature To (typically, Tc - T > 40°C) (6). We present in this paper an investigation of 384

0021-9797/87 $3.00 Copyright© 1987by AcademicPress,Inc. All tightsof reproductionin any formreserved.

Journalof ColloidandInterfaceScience,Vol. 116,No. 2, April 1987

NONIONIC MICELLES the micellar properties of the nonionic amphiphile n-dodecyl octooxyethylene glycol monoether, C12E8, in glycerol-water mixtures at various glycerol fractions and various temperatures. The CMC was determined by surface tension measurements, the aggregation number m by static light scattering, the hydrodynamic radius R by dynamic light scattering, and the cloud curve by a turbidimetric technique. We find that the CMC increases and that m decreases upon addition of glycerol, thus indicating that the amphiphile is less solvophobic in glycerol than in water. The cloud curve, however, is depressed by glycerol, probably because intermicellar interactions are stronger in glycerol than in water. The micellar properties could not be derived for glycerol fractions above 40% because of three concomitant effects: the solution becomes nonideal at room temperature due to the depression of To, the Krafft temperature becomes higher than room temperature, and, finally, the refractive index increment of the solution tends to 0 when water is substituted by glycerol. 2. MATERIALS AND METHODS High purity C12E8 was obtained from Nikko Chemicals, Tokyo, and used without further purification. Electronic grade glycerol was from Carlo Erba, Milan, and from Merck. Water was bidistilled. Measurements were carried out in pure aqueous solution and in solutions with glycerol up to 60% by weight. In mixtures with higher amount of glycerol the amphiphile was no longer soluble at room temperature. The solutions were allowed to stand several hours to reach equilibrium conditions before starting the measurements. The surface tension measurements were carried out with a commercial Lauda tensiometer using the ring method. The rheological properties were determined with a LowShear Viscometer LS 30 Sinus from Contraves. This apparatus allows measurements of viscosity with variable shear rates up to 100 s-1 and also dynamic measurements with angular frequencies up to 10 s-l. Refractive index

385

increments were measured with a Chromatix K M X 16 Differential Refractometer. The cloud point curves were obtained by putting several sealed cells, each filled at a different amphiphile concentration, into a water bath whose temperature was increased at a constant rate. The phase transition temperature was determined by monitoring with a laser beam the sudden variation of the cell turbidity on approaching the phase separation point (7). The light scattering apparatus is equipped with an argon laser operating on the 514.5nm green line, a scattering cell temperaturecontrolled within 1 millidegree over 24 h, and a digital correlator (6). The average scattered intensity I and the intensity correlation function of the scattered light have been measured at a scattering angle 0 = 90 °. The absolute calibration of the scattered intensity was performed according to the method described in previous work (6). 3. RESULTS Surface tension measurements were performed at 25 °C at glycerol concentrations below 40%. Above this value measurements had to be carried out at 35°C because the surfactant precipitates at lower temperature. Figure 1 shows the surface tension data obtained as a function of the surfactant concentration at various compositions of the solvent. The figure also contains the surface tension values for the pure solvents without surfactant. The plots of Fig. 1 show that the surface tension decreases logarithmically with the surfactant concentration at concentrations near the CMC, and remains constant when the concentration is above the CMC. Note that the CMC in water is in good agreement with that for polydisperse C12E8 (8). The obtained CMC values are reported in Fig. 2 as a function of the composition of the solvent. It should be pointed out that we have not applied the Harkins-Jordan correction to the measured surface tension values. The rheological measurements were carried out with the pure solvents and with C12E8 SOJournal of Colloid and Interface Science, Vol. 116, No, 2, April 1987

386

CANTO water

20% vol glycerol -40% vol glycerol

70

60% vol glycerol glycerol o

60 >,

,3

50

o v "~ \v

\\

40

~A-,~,b-A-A--A-A'A-~'4L'~"-A--

30

165

E T AL.

