A new approach to the phase behavior of oppositely charged polymers and surfactants

Article abstract of DOI:10.1021/jp0120458

The complex salt (ionic surfactant + polymeric counterion) cetyltrimethylammonium polyacrylate (CTAPA) has been synthesized, and its aqueous mixtures with cetyltrimethylammonium bromide (CTABr) have been studied. These mixtures differ from conventional oppositely charged polymer/surfactant mixtures in that the conventional counterion of the polyion (usually sodium, for the polyacrylate) is absent, which simplifies the studies and their interpretation considerably. The phase diagram of the CTAPA/CTABr/water system at > 20 wt % water and at 40 ?°C has been established, representing the first truly ternary phase diagram of an oppositely charged polymer/surfactant pair in water. The two dimensions of the phase diagram may be chosen as the water content (in weight percent) and the fraction of bromide counterions, x_Br (in units of charge equivalents). The phase diagram is characterized by a large hexagonal phase (at low water contents and for all values of x_Br), a small cubic phase (at 55 wt % water content and for x_Br < 0.1), a narrow isotropic (micellar) phase (at high water contents and for x_Br > 0.9), and a large multiphase region (at water contents >50 wt %) containing two or three of the cubic, hexagonal, or isotropic phases in coexistence. The cubic and hexagonal phases are connected to the corresponding phases that separate out from aqueous NaPA/CTABr mixtures. The maximum water uptake of the hexagonal phase is remarkably constant at ca. 50 wt % over a large CTAPA/CTABr composition range (x_Br < 0.9). The study confirms previous conclusions that the polyacrylate counterions favor a higher aggregate curvature (leading to smaller aggregates) than do the bromide counterions.

Full text of DOI:10.1021/jp0120458

J. Phys. Chem. B 2002, 106, 1013-1018
1013
A New Approach to the Phase Behavior of Oppositely Charged Polymers and Surfactants
Anna Svensson,† Lennart Piculell,*,† Bernard Cabane,‡ and Philippe Ilekti‡
Physical Chemistry 1, Center for Chemistry and Chemical Engineering, Lund UniVersity, P.O.Box 124,
S-221 00 Lund, Sweden, and Laboratoire PMMH, ESPCI, 10 rue Vauquelin, 75231 Paris Cedex 05, France
ReceiVed: May 30, 2001; In Final Form: October 29, 2001
The complex salt (ionic surfactant + polymeric counterion) cetyltrimethylammonium polyacrylate (CTAPA)
has been synthesized, and its aqueous mixtures with cetyltrimethylammonium bromide (CTABr) have been
studied. These mixtures differ from conventional oppositely charged polymer/surfactant mixtures in that the
conventional counterion of the polyion (usually sodium, for the polyacrylate) is absent, which simplifies the
studies and their interpretation considerably. The phase diagram of the CTAPA/CTABr/water system at >20
wt % water and at 40 °C has been established, representing the first truly ternary phase diagram of an oppositely
charged polymer/surfactant pair in water. The two dimensions of the phase diagram may be chosen as the
water content (in weight percent) and the fraction of bromide counterions, xBr (in units of charge equivalents).
The phase diagram is characterized by a large hexagonal phase (at low water contents and for all values of
xBr), a small cubic phase (at 55 wt % water content and for xBr < 0.1), a narrow isotropic (micellar) phase (at
high water contents and for xBr > 0.9), and a large multiphase region (at water contents >50 wt %) containing
two or three of the cubic, hexagonal, or isotropic phases in coexistence. The cubic and hexagonal phases are
connected to the corresponding phases that separate out from aqueous NaPA/CTABr mixtures. The maximum
water uptake of the hexagonal phase is remarkably constant at ca. 50 wt % over a large CTAPA/CTABr
composition range (xBr < 0.9). The study confirms previous conclusions that the polyacrylate counterions
favor a higher aggregate curvature (leading to smaller aggregates) than do the bromide counterions.
Introduction
In recent years, an increasing number of studies of oppositely
charged polymer/surfactant mixtures have focused on concen-
trated systems. Such systems include the dry complex salts,1-7
the concentrated “precipitates” or “coacervates” that separate
out from aqueous mixtures in certain concentration intervals,8-16
and the collapsed gels that result when polyelectrolyte gels are
immersed in solutions of oppositely charged surfactant.17-24
A
variety of surfactant aggregate structures (micellar, cubic,
hexagonal, lamellar, et al.) have been observed in such systems,
but the understanding of the various factors that determine these
aggregate structures, and their miscibility with water, is still in
its infancy. Apart from the characteristics of the polyion (charge
density, rigidity, hydrophobicity/hydrophilicity) or the surfactant
(chain length, identity of charged group), a key question
concerns the effects of simply varying the concentrations of
the various ionic species (the polyion, the surfactant ion, the
simple counterions) contained in the mixture. To what extent
may the structures and concentrations of the concentrated phase
be manipulated by the choice of the initial content of surfactant,
polyelectrolyte, (external salt), and water in the mixture from
which phase separation occurs?
