Micelle-to-vesicle transition induced by organic additives in catanionic surfactant systems

Article abstract of DOI:10.1002/chem.200501053

A micelle-to-vesicle transition (MVT) induced by the addition of a series of apolar hydrocarbons (n-butylbenzene, n-hexane, n-octane, and n-dodecane) to the catanionic surfactant system n-dodecyltriethylammonium bromide/sodium n-dodecylsulfate (DTEAB/SDS) has been investigated for the first time by means of rheology and turbidity measurements, dynamic light scattering (DLS), and transmission electron microscopy (TEM). Interestingly, a MVT can take place within certain micellar regions, which are dependent on the structure and chain length of the hydrocarbon. However, these hydrocarbons are unable to induce a MVT in another catanionic surfactant system, namely, n-dodecyltriethylammonium bromide/sodium n-dodecylsulfonate (DTEAB/SDSO3), in which the molecular interactions are weaker than in the DTEAB/SDS system. On the other hand, polar additives, such as n-octanol and n-octylamine, exhibit much higher efficiency and activity in inducing MVT than hydrocarbons in both DETAB/SDS and DTEAB/SDSO3. Moreover, DLS, TEM, and time-resolved fluorescence quenching (TRFQ) results demonstrate that the ratio of vesicles to micelles in the system can be actively controlled by addition of polar additives. Possible mechanisms for the above phenomena are presented, and the potential application of controllable micelle/vesicle systems in the synthesis of tailored bimodal mesoporous materials is discussed.

Full text of DOI:10.1002/chem.200501053

FULL PAPER
DOI: 10.1002/chem.200501053
Micelle-to-Vesicle Transition Induced by Organic Additives in Catanionic
Surfactant Systems
[a]
[a]
HaiqingYin, ShengLei, Shengbao Zhu,[a] Jianbin Huang,*[a] and Jianpin Ye[b]
Abstract: A micelle-to-vesicle transi-
tion (MVT) induced by the addition of
a series of apolar hydrocarbons (n-bu-
tylbenzene, n-hexane, n-octane, and n-
dodecane) to the catanionic surfactant
length of the hydrocarbon. However,
these hydrocarbons are unable to
induce a MVT in another catanionic
surfactant system, namely, n-dodecyl-
triethylammonium bromide/sodium n-
dodecylsulfonate (DTEAB/SDSO3), in
which the molecular interactions are
weaker than in the DTEAB/SDS
system. On the other hand, polar addi-
tives, such as n-octanol and n-octyl-
amine, exhibit much higher efficiency
and activity in inducing MVT than hy-
drocarbons in both DETAB/SDS and
DTEAB/SDSO3.
Moreover,
DLS,
TEM, and time-resolved fluorescence
quenching (TRFQ) results demonstrate
that the ratio of vesicles to micelles in
the system can be actively controlled
by addition of polar additives. Possible
mechanisms for the above phenomena
are presented, and the potential appli-
cation of controllable micelle/vesicle
systems in the synthesis of tailored bi-
modal mesoporous materials is dis-
cussed.
system
bromide/sodium
n-dodecyltriethylammonium
n-dodecylsulfate
(DTEAB/SDS) has been investigated
for the first time by means of rheology
and turbidity measurements, dynamic
light scattering (DLS), and transmis-
sion electron microscopy (TEM). Inter-
estingly, a MVT can take place within
certain micellar regions, which are de-
pendent on the structure and chain
Keywords: micelles · surfactants ·
vesicles
Introduction
surfactant systems, much attention has focused on the influ-
ence on the micelle-to-vesicle transition (MVT) or the vesi-
cle-to-micelle transition (VMT) of changes in surfactant
composition,[1a–e] temperature variation,[4] and salt addi-
tion.[5] However, in contrast to the many studies on organ-
ized transitions induced by the solubilization of organic ad-
ditives in single-surfactant systems, little work[6] has been
done on the influence of organic additives in catanionic sur-
factant systems, since organic additives, especially apolar hy-
drocarbons, are usually thought to have less effect in tailor-
ing organized assemblies of catanionic surfactant systems
due to the strong electrostatic attraction in this kind of sys-
tems.
Aqueous mixtures of anionic and cationic surfactants (refer-
red to as catanionic surfactants) exhibit interesting phase be-
havior and properties, which mainly arise from the electro-
static interactions between the oppositely charged head
groups. By adjusting the composition, these interactions can
be tuned to produce various microstructures with character-
istic geometries ranging from spherical to cylindrical to
planar, which have been studied experimentally and theoret-
ically[1] for their potential applications in many fields.
The nature of the transition between micelle and vesicle
is of great importance in a number of practical applica-
tions[2] and of fundamental interest as well.[3] For catanionic
Recently, we found[8] that the addition of some short-
chain alcohols, such as 2-propanol, to some catanionic sur-
factant systems with precipitation not only influenced the
phase behavior, but also transformed the precipitate into
vesicles, which indicates that suitable additives may also
play an important role in the transformation of organized
assemblies in catanionic surfactant systems, in which strong
electrostatic interactions exist. We have now systemically
studied the effects of a series of hydrocarbons (n-butylben-
zene, n-hexane, n-octane, and n-dodecane) and polar organ-
ic additives, such as n-octanol and n-octylamine, on the tran-
[a] H. Yin, S. Lei, S. Zhu, Prof. Dr. J. Huang
State Key Laboratory for Structural Chemistry of Unstable and
Stable Species
College of Chemistry and Molecular Engineering
Peking University, Beijing 100871 (P. R. China)
Fax : (+86)10-6275-1708
E-mail: JBHuang@pku.edu.cn
[b] J. Ye
Key laboratory of Photochemistry
Chinese Academy of Sciences, Beijing 100871 (P. R. China)
Chem. Eur. J. 2006, 12, 2825 – 2835
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2825

sition from micelle to vesicle in the catanionic surfactant
systems n-dodecyltriethylammonium bromide/sodium n-do-
decylsulfate (DTEAB/SDS) and n-dodecyltriethylammoni-
um bromide/sodium n-dodecylsulfonate (DTEAB/SDSO3).