fractions of 0, 0.2, 0.4, and 0.6. We show in Fig. 4 the result obtained with a weight fraction of 0.2, and in Fig. 5 the complete set of cloud curves. The critical temperature Tc goes from 74°C in pure water to 43°C in the mixed solvent with 60% glycerol by weight. The critical concentration grows from 3% in pure water to 5% at the glycerol weight fraction of 0.6. The refractometric data are summarized in Table I. Since the index of refraction of glycerol is very close to that of the pure surfactant, the refractive index increment dn/dc is a strongly decreasing function of the glycerol weight fraction. We also found that, in first approximation, dn/dc for a given solvent is a linearly decreasing function of the temperature. Table I shows the temperature derivative of dn/dc at various solvent compositions. The static light scattering data are used to derive the quantity M', defined by

,. r ,,,J I ,,,J I ,,,J I ,,Ji t ,.,J I ,,,J 164 163 162 161 1 10 100

M'=A(dn/dc)-2(I ' - 1)/(c-CMC)

[1]

C12E8 Concentration (inglem3) FIG. 1. Scmiloganthmic plot of the surface tension for C12Es in water-glycerol mixtures as a function of the surfactant concentration at 25°C (solutions with 0 and 20% glycerol by volume) or 35°C (solutions with 40 and 60% glycerol by volume). ×, water; ©, 20% glyc; &, 40% glyc; V 60% glyc. At the top of the figure the surface tension values for the pure solvents are also shown.

[

I

I

I

I

I

4

lutions at concentrations more than two orders of magnitude above the CMC. The results are

shown in Fig. 3 where the viscosity is plotted as a function of the temperature at various solvent compositions. For all the investigated systems the solutions show Newtonian behavior. The viscosity values of the surfactant solution at temperatures sufficiently below the cloud point are only 10-20% larger than the corresponding values for the pure solvent. As can be seen from Fig. 3, the difference becomes more significant on approaching the cloud point. This effect is particularly evident in the solutions with the higher amount of glycerol because in these solutions the cloud point is considerably depressed. The cloud point curve (lower consolute curve) was measured at the glycerol weight Journal of Colloid and Interface Science, Vol.

116, No. 2, April 1987

i

~T=35"C

/

o 0

,

/

/

/

/

/ .~AJT=35 "C 1 -_..__---/-/- T=25"C T=25"C

I

0

I

20

I

I

40

I

I

E

60 % vol Glycerol

FIO. 2. Plot o f the C M C for solutions o f C l 2 E 8 in waterglycerol m i x t u r e s as a function o f the glycerol fraction.

387

N O N I O N I C MICELLES I

eoAv~

100

I

T (*C)

[] o Z~V<> pure solvent 1.5% wt C12Es

0..,_,....._

3% wt CI2Es

o--. .......

o.......

.--



glycerol

70

_

1(:

~

l-

~"~'D-

- ~

~

6 0 % vol glycero

g

J

water 2 0 % wt

.~ • • """ ...........

I

. _-. . . . . . . . . .

4 0 % wt glycerol

60

glycerol

-'----------~}

40*/, von

g,ycero, 50

vo, glycerol

•- . " ............................

60% wt glycerol

water

40

0 0.111 15

I 25

I 35

FIG. 3. Semilogarithmic plot of the viscosity for solutions of C12E8 in water-glycerol mixtures at various glycerol volume fractions as a function of the temperature.

where A is a calibration constant, I' is the measured scattered intensity normalized to the intensity scattered by the solvent, and c is the surfactant concentration in g/cm. The quantity M' is proportional to the osmotic cornF

I

T ('C)

72

15

FIG. 5. Cloud curves of C~2E8in water-glycerol mixtures at various glycerol concentrations.

pressibility of the solution, and the constant A is calculated in such a way as to express M' in daltons. We call M' the apparent molecular weight of the micelles. M' would be the true molecular weight for an ideal solution. It should be added that Eq. [ 1] is valid only in the limit of zero scattering angle; however, in the investigated range of temperatures and concentrations we never detected any angular dependence of the scattered intensity. We show in Fig. 6 plots of M' as a function of the temperature for various solvent compositions.

