In th e L a b or a t ory

Determination of Critical Micelle Concentration of Some

Surfactants by Three Techniques
Ana Domínguez, Aurora Fernández, Noemí González, Emilia Iglesias,* and Luis Montenegro
Departamento de Química Fundamental e Industrial, Facultad de Ciencias, Universidad de La Coruña
15071-La Coruña, Spain

Surfactants, sometimes called surface-active agents or

detergents , are among the most versatile chemicals avail-
able. They have applications in many areas, including chem-
istry (chemical kinetics or equilibria), biology (as membrane
mimetics), and pharmacy (1). Surfactants are amphiphilic
materials containing both apolar long-chain hydrocarbon
“tail” and polar, usually ionic, “head” groups. In polar sol-
vents, for example water, this dual character of the
amphiphile leads to self-association or micellization: the
surfactant molecules arrange themselves into organized
molecular assemblies known as micelles (Fig. 1). The hydro-
phobic part of the aggregate forms the core of the micelle,
while the polar head groups are located at the micelle–wa-
ter interface in contact with and hydrated by a number of Figure 1. An idealized model for a spherical micelle of sodium
water molecules. Depending on the chemical structure of dodecyl sulfate.
the surfactant, its micelle can be cationic, anionic,
ampholitic (zwitterionic), or nonionic. This unique property
of surfactants makes aqueous surfactant solutions
microheterogeneous media; that is, they are heterogeneous The Critical Micelle Concentration
on a microscopic scale, even though they are often homoge-
neous macroscopically. The concentration (actually an arbi- Micelles are small colloidal particles, relative to the
trary concentration within a narrow range) above which mi- wavelength of light. When micelles form, the aqueous sur-
celles form is called the critical micelle concentration (CMC). factant solution behaves as a microheterogeneous medium.
Above the CMC, monomers and micelles exist in dynamic The value of the CMC can be determined by the change in
equilibrium. the physicochemical properties of the surfactant solution as
Despite their growing importance, microheterogeneous the surfactant concentration increases.
media are not yet considered in most textbooks of general Experimentally, the CMC is found by plotting a graph
or experimental physical chemistry. Several reports illus- of a suitable physical property as a function of surfactant
trate the use of various techniques to measure the CMC (2– concentration. An abrupt change of slope marks the CMC.
5). The experiments described in this article were designed The choice of CMC is never unambiguous, since the change
to familiarize students of physical chemistry with micellar in slope occurs over a more or less narrow range of concen-
solutions and the basic structural parameters of micelles, trations, whose magnitude depends on the physical prop-
and with some common techniques in physicochemical labo- erty being measured and sometimes on the nature of the
ratories used to observe the changes in physical and chemi- data and on the way they are plotted.
cal properties of surfactant solutions when micelles are The CMC can be affected by many variables (6), tem-
formed. The paper presents the CMC determination of some perature and pressure being of relatively minor importance.
surfactants by three methods. The advantages and disad- It decreases with increasing hydrocarbon chain-length of
vantages of each one are indicated. the apolar groups, and for ionic surfactants it also depends
In these experiments, undergraduate students in the on the nature and concentration of counterions in solution.
fourth year in chemistry determined the CMC of a surfac- Added electrolytes decrease the CMC, and the effect in-
tant by measuring a change in (i) the UV-vis spectrum of creases with decreasing charge density of the counterion.
benzoylacetone, (ii) the fluorescence emission spectrum of
pyrene monomers, and (iii) the electrical conductivity of an Materials
ionic surfactant solution as the concentration of the
amphiphile increases. There are several ways of organizing The surfactants sodium dodecyl sulfate (SDS),
the laboratory experiments, depending on the time and CH3(CH2 )11OSO 3Na; tetradecyl trimethyl ammonium bro-
equipment available. Each group of students could deter- mide (TTABr), CH3 (CH2)13N(CH 3)3 Br; polyoxyethylene-9-
mine the CMC of a surfactant, for example sodium dodecyl lauryl ether (C12 E 9), CH 3 (CH 2)11 –(OCH2 CH 2) 9–OH; the
sulfate, by using the three techniques under the same ex- benzoylacetone (BZA); and the pyrene were Aldrich or
perimental conditions. The time required for determination Sigma products and were used as received. (If pyrene shows
of a CMC value by any of the techniques described here is a yellow color it must be purified by column chromatogra-
no longer than two lab sessions; each one takes about four phy using silica gel as stationary phase and cyclohexane as
hours. The accuracy of students’ results and response to the eluent). The additives were all Merck products.
experiments were generally good. CAUTION : Benzoylacetone, pyrene, and the surfactants
are harmful if ingested or inhaled or if they contact the skin;
due precautions must be taken. (Stock solutions of BZA in
*Corresponding author. dioxane and pyrene in methanol could be provided.)

