Structural characterization of micellar aggregates in sodium dodecyl sulfate/aluminum nitrate/urea/water system in the synthesis of mesoporous alumina

Article abstract of DOI:10.1021/jp049482h

The mechanism of templated synthesis of ordered mesoporous alumina using sodium dodecyl sulfate (SDS) as template was investigated by NMR self-diffusion and spin and fluorescence probe techniques. In the precursor solutions, urea triggered a considerable shortening of the long entangled cylindrical aggregates characteristic of the SDS micellar solutions with aluminum salt in large excess. The formation of mesostructured hexagonal alumina was followed using spin probes incorporated in the SDS aggregates. No structural changes were detected during urea hydrolysis, before the occurrence of precipitation. In the precipitate, the surfactant template, its interaction with the aluminum species, and the time evolution of this interaction were observed. Hexagonally structured alumina formed over a broad range of compositions of the starting solution, with the dimension of the unit cell being nearly constant.

Full text of DOI:10.1021/jp049482h

J. Phys. Chem. B 2004, 108, 7735-7743
7735
Structural Characterization of Micellar Aggregates in Sodium Dodecyl Sulfate/Aluminum
Nitrate/Urea/Water System in the Synthesis of Mesoporous Alumina
Agneta Caragheorgheopol,* Horia Caldararu, and Marilena Vasilescu
Romanian Academy, Institute of Physical Chemistry “I. G. Murgulescu”, Splaiul Independentei 202,
0
60021 Bucharest, Romania
Ali Khan and Daniel Angelescu†
UniVersity of Lund, DiVision of Physical Chemistry 1, Center for Chemistry and Chemical Engineering,
2
21 00 Lund, Sweden
Nad eˇ zˇ da Zˇ ilkov a´ and Ji rˇ ´ı Cˇ ejka
J. HeyroVsk y´ Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, Dolej sˇ koVa 3,
1
82 23 Prague, Czech Republic
ReceiVed: February 4, 2004; In Final Form: April 5, 2004
The mechanism of templated synthesis of ordered mesoporous alumina according to the recipe proposed by
Yada et al. (Yada, M.; Machida, M.; Kijima, T. Chem. Commun. 1996, 769) using sodium dodecyl sulfate
(SDS) as template, has been investigated by NMR self-diffusion and spin and fluorescence probe techniques.
In the precursor solutions urea was found to produce a substantial shortening of the long entangled cylindrical
aggregates characteristic of the SDS micellar solutions with aluminum salt in large excess. The formation of
mesostructured hexagonal alumina was followed using spin probes incorporated in the SDS aggregates. No
structural changes were observed during urea hydrolysis, before the onset of the precipitation. In the precipitate,
the presence of the surfactant template is demonstrated as well as its interaction with the aluminum species.
Time evolution of this interaction is also observed. Finally, the influence of the composition of the precursor
system on the structure and thermal stability of as-synthesized alumina has been studied. It appears that
hexagonally structured alumina is formed over a large range of compositions of the starting solution, with the
dimension of the unit cell being almost constant. A model of precipitation is proposed that is based on the
analogy with the formation of the hexagonal crystalline Al-DS precipitate in the ternary SDS/Al(NO
system.
3 3
) /water
Introduction
structure-directing agent. The mechanisms of formation of
ordered mesoporous materials were mostly investigated for
silica-based materials. In the case of an electrostatic route,
the proposed mechanism involves a cooperative organization
of the complex between the organic self-assemblies and the
Since the discovery of mesoporous materials of the M41S
14
1
family, possessing large internal surface areas and narrow pore
size distributions, increasing attention has been paid to meso-
porous materials of various chemical compositions owing to
their potential use as catalysts, molecular sieves, and host
inorganic species, driven by electrostatic interactions. Other
15,16
results
suggested that silicate species polymerize in the bulk
2
-5
materials.
phase and that the growing polymers bind free surfactant ions
until precipitation of the organized polymer/surfactant complex
takes place. In this case, the micelles act as reservoirs of
surfactant.
Such mesoporous molecular sieves can be obtained by the
calcination of their mesostructured precursors synthesized in
the presence of self-assembling amphiphilic compounds. Me-
soporous alumina with disordered, but thermally stable, meso-
structures, has been synthesized mainly in nonaqueous media
starting from aluminum isopropoxide.6-8 Inexpensive aluminum
Sicard et al.17 have studied the formation mechanism of
mesostructured hexagonal alumina in SDS micellar solutions
at 60 °C by using a fluorescence probing method. The authors
found that in the precursor system the micelles are slightly
salts such as aluminum nitrate or chloride were found to lead
to ordered mesostructured alumina.9
-13
Using a urea-based
3+
elongated, and the Al counterions bind strongly to the micelle
surface. A significant micellar growth was, however, not noticed
during urea hydrolysis in the clear isotropic solution, and the
organization of organic/inorganic complex into a mesostructured
hexagonal material is assumed to occur only during, or
immediately before the onset of precipitation.
homogeneous precipitation method the synthesis of hexagonally
structured mesoporous alumina was reported, occurring in an
aqueous sodium dodecyl sulfate (SDS) micellar solution.
However, the sulfate headgroups appear to be embedded in the
inorganic walls, resulting in their collapse after removal of the
In our previous papers18,19 we have undertaken an in-depth
research of the phase behavior of the SDS/Al(NO3)3/H2O system
to understand the conditions under which an ordered hexagonal
*
Corresponding author. E-mail acaraghe@chimfiz.icf.ro.
Permanent address: Institute of Physical Chemistry “I. G. Murgulescu”,
†
Splaiul Independentei 202, 060021 Bucharest, Romania.
1
0.1021/jp049482h CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/12/2004