/ /

/

/

/

TABLEI The Refractive Index Increment (dn/dc) at 25°C and Its Temperature Derivative for CI2Es Water-Glycerol Solutions at Various Glycerol Concentrations X (% by vol)

/ /t

L

71.5

71

I 10

C12E8 Concentration (% by w e i g h t )

I 45

T ('C)

72.5

I 5

/

I

I

5

10

C12E8 Concentration (% by w e i g h t )

FIG. 4. Cloud curve of C~2E8in a water-glycerol mixture at the glycerol concentration of 20% by weight.

X

dn/dc

d(dn/dc)/dT

(%)

(cm3/g)

(cm3/g°C)

0 20 30 40 60

0.134 0. I 10 0.096 0.083 0.054

Journal of Colloid and Interface Science, Vol.

2.40 2.31 2.25 2.16 1.87

× × × × X

10 10 10 10 10

116, N o . 2, A p r i l 1987

388

CANTO ET AL.

All the data have been obtained at a surfactant concentration of 3% by weight which approximately represents the value at which the consolute curves of Fig. 5 show the minimum. We note that all the plots of M' have the same shape and are horizontally shifted one with respect to the other by a temperature interval which closely corresponds to the shifts of T shown in Fig. 5. This confirms the conclusion drawn in earlier papers (6) that the relevant parameter in a nonionic micellar solution is not the temperature itself, but rather the temperature distance from the critical point. As discussed in detail in Refs. (4) and (6), the light scattering data can be safely used to derive the molecular weight M of the nonionic micelle only in the fiat region of the M' plots, where the effect of critical concentration fluctuations has died out. To derive M, we have measured M' at T = 20°C as a function of the surfactant concentration for various solvent compositions. Some results are shown in Fig. 7. M is derived by extrapolating M' to the CMC (in

3 0 % wt

glycerol

0.36

2 0 % wt

glycerol

0.32 0.3

I

I

I

j

I

2 0 % Wt

glycerol

-/ 2.5

/

/

/

3 0 % wt

glycerol

;J

2

1.5 0

t 1

I 2

T 3

C12E8 C o n c e n t r a t i o n

I 4 (% b y w e i g h t )

FIG. 7. Reciprocal of the apparent molecular weight M ' and of the apparent hydrodynamic radius R' of C12Es solutions plotted as functions of the surfactant concentration

at T = 20°C and at various glycerol concentrations. 20

J

i

i

i

i

practice, to c = 0). The obtained values of M, together with the calculated aggregation numbers, are given in Table II. The dynamic light scattering data allow to derive the mass diffusion coefficient of the micellar solution. By using the Einstein-Stokes relation, one can calculate an apparent hydrodynamic radius R' which coincides with the

15 x

10

TABLE II

5

0

I

0

Molecular Weight M, Aggregation Number m, and Hydrodynamic Radius R of the Cl2Es Micelle at 20°C in Water-Glycerol Mixtures at Various Glycerol Concentrations Y (% by vol)

I

20

40 T

('C)

FIG. 6. Apparent molecular weight M ' (proportional to the osmotic compressibility) of C12E8 in water-glycerol mixtures as a function of the temperature. The surfactant concentration is 3% by weight.., water; e , 20% glyc; B, 40% glyc; • 60% glyc. Journal of Colloid and Interface Science, Vol. 116, No. 2, April 1987

Y (%)

M

m

R (nm)