Vol. 74 No. 10 October 1997 • Journal of Chemical Education 1227

In th e L a b or a t ory

Results It is noticed that a remarkable enhancement of enolic

absorption occurs just above the CMC. The concentrations
UV-Absorption Spectroscopy Method corresponding to the break points observed for each surfac-
This method is based on the tautomerism of tant, in the absence and presence of the indicated additive,
benzoylacetone or 1-phenyl-1,3-butadione. Ketones and !- are tabulated in Table 1.
diketones, such as BZA, can exist in two tautomeric forms:
the ketonic and the enolic forms. In solution, BZA enolizes Procedure. To obtain the data shown in Figure 2
practically exclusively to the cis-enolic form, which exists and Table 1, prepare a concentrated solution (5
in a cyclic configuration stabilized by intramolecular hydro- mg/mL) of BZA in dioxane (or another water-mis-
gen bonding: cible nonpolar solvent). From this stock solution
(~0.03 mol L�1) prepare an aqueous BZA solution
H by pipetting 0.40 mL into 25-mL volumetric flasks
O O
O O and diluting to the mark with water. To draw each
C
CH2
C
CH3
KE C
CH
C scan of the spectrum, 0.40 mL of the aqueous BZA
CH3
solution is transferred with a pipet to a 1.0-cm
quartz cell together with the appropriate amount
of surfactant, the additive (if any), and the water
This stabilization of the enol will be more pronounced when volume necessary to give a 3-mL total volume in
intermolecular hydrogen bonding with the solvent does not the cell ([BZA]~7 " 10�5 mol L�1). The reference cell
compete. Hence keto–enol equilibria of BZA are extremely should contain the same concentrations of surfac-
solvent sensitive and the proportion of the enolic form is tant and additive as the sample. The spectrum of
much greater in nonpolar solvents, such as cyclohexane, BZA in the presence of varying surfactant concen-
than in polar or hydrogen-bond donor solvents, such as wa- trations can be measured with any UV-vis spectro-
ter or alcohols. photometer.
BZA exists in water as a mixture of both tautomers (the
percentage of the enol tautomer has been measured as Fluorescence Spectroscopy Method
37.5% [7]). When a surfactant is added to a water solution This method is based on the solvent dependence of vi-
of BZA, the amount of enol increases abruptly when the sur- brational band intensities in pyrene monomer fluorescence.
factant concentration rises above the CMC because the Pyrene is one of the few condensed aromatic hydrocarbons
enolic form is taken up by the micelles, which provide a less that show significant fine structure (vibrational bands) in
polar solvent than the aqueous phase. their monomer fluorescence spectra in solution. The inten-
Figure 2a shows the effect of increasing surfactant con- sities of the various vibrational bands of pyrene showed a
centration on the UV-absorption spectrum of aqueous BZA strong dependence on solvent environment (8). Both solvent
solutions of 7.0 " 10�5 mol L�1. Below the CMC no changes dipole moment and dielectric constants were found to be
are observed in the presence of surfactant monomers; above important in this effect.
the CMC the absorption band centered at # = 312 nm (due Figure 3a presents representative monomer fluores-
to the enolic form [7]) is augmented; and in parallel, the cence spectra for pyrene (in very dilute solutions, [pyrene]
absorption band centered at # = 250 nm (mainly due to the = 2 " 10�6 mol L �1, so that complications due to excimers do
keto form) diminishes. The changes in the absorbance not arise) in aqueous sodium dodecyl sulfate (SDS) solutions
measurements at these two wavelengths with varying con- above and below the CMC. In the absence of micelles (below
centration of surfactant in aqueous solutions are shown in the CMC) pyrene senses the polar environment of water
Figure 2b. molecules; the ratio of fluorescence emission intensities cor-