7736 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Caragheorgheopol et al.
Figure 1. Phase diagram (lower-left corner) of the SDS/Al(NO
3
)
3
/H
2
O ternary system. The compositions corresponding to the synthesis mixtures
3+
studied are marked with the same numbers as in Table 6. Region I denotes the isotropic micellar solution at low Al concentration, II represents
the precipitation region, and III is the micellar phase at high Al3 concentration, the “redissolution” area.
+
CHART 1: Chemical Structure of the Spin and
Fluorescence Probes
alumina structure could be formed. Thus, it is well-known that
dilute aqueous solutions of SDS contain small, spherical
a
micelles. Addition of small amounts of trivalent counterions,
3+
Al (Figure 1, region I) to such a solution does not lead to a
3+
significant micelle growth; at a higher Al concentration,
however (Figure 1, region II), phase separation occurs. One of
the phases is a white precipitate consisting of a complex of
-
3+
dodecyl sulfate (DS ) and Al with a hexagonal structure.
Subsequent addition of the salt results in the complete dissolu-
tion of the precipitate to yield a new micellar phase (L1), which
is highly viscous and contains long, wormlike aggregates (Figure
2
0
1
, region III). Monte Carlo modeling was performed to
understand the origins of the two-phase region in the middle of
an isotropic phase. Simulation of the interactions between two
a
DPCl is the quencher used in the dynamic fluorescence method.
-
spherical DS micelles in the presence of Al(NO3)3 has shown
that the correlation effect between Al3 counterions strongly
adsorbed on the micelle surface results in a purely attractive
force that leads to the destabilization of the isotropic micellar
phase to form the precipitate. These results are important for
understanding the alumina precipitation mechanism, because the
+
SIGMA), and a cationic probe, 4-[N,N-dimethyl-N-(methylene)16]-
ammonium-2,2,6,6-tetramethylpiperidin-1-oxyl iodide (CAT 16,
Molecular Probes). A neutral probe, TEMPO-laurate (C12-NO),
prepared as described was also used. The chemical structures
of these probes are presented in Chart 1.
2
1
composition of the precursor system (without urea) used by
Sample Preparation for the Reference Systems. Samples
for the structural characterization of the SDS/Al(NO3)3/urea/
water system were prepared by weighing the appropriate amount
of each component. The specific spin probes were then added
so as to yield a concentration of (1-2) × 10 M in the final
solution. The procedure used was to evaporate the ethanol
solution containing the probe on the walls of a vial under
nitrogen gas and then to add the examined solution and let the
spin probe dissolve in it. The addition of the fluorescence probe,
Py, and the quencher, DPCl, was realized as described in ref
19.
Yada1
area.
1,13
for alumina synthesis is placed in the redissolution
In the present study we have tried to get a deeper insight
into the formation mechanism of hexagonal alumina by using
magnetic resonance methods. The report is organized as
follows: (i) First, the effect of urea on the micellar aggregation
in the redissolution area of the pseudo-ternary SDS/Al(NO3)3/
H2O system is described. These solutions represent the precur-
sors for alumina synthesis. A number of different spectroscopic
-
4
1
methods (EPR, fluorescence, FT NMR self-diffusion, H NMR)
Synthesis of Mesostructured Alumina. The preparation of
mesostructured alumina is based on the original recipe described
have been used to characterize the surfactant aggregates in this
region and the effect of increasing urea content. (ii) Second,
the changes of these aggregates during the synthesis of meso-
structured alumina are followed using EPR spectra of spin
probes. (iii) Finally, the structural characteristics of mesoporous
alumina precipitated from precursors with various molar com-
positions and their thermal stability are examined using XRD.
1
0,11
by Yada et al,
but various compositions in the starting
mixtures have been used. After the complete dissolution of the
components, the samples were heated at 80 °C in well-closed
vials until the final pH in the range 7.0-7.3 was reached.
Thereafter, the aging was realized by keeping the samples at
5
0 °C for 3 days. Finally, the samples were filtered, washed
with water, and dried at room temperature.
Experimental Section
For two starting mixtures (Y1 and Y2) with the molar ratios
SDS/Al(NO ) /urea/water ) 2/1/30/69 and 0.42/1/21/69, re-
Materials. Sodium dodecyl sulfate (SDS, BDH), aluminum
nitrate nonahydrate (Al(NO3)3‚9H2O, Acros Organics), and urea
3
3
spectively, the changes in the surfactant aggregates during the
synthesis were followed by measuring the ESR spectra of
incorporated spin probes at different stages of the reaction. The
first mixture is the precursor system of the Yada synthesis. After
the precursor was prepared, under stirring, the spin probe was
added as described above, and the ESR measurements were
performed in the following stages: (i) in the initial mixture of
(Sigma) were used without further purification. The deuterated
water (>99.8%) was purchased from Dr. Glaser AG (Basel).
The fluorescence probe, pyrene (Py, Molecular Probes), and
the quencher, 1-decylpyridinium chloride (DPCl, Aldrich),
were recrystallized from ethanol. The spin probes used were
from the series of x-doxyl stearic acids (5- and 16-DSA,

