Catalytic hydrogenation of dimethyl itaconate in non-ionic microemulsions: Influence of the size of micelle

Article abstract of DOI:10.1039/b905063a

The structural dimensions of two different microemulsions were extracted and correlated with the catalytic hydrogenation of dimethyl itaconate (DMI) performed in such media using the water-soluble catalyst complex Rh-TPPTS. The commercial, polyoxyethylene-based non-ionic surfactants Igepal CA-520 and Triton X-100 were used to obtain two different types of microemulsions with cyclohexane and water; pentanol as a cosurfactant was added to the Triton system. Dynamic light scattering (DLS) and small angle neutron scattering (SANS) measurements were used to determine the characteristic sizes of the Igepal and Triton microemulsions, respectively, showing a linear dependence between the initial hydrogenation rate of DMI and the radius of the micelles. The initial hydrogenation rate of DMI in bulk water is exceeded in both microemulsions. Indications of deformation of the originally spherical Triton X-100 reverse micelles upon addition of the water-soluble catalyst complex Rh-TPPTS were found.

Full text of DOI:10.1039/b905063a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Catalytic hydrogenation of dimethyl itaconate in non-ionic
microemulsions: influence of the size of micellewz
Juan Milano-Brusco, Sylvain Prevost, Dersy Lugo, Michael Gradzielski
´
a
bc
b
b
a
and Reinhard Schomacker*
¨
Received (in Montpellier, France) 12th March 2009, Accepted 30th April 2009
First published as an Advance Article on the web 2nd June 2009
DOI: 10.1039/b905063a
The structural dimensions of two different microemulsions were extracted and correlated with the
catalytic hydrogenation of dimethyl itaconate (DMI) performed in such media using the
water-soluble catalyst complex Rh–TPPTS. The commercial, polyoxyethylene-based non-ionic
surfactants Igepal CA-520 and Triton X-100 were used to obtain two different types of
microemulsions with cyclohexane and water; pentanol as a cosurfactant was added to the Triton
system. Dynamic light scattering (DLS) and small angle neutron scattering (SANS) measurements
were used to determine the characteristic sizes of the Igepal and Triton microemulsions,
respectively, showing a linear dependence between the initial hydrogenation rate of DMI and the
radius of the micelles. The initial hydrogenation rate of DMI in bulk water is exceeded in both
microemulsions. Indications of deformation of the originally spherical Triton X-100 reverse
micelles upon addition of the water-soluble catalyst complex Rh–TPPTS were found.
medium. Water in the core may display deviating electrophilic
and nucleophilic properties relative to ‘‘bulk water’’ because of
strong interactions with the head groups of the surfactant,
depending on the size of the droplets. Thus, the effect of this
medium on reaction rates is often dependent on the water/
1
. Introduction
Micelles are supramolecular assemblies in
a colloidal
dimension. These assemblies are based on the amphiphilic
characteristics of surfactants or amphiphiles, which are
molecules with a polar head group and a hydrophobic tail.
Micelles can be formed in water and in non-polar solvents
5
surfactant (o) molar ratio.
We have recently published a kinetic study that compares
(
see Scheme 1), for which the names of micelles and reverse
the catalytic hydrogenation of dimethyl itaconate in a Triton
6
X-100 microemulsion and in a biphasic system. In this study
micelles have been given, respectively. Reverse micelles are
formed by the association of polar head groups of amphiphiles
with colloidal drops of water in an organic medium once
the concentration of the amphiphiles in the interface is
we noticed an important influence of the microemulsion
formulation on the initial rate of the hydrogenation. The
Triton X-100 system used in the earlier study needed the
addition of 1-pentanol as cosurfactant, which also had an
inhibition effect on the hydrogenation. Cosurfactants are used
in order to stabilize the surfactants in the interface. When the
amphiphile molecules have large polar groups, as Triton
1
,2
appropriate. The delimitation of hydrophilic and hydro-
phobic micro-domains is an important feature of reverse micelles
3
favourable to reactions. For this reason they are being
regarded as microreactors, whose capacity for reactants and
variability of solubilization properties are high and of practical
4
interest. Contrary to the case of aqueous micelles, reactions in
7
X-100 has, the interactions between each surfactant molecule
destabilize the micelle. For this reason the addition of
short-chain alkanols, which also have amphiphilic properties
reverse micelles are possible at high concentrations of hydro-
phobic substrates, and in regard to homogeneous reaction
media they can be divided, as observed in Scheme 1, into
three spatial domains: the aqueous core, the hydrophilic–
hydrophobic interface, and the surrounding hydrophobic
and position themselves in between the surfactant molecules,
6,8
cushion the interactions stabilizing the micelles.
The
cosurfactant has also an important effect on the size of
the micelles. Accordingly the determination of the size and
the structural characterization of the microemulsion droplets
is an important aspect in understanding catalytic reactions in
a
Department of Chemistry, TU Berlin, Secr. TC-8, Strasse des 17.
Juni 124, 10623 Berlin, Germany. E-mail: schomaecker@tu-berlin.de
Stranski Laboratorium fu¨r Physikalische und Theoretische Chemie,
b
Department of Chemistry, TU Berlin, Secr. TC-7, Strasse des 17.
Juni 124, 10623 Berlin, Germany
Helmholtz Zentrum Berlin, Glienicker Strasse 100, 14109 Berlin,
c
Germany
w This article is warmly dedicated to Professor Dr G. H. Findenegg on
the occasion of his 70th birthday.
z Electronic supplementary information (ESI) available: Original data
from the cosurfactant partitioning study (Table A, B & C), preliminary
data for the analysis of the SANS measurements (Table D, E & F), and
autocorrelation functions for the DLS measurements (Fig. A & B).
See DOI: 10.1039/b905063a
Scheme 1 Scheme of a micelle and a reverse micelle.
1
726 | New J. Chem., 2009, 33, 1726–1735
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009

View Article Online
microemulsions. Due to the small size of the micelles, whose
9
,10
diameters are typically in the range of 1–100 nm,
micro-
emulsions are usually transparent; this characteristic feature
has rendered them amenable for studies by dynamic light
1
0,11
scattering (DLS).
Such studies are accurate when the
samples are highly diluted, enough to safely neglect inter-
actions between aggregates and multiple scattering. This
Scheme 2 (p-tert-Octylphenoxy)polyethoxyethanol chemical formula.