0 20 30 40

65,000 60,200 56,200 52,400

121 112 104 97

3.3 3.2 2.9 2.8

NONIONIC MICELLES true hydrodynamic radius R only when the intermicellar interactions are negligible. We show in Fig. 7 some plots of R'. The values of R, obtained by extrapolation of R' to c -- 0, are reported in Table II. 4. DISCUSSION It can be seen from Figs. 1 and 2 that the CMC is little affected by the addition of glycerol. Only at the highest glycerol concentration the C M C is appreciably shifted to higher values. This indicates that the solvophobic effect responsible for the micellization process must be smaller in glycerol (9). The curves of Fig. 1 are nearly identical below the CMC, only the curve for 60% glycerol by volume is slightly shifted toward higher concentrations. This means that the surface activity of the surfacrant is smaller in this solvent, a conclusion which is in full agreement with the trend of the C M C values. The fact that micelles are formed in mixed solvent containing a large fraction of glycerol is not surprising, if we consider that glycerol shows a hydrogen-bonded structure somewhat similar to that shown by water. It is worth recalling that Becher (10) has recently proposed a quantitative approach to the question of micellization in nonaqueous or mixed solvents which is based on the representation of the cohesive energy density (CED) of a liquid introduced by Hansen and Beerbower ( 1 1). According to these authors, the CED of a liquid is the addition of three contributions due, respectively, to dispersion, dipolar, and hydrogen-bonding forces. The fraction of CED due to hydrogen bonding is 78% for water and 66% for glycerol which is, among c o m m o n solvents, the more similar to water (10). Our light scattering data show that the aggregation n u m b e r of Cx2E8 micelles far from the cloud point decreases by 20% when the solvent contains 40% by weight of glycerol. The dynamic light scattering results show, however, that the corresponding decrease of the hydrodynamic volume (proportional to the cube of the hydrodynamic radius R) is about

389

45%. Since the hydrodynamic volume contains also the solvent bound to the micelle, our data indicate that micelle solvation is considerably reduced in the mixed solvent. This could be due to a steric effect. Indeed the glycerol molecule is m u c h larger than the water molecule, and therefore the n u m b e r of hydrogen bonds formed by the oxygens of the polyoxyethylene chains with the solvent molecules should be smaller when water is substituted by glycerol. ACKNOWLEDGMENTS This workwas supportedby European EconomicCommunity and Italian Ministry of Public Education Research Grants. REFERENCES 1. Magid, L., in "Solution Chemistry of Surfactants" (K. L. Mittal, Ed.), Vol. I, p. 427. Plenum, New York, 1979. 2. Ray, A., Nature (London) 231, 313 (1971). 3. Shinoda, K., Tamamushi, B., Nakagawa, T., and Isemura, T., "ColloidalSurfactants." AcademicPress, New York, 1963. 4. Degiorgio,V., in "Physics of Amphiphiles. Micelles, Vesicles, and Microemulsions"(V. Degiorgioand M. Corti, Eds.), p. 303. North-Holland, Amsterdam, 1985. 5. Balmbra, R. R., Clunie, J. S., Corkill, J. M., and Goodman, J. F., Trans. Faraday Soc. 58, 1661 (1962); 60, 979 (1964); Nakagawa, T., Inoue, H., Tori, K., and Kuriyama, K., J. Chem. Soc. Jpn. Pure Chem. Sec. 79, 1294 (1958); Attwood, D., J. Phys. Chem. 72, 339 (1968); Muller, N., and Platko, F. E., J. Phys. Chem. 75, 547 (1971). 6. Corti, M., and Degiorgio,V., J. Phys. Chem. 85, 1442 (1981); Corti, M., Degiorgio, V., and Zulauf, M., Phys. Rev. Lett. 48, 1617(1982);Corti, M., Minero, C., and Degiorgio, V., J. Phys. Chem. 86, 309 (1984). 7. Corti, M., Minero, C., Cantfi, L., and Piazza, R., ColloidSurf 12, 341 (1984). 8. Becher, P., J. Phys. Chem. 63, 1675 (1959). 9. Hoffmann, H., and Ulbricht, W., Angew. Chem., in press; Sinanoglu, O., in "Molecular Interactions" (H. Ratajczak and W. J. Orville Thomas, Eds.), Vol. 3, p. 283. 1983. 10. Becher,P., J. Dispersion Sci. Technol. 5, 81 (1984). 11. Hansen, C., and Beerbower,A., in "'Kirk-OthmerEncyclopedia of Chemical Technology," 2nd ed., Supplement, p. 889. Wiley, New York, 1977. Journal of Colloid and Interface Science, Vol. 116, No. 2, April 1987