ELSEVIER

Colloids and Surfaces

A: Physicochemical and Engineering Aspects 131 11998 ) 303 313

COLLOIDS
AND
A
SURFACES

The influence of salicylate counterions on the aggregation

behaviour of a polymerizable cationic surfactant
G. Tuin, F. Candau, R. Zana *
b~stitut Charles Sadron (CRM-EAHP). 6 rue Boussingault, 67083 Strasbourg. Codex. France
Received 2 November 1996: accepted 13 March 1997

Abstract

The aggregation behaviour of the cationic polymerizable surfactant dimethylhexadecyl-(2-acrylamidoethyl )ammonium bromide (AM16) has been studied with time-resolved fluorescence quenching (TRFQ) and viscometry. The
quencher-averaged aggregation number Nq was determined in the absence and presence of strongly binding salicylate
ions using TRFQ. In the absence of sodium salicylate (NaSal) AM 16 behaves very much like other cationic surfactants
with an aggregation number increasing with surfactant concentration and decreasing with increasing temperature. In
the presence of NaSal at constant temperature. Nq first increases slowly with NaSal concentration Cs,~, until the
critical concentration C*sa, above which N a increases sharply. At constant CSa,, with increasing temperature Nq
decreases more strongly at Csa, > C*s~l than at Csa, < C'SaY Viscosity measurements have confirmed the trends observed
with TRFQ. The values of C*Sa~, as determined with the two different techniques, agree well. The copolymerization
of the surfactant with acrylamide leads to water-insoluble copolymers, because of the large compositional drift during
copolymerization. 1998 Elsevier Science B.V.
Keywor&v Polymerizable cationic surfactant: Aggregation behaviour: TRFQ: Viscometry: Copolymerization

1. Introduction

Polymerizable surfactants can be polymerized

to yield polysoaps, which can form hydrophobic
microdomains having properties similar to surfactant micelles, see for example Ref. [1]. The homopolymerization of the polymerizable surfactant
hexadecyldimethyl(vinylbenzyl )-ammonium chloride ( N I 6 ) was studied in detail [2,3]. It was found
that the polymerization of single micelles did not
take place and that high molecular weight polymers were obtained. The resulting homopolymers
were not very soluble in water, probably due to
their high molecular weight and structure [1].

* Corresponding author.
0927-7757,'98/$19.00 ~' 1998 Elsevier Science B.V. All rights reserved.
Pll S0927-7757197100098-8

Polymerizable surfactants can also be copolymerized with water-soluble monomers. This type
of copolymerization has recently attracted much
interest, because the copolymers obtained behave
as associative thickeners [4-10]. The viscosity of
their solutions can be controlled by the adjustment
of the molar ratio of surfactant to monomer. For
example, the free radical copolymerization of N 16
with acrylamide ( A M ) gives rise to copolymers of
high molecular weight and good thickening efficiency [4,11]. This particular process was found
to lead to copolymers with a multiblock structure.
However, during the copolymerization, as the
conversion of monomers increases, a large compositional drift was observed. The polymerizable
surfactant was consumed at a faster rate than AM.
leading eventually to relatively heterogeneous

304

G. Tuin et al. / Colloids Surfaces A: Physicochem. Eng. Aspects 131 (1998) 303-313

copolymers. This compositional drift was ascribed

to the difference in reactivity of the two monomers
involved (different polymerizable groups) and to a
micellar effect inherent to the process (concentration, residence time of the active end of the
propagating radical and exchange dynamics).
The initial molar fraction of surfactant which
could be incorporated into the copolymers without
losing water solubility was limited to around
5mo1%. Recently, it has been shown [8] that
during the copolymerization of AM and the polymerizable surfactant dimethyldodecyl-(2-acrylamidoethyl)ammonium bromide (DAMAB), up to
20 tool% of the surfactant could be incorporated,
leading to water-soluble copolymers. Higher initial
surfactant concentrations led to (partly) insoluble
copolymers. The authors did not study the copolymer composition as a function of conversion.
Instead of using associative thickeners to control
the viscosity of solutions, it is well known that the
addition of specific counterions to micellar solutions can change the micellar size significantly [12].
For a given surfactant concentration, the micellar
size increases slowly upon the addition of salt until
a critical concentration c* above which a much
stronger increase in micellar size is observed.
Lyotropic anions, such as salicylate [13] or thiocyanate [14,15], are known to promote very
efficiently the growth of cationic micelles and to
lower c .
Viscosity measurements have shown that the
addition of counterions, such as salicylate [16-18],
to cationic surfactant solutions results in a strong
increase in the viscosity of the solution which
exhibits a viscoelastic behaviour [18]. These effects
clearly indicate the presence of large micellar structures. These micelles have been investigated by
viscometry [18,19], dynamic light scattering
[20,21] and, to a lesser extent, by time-resolved
fluorescence quenching (TRFQ) [22].
In preliminary experiments we observed that
addition of sodium salicylate (NaSal) to micellar
solutions of N16 gave rise to viscoelastic solutions
with a much higher viscosity than the initial solutions. However, the range over which the micellar
properties of N16 could be changed by the addition
of NaSal proved to be rather limited, as precipitation occurred at ratios [NaSal]/[N16] >0.3.