1
2
condition is often not satisfied with microemulsions. Hence
experiments and data analysis applying the scattering theories
must be performed with special care, and the results must be
regarded with a critical mind and be combined with results
from other techniques. In contrast to DLS, small angle
neutron scattering (SANS) with its higher spatial resolution
can be applied to such systems to corroborate values obtained
for the hydrodynamic radius and add information regarding
the core size of the micelles. Small angle neutron scattering
The surfactants used were the commercially available Triton
X-100 and Igepal CA-520, respectively. Both are based on
tert-octylphenoxypolyethoxyethanol (Scheme 2) with different
number of ethoxyethanol groups. Igepal CA-520 has approxi-
mately half the ethoxyethanol groups (n = 5) of Triton X-100
(
n = 9–10).
Test tubes were prepared with different compositions of
water and cyclohexane (a) but fixed weight fractions of Triton
X-100 and 1-pentanol (g). The sequence in which the compo-
nents were added was: (1) cyclohexane, (2) Triton X-100,
(
SANS) is a well established technique used to characterize
1
3–19
microemulsions in much structural detail,
and it has been
2
0,21
(
3) 1-pentanol, and (4) water. Before and after water addition,
successfully applied to Triton X-100 micellar systems.
the solution was agitated for 5 min. The test tubes were
immersed in a thermostated water bath, and phase behaviour
was observed after 15 min. This procedure was repeated with
increasing temperature of 1 K stepwise. The same procedure
was used for the Igepal CA-520 system. The change in phase
behaviour depending on temperature and composition
followed a pattern typical for non-ionic surfactant-based
microemulsions.
The present study reports the correlation between the size of
reverse micelles obtained by SANS and DLS, and the initial
hydrogenation rate of dimethyl itaconate (DMI) with the water-
soluble catalyst complex Rh–TPPTS using two non-ionic micro-
emulsion systems with different water content (characterized by o)
as dispersive media. In addition, the influence of the cosurfactant–
surfactant mass ratio (d) as a reaction parameter was studied.
2
. Experimental section
2.3 Cosurfactant partition
2
.1 Chemicals
The solvents cyclohexane ( Z 99.5%, Roth) and 1-pentanol
Z 99%, Merck) were degassed and purged under nitrogen
and used without further purification. Cyclohexane-d12
99.5%, Deutero GmbH), deuterium oxide (99.9%, Deutero
Many publications dealing with microemulsion quaternary
systems take into consideration that the cosurfactant, which
in our case is 1-pentanol, partitions itself between the inter-
12,22,23
(
facial film and the continuous oily phase.
We wanted to
start by analysing the percentage of pentanol possibly accu-
mulated in the core of the micelles. For this reason, a partition
study of 1-pentanol in biphasic systems cyclohexane–water
with PEG 400, cyclohexane–water with PEG 1000 and cyclo-
hexane–water alone was made. The compositions for the
Triton X-100 systems shown in Table 1 were used to prepare
the samples, and instead of adding Triton X-100, the same
molar amount of PEG was added. PEG 400 represents a good
approximation for the polar group of the Triton X-100
molecules, and the cyclohexane–water systems with and
without PEG 1000 allows us to observe the influence of
PEG and its ethoxyethanol chain length on the pentanol
partition. The pentanol concentration in the cyclohexane
phase was analysed by gas chromatography (GC) using a
Shimadzu 2010 GC (DB-5HT column, approx. 30 m, d =
(
GmbH), poly(ethylene glycol) 400 and 1000 (100%, Fluka),
the water-soluble ligand TPPTS (30.7 wt% in water, Celanese),
and the surfactants Triton X-100 (100%, Sigma-Aldrich) and
Igepal CA-520 (100%, Sigma-Aldrich) were used as received.
The catalyst precursor [Rh(cod)Cl] (98%, Strem) was kept
2
under nitrogen and used as received.
2
.2 Microemulsion preparation
The following common terms were used to describe the composi-
tions of the four-component systems; these terms are simplified in
case of using a three-component system (without cosurfactant):
moil
a ¼
moil þ mwater
0
2
.32 mm, 0.4 bar N , 70 1C, FID) obtaining a retention time
msurfactant
g ¼
for 1-pentanol of approximately 3 min.
moil þ mwater þ msurfactant
mcosurfactant
2.4 Catalytic hydrogenation runs
d ¼
o ¼
ð1Þ
msurfactant
The hydrogenation of DMI catalysed by the water-soluble
catalyst complex Rh–TPPTS (Scheme 3) was selected as a
reference for studying the influence of the micelle size on the
reaction kinetics. DMI has a partition coefficient smaller
than 1 at low DMI concentrations (o50 mmol Lꢁ ) in a
cyclohexane–water biphasic system, but by increasing the
nwater
nsurfactant
nwater
1
oT ¼
nsurfactant þ ncosurfactant
ꢁ1
where m indicates the mass and n the number of moles.
DMI concentration to higher values (450 mmol L ) more
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1726–1735 | 1727

View Article Online
Table 1 Composition of the different microemulsions for both
systems: [Triton X-100–1-pentanol]–cyclohexane–water and Igepal
CA-520–cyclohexane–water
subsequent stirring at 800 rpm. A decreasing and finally
expiring hydrogen flux indicated the end of the reaction. The
complete hydrogenation of DMI to dimethyl methylsuccinate
(DMS) was confirmed by gas chromatography (GC) using a
HP 5710A instrument (Lipodex E capillary column, ca. 25 m,
a
Sample
Surfactant
a
g (%)
o
o
T
d
ME 1
ME 2
ME 3
ME 4
ME 5
ME 6
ME 7
ME 8
ME 9
ME 10
ME 11
ME 12
ME 13
ME 14
ME 15
ME 16
ME 17
ME 18
ME 19
ME 20
ME 21
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Triton X-100
Igepal CA-520
Igepal CA-520
Igepal CA-520
Igepal CA-520
Igepal CA-520
Igepal CA-520
Igepal CA-520
Igepal CA-520
Igepal CA-520
0.94
0.95
0.97
0.94
0.95
0.97
0.94
0.95
0.97
0.94
0.95
0.97
0.956
0.956
0.956
0.956
0.956
0.956
0.960
0.964
0.968
10
10
10
10
10
10
10
10
10
10
10
10
3.36
3.70
4.00
4.40
5.00
3.30
3.30
3.30
3.30
20.0
15.0
10.0
20.0
15.0
10.0
20.0
15.0
10.0
20.0
15.0
10.0
30.0
27.5
25.0
22.5
20.0
30.6
27.5
25.0
22.5
3.10
2.31
1.54
2.42
1.80
1.20
1.98
1.47
0.98
1.68
1.25
0.83
—
—
—
—
—
—
—
—
—
0.75
0.75
0.75
1.00
1.00
1.00
1.25
1.25
1.25
1.50
1.50
1.50
—
—
—
—
—
—
—
—
—
d = 0.25 mm, 0.6 bar N , 90 1C, FID) obtaining a retention
2
time of DMS of approximately 20 min. Microemulsion
samples were separated into organic and aqueous phases by
addition of water. After phase separation the organic phase
was analysed. Table 1 shows the compositions of the different
microemulsions used as media for the hydrogenation
experiments.