Synthesis of Mesoporous Alumina on SDS
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7737
reagents, after the dissolution of the components; (ii) in solution,
at different times before the onset of precipitation; (iii) for a
sample of the precipitate, the reaction being stopped after 6 h,
after reaching pH 6.1; (iv) for a sample of the precipitate, the
reaction being stopped after about 20 h, with pH 7.3; and finally
(v) for a sample of the precipitate, after 20 h heating at 80 °C
and further 3 days of aging at 50 °C.
Calcination Procedure. The calcination was performed in
-
1
air, at a heating rate of 0.3 C min . The final temperature
selected was either 350 °C, where only the alkyl chain is
removed, or 550 °C, where both parts of the surfactant, alkyl
chain and sulfate group, are eliminated.12 The final temperatures
were maintained for 3 h before cooling the samples to room
temperature.
Ion Exchange. Attempts were made to remove the surfactant
from as-synthesized solids by anionic exchange. A 0.1 g sample
of the solid (as-synthesized alumina) was stirred with 7 mL of
a 50 mM CH3CO2Na/EtOH solution at room temperature for 1
h. The product was collected by filtration and washed with
ethanol and hot water.
EPR Measurements. The EPR spectra were recorded on a
RE1X JEOL spectrometer with 100 kHz field modulation using
X-band frequency. An approximate value of the rotational
2
2
correlation time, τC, was calculated according to the formula
Figure 2. ESR spectra of 5-DSA in SDS/Al(NO ) /water/urea mixtures
3
3
with increasing urea content. Samples are denominated as in Table 1,
where the corresponding compositions are given: (a) S; (b)SA; (c)
SAU2; (d) SAU 4; (e) SAU6.
-
10
1/2
τC ) (6.51 × 10 )∆H(0){[h(0)/h(-1)]
+
1/2
[
h(0)/h(1)] - 2} s (1)
the experimental setup being described in detail earlier.19 The
wavelengths for excitation and emission were λex ) 325 nm
and λem ) 394 nm, respectively.
where ∆H(0) is the line width (in Gauss) of the central line,
and h(-1), h(0), and h(1) are the peak-to-peak heights of the
M ) -1, 0 and +1 derivative lines, respectively. τC is connected
to the local viscosity, η, by the Debye-Stokes-Einstein
equation:
NMR Self-Diffusion Measurements. The experiments were
carried out on a Bruker DMX200 spectrometer, operating at
1
the H resonance frequency of 200 MHz. The self-diffusion
coefficients of the surfactant, water and urea were measured
using the Fourier transform pulsed field gradient stimulated
spin-echo (FT-SPGSE) technique. For all the samples, the
echo intensities decayed monoexponentially.
3
τ ) 4πηR /3kT
(2)
C
2
5
where R is the hydrodynamic radius of the tumbling entity. The
τC values obtained were used to follow, in a qualitative manner,
significant changes in the microenvironment of the probes. The
Results
23
order parameter, S, is defined as S ) (A| - A⊥ )/[Azz - (Axx +
Ayy)/2], where Azz, Axx, and Ayy are the principal elements of the
A tensor in the absence of molecular motion and A| and A⊥ are
derived from experimental spectra (see insert in Figure 2). The
order parameters S were calculated using the following param-
eters reported for doxyl probes:24 Azz ) 33.5 G, Axx ) 6.3 G,
and Ayy ) 5.8 G. To correct for the polarity difference between
the studied sample and the sample whose Azz, Axx, and Ayy values
are used, the S value calculated above is multiplied with the
inverse ratio of the corresponding isotropic nitrogen hfs values,
aN, calculated as the arithmetical means of the tensor compo-
nents.23
Small-Angle X-ray Diffraction. The scattering experiments
on as-synthesized and calcined materials were performed on a
Kratky compact small-angle system equipped with a position
sensitive detector containing 1024 channels of 51.3 µm width.
Cu KR radiation of wavelength 1.542 Å was provided by a
Seifert ID300 X generator, operating at 50 kV and 40 mA. A
1. Structural Characterization of Precursor SDS/Al(NO3)3/
Water/Urea Micellar Solutions. This part is devoted to the
description of the effect of urea on the SDS aggregates. The
spherical micelles from the isotropic micellar solution of the
2
6
SDS/water binary system and the wormlike aggregates
obtained in the case of the pseudoternary SDS/Al(NO3)3/H2O
1
8
system have been chosen as references.
i. EPR Measurements. The effects of increasing urea content
have been followed on a series of samples starting from a SDS/
water sample (S, Table 1) and from a SDS/Al(NO3)3/water
solution (SA, Table 1), in which the concentrations of the
components were those used in the alumina synthesis reaction
mixture (see below). Three spin probes (5-DSA, 16-DSA, and
C12-NO) were used to report on local viscosity (τc) and order
parameter (S) in different regions of the surfactant aggregates.
The ESR line shapes of amphiphilic spin probes strongly
depend on their dynamics. When dissolved in surfactant
aggregates and aligned with the surfactant molecules, their local
motion is dominated by the rapid rotation around the long
molecular axis, resulting in anisotropic effects. Superimposed
motions are the tumbling of the entire aggregate and the lateral
diffusion of the probe molecule over the aggregate surface.
The rotational correlation time of the aggregate tumbling (see
1
0 µm thick Ni filter was used to remove the Kâ radiation. The
distance between sample and detector was 277 mm, and the
sample to detector volume was kept under vacuum to minimize
the background scattering.
Dynamic Fluorescence Measurements. The fluorescence
decay curves of pyrene in the absence and in the presence of a
quencher were recorded by the single photon counting technique,
-
8
eq 2) is smaller than 5 × 10 s for micelles with R < 20-24
Å at 20 °C in water. Such micelles give rise to motionally