In the present study, we have synthesized a new

polymerizable cationic surfactant, dimethylhexadecyl (2-acrylamidoethyl)ammonium
bromide
(AM16), in which a C16-tail (favourable to the
occurrence of viscoelastic properties) is combined
with an AM headgroup (for the purpose of copolymerization with AM [8]). In order to get more
insight into the aggregation behaviour of this
surfactant in the absence and presence of NaSal,
TRFQ and viscosity measurements were performed. The copolymerization of AM16 with AM
was also investigated.

2. Experimental
2.1. Materials

Unless otherwise stated, all solvents were of

analytical grade.
N,N'-Dimethylethylenediamine (Merck, purity
99%), sodium salicylate (Aldrich, purity 99%),
sodium carbonate (Prolabo, p.a.), sodium chloride
(Merck, p.a.), acryloyl chloride (Fluka, purity
99%) and hexadecylbromide (Aldrich, purity 97%)
were used as received.
The fluorescence probe was a ruthenium(II)
derivative: Ru(bipy)2(bipy2Clj2+2Cl-, where
bipy stands for bipyridine and bipy2C17 is a bipyridine to which two C~v-groups have been covalently
bound, in order to make it more hydrophobic.
This probe has already been used by Cochin et al.
[2] and was obtained from Professor M. Gr~itzel
(EPFL, Lausanne, Switzerland). The quencher
9-methylanthracene (9-MeA) (Aldrich) was recrystallized from ethanol.
AM (Merck) was recrystallized twice from chloroform and stored in the dark at 4C until required.
The initiator 4,4'-azobis(4-cyanovaleric acid)
(ACVA) (Aldrich) was used without purification.
Water was Milli-Q quality water (conductivity
< 1 pS cm- 1).
2.2. Synthesis o f A M 1 6

The surfactant AM 16 was synthesized as shown

in Fig. 1. In the first step, N,N'-dimethylethylenediamine (5.14g, 0.0583 mol) was dissolved in

G. Tuin et aL/ Colloids Surfaces A. Physicochem. Eng. Aspects 13l (1998) 303 313
o
II
CH2=CH-C-CI

CH 3

CHaCI2

O H

II

H2N-CHaCH2- N

CHa(CH2)14CH2Br

CH3

CH2=CH-C-N-CH2CH2-N

CH a

305

Na2COJH20

CH a

O H
CH3
II I
I+
CH2= CH - C - N - CHzCH 2- ? - CH2(CH2)14CH3
Br-

CH3

AM16

Fig. 1. Synthetic route of the AM 16 surfactam.

80ml anhydrous CH2C12. After adding 6.75g

(0.0637 tool) powdered Na2CO 3 and a few drops
of nitromethane, the solution was cooled and
purged with nitrogen. A solution of 5.40g
(0.060tool) acryloyl chloride in 50ml of anhydrous CHeC12 was added while keeping the temperature of the solution below 5C. After this
addition, the solution was refluxed for another 3 h.
The organic layer was washed twice with concentrated aqueous Na2CO3 solution. The combined
water layers were saturated with NaC1 and
extracted three times with CHzCI> The combined
organic layers were dried o n N a z S O 4. After evaporation of the solvent, 7.1 g (85%) of a yellow oil
was obtained.
The crude N,N'-dimethylethylacrylamide was
purified by vacuum distillation in the presence of
a small amount of CuC12 as inhibitor of polymerization. A slightly yellow oil was collected at
80 84:C under 0.7mmHg. Yield: 4.58g (55%).
1H N M R (CDC13): 5 (ppm) 2.24 (s, 6H), 2.43 (t,
2H ), 3.38 (t, 2H ), 5.65 (d, 2H), 6.05-6.22 (m, 3H ).
In the second step, 4.50g (0.0317mol) N,N'dimethylethylacrylamide and 10.0g hexadecylbromide (0.0329mol) were added to 100ml of
acetone and this solution was heated at 50~C for
72 h. After cooling to room temperature, the
solvent was evaporated and the resulting material
was recrystallized twice from an acetone/ether
mixture and twice from ethylacetate. The white
crystals were dried under vacuum at room temperature for 2 days. Yield 9.21 g (63%).