For the first time, it was found that apolar hydrocarbons can
induce a MVT in the DTEAB/SDS system within certain
ranges of surfactant composition, while they are unable to
induce a MVT in the DTEAB/SDSO3 system. Moreover,
the polar organic additives are much more efficient in induc-
ing a MVT than the hydrocarbons in both systems and can
actively adjust the ratio of vesicles to micelles. Possible
mechanisms for the above transitions are presented, based
on the theory of the aggregate packing parameter, and the
different effects of the additives on a MVT are explained.
Furthermore, potential application of controllable micelle/
vesicle systems in the synthesis of tailored bimodal mesopo-
rous materials is discussed.
behaviors of various organic additives were studied in these
regions.
MVT induced by addition of n-butylbenzene: The micellar
DTEAB/SDS system with Ctotal =10 mmol and XDTEAB =0.25
was selected for our study. Before addition of n-butylben-
zene, the sample showed obvious non-Newtonian behavior
in the steady-state rheological curve (Figure 2a): shear
thickening at lower shear rates and shear-thinning at higher
shear rates. This behavior is a typical property of rodlike mi-
celles in dilute solutions and has been well investigated in
other work.[9] On addition of n-butylbenzene to the mixed
system, the non-Newtonian behavior becomes progressively
weaker, and this suggests that the shape of aggregates be-
comes more symmetrical. In addition, a notable increase in
absorbance at 500 nm on the addition of n-butylbenzene
Results
Microstructures of the DTEAB/SDS system: The micro-
structures of the DTEAB/SDS system at a fixed total surfac-
tant concentration Ctotal of 10 mmol were determined from
the results (not shown) of turbidity measurements, dynamic
light scattering (DLS), and transmission electron microscopy
(TEM). A partial phase diagram for the variation of XDTEAB
(XDTEAB =CDTEAB/Ctotal) is shown in Figure 1. The vertical
axis is the corresponding turbidity at 500 nm of each investi-
gated sample. In Figure 1, transitions from micelle to vesicle
and finally to precipitate can be observed as XDTEAB ap-
proaches 0.5, which can be understood as the result of the
neutralization of oppositely charged head groups. The micel-
lar regions are at the two edges of the diagram with XDTEAB
ranges from 0 to 0.33 and from 0.85 to 1. The solubilization
Figure 2. a) Steady-state rheological curves and b) variation of absorb-
ance at 500 nm (the solid line is a guide to the eye) for the system
DTEAB/SDS (Ctotal =10 mmol, XDTEAB =0.25) with addition of n-butyl-
benzene.
Figure 1. Partial phase diagram of DTEAB/SDS with Ctotal =10 mmol and
varying XDTEAB. (XDTEAB =CDTEAB/Ctotal; A500 is the turbidity at 500 nm of
each investigated sample). M=micelle, V=vesicle, P=precipitate.
2826
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 2825 – 2835

Micelle-to-Vesicle Transition in Catanionic Surfactant Systems
FULL PAPER
(Figure 2b) indicates that larger aggregates are formed in
the system.
The microstructural transformation induced by addition
of n-butylbenzene to the system was also investigated by
DLS and TEM. Before addition of n-butylbenzene, the plot
aggregates when 1.5 mmol of n-butylbenzene is added. Thus,
MVT by solubilization of n-butylbenzene is confirmed.
Investigation of other micellar DTEAB/SDS systems with
C
total =10 mmol showed MVT by solubilization of n-butyl-
benzene can take place within certain surfactant composi-
tions in Figure 1, that is, from XDTEAB =0.25 to 0.33 and
from XDTEAB =0.86 to 0.96. Hereafter, such regions are re-
ferred as the Reff (effective MVT regions) of the additive.
of the apparent hydrodynamic radius distribution f(Rh,app
)
calculated by the CONTIN method (see Experimental Sec-
tion), has a single peak with an average apparent hydrody-
namics radius hRh,appi of 8 nm (Figure 3a), which reflects a
typical equivalent size of rodlike surfactant micelles. When
1.5 mmol of n-butylbenzene is introduced, the peak in the
DLS plot shifts to the right, hRh,appi increases from 8 to
83 nm (Figure 3a), and corresponding spherical vesicles with
a diameter of 100–200 nm are observed by TEM (Fig-
ure 3b). In this case, the solution becomes a Newtonian
fluid (Figure 2a) and the apparent viscosity ha of the solu-
tion is 0.93 mPas at 258C (Figure 2a), which is quite close
to the viscosity of water (0.89 mPas). The above results sug-
gest that spherical vesicles probably become the dominant
MVT induced by addition of alkanes: When n-octane (C8) is
gradually added to the micellar DTEAB/SDS system with
C
total =10 mmol and XDTEAB =0.25, the solution remains
transparent and homogeneous until small droplets of oil
appear that suggest the saturation limit of n-octane solubili-
zation. Combined with the results of rheology measurements
(not shown), it can be concluded that the rodlike micelles
are transformed into spherical swollen micelles by solubili-
zation of C8, like similar reports for other surfactant sys-
tems.[10] However, the situation is different for the DTEAB/
SDS micellar system with Ctotal =10 mmol and XDTEAB =0.33.
A sharp increase in absorbance is observed as C8 is added to
the solution (Figure 4a). The DLS results also show that
hRh,appi increases remarkably from 25 to 95 nm on addition
of 2 mmol of C8 (Figure 4b). Correspondingly, spherical
vesicles with a diameter of 100–250 nm are observed by
TEM (Figure 5). These results demonstrate that MVT takes
place on addition of C8 to the system. Further investigations
revealed that the Reff of C8 is from XDTEAB =0.29 to 0.33 and
X
DTEAB =0.86 to 0.90, which is narrower than that of n-butyl-
benzene.
n-Hexane (C6) and n-dodecane (C12) were also investigat-
ed. Both can induce MVT within certain ranges of surfactant
composition. The Reff values of the alkanes are listed in
Table 1. The Reff decreases with increasing hydrocarbon
chain length.
Occurrence of a MVT and adjustment of the ratio of vesi-
cles to micelles by addition of polar organic additives: Be-
sides apolar hydrocarbons, the effects of two kinds of polar
organic additives, namely, n-octanol and n-octylamine, on a
MVT were also studied in the DTEAB/SDS system. A
MVT by addition of n-octanol (C8OH) can take place over
the entire micellar region of Figure 1; this suggests that
C8OH is much more efficient in inducing a MVT than the
above hydrocarbons. The ratio of vesicles to micelles can be
conveniently controlled by addition of a small amount of
C8OH.