2.5 Dynamic light scattering
All samples were measured without catalyst. Dynamic light
scattering (DLS) was employed in order to study the size of the
micelles of the Igepal CA-520 systems in terms of the hydro-
dynamic radius. Correlation functions were recorded at a
scattering angle of 901 using a typical ALV goniometer setup
with a Nd:YAG-laser as light source (wavelength l = 532 nm).
The constant output power was 150 mW. All measurements
were done at 25 1C controlled by a toluene matching bath. The
correlation functions were generated using an ALV-5000/E
multiple t digital correlator and finally analysed by inverse
a
For the reactions, all the microemulsions were prepared with cyclo-
hexane and water. For the Triton X-100 system, 1-pentanol was used
as cosurfactant.
2
4
Laplace transformation (CONTIN).
2
.6 Small angle neutron scattering
SANS spectra were recorded on the instrument V4 at the BER
reactor of the Helmholtz Zentrum Berlin, Germany. Data
were recorded on a 64 ꢂ 64 2-dimensional gas detector
Scheme 3 Hydrogenation of dimethyl itaconate formal reaction.
˚
DMI accumulates in the cyclohexane phase and the partition
coefficient becomes higher than 1. This partition behaviour
can be influenced by the presence of surfactants in the
solution, changing the apparent order of the reaction from
(128 ꢂ 128 pixels), at a constant wavelength of 6.3 A (FWHM
18%). Samples were filled into a quartz cuvette (QS, Hellma)
with a neutron path length of 1 mm and thermostated at
25.3 1C ꢃ 0.4. Three sample-detector distances were used: 1, 4,
12 m, with collimation at 2, 4 and 12 m, respectively. Data
reduction was performed using the software package
BerSANS, with correction from the scattering contribution
of the empty cell, and deviations in the pixel efficiency
accounted for using the incoherent scattering of a 1 mm pure
water sample. Electronic and ambient background was
accounted for by the measurement of a cadmium plate. The
transmission of water was used to obtain the instrumental
coefficient, assuming the ideal case where non-transmitted
neutrons are scattered uniformly over the full solid angle
(4p). Finally data were azimuthally averaged, and intensities
at different configurations but corresponding to the same
sample were merged without the need of a scaling factor.
Due to the fact that the interactions of neutrons with hydrogen
and deuterium are widely different, labelling the micro-
domains of the micelles with deuterated compounds is
6
zero in a biphasic system to one in microemulsions.
First, the catalyst complex was prepared by mixing 22 mg of
the catalyst precursor [Rh(cod)Cl]2 (0.089 mmol Rh) with
1
240 mg aqueous 30% TPPTS solution (380.4 mg, 0.67 mmol
TPPTS) under nitrogen. This mixture was stirred under
nitrogen at ambient temperature for 24 h before it was used
in the hydrogenation reaction. The high molar ratio of
TPPTS/Rh (7.5) is used to ensure that the hydrogenation
starts at a maximal reaction rate and to avoid the formation
6
of rhodium metal. A thermostated double wall 200 ml glass
reactor equipped with a gas dispersion stirrer was used in this
study. Semi-batch reactions were performed under a constant
hydrogen pressure of 1.1 bar. The reaction rates were
calculated using the monitored hydrogen flow rate for keeping
the pressure at a constant level.
The microemulsions were prepared one day before the
experiment and agitated overnight. The reactor was evacuated
at 150 mbar and refilled with nitrogen 3 times after intro-
ducing the solvent and again after injection of the catalyst
solution (1.2 g) and 2 g of DMI, respectively. The mixture was
stirred at 400 rpm and 40 1C for 30 min and afterwards at
reaction temperature for 30 min. The reaction was initiated
after evacuating the reactor to 150 mbar, followed by an
increase in the pressure to 1.1 bar with hydrogen gas and
2
5
possible. For this reason, different contrast conditions were
used: microemulsions were prepared from heavy water,
Triton X-100, 1-pentanol, and either hydrogenated or
perdeuterated cyclohexane. The former case provides an inner
2
look at the D O-based core of the micelles, while in the
latter case the hydrogenated interfacial film is probed. Two
different methods were used to calculate the hydrodynamic
radius, both using the parameters obtained by the Guinier
1
728 | New J. Chem., 2009, 33, 1726–1735
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009

View Article Online
1
5
approximation.
following form:
The Guinier approximation has the
3. Results and discussion
3
.1 Microemulsions
"
#
q2R
2
g
Since Kahlweit and Strey presented for the first time the phase
IðqÞ ¼ I0 exp ꢁ
ð2Þ
2
6
3
prism for ternary water–oil–surfactant systems, built up
from different Gibbs diagrams of the ternary system at
different temperatures, studies in this area that dealt with
different applications of microemulsions have mainly used
two vertical sections of this prism, as seen in Fig. 1, which
where I
0
is the absolute intensity and R
g
is the radius of
gyration. R
g
is determined with linear regressions on Guinier
2
plots (ln(I) vs. q ) of the SANS data, and the radius of the
corresponding homogeneous spheres is estimated following
the form:
2
7,28
result in pseudo-binary phase diagrams.