~H N M R (CDCI3): 6 (ppm) 0.85 (t, 3H), 1.25

(br, 28H), 1.73 (m, 2H), 3.35 (s, 6H), 3.78 (m,
4H), 5.62 (m, 1H), 6.35 (m, 2H), 8.72 (br, IH ).

2.3. 1HNMR
1H N M R spectra were recorded on a Bruker
AC200 spectrometer using deuterated chloroform
as solvent.

2.4. Comtuctivitv measurements

The conductances of the surfactant solutions
were measured with an automated conductivity
bridge ( W a y n e - K e r r B905) operating at 1 kHz
equipped with a conductivity cell (Tacussel
CM02), for a temperature range from 15 to 5 0 C .
The critical micelle concentration (cmc) was
determined from the intercept of the two straight
lines, fitted to data points before and after the cmc.
An estimate of the micelle degree of ionization u,
was obtained as the ratio of the slopes of the lines
after and bel\~re the cmc.
2.5.

Vi,~;COltl~.'IrF

Viscosity measurements were performed at

25.5 'C using a Contraves LS 30 low shear viscometer equipped with a cylinder geometry. The solutions containing the required amounts of AM16
and NaSal u'ere for stirred 24 h at a low stirring
speed at room temperature and left to equilibrate

306

G. Tuin et al. / Colloids Surfaces A: Physicochem. Eng. Aspects 131 (1998) 303-313

for another 24 h without stirring. For a given

AM16/NaSal mixture, apparent viscosities ~1were
measured as a function of time at a given shear
rate until a constant viscosity was reached. When
~1is plotted vs. shear rate, two regimes are distinguished: at low shear rates the viscosity is nearly
independent of shear rate (Newtonian behaviour),
whereas at higher shear rates the viscosity
decreases with shear rate (shear thinning
behaviour).

2.6. Copolymercomposition
The incorporation of AM16 in the AM copolymer was determined by bromide ion elemental
analysis.

In the absence and in the presence of quencher

the decay curves are described by Eqs. (1) and (2)
respectively:
I, =I0 exp(-t/z)

(1)

/t =I0 e x p { - A z t - A 3 [ 1 - e x p ( - A 4 t ) ]}

(2)

where Io and It are the intensity at times 0 and t.

The fitting parameters A2, A3 and A4 are independent of time.
When the probe and quencher distributions do
not vary over a time long compared with v, then

A2=l/r A3=[Q]/[M ] A4=]o

The quencher-averaged aggregation number Nq is

given by

2.7. TRFQ

[C-cmc]
Nq --

TRFQ [22,23] was used to determine the aggregation number of AM16 micelles in the absence
and presence of NaSal. The fluorescence decay
curves were recorded using the single photon
counting technique [24-27]. The fluorescence was
excited at a wavelength of 450 nm by a flashlamp
after which the decay curves were recorded over
about four lifetimes r of the ruthenium probe
( ~ 6 5 0 ns at 25C). The decay curves were analysed using a non-linear weighted least-squares
fitting procedure.
The TRFQ method applies to systems where the
value ~'kQ > 1 (where kQ is the intramicellar quenching rate constant [28,29]). This condition is fulfilled in most of the experiments. The overall error
on the aggregation number is estimated to be
+ 10%. However, in the systems where the micelles
were very large (high NaSal concentrations) the
asymptotic part of the decay curves was sometimes
not reached, which led to larger errors (estimated
to be + 15-20%).
The measurements were performed in the range
15-50~'C, using solutions thoroughly deoxygenated
by repeated freeze-pump-thaw cycles. For each
surfactant solution two fluorescence decay curves
were recorded: one in the absence and one in the
presence of quencher. In all TRFQ measurements
care was taken that [Q]/[M]~ 1 (where [Q] is the
quencher concentration and [M] is the micellar
concentration) and [P]/[M] <0.05 (where [P] is the
probe concentration).