As mentioned above, the solution of DTEAB/SDS with
C
total =10 mmol and XDTEAB =0.25 is composed of rodlike
micelles with an average Rh,app of 8 nm by DLS (Figure 6a).
After addition of 0.50 mmol of C8OH, two distinct peaks
can be clearly observed in the plot of f(Rh,app) versus Rh,app
(Figure 6b), which correspond to fast and the slow diffusion
modes, respectively. The fast mode has a hRh,appi of 5 nm,
which is close to that of the original rodlike micelles, and it
is reasonable to attribute this mode to diffusion of the mi-
celles. The slow mode has an hRh,appi of 220 nm. Correspond-
Figure 3. a) Plots of f(Rh,app) vs Rh,app by DLS before and after addition of
1.5 mmol of n-butylbenzene and b) micrograph of vesicles by TEM after
addition of 1.5 mmol of n-butylbenzene for the system DTEAB/SDS
(Ctotal =10 mmol, XDTEAB =0.25).
Chem. Eur. J. 2006, 12, 2825 – 2835
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2827

J. Huang et al.
Table 1. Reff for different hydrocarbons in the system DTEAB/SDS
(Ctotal =10 mmol).
Alkanes
Reff
C6
C8
C12
0.26–0.33, 0.86–0.92
0.29–0.33, 0.86–0.90
0.32–0.33, 0.86–0.87
duced that the slow mode reflects the formation of vesicles
in the solution. Therefore, the two peaks are designated as
the “micelle peak” and the “vesicle peak” respectively. Fur-
ther addition of C8OH causes the micelle peak to shrink
while the vesicle peak grows gradually (Figure 6c and d).
Moreover, the contribution of the vesicle peak to the total
scattering intensity of the sample Iv%, which is calculated
from the weight of the vesicle peak area in the plot of f-
(Rh,app) versus Rh,app, also increases on addition of C8OH
(Figure 8a). These results indicate growing vesicle popula-
tion and increasing ratio of vesicles to micelles. This tenden-
cy is further confirmed by TEM (Figure 7b and c). The var-
iations of hRh,appi for micelle and vesicle peaks as a function
of C8OH concentration are shown in Figure 8b. For the vesi-
cles, hRh,appi gradually decreases with increasing C8OH con-
centration. When the amount of C8OH reaches 1.5 mmol,
the micelle peak almost disappears in the DLS plot (Fig-
ure 6e), and the vesicle peak has hRh,appi of 79 nm. In this
case, the rodlike micelles in the system are probably com-
pletely transformed into spherical vesicles.
The adjustment of the ratio of vesicles to micelles on
C8OH addition was also characterized by time-resolved fluo-
rescence quenching (TRFQ) experiments. The time-depen-
dent fluorescence intensity of a probe in the micelle in the
presence of a quencher is thought to obey the Infelta–Ta-
chiya equation [Eq. (1)],[11]
Figure 4. a) Variation of absorbance at 500 nm on addition of C8 (the
solid line is a guide to the eye) and b) plots of f(Rh,app) vs Rh,app by DLS
before and after addition of 2.0 mmol of C8 for the system DTEAB/SDS
(Ctotal =10 mmol, XDTEAB =0.33).
IðtÞ ¼ Ið0ÞexpfÀA2tÀA3½1ÀexpðÀA4tފg
ð1Þ
where A2 and A4 are independent of quencher concentration
and A3 is defined as Equation (2),
A3 ¼ N½QŠ=Cm
ð2Þ
where Cm is the concentration of surfactant in micellar form,
[Q] the concentration of quencher, and N the average aggre-
gation number of the micelles. In vesicles and other large
aggregates, the time-dependent fluorescence intensity of a
probe in the presence of a quencher is thought to obey the
modified Stern–Volmer equation [Eq. (3)],[12]
IðtÞ ¼ Ið0ÞexpfÀA5tÀA6t1=2
g
ð3Þ
Figure 5. Micrograph of vesicles observed by TEM in the DTEAB/SDS
system (Ctotal =10 mmol, XDTEAB =0.33) on addition of 2 mmol of n-
octane.
where A5 and A6 both depend on quencher concentration.
When micelles and vesicles coexist in one system, the
quenching kinetics will be governed by a linear combination
of Equations (1) and (3). Unfortunately, the large number of
parameters involved in such a combination makes quantita-
tive analysis intractable. However, if the decay curves are
ingly, a few spherical vesicles with a diameter of 200–500 nm
were observed by TEM (Figure 7a). Therefore, it can be de-
2828
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 2825 – 2835

Micelle-to-Vesicle Transition in Catanionic Surfactant Systems
FULL PAPER
Figure 6. Distribution of apparent hydrodynamic radii f(Rh,app) of the system DTEAB/SDS (Ctotal =10 mmol, XDTEAB =0.25) with different concentrations
of C8OH. a) 0, b) 0.50, c) 0.70, d) 1.0, and e) 1.5 mmol.
forced to fit the form of Equation (1), A4 shows dependence
on quencher concentration and the ratio of vesicles to mi-
celles. By this method, TRFQ can be used to track the tran-
sition between micelles and vesicles. The TRFQ data for
DTEAB/SDS mixtures with Ctotal =10 mmol, XDTEAB =0.25,
and different concentrations of C8OH were measured. The
variation of A4 (obtained by fitting the data to [Eq. (1)]) as
a function of h (h=[Q]/Ctotal) for these solutions is shown in
Chem. Eur. J. 2006, 12, 2825 – 2835
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2829

J. Huang et al.
Figure 7. Micrographs of vesicles (negative-staining technique) for the system DTEAB/SDS (Ctotal =10 mmol, XDTEAB =0.25) on addition of a) 0.50,
b) 1.0, and c) 1.5 mmol of C8OH.