Recently, Ilekti et al.8,9 investigated the mixtures of sodium
polyacrylate (NaPA), cetyltrimethylammonium bromide (CTABr),
and water in some detail with the questions above in mind. A
rich variety of structures (micellar, hexagonal, cubic) and water
contents were found, depending on both the surfactant/poly-
electrolyte ratio and the overall water content of the initial
Figure 1. A three-dimensional pyramid phase diagram (schematic)
that fully describes the phase behavior of aqueous mixtures of a charged
surfactant with an oppositely charged polyelectrolyte. See text.
mixture. Indeed, it was possible to vary both the structure and
the concentration of the concentrated phase(s) simply by
successively diluting a stoichiometric mixture of polyelectrolyte
and surfactant with water.
If the results were interesting and promising, the studies by
Ilekti et al.,8,9 as well as previous less-detailed studies,25,11 also
clearly illustrated the inherent difficulty in studying the phase
behavior of conventional mixtures of polyelectrolyte, surfactant,
and water. Such systems are four-component systems, and their
full description requires a three-dimensional phase diagram, such
as the pyramid phase diagram proposed by Thalberg et al.25
(Figure 1). The pyramid has water at the top and the four
possible combinations of the four ions into neutral salts at the
* Corresponding author. E-mail: Lennart.Piculell@fkem1.lu.se.
† Lund University.
‡ ESPCI.
10.1021/jp0120458 CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/12/2002

1014 J. Phys. Chem. B, Vol. 106, No. 5, 2002
Svensson et al.
corners of its quadratic base. Here, as in refs 8 and 9, we have
chosen to use the following terminology for the various ions
and neutral salts. The surfactant is the salt composed by the
surfactant ion and its (normal) simple counterion, whereas the
polyelectrolyte is the salt composed by the polyion and its simple
counterion. The complex salt is then composed by the surfactant
ion and the polyion, whereas the simple salt, finally, is the
combination of the two simple counterions from the surfactant
and the polyelectrolyte.
Oppositely charged polymer/surfactant mixtures are generally
prepared as mixtures of the surfactant with the polyelectrolyte;
this defines the conVentional mixing plane indicated in Figure
1. A two-dimensional map of the phase behavior of conventional
mixtures can be drawn in this plane, defining one-phase, two-
phase, and multiphase areas. The problem with this description,
however, is that the compositions of the coexisting phases in
the two- or multiphase areas are not situated within the
conventional mixing plane, that is, their compositions cannot
be described as combinations of polyelectrolyte, surfactant, and
water. This is schematically illustrated by the tie-line in Figure
1, which shows that, quite naturally, the concentrated phase that
separates out is enriched in the complex salt, whereas the dilute
phase (often) essentially contains only simple salt and water.
Apart from the difficulties to describe the phase diagram
graphically, it is clear that a full description of the pyramid is
cumbersome, to say the least.
same batch) as that used in refs 8 and 9. The batch in refs 8
and 9 was examined through size-exclusion chromatography
coupled with low-angle light scattering. (See the references for
experiment details.) The HPA had a number average molar mass
Mn of 2800 g/mol and a weight average mass Mw of 4700 g/mol.
The polymer batch used in this study was purified by dialysis
for 5 days against Millipore water, followed by freeze-drying.
1H NMR revealed a small amount of an organic impurity in
the polymer, which remained even after the dialysis procedure.
Two different batches of HPA both showed the presence of the
impurity, which we tentatively ascribe to heterounits at the ends
of the short polymer, originating from the synthesis procedure.
Titration of HPA with NaOH showed that the equivalent molar
mass of the polymer was 89.3 g/(mol carboxylic acid), as
compared with the theoretical value of 72 g/mol for a repeating
unit of HPA. Presumably, this difference is caused primarily
by the impurity in the polymer. The contribution from water to
the equivalent molar mass should be small. The water uptake
of freeze-dried HPA exposed to ambient air was 2 wt % after
1 h and 5 wt % after several days.