One section with
the temperature as ordinate is obtained by keeping the oil-to-
water ratio constant (usually at a = 0.5), the other by keeping
the amount of surfactant constant. This last diagram shows
two biphasic system areas, separated by a horizontal one
phase canal. Fig. 2 shows a part of the one phase canal for
two systems: (a) [Triton X-100–1-pentanol]–cyclohexane–
water and (b) Igepal CA-520–cyclohexane–water. Both
systems were studied at surfactant concentrations (g) of 5
and 10%, respectively, showing the typical behaviour for such
systems of a wider one phase canal when using higher
concentrations of surfactants. 1-Pentanol was added to the
2
gSph
3
¼ R
5
2
R
ð3Þ
Small angle neutron scattering theory allows for another basic
way of determining the radius of the micelles, namely the
coherent macroscopic scattering cross section from an
assembly of particles dispersed in a solvent. The macroscopic
scattering cross section is represented by the expression given
2
5
below:
2
9
2
Triton X-100 system in order to stabilize the reverse micelles,
I(q) = fDp V F(q)S(q)
(4)
p
as the larger EO group of the Triton X-100 yields a smaller
packing parameter that is increased by the addition of the
pentanol. In addition the presence of pentanol, as will be
shown further, also has an important influence on the DMI
hydrogenation catalysed by the water-soluble catalyst complex
Rh–TPPTS included in the reverse micelles.
where f is the volume fraction of particles, V
p
their volume,
Dp the contrast, F(q) is the form factor accounting for the
shape and size of particles and S(q) is the structure factor
accounting for the spatial correlation between particles.
By considering the micelles to be non-interactive in a dilute
solution (S(q) = 1) and extrapolating to zero q (F(q) = 1),
eqn (4) can be re-arranged and gives:
3
.2 DMI hydrogenation in microemulsion
I0
fDp2
Using the compositions from Table 1 for 100 mL of micro-
emulsion, and hydrogenating 2 g of DMI with 20 mg of
Vp ¼
ð5Þ
[
Rh(cod)Cl]
2
and 346 mg of TPPTS for each microemulsion,
ꢁ1
a course of hydrogen consumption rate ([dV(H
2
) dt ],
f is calculated with the volume fractions of the polar
components:
ꢁ1
mL min ) versus time (min) is obtained. The maximum
hydrogenation rate observed after stabilization of the pressure
control system is taken as the initial rate of the hydrogenation
f = fEO10 + f
+ f
+ (x f )
C H12O C H12O
5 5
(6)
H O
D O
2
2
ꢁ ꢁ1
[dV(H ) dt ]0, mL min ). Fig. 3 shows the monitored
2
1
(
where fEO10, fH O, fD O and fC5H12O are the volume fractions
2
2
hydrogen consumption rate when hydrogenating 2 g of DMI
in microemulsion ME 1 containing Triton X-100; the initial
of the polar part of Triton X-100, water attached to the ethoxy
chain of the surfactant, deuterium oxide and pentanol, respec-
tively, and xC5H12O is the fraction of pentanol inside the core
of the micelles. The contrast is the difference between the
scattering length densities (SLD) of the medium composed of
cyclohexane-d12 and a fraction of pentanol and the mostly
hydrogenated particles using the following equation:
ꢁ
1
hydrogenation rate is 5.9 mLH2 min
.
By increasing the amount of water inside the reverse
micelles relative to the amount of surfactant (o) the initial
P
P
SLDpolar f
SLDnonpolar f
i
polari
i
nonpolari
Dp ¼
P
ꢁ
P
ð7Þ
f
f
polari
nonpolari
The software package SASfit, written by Joachim Kohlbrecher
Paul-Scherrer Institute, Switzerland), was used to analyse the
SANS spectra.
(
Fig. 1 Schematic phase prism of a ternary mixture of water–oil–
surfactant including two characteristic sections at constant a and
constant g, and whose distinctive regions of one phase microemulsion
(ME), biphasic (2f) and three-phase system (3f) are observed as a
function of the temperature (T).
2
.7 Conductivity
A commercially available compact conductivity meter cond
30i (WTW, Weilheim, Germany) was used. All conductivity
measurements were done at 25 1C.
3
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1726–1735 | 1729

View Article Online
Fig. 2 Phase prism section (constant g) with the respective one phase
microemulsion (ME) and biphasic system (2f) regions of: (a) the
quaternary system [Triton X-100–1-pentanol]–cyclohexane–water
with two constant weight fractions of Triton X-100 and 1-pentanol,
and (b) the ternary system Igepal CA-520–cyclohexane–water with
two constant weight fractions of Igepal CA-520.
Fig. 5 Initial hydrogenation rate of the catalytic hydrogenation of
DMI (2 g in 100 ml of microemulsion) as a function of g in Igepal
CA-520 microemulsions with a as a constant and as a variable, at
2
2
5 1C and 1.1 bar, using 20 mg of [Rh(cod)Cl] and 357 mg of TPPTS.
concentration is noticeable. As Fig. 5 shows, there are two
different patterns described in this figure. First, by decreasing
the concentration of Igepal CA-520 (g) with a constant ratio
cyclohexane-to-water (a = 0.956), the initial hydrogenation
rate increases. Second, by keeping the concentration of
Igepal CA-520 constant (g = 0.033) with decreasing ratio
cyclohexane-to-water (a), the initial hydrogenation rate also
increases. Both patterns are proof of an increase of the initial
hydrogenation rate with increase of the molar ratio of water to
surfactant (o). Fig. 5 also shows that by comparing the initial
hydrogenation rate of DMI in Igepal microemulsions of equal
o from both patterns, the Igepal microemulsions with lower
concentration of surfactant (g = 0.033) allowed for higher
initial hydrogenation rates. By keeping the ratio o constant,
with higher concentrations of Igepal, we expect an increase in
the concentration of micelles of equal size, so the catalyst
should distribute itself over more micelle cores. For the
Triton X-100 microemulsions, by increasing the amount of
cosurfactant in relation to the amount of surfactant (d), the
initial hydrogenation rate for each o decreases, resulting in a
displacement of the linearly dependent curves to smaller initial
hydrogenation rates, at constant slope (except for d = 1). This
behaviour indicates a direct proportionality between o and the
initial hydrogenation rate within an even wider o range than
shown in Fig. 4 (o = 10–20), where adding pentanol means an
additional amount of surfactant is available for stabilizing the
reverse micelles. For this reason the pentanol was also taken
into account for the correlation of the initial hydrogenation
Fig. 3 Hydrogen flux vs. time and conversion vs. time in the hydro-
genation of 2 g of DMI in a microemulsion system constituted by
4
5
2
.27 g of water, 64.23 g of cyclohexane, 7.61 g of Triton X-100 and
.71 g of 1-pentanol (ME 1 in Table 1), at 25 1C and 1.1 bar, using
0 mg of [Rh(cod)Cl] and 357 mg of TPPTS.
2
hydrogenation rate of DMI increases. This influence is linearly
dependent for all systems as can be seen in Fig. 4. For the
Igepal CA-520 systems, also an influence of the surfactant
rate of DMI vs. o (Fig. 6).