Figure 9. Before C8OH addition, A4 is independent of h,
that is, the solution is a pure micellar phase. The average ag-
gregation number calculated from Equation (2) is 274Æ15,
which is also a typical value for rodlike micelles. On addi-
tion of 0.50 mmol of n-octanol, the slope of A4 versus h
begins to deviate from zero, and this suggests the formation
of vesicles. Nevertheless, the dependence of A4 on h is indis-
tinctive, that is, the micelles are still the major component in
the solution. With further addition of C8OH, the slope of
A4/A4,1 (A4,1 is the value of A4 when h=0.002) versus h be-
comes increasingly positive (Figure 9), that is, the ratio of
vesicles to micelles increases, as was proved by other stud-
ies.[7,13] Although the exact ratio of vesicles to micelles is dif-
ficult to obtain from the above measurements, it is con-
firmed that addition of C8OH is an effective way of adjust-
ing the ratio of vesicles to micelles in the DTEAB/SDS
Figure 8. a) Variation of the contribution of the vesicle peak to the total
scattering intensity of the solution Iv% as a function of the concentration
of added C8OH and b) hRh,appi for the micelle and vesicle peaks as a func-
Figure 9. Normalized value of the parameter A4 as a function of h (h=
[Q]/Ctotal) with various concentrations of added C8OH (A4,1 is the value
of A4 when h=0.002) for the system DTEAB/SDS (Ctotal =10 mmol,
tion of C(C8OH) for the system DTEAB/SDS (Ctotal =10 mmol, XDTEAB
=
0.25). The solid lines are visual aids.
XDTEAB =0.25). The solid lines are visual aids.
2830
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 2825 – 2835

Micelle-to-Vesicle Transition in Catanionic Surfactant Systems
FULL PAPER
system, and thus vesicle/micelle solutions with variable ratio
of vesicles to micelles can be easily obtained.
cause a MVT in the entire micellar region. Moreover, the
ratio of vesicles to micelles can also be adjusted by addition
of the polar additives. The solution of DTEAB/SDSO3 with
n-Octylamine (C8NH2) has the same hydrocarbon chain
length as C8OH but a different polar head group. Its Reff
also covers the entire micellar region of the DTEAB/SDS
system with Ctotal =10 mmol. For the DTEAB/SDS system
(Ctotal =10 mmol, XDTEAB =0.25), only 1.0 mmol of C8NH2 is
required for complete MVT. In addition, similar to C8OH,
micelle/vesicle solutions can also be easily realized and ac-
tively controlled by addition of C8NH2.
C
total =10 mmol and XDTEAB =0.25 is selected as an instance.
The micelles in this system have hRh,appi=2 nm by DLS and
an average aggregation number of 164Æ10 by TRFQ. Ac-
cording to the DLS results (Figure 10) and TEM
(Figure 11), the ratio of vesicles to micelles gradually in-
creases on addition of C8OH, and 2.5 mmol C8OH is re-
quired for the complete MVT.
MVT induced by organic additives in the DTEAB/SDSO3
system: The solubilization behavior of the above organic ad-
ditives was also studied in the n-dodecyltriethylammonium
bromide/sodium n-dodecylsulfonate (DTEAB/SDSO3) cat-
anionic surfactant system with Ctotal =10 mmol. None of the
apolar hydrocarbons can induce a MVT in this system. How-
ever, the addition of C8OH or C8NH2 can still efficiently
Discussion
Mechanism of hydrocarbon-induced MVT: As far as we
know, this study is the first report that solubilization of a hy-
drocarbon (alkylbenzene or alkane) can induce a MVT in a
surfactant system. It is known that the location, distribution,
Figure 10. The apparent hydrodynamic radius distribution f(Rh,app) of the system DTEAB/SDSO3 (Ctotal =10 mmol, XDTEAB =0.25) with different concen-
trations of added C8OH: a) 0, b) 1.0, c) 2.0, and d) 2.5 mmol.
Chem. Eur. J. 2006, 12, 2825 – 2835
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2831

J. Huang et al.
Figure 11. Micrographs of vesicles (negative-staining technique) in the system DTEAB/SDSO3 (Ctotal =10 mmol, XDTEAB =0.25) on addition of a) 1.0,
b) 2.0, and c) 2.5 mmol of C8OH.
and orientation of the solubilized molecules can significantly
influence the morphology of the aggregate.[14] Possible
mechanisms for these phenomena are discussed below.
usually difficult to obtain the accurate value of a0, which is
quite dependent on aggregate composition and other envi-
ronmental factors, such as counterion concentration. Al-
though the value of a0 is often substituted by the value of
the area per molecule occupied at the air/water interface
(obtained from adsorption data) when the surfactant con-
centration is near the critical micelle concentration (CMC),
this approximation may cause considerable errors when the
concentration of the investigated system is much higher
than CMC (e.g., the total surfactant concentration in our
studied systems is around 100 times CMC).
Previous studies have shown that the alkyl chain of alkyl-
benzene tends to dissolve in the palisade of the surfactant
micelle with the benzene group located near the water/hy-
drocarbon interface,[15] and this increases the volume of the
surfactant hydrocarbon tail. The well-known theory of the
packing parameter p, proposed by Israelachvili et al.,[16] has
been widely and successfully used to explain the transforma-
tions of organized assemblies in dilute surfactant solutions:
0ꢀpꢀ1/3 for spherical micelles, 1/3ꢀpꢀ1/2 for cylindrical
micelles, and 1/2ꢀpꢀ1 for bilayer structures (p is defined
as v/a0lc, where v is the surfactant tail volume, lc is the tail
length, and a0 is the equilibrium area per molecule at the ag-
gregate surface).[17] It is conceivable that, on solubilization
of n-butylbenzene in the DTEAB/SDS system, the variation
of a0 and lc will be insignificant. Therefore an increase in the
aggregate parameter p can be expected, and formation of
larger aggregates with lower curvature is more favored. In
fact, spherical micelles were found to undergo transforma-
tion to rodlike micelles in some ionic surfactant systems on
addition of alkylbenzenes, but no further transformation
from rodlike micelles to vesicles was observed in these sys-
tems, probably because strong electrostatic repulsions be-
tween ionic head groups prevent further growth of the mi-
celles. In catanionic surfactant systems, such as DTEAB/
SDS mixtures, the electrostatic repulsions between identical-
ly charged molecules can be partly screened if not totally,
depending on the mixing ratio of the two surfactants in the
aggregate, by electrostatic attraction between the oppositely
charged molecules, which makes the packing parameter p of
the micelles larger than that of single SDS or DTEAB mi-
celles. Therefore, at appropriate surfactant composition, the
additional increase in p induced by the solubilization of n-
butylbenzene may satisfy the requirements for vesicle for-
mation and induce MVT in the system. Note that a qualita-
tive rather than quantitative explanation based on the
theory of the packing parameter is presented here, since it is
Based on our study, the Reff of n-butylbenzene (XDTEAB
=
0.25–0.33 and 0.86–0.96) is close to the vesicular region on
both the DTEAB-rich and SDS-rich sides (see Figure 1);
this suggests that the micellar phases within Reff have larger
p parameters closer to those of vesicles than those outside
Reff. Therefore, it is possible that the capability of n-butyl-
benzene to increase the packing parameter is limited, so
that for micellar phases outside Reff the contribution of n-bu-
tylbenzene to increasing the p parameter cannot meet the
requirement for the formation of vesicles, and no MVT is
observed in these solutions.