CTABr was purchased from Merck and used without further
purification. The molar mass of CTABr is 364.5 g/mol. The
molar mass of the bromide ion, 79.9 g/mol, is close to the molar
mass per charged unit of PA. Thus, the equivalent molar masses
(the molar masses per ion pair) of CTAPA and CTABr are
almost the same (within 2%). The (equivalent) fraction of
In this study, we have circumvented the problem above by a
seemingly obvious strategy that, nevertheless, seems not to have
been used previously. We have eliminated one of the simple
ions from our mixtures. Thus, we have chosen to study a truly
ternary mixture (neglecting the polydispersity of the polyion)
of the complex salt, the surfactant, and water. This mixing plane
corresponds to the front face of the pyramid in Figure 1. By
performing this study, we are thus asking the question: How
does the phase behavior of an ionic surfactant in water change
when the counterions are changed from pure simple counterions,
via mixtures of simple counterions and polyions, to pure
polyions? As far as we are aware, this simple and basic question
has not previously been addressed in a clean systematic study.
For this first study of the phase behavior of a ternary
surfactant/complex salt/water mixture (additional studies are in
progress), we have chosen CTABr as the surfactant and
cetyltrimethylammonium polyacrylate (CTAPA) as the complex
salt. Thus, the results here belong to the same pyramid as the
mixtures of CTABr and NaPA that were studied in detail
previously.8,9 The polyion, PA-, is quite hydrophilic; there is
no strong hydrophobic attraction between the polyion and the
surfactant aggregate, and hence the polyion-surfactant attraction
is essentially electrostatic. This is evidenced because the
complex CTAPA salt dissolves in water on the addition of
simple salt caused by electrostatic shielding.
bromide counterions in a sample will be expressed as xBr
(equivalents of bromide)/(total equivalents of bromide +
polyacrylate).
)
The complex salt, CTAPA, was prepared by titrating the
hydroxide form of the surfactant with the acid form of the
polymer. The first step was to convert CTABr into CTAOH by
ion exchange.26 The ion-exchange resin (Dowex SBR, dry mesh
20-50, from Sigma) was charged by stirring in an excess
amount of 1 M NaOH for 2 h and then rinsed with Millipore
water until the pH of the water reached 7. CTABr (40 g) was
then dissolved in a plastic beaker containing 200 g of the
charged ion-exchange resin (a large excess of hydroxide ions)
and 300 mL of Millipore water. The solution was stirred until
all CTABr was dissolved. The slurry was filtered and the filtrate
rinsed with Millipore water into a new batch of 200 g resin
and 300 mL water, which was stirred for another 2 h. The last
step was repeated once more with a third fresh batch of resin.
The alkaline solution now contained CTAOH at a concentration
of approximately 0.05 M. A solution of HPA (0.5 M) was
titrated drop by drop into the freshly prepared solution of
CTAOH under stirring. This was done immediately to avoid
Hofmann elimination of the quaternary ammonium hydroxide
group of the surfactant in the basic solution.27 (Any possible
uptake of carbon dioxide from the air should have been driven
back by the addition of acid to the solution.) The pH was
measured by using a standard pH electrode. A white precipitate,
the complex salt was formed at the start of the titration. The
titration was continued until the equivalence point was reached
in the solution. The latter point was taken as the inflection point
(pH ) 8.6) in the pH-titration curve, as determined in a separate
measurement. After equilibrating overnight, the solution with
the precipitate was freeze-dried. The complex salt was then
obtained as a white, hygroscopic powder, which was put to
storage over a silica gel in a desiccator. Titrimetric analysis gave
a bromide content below the detection limit (0.3 wt %) in the
complex.
An interesting conclusion from the previous studies of
CTABr/NaPA mixtures was that the curvature and geometry
of the surfactant aggregate was influenced not only by the
change from monomeric to polymeric counterions per se, but
also by the chemical identity of the charged group (bromide vs
carboxylate). Another objective of our study is to elucidate this
specific point further. As in the previous studies, both phase
compositions and phase structures have been determined, the
latter by small-angle X-ray scattering (SAXS) experiments.
Materials and Methods
Materials. Poly(acrylic) acid (HPA; 2 000 g/mol), was
purchased from Aldrich. This is the same product (but not the
Weighing experiments indicated a weight increase of the
complex salt, through water uptake, by approximately 10 wt %

Phases of Oppositely Charged Polymers/Surfactants
J. Phys. Chem. B, Vol. 106, No. 5, 2002 1015
in the desiccator, and by 20 wt % after prolonged storage in
air. Care was taken to minimize the exposure of the complex
salt to air during sample preparation, and a water content of 10
wt % in the complex salt was assumed when calculating the
sample compositions. The water uptake of the surfactant,
CTABr, was less than 1 wt % in air.