Table 1. Values of the CMC of Some Surfactants

CMC m1 m2
Surfactant Method Conditionsa %
(mol L�1) ($�1 cm2 mol�1) ($�1 cm2 mol�1)
SDS conductance no additives 8.0 " 10�3 73.5 ± 0.6 25.7 ± 0.2 0.35
SDS conductance [TMU] = 1 mol L�1 1.4–1.5 " 10�2 b 52.5 ± 0.3 39.7 ± 0.6 0.76
SDS conductance [HCl] = 6 mmol L�1 5.5 " 10�3 60.5 ± 0.4 — —
SDS BZA-abs no additives 7.8 " 10�3 — — —
SDS BZA-abs [HCl] = 33 mmol L�1 2.8 " 10�3 — — —
SDS fluorescence no additives 7.4 " 10�3 — — —
TTABr conductance no additives 3.9 " 10�3 92.3 ± 0.3 22.6 ± 0.1 0.24
TTABr BZA-abs no additives 3.5 " 10�3 — — —
TTABr BZA-abs [HCl] = 33 mmol L�1 0.8 " 10�3 — — —
TTABr fluorescence no additives 3.5 " 10�3 — — —
TTABr fluorescence [NaBr] = 0.2 mol L�1 2.3 " 10�4 — — —
TTABr fluorescence [TMU] = 0.56 mol L�1 5.9 " 10�3 — — —
C12E9 BZA-abs no additives 1.6 " 10�4 — — —
C12E9 fluorescence no additives 1.7 " 10�4 — — —
C12E9 fluorescence 15% CH3OH 3.4 " 10�4 — — —
aAll
experiments were conducted at 25 °C.
b
The value 1.5 " 10�2 was determined from the intercept of eq 3. In the other cases, the conductance technique yielded the same value of the
CMC by both the cross point and the intercept method.

1228 Journal of Chemical Education • Vol. 74 No. 10 October 1997

 In th e L a b or a t ory

responding to the first and third vibrational peaks (II/IIII)

is high. Above the CMC, when micelles are present, and
owing to the high hydrophobicity of pyrene molecules, these
are solubilized in the interior micellar phase. This is a hy-
drocarbon-like solvent, so the environment sensed by
pyrene is less polar; and therefore the ratio II/IIII decreases.
Figure 3b shows the variation of the ratio of fluorescence
emission intensities as a function of SDS concentration. The
curves show a sharp decrease in the quotient intensities of
both vibrational peaks when micelles are formed (i.e., at the
CMC). The surfactant concentrations corresponding to the
sharp decrease are listed in Table 1 as the CMC values.
(a) (b)
Procedure. Prepare a pyrene solution by dissolv-
ing 5 mg in 10 mL of methanol. Dilute this solu-
tion 20 times (pipet 0.5 mL into 10-mL volumetric
flasks). To prepare the working mixtures for fluores-
cence measurements, a small aliquot (50 µL) of the
Figure 2. (a) Absorption spectrum of 7.0 " 10�5 mol L�1 BZA. 1, 2, 3: latter methanolic pyrene solution was transferred
[TTABr] below the CMC are; 4: [TTABr] = 4.4 mmol L�1 ; 5: [TTABr] with an automatic pipet to a quartz fluorescence
= 6.6 mmol L �1; 6: [TTABr] = 8.8 mmol L�1; 7: [TTABr] = 13.2 cell and mixed with the appropriate volumes of
mmol L�1; 8: [TTABr] = 17.6 mmol L�1; 9: [TTABr] = 22 mmol L�1 . water, surfactant solution (~0.03 mol L�1), and ad-
Arrows indicate the isobestic points. (b) Absorbance of BZA as a ditive (if any) to give a total final volume of 3 mL,
function of [TTABr] at 312 nm (�) and at 250 nm ( �). and to obtain the final surfactant concentrations
shown in Figure 3. (If it is not possible to measure
a small aliquot of the methanolic solution, the sol-
vent should be allowed to evaporate before prepar-
ing the sample solutions.) The fluorescence emis-
sion was measured with an Aminco-Bowman
spectrofluorimeter. Pyrene was excited at 334 nm
and its emission was recorded at 373 and 384 nm,
which correspond to the first and third vibrational
peaks, respectively, and with the use of excitation
and emission slits of 8 and 2 nm, respectively.