(3)

A3

A3 -

[Q]

(4)

~1

where q=[Q]/[C-cmc] and C is the surfactant

concentration.
Eq. (3) assumes monodisperse micelles, which is
normally the case for surfactant solutions close to
the cmc or in the absence of added electrolyte [3032]. At higher surfactant concentrations, or in the
presence of added electrolyte, the micelles tend to
grow and may become more polydisperse [33,34].
The existence of miceile polydispersity leaves
Eq. (2) unchanged [35, 36], but the apparent aggregation number becomes dependent on the
quencher concentration. The quencher-averaged
aggregation number Nq is then given by [23]
~72rl

Nq =N,~

A q2

(5)

where Nw is the weight-averaged aggregation

n u m b e r , a 2 is the mean square deviation and A is
the skewness of the micellar distribution function.
The analysis of the fluorescence decay curves by
using Eq. (2) yields the parameters Io, A2, A3 and
A 4. Each of the measured curves has been individually fitted and the fit parameters were then used
to calculate the micellar aggregation number Nq
by using Eq. (4). In this way, values of the apparent aggregation number Nq are obtained. The
value of q was kept nearly constant in order to
minimize the influence of polydispersity. The

G. Tuin et al. / Colloids Surfaces A. Physicochem. Eng. Aspects 131 (1998) 303 313

[Q]-dependence of N q w a s only investigated for

one surfactant concentration.

2.8. Copolymerizations
Copolymerizations of the surfactant AM 16 with
AM were performed at 50C using ACVA as
initiator. During copolymerization no NaSal was
added. The total monomer concentration was
2 wt,%. The initial percentage of AM16 in the
monomer feed was 2mo1% or 3mo1% (AM16
concentrations of 5.1 mM and 7 . 3 m M respectively). These concentrations are above the cmc,
which is 0.83 m M at 50~C. Further experimental
details can be found elsewhere [4,11].
Conversion was measured by taking samples at
regular intervals and precipitating the copolymers
formed in acetone. The polymers were washed
three times with acetone and dried for 3 days at
40~C under vacuum.

3. Results and discussion

3.1. d'm~'s

Fig. 2 shows the variation of the cmc and micellar degree of ionization ~ with temperature. The
values are listed in Table 1.
The change of the cmc and ~ with temperature
are qualitatively similar to those reported for ionic
surfactants [37 40].
,'

307

Table 1
Cmc and micelle degree of ionization ,~ at different temperatures
T(C)

Cmc(mM)

14.8
24.8
34.7
39.2
42.5
50.1

0.59
0.61
0.67
0.69
0.72
0.83

0.32
o.33
0.36
0.37
o.38
(i.39

3.2. TRFQ

In Fig, 3 typical fluorescence decay curves are

shown for different quencher concentrations at
constant surfactant concentration. In the absence
of quencher, the plot of the emission intensity vs.
time displays a good linearity in the semi-logarithmic scale used. The decay curves at different
quencher concentrations exhibit the same monoexponential decay at long times with nearly the
same slope (A2) a s for the quencher-free solutions
(1/:), within the experimental error.
Fig. 4 shows that although the values v and
A2 ~ depend strongly on temperature, they fall on
a single curve in the temperature range investigated. This has been found for most fluorescence
experiments performed, and indicates that there is
no probe or quencher migration taking place on
the fluorescence time scale. These results justify
the use of Eq. (3) for the simple case given above
for the calculation of micellar aggregation
numbers.

!.'OI J

.-!
G.SO
i

O.GO
O

E
O

C).~'-'U
i,

z,

1:5

-',-:

3~-,

40

50

G.GO

- -

4~2:~

),f ,

': "

~ ,~ '

' ' '

G<O
time

(ns)

(C)

Fig. 2. Dependence of the degree of ionization ~ ( )

(O} on temperature for AM16.

and cmc

Fig. 3. Normalized fluorescence decay curves for a 15.(]5 m M

AM16 solution at different quencher to surfactant ratios ~1 (0,
0.0057, 0.0093 and 0.0139 from top to b o n o m ) at 25.5'C.

G. Tuin et al. / Colloids Surfaces A. Physicochem. Eng. Aspects 131 (1998) 303 313

308

120

750

I
~

~
------~__._

650

!00

\.

-\

\,

550

45Q

'
0

10

20

..,

30

40

80
\

"\

b O

10

20

(C)

30

40

50

T (C)

Fig. 4. Values of ~ (O) and A~-1 ( i ) for a 14.87 mM AM16

solution at different temperatures. The line is a guide to the eye.

3.3. Micellar aggregation numbers in pure water

In the absence of NaSal, aggregation numbers

were determined for various AM 16 concentrations
and different temperatures. The results are presented in Fig. 5.
For a fixed temperature, an increase in surfactant
concentration results in an increase of the aggregation number, as found for other ionic surfactants
[33,41].
Fig. 6 shows that for a given surfactant concentration, an increase in temperature results in a
linear decrease of the micellar aggregation number,
as found for other ionic systems studied by TRFQ

120

'

50

Fig. 6. Quencher-averaged aggregation number Nq as a function

of temperature for different AM 16 concentration: 6.04 mM,
& 14.87 mM, 38.95 mM, 96.32 mM.