T
A model of the micelle is represented in Scheme 4. The three
domains observed in this scheme should not be well defined in
our microemulsions as the polar chains of Triton X-100 and
Igepal CA-520 are rather long.
The results from the partitioning study of pentanol between
cyclohexane and water, cyclohexane and an aqueous solution
of PEG-400, and the same oil and an aqueous solution of
PEG-1000 are shown in Table 2, and the original data for the
partitioning study of pentanol is presented in Tables SA, SB
Fig. 4 Initial hydrogenation rate of the catalytic hydrogenation of
DMI (2 g in 100 ml of microemulsion) as a function of o in Triton
X-100 microemulsions with different d, and in Igepal CA-520 micro-
emulsions with g as a constant and as a variable, at 25 1C and 1.1 bar,
2
using 20 mg of [Rh(cod)Cl] and 357 mg of TPPTS.
1
730 | New J. Chem., 2009, 33, 1726–1735
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009

View Article Online
the surfactant Triton X-100 to introduce pentanol inside the
core of the micelles is complex, because it is limited by the
solubility of the aliphatic chain of 1-pentanol in the interfacial
layer and the continuous phase of the microemulsion. For this
reason and in order to simplify our pentanol partition calcula-
tions, in the following we assume those values in Table 2
(
for PEG 400) to represent the amount of pentanol ascribed to
the interfacial region, acting as a cosurfactant; the remaining
part is therefore diluted in the continuous phase. As we show
further in the paper, with this assumption results of micellar
volume using the macroscopic scattering cross section are
consistent with the micellar volume calculated using the
Guinier approximation for ellipsoidal micelles. In Fig. 6, the
curves with d r 1 are overlapped and require higher ratios o ,
T
to achieve similar initial hydrogenation rates to the curves with
d 4 1, which also are overlapped. By adding a smaller amount
of surfactant and controlling the water percentage similar
initial hydrogenation rates can be achieved. The two groups
of overlapped curves show two different patterns, which could
be explained by the size and the geometry of the micelles. The
extrapolation of the four curves reaches the same point of zero
Fig. 6 Initial hydrogenation rate of the catalytic hydrogenation of
DMI (2 g in 100 ml of microemulsion) as a function of o in Triton
X-100 microemulsions with different d, at 25 1C and 1.1 bar, using
0 mg of [Rh(cod)Cl] and 357 mg of TPPTS.
T
2
2
ꢁ1 ꢁ1
2 0
initial hydrogenation rate ([dV(H ) dt ] = 0 mLH2 min ) at
a o of 1.5. This result was experimentally confirmed by
T
hydrogenations using the four systems with different d, which
exhibit zero activity.
3
.3 Igepal CA-520 micelles—structural characterization
DLS measurements of the Igepal CA-520 microemulsions of
compositions reported in Table 1 could be performed success-
fully for g down to 3.36%. With g = 3.30% results were
poorly reproducible, and as one observes in Table 3
(
ME 18–21), the measured radii of the micelles fluctuate
without any significant pattern. This behaviour could be
explained by the closeness to the cloud point of the system
and for this reason they are not taken into account in the
following discussion. For higher values of g the micro-
emulsions are relatively monodisperse. A linear dependence
of the initial hydrogenation rate to the hydrodynamic radius
of the micelles was obtained, which can be represented by the
following equation of a line:
Scheme 4 Ellipsoidal micelle model.
and SC of the ESI.z As we observe in Table 2, the percentage
of pentanol in the water phase increases when PEG is added
and when the ethoxy chain length is increased. This effect of
PEG on the partition of pentanol in cyclohexane–water
biphasic systems is a simple representation of the effect of
the surfactant in our system. The tendency of the polar part of
ꢀ
ꢁ
dVðH2Þ
dt
¼
1:07 þ 24:27r
ð8Þ
0
Table 2 Cosurfactant partitioning: molar percentage (%) of 1-pentanol concentrated in the aqueous phase of a biphasic system PEG400–
cyclohexane–water at 25 1C, using the compositions from Table 1 for Triton X-100 microemulsions with PEG 400 instead of Triton X-100
a
a
a
Sample
C
5
H
12
O
without PEG (%)
C
5
H
12
O
PEG-400 (%)
5 12
C H OPEG-1000 (%)
ME 1
ME 2
ME 3
ME 4
ME 5
ME 6
ME 7
ME 8
ME 9
ME 10
ME 11
ME 12
11.67
9.49
14.62
12.65
13.89
16.43
15.00
16.65
18.75
19.63
19.77
19.42
19.04
20.27
20.56
22.85
28.25
27.34
25.08
27.79
21.42
21.10
29.09
24.94
21.76
30.01
10.65
13.93
12.11
13.39
14.87
17.09
15.25
18.45
18.40
16.49
a
Concentration of pentanol in the aqueous phase.
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1726–1735 | 1731

View Article Online
Table 3 Radii of the Triton X-100 micelles, volume of the micelles
Sph) calculated with the Guinier approximation for homogeneous
spheres and (VEllip) for homogeneous ellipsoids using SANS scattering
approximation, which generally gives estimates which are
accurate to within 5–10% if the analysis is performed using
(
V
data collected within the range where qR E 1, and provided
g
measurements of the Triton X-100 microemulsions, absolute intensity
that the particles studied are still so dilute that their interaction
1
can be neglected.
0
(I ), volume of the Triton X-100 micelles (VI0) calculated with the
9
coherent macroscopic scattering cross section assuming all the
cosurfactant is concentrated in the continuous phase, and radii of
the Igepal CA-520 micelles estimated from DLS measurements of the
Igepal CA-520 microemulsions at 25 1C
3
.4.1 Scattering curves. Fig. 7 presents SANS scattering
curves for four Triton microemulsions with o = 20 and
different amounts d of pentanol contained. All the curves
show the existence of non-interactive particle systems and by
comparing them one could suggest that the change of
maximum height that increases with decreasing d points to
the increasing micellar size. The points of the scattering
curves within the marked q range were used for Guinier
approximation.