Sample Preparation. Appropriate amounts of the complex
salt, the surfactant and water were weighed and put in glass
tubes. After mixing with a Vortex vibrator, the tubes were flame-
sealed. The mixing was continued in a centrifuge during 6 h at
4 000 rpm and 40 °C. The tubes were turned end over end every
15 min. The samples were left to equilibrate in an oven at 40
°C for several weeks. This temperature was chosen to exceed
the Krafft point of CTABr (23 °C); it was also the temperature
chosen in the earlier studies of CTABr/NaPA/water mixtures.8,9
Methods. The samples were investigated by visual inspection
in normal light and between crossed polarizers to detect optically
anisotropic phases (in the present case, the hexagonal phase).
SAXS measurements were performed with two different
setups. At the D 43 instrument at L.U.R.E. in Orsay, France, a
parallel monochromatic beam of wavelength 1.445 Å was
focused with point collimation on the sample. The sample was
contained in a flat cell with mica windows kept at 40 °C. The
sample-to-detector distance was 395 or 377 mm. A Kratky
compact small-angle system with linear collimation was used
at the Lund laboratory. The X-rays were detected with a positive
sensitive detector. The wavelength was 1.54 Å, and the sample-
to-detector distance 277 mm. The sample cell had mica windows
and was maintained at 40 °C.
Figure 2. Experimental phase diagram of the complex salt CTAPA,
the surfactant CTABr, and water at 40 °C. The mixing plane
corresponds to the front face of the pyramid in Figure 1.
The bromide content of dilute phases was determined by a
titrimetric analysis. The sample was titrated with a mercuric
nitrate solution with diphenyl carbazone as indicator.
Results and Discussion
Overview of the Phase Diagram. The phase diagram of the
ternary system CTAPA/CTABr/water at 40 °C is presented in
Figure 2. Sample compositions are given as weight percent of
the components. Owing to the closely similar equivalent molar
masses of CTAPA and CTABr (cf. the materials section), the
total equivalent concentration of salt at a given weight percent
of solids is very nearly independent of the counterion; moreover,
the equivalent ratio of CTABr to CTAPA is almost equal to
the mass ratio.
The phase diagram is based on analyses of almost 150
samples. It contains, in the investigated composition region (>20
wt % water) three single-phase regions, three two-phase regions,
and one three-phase triangle. The dominating parts of the phase
diagram are the single hexagonal phase (which extends all the
way across, from pure CTAPA, via all mixing ratios, over to
pure CTABr), and the large two-phase area with coexisting
micellar and hexagonal phases. The other single-phase regions,
the cubic phase and the micellar phase, are limited to the almost
pure ion forms CTAPA and CTABr, respectively. Note that the
phase diagram, being a true ternary phase diagram, contains
explicit tie lines connecting coexisting compositions of the
single-phase regions.
Figure 3. SAXS spectrum of the hexagonal phase containing 60 wt
% CTAPA and 40 wt % water at 40 °C. Relative locations of visible
peaks: 1, 31/2, 41/2
.
water content is ca. 55 wt % of the total mass, corresponding
to 25 water molecules per ion pair of CTAPA. Further added
water ends up as a separate phase of essentially pure water (cf.
the biphasic area on the water-CTAPA axis in Figure 2).
Below 55 wt % water, different structures of CTAPA are
found, depending on the water content. The dry complex salt
is a white and hygroscopic powder. It readily absorbs humidity
from air up to 20 wt % water (5 water molecules per ion pair
of CTAPA). SAXS measurements on a sample that contained
ca. 10 wt % water indicated a mixture of lamellar structures.
Previous investigators have also found a lamellar structure for
dry CTAPA.6,14,29 A detailed investigation of the dry or very-
water-poor systems is beyond the scope of this investigation.
At water contents between ca. 30 and 50 wt %, the CTAPA/
water mixtures are translucent, solidlike, and birefringent. The
SAXS measurements show hexagonal patterns (Figure 3), and
this region may thus be identified as a hexagonal single-phase
region. At about 50 wt % of water, a sharp transition occurs;
Below, we will discuss in some detail the trends that are
observed as the compositions of the mixtures are varied along
different paths in the phase diagram.
The Binary Mixtures. As is well-known, CTABr is highly
water soluble, forming a micellar or a hexagonal phase at high
or low water contents, respectively.28 By contrast, as expected,
CTAPA only takes up a limited amount of water. The maximum

1016 J. Phys. Chem. B, Vol. 106, No. 5, 2002
Svensson et al.
not succeed in separating the cubic and hexagonal phases
macroscopically; hence, they could not be analyzed separately.