As mentioned above, alkanes were usually considered to
be solubilized in the micellar core of surfactant micelles and
induce formation of swollen spherical micelles.[10] However,
induction of a MVT by alkanes in the DTEAB/SDS system
can not be satisfactorily explained if alkane molecules are
merely solubilized in the micellar core. It was proposed that
alkane molecules that have shorter hydrocarbon chains than
the surfactant are not restricted to the interior of the micelle
and can partially penetrate into the palisade layer of the mi-
celle,[18] which will also promote an increase in the effective
hydrocarbon tail volume v in the hydrophobic region.
Hence, the solubilization of alkane, similar to the effect of
n-alkylbenzene, may increase the packing parameter, and
the MVT may then take place within certain ranges of sur-
factant composition in the catanionic surfactant system. In
addition, the Reff of the alkanes are generally narrower than
that of n-butylbenzene, probably because the alkane mole-
2832
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 2825 – 2835

Micelle-to-Vesicle Transition in Catanionic Surfactant Systems
FULL PAPER
cules are not completely solubilized in the micelle palisade
and are still located close to the interior of the micelle, so
that the ability of alkanes to increase the packing parameter
is weaker than that of n-butylbenzene. Moreover, it was pro-
posed that alkane molecules with longer chains have a
stronger tendency to be solubilized in the hydrophobic core
instead of the palisade of the surfactant micelle;[18a,19] hence,
Reff of alkanes becomes narrower with increasing hydrocar-
bon chain length (see Table 1). Thus, it is reasonable that
the order of the ability of hydrocarbons to induce a MVT is
n-alkylbenzene>n-hexane>n-octane>n-dodecane.
However, these hydrocarbons are not generally effective
in inducing MVTs in various catanionic surfactant systems.
None of the above hydrocarbons can induce a MVT in the
micellar phase of DTEAB/SDSO3 (Ctotal =10 mmol), no
matter how the surfactant ratio is varied, although the mo-
lecular structure of SDSO3 is very similar to that of SDS
except for the head group. The contrasting solubilization be-
haviors of the hydrocarbons may be correlated to the differ-
ent molecular interactions between surfactant molecules in
the two systems. Recently, we demonstrated[4b] that the
DTEAB/SDSO3 system exhibits weaker intermolecular in-
teractions than the DTEAB/SDS system by comparison of
the b parameters based on regular solution theory.[20] There-
fore, it can be expected that the surfactant molecules in the
DTEAB/SDSO3 micelle are more loosely packed than those
in the DTEAB/SDS micelle,[20b,c] and this may make the
packing parameter and aggregation number smaller than
those of the DTEAB/SDS micelle at similar surfactant com-
position. In fact, our TRFQ experiments have shown that at
similar surfactant composition (e.g., Ctotal =10 mmol and
group. Besides the contribution of the hydrocarbon chain to
increasing effective hydrocarbon tail volume v, the effect of
the head group should also be considered. Previous work
showed that solubilization of certain alcohols in lipid solu-
tion can significantly reduce the bending energy of the lipid
multilayers and make the membranes more flexible.[21]
Therefore, it is reasonable to think that the presence of the
polar head group of the polar additive probably significantly
alters the interactions between surfactant head groups in
catanionoic surfactant systems and favors the formation of
vesicles. Thus, the polar additives would have much higher
efficiency and activity in inducing a MVT than the apolar
hydrocarbons. It is also noteworthy that for the DEAB/SDS
system (Ctotal =10 mmol, XDTEAB =0.25), a lower concentra-
tion of C8NH2 (1.0 mmol) is required for a complete MVT
compared with that of C8OH (1.5 mmol). This difference
+
may come from the partial protonation of C8NH2 to C8NH3
in water,[22] which can neutralize the excess negative charges
in the aggregate and make an extra contribution to the in-
crease in the packing parameter than the neutral C8OH mol-
ecule. Hence C8NH2 will exhibit greater efficiency in induc-
ing a MVT in micellar phases that are rich in anionic surfac-
tant.
The ratio of vesicles to micelles can be adjusted by addi-
tion of relatively small amounts of polar organic compounds
in both catanionic surfactant systems, and this may have po-
tential applications in templated synthesis of materials. Re-
cently, the synthesis of bimodal mesoporous materials has
attracted special interest for their higher activities and
better control of selectivity over monomodal porous materi-
als.[23] Bimodal mesoporous materials with a well-defined
pore size distribution on two different scales can be easily
synthesized in one step by a dual-templating method[24] in
which two different kinds of organized supermolecular as-
semblies simultaneously coexist in one phase (e.g., vesicle/
micelle system) and separately control the bimodal pores.