3
3
ꢁ1
3
Sample
rRg/nm
V
Sph/nm
VEllip/nm
I
0
/cm
VI0/nm
ME 1
ME 2
ME 3
ME 4
14.4
7.0
4.3
9.2
13.0
7.1
12588.94
1429.35
332.39
3216.55
—
2670.19
947.79
323.63
1437.86
—
138.61
37.13
18.42
55.10
—
2046.17
461.91
191.24
952.51
—
a
ME 4
ME 4
b
—
—
—
—
ME 5
ME 6
ME 7
ME 8
ME 9
ME 10
ME 11
ME 12
6.5
4.2
8.2
6.5
3.5
7.2
5.7
4.5
23.5
19.0
16.3
14.3
12.0
18.9
35.5
43.0
38.9
1152.32
306.02
2332.92
1125.36
183.01
1575.70
754.56
383.53
—
—
—
—
—
817.48
289.13
1035.23
767.15
170.00
752.19
588.35
282.86
—
—
—
—
—
28.97
15.15
43.25
25.23
9.45
34.60
21.76
11.38
—
—
—
—
—
412.88
177.14
884.29
411.91
124.41
823.00
408.25
168.28
—
—
—
—
—
3
.4.2 Guinier approximation. Linear regressions on
2
Guinier plots (ln(I) vs. q ) of the SANS data give access to
the radius of the micelle, as shown in Fig. 8. The radius of
gyration of the micelle was extracted from eqn (2) and it is
pffiffiffiffiffiffi
equal to 3m, where m is the slope of the curve. The radius of
the corresponding homogeneous sphere was obtained by using
eqn (3). Fig. 9 summarizes the resulting radii for all the Triton
X-100 microemulsions shown in Table 1. In this case the
continuous and the internal phase were deuterated, so
the sizes determined are the overall radii of gyration of the
micelles. The four systems show tendencies which surpass the
initial hydrogenation rates obtained with the biphasic system
c
ME 13
ME 14
ME 15c
c
c
ME 16
ME 17
c
c
ME 18
ME 19
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
c
c
ME 20
ME 21
c
ꢁ1 6
a
ꢁ1Rh
(4 mLH2 min ). When comparing the four systems with the
same o, the gyration radii of the micelles with different d have
similar tendencies. Only one system (ME 1 with o = 20,
d = 0.75) stands out of the standard behaviour, with smaller
amount of pentanol but larger amount of deuterium oxide.
The comparison of the scattering curves for the four systems
with o = 20, displayed in Fig. 7, gives a hint. By comparing
the scattering curves of the four systems with o = 20, it is
possible to identify the difference of this system, which shows a
ME 4 with D
b
2
O and cyclohexane-d12, 0.8 mmol L
c
using a P/Rh
2
ratio of 7.5. ME 4 with D O. Radii were measured using DLS.
ꢁ
1
where [dV(H ) dt ] is the initial hydrogenation rate in
2 0
ꢁ1
mL min and r is the hydrodynamic radius in nm.
As observed in Table 3, the obtained radii are significantly
larger than the length of the fully extended Igepal CA-520
molecule containing 14 C–C bonds, 11 C–O bonds, a benzene
ring, and an O–H bond. The radii are larger than 12 nm, hence
this could mean that a water pool is formed. Another possible
explanation is that, because DLS measurements have no
means of determining geometrical deformations and it only
allows for the determination of an effective hydrodynamic
radius, the micelles could be elongated. By doubling the radius
from 12 to 24 nm the initial hydrogenation rate of DMI almost
32,33
distinguishable characteristic curve for ellipsoidal micelles,
in which a smoother decrease of the intensity at mid q is
observed. This could be caused by the penetration of solvent
(
cyclohexane or pentanol), originating a Gaussian distribution
ꢁ
1
doubles as well: from 2 to 3.5 mLH2 min . Investigations of
AOT microemulsions have shown that the properties of the
water are modified as the water content in the microemulsion
varies, reaching properties similar to bulk water for high
3
0,31
values of o in the microemulsions.
the initial hydrogenation rate of DMI in a biphasic system
We found earlier that
cyclohexane–water, in which the water properties are those of
ꢁ
1 6
bulk water, is approximately 4 mLH2 min
.
Plots of
the autocorrelation functions g (t) are available as ESIz
Fig. SA and SB).
2
(
3
.4 Triton X-100 micelles—structural characterization
As for the Igepal CA-520 systems, our first aim for the Triton
X-100 systems was to find out if the size of the micelles could
be easily correlated to the initial hydrogenation rate of DMI.
The SANS measurements were analysed using the Guinier
Fig. 7 SANS spectra of Triton X-100 microemulsions with o = 20
and different d, and influence of the water-soluble catalyst complex
Rh-TPPTS on the spectra.
1
732 | New J. Chem., 2009, 33, 1726–1735
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009

View Article Online
Fig. 10a, the volume of the spherical micelles is similar to
V
I
I0 when o is smaller. We also calculated the ratio VEllip/V 0.
The volume of the ellipsoidal micelles (VEllip) was calculated
using the following equations for Guinier approximations:
2
1
5
2
2
4
2
3
Rg
¼ ð2a þ b Þ VEllip ¼ a b
ð9Þ
Ellip
where a and b are the short and long cross sections of the
ellipsoidal micelle.
Guinier approximations are generally used at low q to
obtain an estimation of the radius of the particles. Generally,
the decay of the intensity at mid q for spherical particles is
steeper than the one for ellipsoidal particles. This is because at
mid q the shorter cross section of the ellipsoidal particles can
be looked at. For this reason, a Guinier approximation at mid
q for ellipsoidal particles allows for an estimation of the
smaller radius and it was used to determine a. Using eqn (9),
b and VEllip were obtained.
2
Fig. 8 Guinier plot of ln(I) vs. q for the Triton X-100 microemulsion
constituted by 4.17 g of water, 62.63 g of cyclohexane, 7.43 g of
Triton X-100 and 7.43 g of 1-pentanol (ME 4 in Table 1), and
corresponding fit.
As seen in Fig. 10b, complemented with Table SF of the
ESI,z the volume of the ellipsoidal micelles is similar to V
I0
when o is higher so both diagrams in Fig. 10 corroborate the
tendency to form ellipsoids with higher amounts of water. The
ellipticity was also calculated; it is the ratio between b and a.
Fig. 11 shows a very small influence of the cosurfactant on the
ellipticity of the micelles. A large ellipticity is observed with
small amount of pentanol (d = 0.75) and large amount
of water (o = 20), which reinforces the idea of solvent
penetration.