The limited extension of the cubic phase is a striking
feature: replacing a small amount of PA by Br results in the
formation of a new, hexagonal, phase, which is enriched in Br.
A similar behavior was observed in previous work on the phase
behavior of stoichiometric CTABr/NaPA mixtures with increas-
ing amounts of added water.8 The authors found that an
increasing total water content led to a decreasing bromide
content of the surfactant-rich phase, and that the latter was
hexagonal down to very low bromide fractions (e.g., xBr ) 0.28).
This result agrees with the phase diagram in Figure 2, where
the phase in equilibrium with the dilute phase is hexagonal down
to a bromide fraction of 0.20-0.30. In the CTABr/NaPA
mixtures, the cubic structure of the surfactant-rich phase
developed only when the bromide fraction had decreased further
(e.g., at a total water content of 99 wt %).
Figure 4. SAXS spectrum of the cubic phase containing 45 wt %
CTAPA and 55 wt % water at 40 °C. Relative locations of peaks (with
invisible peaks of the Pm3n structure given in parentheses): (1), 21/2
,
,
Addition of CTAPA to CTABr/Water Mixtures. We saw
above that small additions of CTABr sufficed to make one of
the structures, the cubic phase, disappear. Similarly strong effects
are seen at the other end of the phase diagram, when small
amounts of CTAPA are added to aqueous CTABr. The
appearance of the phase diagram makes it convenient to consider
the effects of added CTAPA separately for three different
concentration regions of aqueous CTABr, corresponding to the
micellar phase, the dilute hexagonal phase, and the intermedi-
ately concentrated hexagonal phase.
41/2, 51/2, 61/2, 81/2, 101/2, 121/2, (131/2), 141/2, 161/2, 171/2, (181/2), 201/2
211/2
.
the samples become optically isotropic and the SAXS spectra
show a cubic structure of the Pm3n type (Figure 4). This cubic
structure is composed of slightly elongated globular micelles
in a body-centered cubic arrangement.30 The same structure was
previously observed for the concentrated phase that separates
out from aqueous CTABr/NaPA mixtures at very high dilution.8
The resolution of the spectrum is much better in the present
study, and 12 of the spectral lines can be identified (Figure 4).
The micellar phase is only stable down to a bromide fraction
of xBr ≈ 0.90. At a larger fraction of CTAPA, a concentrated
hexagonal phase, enriched in CTAPA, separates out from the
mixture.
The stability range of the cubic phase in the binary CTAPA/
water mixture is quite narrow. In our experiments, an increase
of the water content beyond 55 wt % resulted in samples that
were initially whitish, but after prolonged storage (for months)
at 40 °C separated into a clear cubic top phase and a low-viscous
aqueous bottom phase. (The density of aqueous CTAPA is lower
than the density of water.8) The solubility of the complex salt
in water was not determined, but in all probability it is very
low. The most dilute mixture that we prepared (1 wt % of
CTAPA in water) was clearly still biphasic. The concentrated
phases from several samples in the biphasic region were
investigated by SAXS, giving spectra identical (within experi-
mental error) with the one shown in Figure 4.
When a few percent of CTAPA are added to a dilute
hexagonal phase of CTABr in water (up to ca. 35 wt %
surfactant), the sample is transformed into a monophasic micellar
solution. As a result, the micellar phase extends to higher
concentrations when polyacrylate counterions are present. A
similar effect of adding NaPA to aqueous CTABr was observed
in ref 9 and was explained in terms of the effect of short-range
counterion-surfactant interactions on the curvature of the
surfactant aggregate. Polyacrylate, just as the corresponding
simple acetate counterion, favors a higher curvature, and brings
about a tendency toward spherical or finite aggregates. Bromide,
on the other hand, gives a lower curvature and more extended,
rodlike aggregates. The hexagonal-to-micellar transition induced
by replacing bromide in CTABr with polyacrylate thus mirrors
the cubic-to-hexagonal transition that occurs when the poly-
acrylate ions in CTAPA are replaced by bromide (cf. above).
Addition of CTABr to CTAPA/Water Mixtures. The most
striking effect of adding CTABr to CTAPA/water mixtures is
that the cubic phase rapidly disappears. The pure cubic phase
extends only to xBr ) 0.06; a larger fraction of bromide
counterions leads to the formation of a hexagonal phase.