[40,42] and other experimental techniques, for

instance T-jump [43] and light scattering [44,45].
In Fig. 6 it is observed that the decrease of the
aggregation number with temperature is small at
lower surfactant concentrations and becomes more
pronounced at higher C.
Information on the micelle polydispersity was
obtained by measuring the aggregation number
Nq at increasing quencher concentration, for a
given surfactant concentration. Fig. 7 shows the
variation of Nq with ,t=[Q]/[C-cmc] for C=
38.95 mM.
Within the experimental error the results fall on
a straight line. The qZ-term in Eq. (5) was, there150

......

100

100
z~

50
80

0
0.00
10 ~

0.50

10 )

1.50

2.00

102

(-x- 10 2)
C

(mM)

Fig. 5. Quencher-averaged aggregation number Nq vs. AMI6

concentration at different temperatures: O 15.1C, A 25.5C,
35.3C, 44.T'C.

Fig. 7. Quencher-averaged aggregation number Nq vs. quencher

to surfactant ratio q=[Q]/[C-cmc] for a 38.95mM AM16
solution at 25.5C. The line is a weighted linear least-squares
fit according to Eq. (5) without the q2 term.

G. Tuin et al. / Colloids SurJ~tces A." Physicochem. Eng. Aspects 131 (1998) 303 313

fore, omitted in fitting this equation to the data.

This is equivalent to assuming a symmetrical size
distribution function. A weighting factor proportional to [Q] was applied to the data points, which
takes into account the increased accuracy of the
fitted decay parameters, used to calculate Nq, with
increasing [Q]. The fit yielded Nw=94 and or=41.
Thus the micelles show polydispersity even at this
relatively low surfactant concentration.

3.4. ~Iicellar aggregation monbers of AM16 in the

presence of NaSal
The influence of NaSal has been investigated at
the surfactant concentration C = 1 4 . 8 7 m M for
different temperatures. This concentration has
been chosen because in the absence of NaSal it is
well below c* (see Section 3.3). The change in
aggregation numbers in the presence of NaSal are
then only to be attributed to NaSal.
The variation of N q with the NaSal concentration Cs~ is shown in Fig. 8 for the different
temperatures studied.
Two regimes are clearly seen in the dependence
of Nq on CSa~: at CsaJ < CS~l a slow increase in
aggregation number is observed, whereas Nq
increases much more steeply at Cs,] > Cs~. The
aggregation numbers at high Cs~l are subject to a
rather large error, about 20%, because these aggre-

4' ~'Z: f .

309

gation numbers are the largest which could be

measured reliably.
The steep and nearly exponential increase of the
micelle aggregation number at salt concentrations
larger than a threshold value has also been
observed by T R F Q for another surfactant [46].
The results in Fig. 8 have been used to represent
the dependence of the aggregation number on
temperature in Fig. 9 for selected NaSal concentrations, one below C~Sa~and the others above
CSa~.
For Csa I below C~s~l, the decrease of Nq upon
increasing temperature is comparable with the
system in the absence of NaSal. However, if Cs~l
is increased above Csa~ the decrease of Nq is much
more pronounced, in agreement with literature
results [47].

3.5. Rheology of surfactant solutions in the

presence o/' NaSal
Fig. l0 shows the variation of the viscosity at
zero shear rate ~7 vs. Cs~]for three AM 16 solutions.
The three curves all have the same shape. The
viscosity of the solutions remains constant until
Cs~ has been reached. A higher CSa] results in a
fast increase in viscosity until the viscosity passes
over a maximum. Further increase of CSaI results
in a decrease of the viscosity until at very high

S ....::I

1!

t1

!
l

[. . . . .

Csal (raM)
Fig. 8. Quencher-averaged aggregation number Nq vs. Cs,] at
different temperatures for a 14.87mM AM16 solution:
46.7'~C, 36.9':C, C~ 31.9 C, + 25.5~C. The arrow indicates
C~Sai"

Fig. 9. Quencher-averaged aggregation number Nq vs. temperature at different NaSal concentrations l\~r a 14.87 mM A M I 6
solution: 7.01 raM, ~ 8.88 m M , ' ' 10.69 raM, + 12.04 raM.
The lines are guides to the eye.

G. Tuin et al. / Colloids Surfaces A: Physicochem. Eng. Aspects 131 (1998) 303-313

310

1000

s..