3
.4.3 Interface. As shown in Table 3, the radius obtained
for the microemulsion ME 4 with deuterated core (deuterium
oxide) calculated using the slope of the Guinier approximation
for this sample is 7.1 nm. By comparing it to the 9.2 nm radius
obtained for the same microemulsion ME 4 with deuterated
core and cyclohexane-d12 as continuous phase, an acceptable
7
interfacial surfactant layer of 2.1 nm thickness is estimated.
Fig. 9 Initial hydrogenation rate of the catalytic hydrogenation of
DMI (2 g in 100 ml of microemulsion) as a function of the micelle
radius of the homogeneous spheres (determined by Guinier approxi-
mation of the SANS spectra) in Triton X-100 microemulsions with
Dennis et al. calculated the fully extended length of the
octylphenyl group using Corey–Pauling–Koltung models.
They obtained a length smaller than 1 nm, so they assumed
it to be 1 nm. They also calculated the length of the ethoxy
chain, assuming it forms a random coil obtaining a length of
different d, at 25 1C and 1.1 bar, using 20 mg of [Rh(cod)Cl]
57 mg of TPPTS.
2
and
3
1
.6 nm. Adding both parts, results are not so far from ours.
3
4
The differences can be explained by the solvent penetration,
which solvates the aliphatic chain of the surfactant. Because
the interface and the continuous phase are diffuse, a defined
boundary between interface and continuous phase is difficult
to detect.
around the mean value of the radius. This penetration of
solvent is represented in Scheme 4 by molecules of cyclohexane
in the interface of the micelle.
In order to obtain a preliminary study of the shape of the
micelles which could serve as a first step for a deeper analysis
of the micelles using specific models, the ratio VSph/VI0 was
calculated. The volume of the spherical micelles (VSph
)
was calculated with the Guinier approximation and VI0 was
estimated with the coherent macroscopic scattering cross
section (see Tables SD and SE in ESIz). The amount of
pentanol inside the micelles increases the contrast and the
volume fraction, which decreases the calculated volume of the
micelles. For this reason and following the simplifications
discussed from the resulting pentanol partitioning study, we
assume no pentanol to be introduced inside the core of the
micelles. With this assumption the two volumes are more
similar. The results are shown in Table 3. As shown in
Fig. 10 Volume ratios of the Triton X-100 micelles with different d,
calculated with Guinier approximation as a function of o assuming
the micelles are: (a) spheres and (b) ellipsoids.
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1726–1735 | 1733

View Article Online
emulsion ME 4, the SANS spectrum shows a different size,
and similar to the microemulsion ME 1, a steeper decay for
ME 4 at low q is observed (but not so pronounced as observed
in Fig. 7 for ME 1). The radius of the catalyst containing
ꢁ
1
micelles (CRh = 0.8 mmol L ) determined by Guinier
approximation of the SANS spectra is 13.0 nm, which repre-
sents almost one and a half times the size of the empty micelles
(
r = 9.2 nm). This non-proportionate growth of the micelles is
an indicator of the deformation of the micelles. The slight
elongation of the micelles could be a result of the competition
between the polar ethoxylated groups of the Triton X-100
molecules and the water-soluble catalyst complex Rh–TPPTS
for the water molecules. Other important effects could be: the
incorporation of the amphiphilic TPPTS into the interface or
the interaction between the trivalent anion TPPTS and the
ethoxy head groups. This effect should be stronger when d is
smaller: the 1-pentanol molecules distribute between the inter-
facial layers where they substitute Triton molecules and the oil
rich bulk phase, thereby making the amphiphilic layers effec-
tively more lipophilic, and in turn less interactive with the
Fig. 11 Ellipticity of the Triton X-100 micelles as a function of d.
3
.4.4 Conductivity. The difference in geometry was also
identified with the conductivity. The conductivity of a micro-
3
emulsion is influenced by the microstructures of the micelles.
5
3
8,39
Oil-in-water microemulsions have similar conductivities to
bulk water, and in water-in-oil microemulsions, the conductive
polar phase is isolated by a continuous oil phase, therefore
leading to small conductivities. Bicontinuous microemulsions
show high conductivities, which decrease when the water
water molecules.
This behaviour could be the reason for
the different tendencies shown in Fig. 5. A detailed study of the
effect of the water-soluble catalyst complex on the Triton
X-100 reverse micelles is a topic currently under research in
our laboratory.
3
6
content is lowered. Water-in-oil microemulsions with small
water contents show conductivities which decrease with the
3. Conclusions
3
decrease in the surfactant concentration. Independently of
7
what the SANS measurements show, the presence of a maxi-
In order to improve the understanding of the mutual influence
of the micelle microstructure on the catalytic hydrogenation of
DMI when using microemulsions as a reaction medium, the
structural dimensions of two octylphenyl-based micro-
emulsions were studied and correlated with the catalytic
hydrogenation of DMI performed in such media using the
water-soluble catalyst complex Rh–TPPTS. The initial hydro-
genation rate of DMI is proportional to the Rh–TPPTS
concentrated Igepal CA-520 micelle radius, which was
measured with DLS and also proportional to catalyst
concentrated Triton X-100 micelle radius, determined from
SANS spectra using Guinier approximations. The cosurfac-
tant 1-pentanol plays an important role in the micelle size and
shape of the Triton-based microemulsions, which at the
same time influences the initial hydrogenation rate. The
water-soluble catalyst complex Rh–TPPTS showed a complex
influence on the shape and size of the Triton X-100 micelles
and induces an elongation of the aggregates.
ꢁ1
mum conductivity peak of 0.9 mS cm in Fig. 12, when
keeping o = 20 and changing the pentanol to Triton X-100
ratio (d), identifies the existence of a different geometry of the
microstructure, which could be reaching out, interacting with
each other and creating some continuity. The maximum peak
is obtained at d = 0.70. This is in agreement with the SANS
data (Fig. 7) which indicate the presence of elongated micelles
at d = 0.75, and elongated reverse micelles are expected to
exhibit a higher conductivity compared to spherical micelles.
3
.4.5 Effect of the catalyst incorporation. When adding the
water-soluble catalyst complex (Rh–TPPTS) to the micro-
Acknowledgements
This work is part of the Cluster of Excellence "Unifying
Concepts in Catalysis". Financial support by the Deutsche
Forschungs-gemeinschaft (DFG) within the framework of the
German Initiative for Excellence is gratefully acknowledged.
The experiments at BENSC in Berlin were supported by the
European Commission under the 6th Framework Programme
through the Key Action: Strengthening the European
Research Area, Research Infrastructures, Contract No.