Electrical Conductivity Method

This method can only be applied to measure the CMC
of ionic surfactants. The change in the electrical conduc-
tance of aqueous ionic surfactant solutions at the CMC is
due to the different degree of surfactant ionization below
(surfactant monomers behave as strong electrolytes) and
(a) above (micelles are partially ionized) the CMC. On the as-
sumption that the aqueous surfactant solutions obey
Kohlrausch’s law (9), the specific conductivity, &, of surfac-
tant solutions can easily be calculated in terms of the mo-
lar ionic conductivities of ions (#i = ziui�, where z and u are
the charge and mobility of the ion, and � the Faraday con-
stant); ionic charges are omitted in the subscripts for the
sake of simplicity.
Let us consider the case of SDS under two conditions:
1. Below the CMC, it is accepted that no micelles are
formed; then the specific conductivity of aqueous SDS solu-
tion, in mS�cm�1 (S = $�1), is made up of independent contri-
butions of anions CH3(CH2 )11–OSO3 �, SD� anions and Na+
cations:
& = (#Na + #SD) [SDS]t = m1 [SDS]t (1)
where m1 is the slope of the plot of & vs. [SDS]t below the
(b) CMC.
2. Above the CMC, the conductivity of ionic surfactants
such as SDS usually decreases. This is explained by the
inclusion within the micelle of ions of opposite charge (coun-
terions) to the long-chain ions. The percentage of counter-
Figure 3. (a) Fluorescence emission spectrum of pyrene (2 " 10�6 ions in relation to the number of long-chain ions in the mi-
mol L�1) at [SDS] of 0.012 mol L�1 (—) (i.e., above the CMC) and celle (the aggregation number, Nagg [10]) is called the frac-
3 " 10�3 mol L�1 (…) (i.e., below the CMC). (b) Quotient of vibra- tion of micellar charge neutralized, !; then % (= 1 – !) is the
tional band intensities (II/IIII ) as a function of [surfactant]. ([TMU] degree of micellar ionization. Therefore the concentration
= 0.56 mol L�1.) of free counterions (Na+, in the case of the SDS surfactant)

Vol. 74 No. 10 October 1997 • Journal of Chemical Education 1229

In th e L a b or a t ory

is given by [Na+]f = CMC + %[SDS] m, where [SDS]m = [SDS] t Procedure. To obtain the data shown in Figure 4, a sur-
– CMC. The specific conductivity of an aqueous SDS solu- factant solution of the higher concentration (0.05 mol L�1)
tion at concentrations above the CMC may be regarded as was prepared. To measure the specific conductivity at
divisible into three components: that due to the single ions each [surfactant] we followed the step-by-step dilution–
extraction method (12). The conductivity measurements
(Na+ and SD�) at the CMC, that due to the micellar ions,
were carried out in a jacketed beaker thermostated at
and that due to the Na+ ions in excess. 25 °C by circulating water, and by using a Knick micro-
The specific conductivity of an SDS micellar solution processor conductimeter provided with a four-pole mea-
may be written in the form suring cell (factor cell 1.1 cm�1).