100

12.
10

0.5
10

10 ~

.02

103

C,~l (mM)

Fig. 10. Zero shear viscosity t/ vs. NaSal concentration for

three constant A M 1 6 concentrations at 25.5C: 14.87 m M ,
29.61 m M , + 79.92 m M . The lines are guides to the eye.

values of Cs~ the viscosity has reached the same

value as before the critical concentration.
The maximum value of t/ depends strongly on
the surfactant concentration. This value increases
by a factor of 30 as the surfactant concentration
increases only by a factor of 3. The increase in
viscosity with increasing Cs~l in the AM16-NaSal
system is much less pronounced than for similar
systems studied [16, 17,19]. In the system studied
by Hoffmann [19], cetylpyridinium chloride
(CpyC1) with NaSal, maximum values of the viscosities as high as 105 mPa s were found for similar
surfactant concentrations. These values are two
orders of magnitude higher than in the
AM16-NaSal system. In addition, the shape of
the viscosity vs. Csa~ curves are different for the
CpyC1-NaSal and AM 16-NaSal systems. Thus, in
the CpyC1-NaSal system [19] the viscosity goes
through a maximum, a minimum and a second
maximum with increasing Csa~. In the
AM16-NaSal system the second maximum is
replaced by a shoulder (see Fig. 10).
In the CpyCl-NaSal system [19] the maximum
in viscosity is found at the same [NaSal]/[CPyC1]
ratio with increasing CPyC1 concentration. This is
in contrast to our results, where the ratio
[NaSal]/[AM16] changes from 1.15 to 0.6 when
the surfactant concentration increases from
14.87 mM to 79.92 mM.
Preliminary experiments with the NaSal-N16

system have shown that the increase in viscosity

with increasing Csal was comparable to that of the
CPyC1-NaSal system and was much more pronounced than in the NaSal-AM16 system. The
NaSal-N 16 system had the disadvantage that precipitation occurred at [NaSal ]/[N 16 ] > 0.3. It was,
therefore, not possible to observe a maximum
value in the viscosity as a function of Csa~.
It is clear from the above results that the type
of surfactant plays an important role for the
preparation of viscoelastic solutions. The surfactants described above all have the same alkyl
chain. Therefore, the length of the surfactant tail
is not the predominant factor determining the
viscoelastic properties of these solutions. A more
important factor determining these properties is
the decrease of micelle charge density. This is done
very efficiently by salicylate ions, because they are
adsorbed strongly at the micellar surface [48,49].
In the AM16-NaSal system the salicylate ions
compete with bromide ions for binding to the
micelles. In the CpyC1-NaSal system, as well as in
the N16 NaSal system, the salicylate ions compete
with chloride ions for binding to the micelles. It is
known that bromide ions are bound in larger
number to cationic micelles than chloride ions
[14,15]. The binding of salicylate ions in the
presence of bromide ions may thus be less than in
the presence of chloride ions. The decrease of
micelle charge density by salicylate ions binding
may, therefore, be less in the AM16-NaSal system
than in the other systems. This may explain the
viscosity behaviour of the solutions.
Fig. 11 shows the variation of the viscosity at
zero shear rate ~/0 and of the quencher-averaged
aggregation number Nq with Cs~1 for one AM16
solution.
It is seen that the increase in Nq is accompanied
by an increase of the viscosity at zero shear rate
~/o. The values of the critical concentration C~s~,
as determined by the two types of measurement,
are in good agreement.

3.6. Copolymerization of AM16 and A M

During copolymerization of AM 16 and AM the
initial clear solutions became progressively turbid,
indicating the formation of water-insoluble regions

G. Tuin et al. / Colloids Surfaces A. Physicchem. Eng. Aspects 131 (1998) 303 313
o

400

300
o

EL

200

100

I
C,

i%

0:

0
<!c

101
Csan (mM)

Fig. 11. Quencher-averaged aggregation number Nq ( 0 ) and

zero shear viscosity q (01 vs. NaSal concentration for a
14.87 mM A M I 6 solution at 25.5C.

in the copolymers. After precipitation and drying

of the copolymers, it was not possible to dissolve
these copolymers in water. Instead of the expected
clear and viscous solutions, as observed for the
N16-AM copolymers [4, 11], turbid and low-viscous solutions were obtained. No attempt was
made to solubilize the polymers in organic solvents
such as dioxane or THF.
Fig. 12 shows the composition of the copolymers
as function of conversion.
Although low initial concentrations of surfactants are used, and in spite of the fact that the
monomers bear the same acrylamido polymeriza-

2
0

,:,

t.ff

"

::
~'~

....... :

0 L
__P~
total

monomer

GO

7~,
J

conversion

105 )
(%)

Fig. 12. Copolymer composition vs. total monomer conversion

for initial concentrations of AM16: 3 mol% (solid line) '7)
2 mol% ( broken line).