RII3-CT-2003-505925 (NMI3). A special thanks to Verena
Strempel for all the hard work invested in the experiments.
Fig. 12 Conductivity of 10% Triton X-100 microemulsions as a
function of d keeping o = 20 at 25 1C.
1
734 | New J. Chem., 2009, 33, 1726–1735
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009

View Article Online
J.M. is also grateful to the Deutscher Akademischer
Austauschdienst (DAAD) and to the Fundacion Gran
18 B. Abe
´
cassis, F. Testard, L. Arleth, S. Hansen, I. Grillo and
T. Zemb, Langmuir, 2006, 22, 8017–8028.
´
1
2
2
2
9 K. S. Freeman, N. C. B. Tan, S. F. Trevino, S. Kline,
L. B. McGown and D. J. Kiserow, Langmuir, 2001, 17, 3912–3916.
0 J. Oberdisse, O. Regev and G. Porte, J. Phys. Chem. B, 1998, 102,
Mariscal de Ayacucho (Fundayacucho) for receiving a
doctoral scholarship.
1
102–1108.
1 G. Verma, V. K. Aswal, S. K. Kulshreshtha, P. A. Hassan and
E. W. Kaler, Langmuir, 2008, 24, 683–687.
2 E. Caponetti, A. Lizzio, R. Triolo, W. L. Griffith and
J. S. Johnson, Langmuir, 1992, 8, 1554–1562.
References
1
G. Oehme, in Aqueous-Phase Organometallic Catalysis,
ed. B. Cornils and W. A. Herrmann, Wiley-VCH, Weinheim,
23 L. Garcia-Rio and P. Hervella, Chem.–Eur. J., 2006, 12,
8284–8295.
24 S. W. Provencher, Comput. Phys. Commun., 1982, 27, 213–227.
25 B. Hammouda, Probing Nanoscale Structures—The Sans Toolbox,
http://www.ncnr.nist.gov/staff/hammouda/the_SANS_toolbox.pdf
(November 2008).
2
004, pp. 256–271.
D. Langevin, Annu. Rev. Phys. Chem., 1992, 43, 341–369.
J. Sjoblom and S. E. Friberg, in Handbook of Microemulsion
2
3
¨
Science and Technology, ed. P. Kumar and K. L. Mittal, Marcel
Dekker, Inc, New York, 1999, pp. 833–842.
4
5
6
7
8
9
M. J. Schwuger, K. Stickdorn and R. Schoma
995, 95, 849–864.
T. Dwars, E. Paetzold and G. Oehme, Angew. Chem., Int. Ed.,
005, 44, 7174–7199.
J. S. Milano-Brusco, M. Schwarze, M. Djennad, H. Nowothnick
and R. Schomacker, Ind. Eng. Chem. Res., 2008, 47, 7586–7592.
R. J. Robson and E. A. Dennis, J. Phys. Chem., 1977, 81,
075–1078.
S. M. Andrade and S. M. B. Costa, Photochem. Photobiol. Sci.,
002, 1, 500–506.
K. Wormuth, O. Lade, M. Lade and R. Schoma
¨
cker, Chem. Rev.,
26 M. Kahlweit and R. Strey, Angew. Chem., Int. Ed. Engl., 1985, 24,
654–668.
27 K. V. Schubert and E. W. Kaler, Ber. Bunsen-Ges. Phys. Chem.,
1996, 100, 190–205.
28 M. Lade, H. Mays, J. Schmidt, R. Willumeit and R. Schomacker,
Colloids Surf., A, 2000, 163, 3–15.
29 S. P. Moulik and B. K. Paul, Adv. Colloid Interface Sci., 1998, 78,
99–195.
30 L. Garcia-Rio, J. C. Mejuto and M. Perez-Lorenzo, New J. Chem.,
2004, 28, 988–995.
31 T. K. Jain, M. Varshney and A. Maitra, J. Phys. Chem., 1989, 93,
7409–7416.
32 O. Glatter, in Neutrons, X-rays and Light: Scattering Methods
Applied to Soft Matter, ed. P. Lindner and T. Zemb, Elsevier
Science B.V., Amsterdam, 2002, pp. 73–103.
33 C. Moitzi, N. Freiberger and O. Glatter, J. Phys. Chem. B, 2005,
109, 16161–16168.
34 R. Strey, Colloid Polym. Sci., 1994, 272, 1005–1019.
35 M. Jonstromer, B. Jonsson and B. Lindman, J. Phys. Chem., 1991,
¨ ¨
95, 3293–3300.
36 C. Cabaleiro-Lago, L. Garcia-Rio and P. Hervella, Langmuir,
2007, 23, 9586–9595.
37 H. F. Eicke and J. C. Shepherd, Helv. Chim. Acta, 1974, 57,
1951–1963.
38 M. Kahlweit, J. Phys. Chem., 1995, 99, 1281–1284.
39 Z. H. Luo, X. L. Zhan and P. Y. Yu, Chin. Chem. Lett., 2004, 15,
1101–1104.
1
2
¨
1
2
¨
cker, in Handbook
of Applied Surface and Colloid Chemistry, ed. K. Holmberg, John
Wiley & Sons, West Sussex, 2001, pp. 605–627.
0 D. M. Zhu, K. I. Feng and Z. A. Schelly, J. Phys. Chem., 1992, 96,
1
2382–2385.
1
1
1 D. M. Zhu, X. Wu and Z. A. Schelly, Langmuir, 1992, 8, 1538–1540.
2 M. Dvolaitzky, M. Guyot, M. Lagues, J. P. Lepesant, R. Ober,
C. Sauterey and C. Taupin, J. Chem. Phys., 1978, 69, 3279–3288.
3 S. H. Chen, Annu. Rev. Phys. Chem., 1986, 37, 351–399.
4 S. H. Chen, J. Rouch, F. Sciortino and P. Tartaglia, J. Phys.:
Condens. Matter, 1994, 6, 10855–10883.
5 M. Gradzielski, D. Langevin and B. Farago, Phys. Rev. E: Stat.
Phys., Plasmas, Fluids, Relat. Interdiscip. Top., 1996, 53, 3900–3919.
6 M. Gradzielski, D. Langevin and B. Farago, Prog. Colloid Polym.
Sci., 1996, 100, 162–169.
1
1
1
1
1
7 M. Gradzielski, D. Langevin, T. Sottmann and R. Strey, J. Chem.
Phys., 1997, 106, 8232–8238.
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1726–1735 | 1735