Although mixed vesicular/micellar phases exist in many cat-
anionic surfactant systems and are quite promising for use
as dual templates, the region of coexisting vesicle/micelle
phases in the phase diagram is usually narrow.[1b,25] There-
fore, it is difficult to actively control the ratio of vesicles to
micelles by varying the surfactant ratio in catanionic surfac-
tant systems, and so far few attempts[26] have been made to
synthesize controlled bimodal mesoporous materials with
this kind of dual templates. Our work may provide a novel
and universal method to realize such goal. Micelle/vesicle
solutions of catanionic surfactant systems generated by addi-
tion of polar additives could act as templates for synthesis
of bimodal porous materials in which the two pore size dis-
tributions are separately controlled by the micelles and vesi-
cles. Moreover, a series of dual-template systems with differ-
ent ratios of vesicle to micelle can be easily obtained by
gradual addition of polar additives, and thus the connection
between the dual-template solutions and the structures, as
well as the properties of synthesized materials, can also be
systemically investigated, which may help to understand the
complex template mechanism and find suitable templates
X
DTEAB =0.25), the aggregation number of the DTEAB/
SDSO3 micelle is clearly smaller than that of the DTEAB/
SDS micelle. Considering the limited contribution of the hy-
drocarbons to the increased packing parameter, it is reason-
able to assume that the increase in the total aggregate pack-
ing parameter due to the presence of these hydrocarbons
can not meet the requirements for vesicle formation in all
micellar solutions of the DTEAB/SDSO3 system with
C
total =10 mmol. Thus, it can be concluded that MVT cannot
be induced by addition of hydrocarbons to catanionic sys-
tems in which the interactions between surfactant molecules
are relatively weak.
High efficiency and activity of polar organic additives in in-
ducinga MVT and adjustingthe ratio of vesicles to micelles :
The Reff of both C8OH and C8NH2 are much broader than
those of the hydrocarbons in the DTEAB/SDS system
(Ctotal =10 mmol), and the polar additives can induce a
MVT even in the DTEAB/SDSO3 system (Ctotal =10 mmol),
whereas the hydrocarbons fail to do so under the same con-
ditions, which demonstrates the superiority of polar organic
additives over hydrocarbons in inducing MVTs. Polar organ-
ic additives, such as C8OH, can preferentially penetrate into
the palisade of surfactant micelles with their polar head
group (OH group of C8OH) towards the surface and thus
act as a kind of co-surfactant with a rather small head
Chem. Eur. J. 2006, 12, 2825 – 2835
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2833

J. Huang et al.
1
Z
for other organized assemblies. Further study is ongoing in
our laboratory.
gð1Þðq,tÞ ¼
GðG,qÞexpðÀGtÞdG
ð5Þ
0
Laplace inversion of Equation (5) with the CONTIN program for peak-
constrained analysis[28] supplied with the ALV-5000 digital time correlator
yields the spectrum of decay rates for the various processes in the scatter-
ing samples as a series of amplitudes G(G) along a grid of G values. Equa-
tion (5) can be modified as Equation (6),
Conclusion
It can be concluded that the induction of MVT in catanionic
surfactant systems by addition of organic additives is closely
related to the structure and polarity of the additive, the sur-
factant composition, and the interactions between the sur-
factant molecules. Addition of a nonpolar hydrocarbon with
a certain chain length, such as n-butylbenzene or n-octane,
can induce MVT at proper surfactant compositions in a cat-
anionic surfactant system with relatively strong intermolecu-
lar interactions and high micelle aggregation number. Polar
additives, such as n-octanol, are much more efficient in in-
ducing MVT even in the system with weaker intermolecular
interactions. Moreover, solubilization of the polar additive
provides a series of two-template systems with variable
ratios of vesicles to micelles for the synthesis of controlled
bimodal mesoporous materials. This work may advance fur-
ther understanding of the controlled transformation of or-
ganized assemblies in surfactant solutions and promote its
applications in other fields.
1
Z
gð1Þðq,tÞ ¼
GðG,qÞexpðÀGtÞdlgðG=q2Þ
ð6Þ
0
so that the plot is in an equal-area representation and the area under
each G(G) peak corresponds to the weight of the peak. The effective dif-
fusion coefficient Deff of the aggregates is defined by Deff =G/q2. There-
fore, G(G) obtained can be transformed into an effective coefficient dis-
tribution G(Deff). Furthermore, G(Deff) can be converted to the apparent
hydrodynamic radius distribution f(Rh,app) by the well-known Stokes–Ein-
stein equation Rh,app =kBT/(6phDeff), where kB is the Boltzmann constant,
h the viscosity of the solvent, and T the absolute temperature of the solu-
tion. The Stokes–Einstein equation is only strictly valid for spherical par-
ticles. For asymmetrical particles, Rh,app obtained from the Stokes–Ein-
stein equation is the equivalent value corresponding to a sphere of the
same diffusion coefficient.
Time-resolved fluorescence quenching(TRFQ) : Pyrene was used as fluo-
rescence probe, and dodecylpyridinium chloride (C12PyCl) as quencher of
the fluorescence probe. The desired amount of pyrene in ethanol and
C12PyCl in water were added to the surfactant solution directly, and then
the solution was stirred vigorously for 48 h. The concentration of pyrene
was kept low (1”10À6 m) to prevent excimer formation. The amount of
ethanol in the surfactant solution was controlled to less than 1 vol%. All
of the solutions were degassed by nitrogen for 15 min before the mea-
surement to eliminate the influence of oxygen. Pyrene fluorescence
decay curves were monitored by an Edinburgh FLS920 time-resolved
fluorescence spectrophotometer (excitation at 337 nm and emission at
385 nm). The fluorescence decay of pyrene was fitted by using modified
DECAN 1.0 software from the Key Laboratory of Photochemistry, CAS.
Experimental Section
Materials: n-Dodecyltriethylammonium bromide (DTEAB) was pre-
pared by reaction of 1-bromododecane and triethylamine and was recrys-
tallized five times from ethanol/acetone. The purity of the surfactant was
examined, and no minimum was found in the surface tension curve.
Sodium n-dodecyl sulfate (SDS) and sodium n-dodecyl sulfonate
(SDSO3) were purchased from Acros Organics Co. and were used as re-
ceived. The water used was redistilled from potassium permanganate.
The other reagents were products of A.R. grade.
Absorbance measurements: Absorbance measurements were carried out
with a Varian CARY 1E spectrophotometer at 500 nm.
Sample preparation: Catanionic surfactant mixtures were prepared by
mixing the single surfactant solutions directly in a test tube. Then the de-
sired amount of organic additive was added to the tube by microsyringe.
After stirring for several minutes, the sample was held thermostatically at
258C for at least one week to reach equilibrium.
Transmission electron microscopy (TEM): Samples for TEM were pre-
pared by negative-staining technique with aqueous uranyl acetate solu-
tion. A JEM-100CX electron microscope was employed.