& = (#Na + # SD) CMC + #Na%[SDS]m + #mic[micelles] (2) Discussion

Taking into account that [micelles] = ([SDS] t – CMC)/Nagg From Figure 2a it can be seen that BZA shows two ab-
and assuming that the contribution of the micelle to the sorption bands in the UV region: one centered around 250
conductance is the same as that of an equivalent number nm (due mainly to the ketonic form) and the other centered
of monomeric ions, the sum of whose charges equals the at 312 nm (due to the enolic form). Also seen is the pres-
micellar charge, then #mic = %Nagg#SD and eq 2 becomes eq 3: ence of two well-defined isosbestic points in the absorption
spectrum of BZA as a function of [surfactant]. The students
& = (#Na + #SD) CMC(1 – %) + (#Na + #SD)%[SDS] t = are required to relate the two absorption bands with the
&o + m2 [SDS]t (3) possible electronic transitions (' ( '* and n ( '*) in both
keto and enol tautomers of BZA, and to explain the mean-
where m2 (= m1%) is the slope of the linear plot of & vs. [SDS]t ing of the isosbestic points in terms of the keto–enol equi-
above the CMC, and &o is the corresponding intercept. librium.
The CMC value can then be determined either as the It is seen that #max in the case of BZA (see Fig. 2a) does
cross point of the two straight lines defined by eqs 1 and 3, not change in the presence of micelles, whereas in the fluo-
or from the intercept of the straight line obtained above the rescence spectrum of pyrene (see Fig. 3a) # max shifts toward
CMC (eq 3) together with the values of the slope m1 and
the fractional micellar ionization %, which can be obtained
as the ratio of the slopes of conductance vs. [SDS]t above
and below the CMC; that is, % = m2 /m 1. (a)
Figure 4a shows results obtained working with aque-
ous solutions of SDS in the presence and absence of 1
mol L �1 of tetramethylurea (TMU). The following parameter
values were obtained: in the absence of TMU, m 1 = 73.5 ±
0.6 and m 2 = 25.7 ± 0.2 $ �1 cm2mol�1 , and % = 0.35; in the
presence of TMU, m 1 =52.5 ± 0.3 and m 2 = 39.7 ± 0.6
$�1cm2 mol�1 and % = 0.76. The CMC values obtained in each
case are listed in Table 1. Both CMC and % values increase
in the presence of TMU. This is because, in the first case,
TMU decreases the dielectric constant of water mixtures,
and thus the solubility in water of the hydrophobic mono-
mers of SDS increases. In the second case, the inclusion of
TMU molecules within the micelle decreases the electrical
repulsions between the ionic head groups (–OSO3� in SDS)
and the degree of micellar ionization, %, increases.
The same treatment can be applied to any other ionic
surfactant to determine CMC and % values.
Figure 4b shows the specific conductivity of aqueous SDS
solutions in the presence of a constant concentration of HCl. (b)
Below the CMC the conductivity increases, owing to the in-
dependent contributions of Na +, SD�, H+, and Cl� ions:

& = (#H + #Cl)[HCl] + (#Na +#SD ) [SDS]t (4)

Above the CMC part of the Na+ ions included within the
micelle are exchanged with H+ ions: Nam+ + Hw+ Na+w + Hm+,
with KI being the corresponding equilibrium constant whose
value are in the range of 0.7–1.0. The subscripts m and w
refer to micelle and aqueous pseudophases, respectively.
Then the specific conductivity of the solution is given by

& = &o + (#H +# Cl) [HCl] + (#Na – #H) [H +]m

+ % (#Na+#SD ) [SDS] t (5)

but as #H = 349.8 $�1cm2mol�1 and #Na = 50.1 $ �1cm2 mol�1

(11) the observed specific conductivity of the solution de- Figure 4. Electrical conductivity, in mS cm�1 (S = $ �1). (a) As a
creases at low [SDS] because the third (and negative) term function of [SDS] with no additives (� ), and [TMU] = 1 mol/dm3
predominates over the fourth. (� ). (b) In the presence of [HCl] = 6.0 " 10�3 mol L�1.