Evidently, (cf. also below) the curvature of the surfactant
aggregate decreases when polyacrylate counterions are replaced
by bromide counterions. This decrease in curvature induces a
transition from essentially globular to cylindrical surfactant
aggregates. The difference between the two counterions may
be attributed to the fact that bromide ions bind “chemically” to
cationic surfactant aggregates, whereas carboxylate ions, such
as the simple acetate ion, remain fully dissociated.31 The
hexagonal and the cubic phases coexist up to a bromide fraction
of 0.20-0.30, and at high water contents, a three-phase
coexistence exists between these two phases and excess water.
At still higher bromide fractions, the large two-phase area of
hexagonal/micellar coexistence appears.
For hexagonal CTABr samples at intermediate concentrations
(35-50 wt %), the successive replacement of bromide coun-
terions with polyacrylate eventually results in a phase separation
into a concentrated (50 wt %) hexagonal phase and a more dilute
micellar phase. This part of the wide hexagonal/micellar two-
phase region is the most difficult region to investigate in the
entire phase diagram. The reasons for this are the following.
First, both of the coexisting phases are quite concentrated in
this region. As a result, the development of a macroscopic phase
separation is very slow. Another consequence of the concen-
trated viscous phases is a possibility that a shear-induced
orientation of the micellar phase, as a result of the introduction
of the sample into the sample cell, could persist for a very long
time. Whereas some of the micellar phases in this region showed
an anisotropic two-dimensional diffraction pattern, which could
indicate a nematic phase, we cannot exclude that this anisotropy
was actually caused by such shear-induced ordering.
Samples from all multiphase areas were checked visually,
and their multiphase nature was confirmed by SAXS analyses,
showing superposed cubic/hexagonal structures in the cubic/
hexagonal two-phase area and in the three-phase area. We did

Phases of Oppositely Charged Polymers/Surfactants
J. Phys. Chem. B, Vol. 106, No. 5, 2002 1017
Addition of Water to the Hexagonal Phase. We will end
our tour around the phase diagram by considering the effect of
adding water to the hexagonal phase at different ratios of the
counterions. We have already noted the transition to a cubic
phase at low bromide contents. What remains to be discussed
in more detail is the transition from the pure hexagonal phase
into hexagonal-micellar coexistence. At high bromide contents,
the behavior is that of the regular hexagonal phase of CTABr.
A repulsion occurs between the surfactant aggregates, and the
hexagonal phase therefore swells continuously when water is
added. Eventually (at 70 wt % water) the long-range order is
lost, and across a narrow two-phase region a transition occurs
to an isotropic micellar phase. By contrast, when water is added
to a hexagonal phase containing a sufficient (small) fraction of
polyacrylate counterions, a maximum water content exists above
which the hexagonal phase refuses to swell. Further added water
ends up in a more dilute micellar phase. Clearly, the polyacrylate
counterions give rise to an attractive force that keeps the
surfactant aggregates together. This attraction may have two
causes: polymer bridging, or an electrostatic attraction between
surfactant aggregates with polyions condensed on their surfaces.
Remarkably, the phase diagram shows that the maximum
water uptake of the hexagonal phase is almost constant at about
50 wt % across a large range of Br/PA ratios in the middle of
the phase diagram. This indicates that the minimum in the
interaction potential is located at approximately the same
distance between the surfactant aggregates for all counterion
compositions below xBr ) 0.90. We believe that the underlying
reason for this is that the polyion-bridging attraction is short-
ranged. As the proportion of polymeric counterions is decreased,
the depth of the potential minimum may be strongly affected,
but not its position. At sufficiently low proportions of polymeric
counterions, however, the minimum disappears, and the interac-
tion force becomes repulsive. Evidently, this is what happens
in the experiments above xBr ) 0.90.
Figure 5. Values of q1, the first hexagonal peak in the SAXS spectrum,
for a series of aqueous CTAPA/CTABr samples at 60 wt % water and
with different bromide contents. Samples at xBr < 0.9 are biphasic
(hexagonal + micellar).
A second reason for the difficulties to analyze biphasic
micellar/hexagonal samples at high bromide contents is that a
density inversion occurs somewhere in this region. As we have
pointed out above, the mass per unit volume of CTAPA/water
mixtures decreases with increasing CTAPA content, whereas
the reverse is true for aqueous CTABr. Thus, with mixed
counterions an intermediate situation must exist in which the
less concentrated (micellar) phase has exactly the same density
as the more concentrated (hexagonal) phase. In the region of
density inversion, centrifugation obviously does not help to
achieve a macroscopic phase separation.