Rheology measurements: The rheological properties of samples were
measured with a ThermoHaake RS300 rheometer. A double-gap cylindri-
cal sensor system was used with an outside gap of 0.30 mm and an inside
gap of 0.25 mm.
Dynamic light scattering (DLS): The hydrodynamic properties of the sur-
factant aggregates can be characterized by DLS. DLS measurements
were performed with a spectrometer (ALV-5000/E/WIN Multiple Tau
Digital Correlator) and a Spectra- Physics 2017 200 mW Ar laser (l=
514.5 nm). The scattering angle was 908. All solutions were filtered
through a 0.4 mm filter into the cylindrical scattering cell before the ex-
periment. Experimentally, the normalized intensity autocorrelation func-
tion g(2)(q,t) at delay time t and scattering vector q is measured first. q is
defined as q=(4pn0/l0)sin(q/2), where n0 is the refractive index of the
solvent and l0 is the wavelength of the incident light in vacuum. g(2)(q,t)
Acknowledgements
This work was supported by the NSFC and doctoral program of MOE of
China.
can then be converted to the electric field autocorrelation function g(1)
(q,t) by means of a Siegert relationship [Eq. (4)],[27]
-
[1] a) E. W. Kaler, A. K. Murthy, B. E. Rodriguez, J. A. N. Zasadzinski,
Science 1989, 245, 1371; b) E. W. Kaler, K. L. Herrington, A. K.
Murthy, J. A. N. Zasadzinski, J. Phys. Chem. 1992, 96, 6698; c) K. L.
Herrington, E. W. Kaler, D. D. Miller, J. A. Zasadzinski, S. Chiruvo-
lu, J. Phys. Chem. 1993, 97, 13792; d) A. Caria, A. Khan, Langmuir
1996, 12, 6282; e) E. F. Marques, O. Regev, A. Khan, M. G. Miguel,
B. Lindman, J. Phys. Chem. B 1998, 102, 6746; f) K. Horbaschek, H.
Hoffmann, J. C. Hao, J. Phys. Chem. B 2000, 104, 2781; g) T. Kato,
H. Takeuchi, T. Seimiya, J. Phys. Chem. 1992, 96, 6839; h) M. B.
Sjçbom, H. Edlund, Langmuir 2002, 18, 8309; i) R. D. Koehler, S. R.
2
gð2Þðq,tÞ ¼ Bf1 þ bjgð1Þðq,tÞj g
ð4Þ
where B is the (infinite time) baseline and b is an optical constant. g(1)
(q,t) can be written as the Laplace transformation of the distribution of
relaxation rates G(G) [Eq. (5)],
-
2834
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 2825 – 2835

Micelle-to-Vesicle Transition in Catanionic Surfactant Systems
FULL PAPER
Raghavan, E. W. Kaler, J. Phys. Chem. B 2000, 104, 11035; j) M.
Bergstrçm, Langmuir 1996, 12, 2454.
ceedings of the 2nd International Congress of Surface Activity, Vol. 1,
Butterworth, London, 1957, p. 311.
[2] a) A. Helenius, K. Simons, Biochim. Biophys. Acta 1975, 415, 29;
b) J. Fendler, Membrane Mimetic Chemistry, Wiley, New York, 1983.
[3] a) P. Fromherz, Chem. Phys. Lett. 1983, 94, 259; b) D. D. Lasic, Bio-
chem. J. 1988, 256, 1; c) M. M. Stecker, G. B. Benedek, J. Phys.
Chem. 1984, 88, 6519.
[4] a) H. Q. Yin, Z. K. Zhou, J. B. Huang, R. Zheng, Y. Y. Zhang,
Angew. Chem. 2003, 115, 2238; Angew. Chem. Int. Ed. 2003, 42,
2188; b) H. Q. Yin, J. B. Huang, Y. Y. Lin, Y. Y. Zhang, S. C. Qiu,
J. P. Ye, J. Phys. Chem. B 2005, 109, 4104.
[5] L. L. Brasher, K. L. Herrington, E. W. Kaler, Langmuir 1996, 12,
4267.
[6] M. Mao, J. B. Huang, B. Y. Zhu, J. P. Ye, J. Phys. Chem. B 2002, 106,
219.
[7] a) M. Singh, C. Ford, V. Agarwal, G. Fritz, A. Bose, V. T. John, G. L.
McPherson, Langmuir 2004, 20, 9931; b) H. Hoffmann, C. Thunig,
D. Miller, Colloids Surf. A 2002, 210, 147; c) H. Hoffmann, K. Hor-
baschek, F. Witte, J. Colloid Interface Sci. 2001, 235, 33; d) R. Oda,
L. Bourdieu, J. Phys. Chem. B 1997, 101, 5913; e) S. Chiruvolu,
H. D. Warriner, E. Naranjo, S. H. J. Idziak, J. O. Rꢁdler, R. J. Plano,
J. A. Zasadzinski, C. R. Ssfinya, Science 1994, 266, 1222; f) H. Hoff-
mann, G. Ebert, Angew. Chem. 1988, 100, 933; Angew. Chem. Int.
Ed. Engl. 1988, 27, 902.
[8] a) C. Z. Wang, S. H. Tang, J. B. Huang, X. R. Zhang, H. L. Fu, Col-
loid Polym. Sci. 2002, 280, 770; b) X. R. Zhang, J. B. Huang, M.
Mao, S. H. Tang, B. Y. Zhu, Colloid Polym. Sci. 2001, 279, 1245.
[9] a) R. Bruinsma, W. Gelbart, A. B. Shaul, J. Chem. Phys. 1992, 96,
7710; b) S. Q. Wang, J. Phys. Chem. 1990, 94, 8381; c) S. Hofmann,
H. Hoffmann, J. Phys. Chem. B 1998, 102, 5614; d) Cl. Oelschlaeger,
G. Waton, S. J. Candau, M. E. Cates, Langmuir 2002, 18, 7265; e) R.
Oda, P. Panizza, M. Schmutz, F. Lequeux, Langmuir 1997, 13, 6407.
[10] a) R. Nagarajan, Curr. Opin. Colloid Interface Sci. 1996, 1, 391;
b) O. Bayer, H. Hoffmann, W. Ulbricht in Surfactants in Solution,
Vol. 4, Plenum, New York, 1987, Part I, p. 343.
[16] a) J. N. Israelachvili, D. J. Mitchell, B. W. Ninham, J. Chem. Soc.