311

ble group, a large compositional drift occurs

during copolymerization. This drift is even larger
than that reported by Renoux and coworkers
[4, 11] for the copolymerization of AM and N16.
This leads to high concentrations of AM I6 in the
copolymers in the early stage of the polymerization: these regions of high AM16 concentration
are very insoluble in water, likely resulting in the
water insolubility of the copolymers formed.
Recall that the copolymerization of AM and
dimethyldodecyl(2-acrylamidoethyl)-ammonium
bromide (DAMAB), which differs from our surfactant AM16 only in the length of the hydrocarbon tail, leads to water-soluble copolymers even
with an initial DAMAB concentration of 20 mol%
[8]. In the AM16-AM copolymers, an initial
AM16 concentration as low as 1 mol% already
results in water-insoluble copolymers and a large
drift in composition is observed. As the two systems studied differ only in the surfactant chain
length, the water-solubility of the copolymers
seems to be determined by the hydrophobic character of the surfactant used. More studies are
required in order to check this conclusion.

4. Conclusions

This paper describes the aggregation behaviour

of a polymerizable cationic surfactant in the
absence and presence of strongly binding salicylate
ions. The surfactant combines a cationic (quaternary ammonium) headgroup with a long aliphatic
C16-tail and a polymerizable (acrylamido) group
of high reactivity.
In the absence of NaSal the aggregation number
increases with increasing surfactant concentration
at constant temperature. At constant surfactant
concentration the aggregation number decreases
upon increasing temperature.
In the presence of NaSal at constant temperature, the aggregation number increases slowly
with increasing NaSal concentration, until reaching the critical concentration C~salabove which the
aggregation number increases sharply. At constant
NaSal concentration the aggregation number
decreases more strongly with increasing temper-

312

G. Tuin et al. / Colloids Surfaces A: Physicochem. Eng. Aspects 131 (1998) 303 313

ature at NaSal concentrations above CSaI than

below C~SaV
Viscosity measurements have confirmed the
trends observed with TRFQ. However, it is
observed that the viscosity of the micellar solutions
in the presence of NaSal is much lower than for
similar systems described in literature, probably
because of the difference in nature of the surfactant
counterion used.
The values of the critical concentration Cs*~
determined with TRFQ and viscometry agree well,
even though the two techniques are based on
totally different principles: TRFQ measures a
microscopic property, whereas viscosity measures
a more macroscopic property.
The copolymerization of AM 16 with AM results
in the formation of water-insoluble copolymers for
the concentrations of AM16 studied, owing to the
large compositional drift observed during the
reaction.

Acknowledgment
One of us (G.T.) thanks HCM Research
Network No. CHRX-CT93-0273 for the provision
of a postdoctoral fellowship.

References
[ 1] A. Easchewsky, Adv. Polym. Sci. 124 (1995) 1.
[2] D. Cochin, F. Candau, R. Zana, Macromolecules 26
(1993) 5755.
[3] D. Cochin, R. Zana, F. Candau, Macromolecules 26
(1993) 5765.
[4] D. Renoux, J. Selb, F. Candau, Prog. Colloid Polym. Sci.
97 (1994) 213.
[5] E. Volpert, J. Selb, F. Candau, Macromolecules 29
(1996) 1452.
[6] D.G. Peiffer, J. Polym. Sci. A: 28 (1990) 619.
[7] D.G. Peiffer, Polymer 31 (1990) 2353.
[8] Y. Chang, C.L. McCormick, Macromolecules 26 (1993)
6121.
[9] S. Biggs, J. Selb, F. Candau, Polymer 34 (1993) 580.
[10] A. Hill, F. Candau, J. Selb, Macromolecules 26 (1993)
4521.
[11] D. Renoux, Ph.D. Thesis, University Louis Pasteur,
Strasbourg, France, 1995.
[12] J. Appelk G. Porte, Y. Poggi, J. Colloid Interface Sci. 87
(1982) 492.