To illustrate the transitions occurring in this bromide-rich
region at 50-65 wt % water, we investigated by SAXS a series
of samples with different Br/PA ratios, including pure CTABr,
at a constant water content of 60 wt %. The results are shown
in Figure 5. As Figure 5 shows, the q value of the first hexagonal
peak stays at about 0.11 Å-1 in the single-phase hexagonal
region (at high Br contents), then it suddenly jumps to ca. 0.14
A final feature to note regarding the hexagonal-micellar
coexistence is the direction of the tie lines, showing that the
bromide fraction is always higher in the micellar phase than in
the coexisting hexagonal phase. (As pointed out above, the
minimum bromide content of the micellar phase is rather
constant at xBr ≈ 0.90.) This is, of course, entirely expected;
the monomeric counterions are enriched in the more dilute
phase.
Å-1 as the two-phase region is entered (approximately xBr
≈
0.90). Within the two-phase region, nothing much happens to
the hexagonal spacing as the bromide content is lowered further.
This result is in accordance with the phase diagram in Figure
2, which shows that the maximum water content of the
hexagonal phase is constant across a wide range of compositions.
The Hexagonal Phase. The hexagonal phase is the only
single phase that extends across the phase diagram, regardless
of the PA/Br ratio. At a fixed water content, the characteristic
dimensions of the hexagonal phase vary only weakly with the
nature of the counterion. The radius of the cylindrical aggregates,
R , was calculated from the first diffraction line, q1, of the
hexagonal spectrum and the volume fraction, fs, of surfactant
ions (assuming a specific volume of 1 cm3/g) by the following
set of equations.
Concluding Remarks
In this work we have demonstrated that the study of the phase
behavior of oppositely charged polymer-surfactant systems is
simplified immensely if one of the simple ions is eliminated
from the mixture. The strategy involves preparing the complex
polyion-surfactant ion salt and studying its binary mixtures with
water, and ternary mixtures including, in addition, either the
surfactant or the polyelectrolyte. Here, we have chosen to study
mixtures with the surfactant, addressing the question as to the
effects on the surfactant phase behavior of gradually exchanging
the counterions from 100% polymeric to 100% monomeric. We
note several interesting results of our study.
• Surfactant mesophases with polymeric counterions have a
limited swelling in water. This reveals the presence of an
attractive force that keeps the surfactant aggregates together.
An interesting problem for future research is to discover the
nature of this force, and the balance of forces that determines
the maximum swelling.
2 2π
a )
(1)
(2)
q
x
3
φs
x
R ) a
3
x2π
At 35 wt % water R increases slightly, from 16 to 20 Å, as the
counterion is exchanged from polyacrylate to bromide. This
effect is in line with the findings in ref 9, and it is consistent
with the general trend that the carboxylate group of polyacrylate
induces a higher curvature of the aggregates than bromide.
• The ternary CTABr/CTAPA/water phase diagram contains
both of the mesophases (cubic and hexagonal) that have

1018 J. Phys. Chem. B, Vol. 106, No. 5, 2002
Svensson et al.
previously been found to separate out from dilute CTABr/NaPA/
water mixtures. The stability of those phases, with respect to
the PA/Br ratio and the overall water content, has been clarified.
The cubic structure exists essentially for pure CTAPA at its
maximum water content. The marginal stability of this structure
with respect to changes in either water or bromide content is
intriguing. A single hexagonal phase has been shown to extend
through all possible mixing ratios of polyacrylate and bromide
counterions. This result confirms that the hexagonal phase
separating from CTABr/NaPA mixtures in water is connected
to the “normal” hexagonal phase of CTABr, as speculated in
ref 8.
• The micellar phase of CTABr can only accept a limited
“contamination” with polymeric counterions before a hexagonal
phase separates out. This is a consequence of the same attractive
forces that limit the maximum water uptake of the hexagonal
phase, that is, the bridging of the surfactant aggregates by the
polyions.
• The importance of short-range interactions between the
counterions and the surfactant aggregates has been confirmed
and clarified by this study, as well as the previous conclusion9
that the polyacrylate counterion favors a higher aggregate
curvature than does the bromide counterion. The latter conclu-
sion is supported by (i) the cubic-to-hexagonal transition as PA-
is replaced by Br-, (ii) the hexagonal-to-micellar transition as
Br- (in the dilute hexagonal region) is replaced by PA-, and
(iii) the decreased radius of the rodlike micelles in the hexagonal
phase as Br- is replaced by PA-.
References and Notes
(1) Schwartz, H.-H.; Richau, K.; Apostel, R. Macromol. Symp. 1997,
126, 95.
(2) Macromolecules 1994, 27, 6007.
(3) Antonietti, M.; Burger, C.; Effing, J. Materials 1995, 8, 751.