Faraday Trans. 2 1976, 72, 1525; b) J. N. Israelachvili, D. J. Mitchell,
B. W. Ninham, Biochim. Biophys. Acta 1977, 470, 185; c) J. N. Israel-
achvili, S. Marcelja, R. G. Q. Horn, Q. Rev. Biophys. 1980, 13, 121.
[17] a) R. Nagarajan in Structure-Performance Relationships in Surfac-
tants (Eds.: K. Esumi, M. Ueno), Marcel Dekker, New York, 1997,
Chap. 1, pp. 1–89; b) K. A. Dill, P. J. Flory, Proc. Natl. Acad. Sci.
USA 1980, 77, 3115; c) A. Ben-Shaul, I. Szleifer, W. M. Gelbart,
Proc. Natl. Acad. Sci. USA 1984, 81, 4601; d) S. Puvvada, D.
Blankschtein, J. Phys. Chem. 1992, 96, 5567.
[18] a) B. Bumgdad, J. Eastoe, R. K. Heenan, J. R. Lu, D. C. Steytler, S.
Egelhaad, J. Chem. Soc. Faraday Trans. 1998, 94, 2143; b) L. E.
Scriven, J. Chem. Phys. 1983, 79, 11; c) F. Lichterfeld, T. Schmeling,
R. Strey, J. Phys. Chem. 1986, 90, 5762.
[19] J. Penfold, E. Staples, I. Tucker, J. Phys. Chem. B 2002, 106, 8891.
[20] a) B. Jonsson, B. Lindman, K. Holmberg, B. Kronberg, Surfactants
and Polymers in Aqueous Solution, Wiley, New York, 1998;
b) Mixed Surfactant Systems (Eds.: P. M. Holland, D. N. Rubingh)
ACS Symposium Series 501, American Chemical Society, Washing-
ton, DC, 1992, Chaps. 1 and 2; c) S. R. Raghavan, G. Fritz, E. W.
Kaler, Langmuir 2002, 18, 3797.
[21] C. R. Safinya, E. B. Sirota, D. Roux, G. S. Smith, Phys. Rev. Lett.
1989, 62, 1134.
[22] Kabir-ud-Din, S. Kumar, Kirti, P. S. Goyal, Langmuir 1996, 12,
1490.
[23] a) M. Almgren, J. E. Lçfroth, J. Colloid Interface Sci. 1981, 81, 486;
b) M. Tachiya, Chem. Phys. Lett. 1975, 33, 289; c) M. H. Gehlen,
F. C. De Schryver, Chem. Rev. 1993, 93, 199; d) F. Caruso, F. Grieser,
A. Murphy, P. Thistlethwaite, R. Urquhart, M. Almgren, E. Wistus,
J. Am. Chem. Soc. 1991, 113, 4838.
[24] a) D. M. Antonelli, Microporous Mesoporous Mater. 1999, 33, 209;
b) J. L. Blin, A. Lꢃonard, Z. Y. Yuan, L. Gigot, A. Vantomme, A. K.
Cheetham, B. L. Su, Angew. Chem. 2003, 115, 2978; Angew. Chem.
Int. Ed. 2003, 42, 2872.
[11] a) M. Almgren, J. E. J. Lçfroth, J. Colloid Interface Sci. 1981, 81,
486; b) M. Tachiya, Chem. Phys. Lett. 1975, 33, 289; c) M. Almgren,
Adv. Colloid Interface Sci. 1992, 41, 9; d) M. H. Gehlen, F. C. De
Schryver, Chem. Rev. 1993, 93, 199.
[12] a) F. Caruso, F. Grieser, A. Murphy, P. Thistlethwaite, R. Urquhart,
M. Almgren, E. Wistus, J. Am. Chem. Soc. 1991, 113, 4838; b) B.
Medhage, M. Almgren, J. Fluoresc. 1992, 2, 7.
[25] a) M. T. Yatcilla, K. L. Herrington, L. L. Brasher, E. W. Kaler, S.
Chiruvolu, J. A. Zasadzinski, J. Phys. Chem. 1996, 100, 5874; b) A. J.
OꢄConnor, T. A. Hatton, A. Bose, Langmuir 1997, 13, 6931; c) M.
Bergstrçm, J. S. Pedersen, Langmuir 1999, 15, 2250.
[26] a) J. G. C. Shen, J. Phys. Chem. B 2004, 108, 44; b) A. Lind, B.
Spliethoff, M. Lindꢃn, Chem. Mater. 2003, 15, 813.
[13] a) D. D. Miller, L. J. Magid, F. D. Evans, J. Phys. Chem. 1990, 94,
5921; b) O. Sonderman, K. L. Herrington, E. W. Kaler, D. D. Miller,
Langmuir 1997, 13, 5531.
[27] a) B. J. Berne, R. Pecora, Dynamic Light Scattering, Wiley, New
York, 1976; b) B. Chu, Laser Light Scattering, 2nd ed., Academic
Press, New York, 1991.
[14] a) H. Hoffmann, W. Ulbricht, J. Colloid Interface Sci. 1989, 129, 388;
b) P. Mukerjee, J. R. Cardinal, J. Phys. Chem. 1978, 82, 1620; c) R.
Nagarajan, M. A. Chaiko, E. Ruchenstein, J. Phys. Chem. 1984, 88,
2916.
[28] a) S. W. Provencher, J. Hendrix, L. D. Maeyer, J. Chem. Phys. 1978,
69, 4273; b) S. W. Provencher, Makromol. Chem. 1979, 180, 201;
c) S. W. Provencher, Comput. Phys. Commun. 1982, 27, 213.
[15] a) S. Kumar, A. Z. Naqvi, Kabir-ud-Din, Langmuir 2001, 17, 4787;
b) N. Hedin, R. Sitnikov, I. Furꢂ, U. Henriksson, O. Regev, J. Phys.
Chem. B 1999, 103, 9631; c) P. M. Lindemuth, G. L. Bertrand, J.
Phys. Chem. 1993, 97, 7769; d) M. B. Smith, A. E. Alexander, Pro-
Received: August 28, 2005
Revised: November 7, 2005
Published online: January 17, 2006
Chem. Eur. J. 2006, 12, 2825 – 2835
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2835