[13] S. Gravsholt, J. Colloid Interface Sci. 57 (1976) 575 and

references cited therein
[14] E. Anacker, H. Ghose, J. Phys. Chem. 67 (1963) 1713.
[15] E. Anacker, H. Ghose, J. Am. Chem. Soc. 90 (1968) 3161.
[16] H. Rehage, H. Hoffmann, Mol. Phys. 74 (1991) 933.
[17] H. Hoffmann, G. Platz, H. Rehage, W. Schorr, Adv.
Colloid Interface Sci. 17 (1982) 275.
[18] F. Kern, R. Zana, S.J. Candau, Langmuir 7 (1991) 1344.
[19] H. Hoffmann, in: C.A. Herb, R.K. Prud'homme (Eds.),
Structure and Flow in Surfactant Solutions, ACS
Symposium Series, vol. 478, American Chemical Society,
1994, Washington D.C. pp. 2 31.
[20] A. Koike, T. Yamamura, N. Nemoto, Colloid Polym. Sci.
272 (1994) 955.
[21] N. Nemoto, M. Kuwahara, Langmuir 9 (1993) 419.
[22] R. Zana, J. Lang, Colloids Surf. 48 (1990) 153.
[23] R. Zana, in: R. Zana (Ed.), Surfactant Solutions, New
Methods of Investigation, Dekker, New York, 1987,
pp. 241-294.
[24]J.N. Demas, Excited State Lifetime Measurements,
Academic Press, New York, 1983.
125]W. Binana-Limb~16, Ph.D. Thesis, University Louis
Pasteur, Strasbourg, France, 1988.
[26] J. Lang, N. Lalem, R. Zana, J. Phys. Chem. 95 (1991)
9533.
[27] R. Zana, H. L~vy, D. Papoutsi, G. Beinert, Langmuir 11
(1995) 3694.
[28] M. van der Auweraer, S. Reekmans, N. Boens, F.C. de
Schryver, Chem. Phys. 132 (1989) 99.
[29] M. Almgren, J. Alsins, E. Mukhtar, J. van Stam, J. Phys.
Chem. 92 (1988) 4479.
[30] E. Roelants, E. Gelad~, J. Smid, F.C. de Schryver,
J. Colloid interface Sci. 107 (1985) 337.
[31] P. Lianos, R. Zana, J. Colloid Interface Sci. 84 ( 1981 ) 100.
[32] J. lsraelachvili, J. Mitchell, B. Ninham, J. Chem. Soc.
Faraday Trans. 1I: 72 ( 1976 ) 1525.
[33] A. Malliaris, J. Lang, R. Zana, J. Colloid Interface Sci.
110 (1986) 237.
[34] S.J. Candau, E. Hirsch, R. Zana, J. Phys. (Paris) 45
(1983) 1263.
[35] M. Almgren, J.E. L6froth, J. Chem. Phys. 76 (1982) 2734.
[36] G. Warr, F. Grieser, J. Chem. Soc. Faraday Trans. I: 82
(1986) 1813.
[37] D.F. Evans, M. Allen, B. Ninham, A. Fouda, J. Solution
Chem. 13 (1984) 87.
[38] M. Tanaka, S. Kaneshina, K. Shin-No, T. Okajima,
T. Tomida, J. Colloid Interface Sci. 46 (1974) 132.
[39] M. Tanaka, S. Kaneshina, G. Sugihara, N. Nishikido,
Y. Murata, in: K.L. Mittal, E. Fendler (Eds.), Solution
Behaviour of Surfactants, vol. 1, Plenum, New York,
1992, p. 41.
[40] A. Malliaris, J. Le Moigne, J. Sturm, R. Zana, J. Phys.
Chem. 89 (1985) 2709.
[41 ] P. Lianos, R. Zana, J. Colloid Interface Sci. 88 ( 1982 ) 594.
[42] E. Roelants, F.C. de Schryver, Langmuir 3 (1987) 209.
[43] M. Grubic, R. Strey, M. Teubner, J. Colloid Interface Sci.
80 (1981) 453.

G. Tuin et al. / Colloids Surfaces A: Physicochem. Eng. Aspects 131 (1998) 303 313
[44] M. Jones, J. Piercy, J. Chem. Soc. Faraday Trans. I: 68
(1972) 1839.
[45] N.A. Mazer. G.B. Benedeck, J. Phys. Chem. 80 (1976)
1075.
[46] H. von Berlepsch, K. Stfihler, R. Zana, Langmuir 12
(19%) 5033.

313

[47] M.E. Cates, N.J. Candau, J. Phys.: Condens. Matter 2

(19901) 6869.
[48] J. Ulmius, B. Lindman, G. Lindblom, T. Drakenberg,
J. Colloid Interface Sci. 65 (1978) 88.
[49] U R . K . Rao, C. Manohar, B.S. Valaulikar. R.M. l,,er,
J. Phys. Chem. Ol (1987) 3286.