(4) Ober, C. K.; Wegner, G. AdV. Mater. 1997, 9, 17.
(5) AdV. Mater. 2000, 12, 545.
(6) Antonietti, M.; Maskos, M. Macromolecules 1996, 29, 4199.
(7) Antonietti, M.; Conrad, J. Angew. Chem., Int. Ed. Engl. 1994, 33,
1869.
(8) Ilekti, P.; Piculell, L.; Tournilhac, F.; Cabane, B. J. Phys. Chem. B
1997, 102, 344.
(9) Ilekti, P.; Martin, T.; Cabane, B.; Piculell, L. J. Phys. Chem. B
1999, 103, 9831.
(10) Everaars, M. D.; Nieuwkerk, A. C.; Denis, S.; Marcelis, A. T. M.;
Sudho¨lter, E. J. R. Langmuir 1996, 12, 4042.
(11) Langmuir 1993, 9, 2933.
(12) Chen, L.; Yu, S.; Kagami, Y.; Gong, J.; Osada, Y. Macromolecules
1998, 31, 787.
(13) Kim, B.; Ishizawa, M.; Gong, J.; Osada, Y. J. Polym. Sci., Part A:
Polym. Chem. 1999, 37, 635.
(14) Wakita, M.-a.; Edwards, K. A.; Regen, S. L. J. Am. Chem. Soc.
1988, 110, 5222.
(15) Langmuir 2000, 2000, 10061.
(16) Merta, J.; Torkkeli, M.; Ikonen, T.; Serimaa, R.; Steniua, P.
Macromolecules 2001, 34, 2937.
(17) Khandurina, Y. V.; Rogacheva, V. B.; Zezin, A. B.; Kabanov, V.
A. Polym. Sci. 1994, 36, 184.
(18) Narita, T.; Hirota, N.; Gong, J. P.; Osada, Y. J. Phys. Chem. B
1999, 103, 6262.
(19) Langmuir 1998, 14, 2269.
Further studies of other ternary mixtures involving the
complex salt, CTAPA, are in progress in our laboratory. These
include aqueous mixtures with the polyelectrolyte (the CTAPA/
NaPA/water system) and mixtures with the acetate (Ac) salt of
the surfactant (the CTAPA/CTAAc/water system). The former
mixtures correspond to another face of the phase pyramid in
Figure 1. The mixtures with CTAAc are of special interest
because the acetate ion may be viewed as one repeating unit of
polyacrylate. Experiments with longer polyacrylates are also in
progress.
(20) Hansson, P. Langmuir 1998, 14, 4059.
(21) Sokolov, E.; Yeh, F.; Khokhlov, A.; Grinberg, V. Y.; Chu, B. J.
Phys. Chem. B 1998, 102, 7091.
(22) Mironov, A. V.; Starodoubtsev, S. G.; Khokhlov, A. R.; Dembo,
A. T.; Yakunin, A. N. Colloids Surf. A 1999, 147, 213.
(23) Khandurina, Y. V.; Rogacheva, V. B.; Zezin, A. B.; Kabanov, V.
A. Polym. Sci. 1993, 36, 195.
(24) Zhou, S.; Yeh, F.; Burger, C.; Chu, B. J. Phys. Chem. B 1999,
103, 2107.
(25) Thalberg, K.; Lindman, B.; Karlstro¨m, G. J. Phys. Chem. 1991,
95, 6004.
(26) Nyde´n, M.; So¨derman, O. Langmuir 1995, 11, 1537.
(27) Morrison, R. T.; Boyd, R. N. Organic Chemistry, 4th ed.; Allyn
and Bacon Inc.: Newton, MA, 1983.
(28) Wa¨rnheim, T.; Jo¨nsson, A. J. Colloid Interface Sci. 1988, 125, 627.
(29) Khandurina, Y. V.; Dembo, A. T.; Rogacheva, V. B.; Zezin, A.
B.; Kabanov, V. A. Polym. Sci. 1994, 36, 189.
(30) Fontell, K. Colloid Polym. Sci. 1990, 268, 264.
(31) Pashley, R. M.; McGuiggan, P. M.; Ninham, B. W.; Brady, J.;
Evans, D. F. J. Phys. Chem. 1986, 90, 1637.
Acknowledgment. This project was funded by the Swedish
National Graduate School of Colloid and Interface Technology
(A.S.) and the Swedish Research Council for Engineering studies
(L.P.). Dominique Durand is acknowledged for help with the
X-ray scattering experiments at L.U.R.E. Per Hansson; Bo
Jo¨nsson and Håkan Wennerstro¨m are acknowledged for valuable
discussions.