1230 Journal of Chemical Education • Vol. 74 No. 10 October 1997

 In th e L a b or a t ory

red in the presence of micelles. Again, the students are re- factants. The results obtained in the presence of high elec-
quired to relate these effects with the possible location of trolyte concentrations are not good, as a consequence of the
both probes (BZA and pyrene) within the micelle and to residual conductance by the electrolyte. Another problem
draw conclusions about the polarity of the different regions arises when an ion of the salt can exchange with surfac-
in a micelle. tant counterions, as is the case of SDS in the presence of
In Table 1 one can observe the opposite effects on the H+ ions or of TTABr in the presence of OH� ions.
CMC of electrolytes (NaCl, HCl) and nonelectrolytes
(tetramethylurea, methanol), which are a consequence of Acknowledgment
the polarity of the water–electrolyte and water–nonelectro-
lyte media. The presence of TMU increases the degree of Financial support from the Xunta de Galicia (project
ionization of SDS micelles. This is due to the solubilization XUGA 10302A95) is gratefully acknowledged.
of TMU in the Stern layer of the micelle, which separates
the ionic head groups and consequently decreases the re- Literature Cited
pulsion between them.
1. Fendler, J. H. M e m b r a n e M i m e t i c C h e m i s t r y ; Wiley
Finally, it is important to recognize that there may be
Interscience: New York, 1982; Fendler, J. H.; Fendler, E. H.
systematic differences among methods for determining the Catalysis in Micellar and Macromolecular Systems; Academic:
CMC. The solubilization of a dye (or any hydrophobic spe- New York, 1975; Bunton, C. A.; Nome, F.; Quina, F. H.; Romsted,
cies, such as BZA) in micelles depends mainly on the total L. S. Acc. Chem. Res. 1991, 24, 357.
amount of micelles in a system. Therefore, in general, the 2. Roessler, N. J. Chem. Educ. 1979, 56, 675.
method of keto–enol tautomerism of BZA is not applicable 3. Furton, K. G.; Norelus, A. J. Chem. Educ. 1993, 70, 254.
to determine the CMC of surfactants with low CMC values 4. Rujimethabhas, M.; Wilairat, P. J. Chem. Educ. 1978, 55, 342.
(less than ~5 " 10�4 mol L�1). Nevertheless, this method is 5. Worley, J. D. J. Chem. Educ. 1992, 69, 678.
satisfactory for the determination of CMC of anionic, cat- 6. Myers, D. Surfaces, Interfaces, and Colloids; VCH: New York,
1991; p 317.
ionic, or nonionic surfactant under any experimental condi-
7. Iglesias, E. J. Phys. Chem . 1996, 100, 12592.
tions of electrolyte or nonelectrolyte concentration. The fluo- 8. Kalyanasundaram, K. In Photochemistry in Organized and
rescence method is the most versatile and precise. A vari- Constrained Media; Ramamurthy, V., Ed.; VCH: New York,
ety of probe molecules have been used to determine the 1991; p 54; Bohne, C.; Redmond, R. W.; Scaiano, J. C. In Pho-
CMC of micelles, most of which have fluorescence param- t oc h e m i s t r y i n O r g a n i z e d a nd C o n s t r a i n e d M e d i a ;
eters sensitive to solvent polarity, such as pyrene. To deter- Ramamurthy, V., Ed.; VCH: New York, 1991; p 100.
mine correct values for the CMC, the probes employed must 9. Vemulapalli, G. K. P h y s i c a l C h e m i s t r y ; Prentice Hall:
be mostly solubilized in the micellar phase (high partition- Englewood Cliffs, NJ, 1993; p 946.
ing coefficient) and special care must be taken to ensure 10. Rodríguez-Prieto, M. F.; Rios-Rodríguez, M. C.; Mosquera-
González, M.; Rios-Rodríguez, A.; Mejuto-Fernández, J. C. J.
that no aggregation between the probe and surfactant
Chem. Educ. 1995, 72, 662.
monomers occurs below the CMC. The method of the vibra- 11. Sime, R. J. Physical Chemistry. Methods, Techniques, and Ex-
tional emission spectrum of pyrene reported here can also periments ; Saunders: Philadelphia, 1990; p 561.
be applied to test for polarity in micelles. The electrical con- 12. Jover, A.; Meijide, F.; Mosquera, V.; Vázquez-Tato, J. J. Chem.
ductance method, obviously, can only be applied to ionic sur- Educ. 1990, 67, 530.

Vol. 74 No. 10 October 1997 • Journal of Chemical Education 1231