7738 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Caragheorgheopol et al.
TABLE 1: ESR Parameters of Spin Probe Spectra in SDS/Al(NO
3
)
3
/Urea/Water Solutions
composition (mol‚kg-1)
5-DSA
1
6-DSA
C12-NO
10 τ (s)
C
1
0
10
sample
SDS
Al(NO
3
)
3
H
2
O
urea
A
|
(G)
A
⊥
(G)
S
10 τ
C
(s)
S
SA
0.42
0.42
0.42
0.42
0.42
0.42
0.42
0.42
0
1
1
1
1
1
1
1
69
69
69
69
69
69
69
69
0
0
5.0
9.6
11.8
12.0
12.7
13.5
14.7
2.5
4.2
6.2
7.5
8.7
8.8
8.7
10.0
SAU1
SAU2
SAU4
SAU6
SAU8
SAU13
2.3
4.5
9.0
13.5
18.0
30.0
24.9
25.8
25.9
26.3
10.3
10.1
10.4
8.9
0.53
0.57
0.55
0.65
narrowed three-line spectra. In a similar way, lateral diffusion
contributes to averaging of anisotropic features if the transla-
tional diffusion coefficient on the aggregate surface, D > 3 ×
-
11
2
27,28
1
0
m /s.
The ESR spectrum of 5-DSA in sample S containing small,
spherical micelles, consists of three asymmetric, rather broad
lines (Figure 2a), characteristic of averaging of the anisotropic
components of the A and g tensors, but with the probe tumbling
at the limit of rapid toward slow motion. This happens when
the rotational correlation time of the probe, τc, is higher than
-
9
1
0
s. In this case τc cannot be correctly calculated with eq 1.
The same type of ESR spectrum was found in solution SA
Figure 2b), where the SDS aggregates are long, entangled
(
18
cylinders (“wormlike” aggregates ). However, increasing quan-
tities of urea gradually change the spectrum of 5-DSA in the
SAU series (see Table 1), so that starting from sample SAU4
only the local, anisotropic motion determines the line shape
(Figure 2d,e). In these cases an order degree, S, could be
measured. This type of ESR spectrum could arise from three
different structures:27 (1) discoidal micelle (oblate ellipsoid)
large enough so that tumbling is slow on the ESR time scale;
3 3
Figure 3. ESR spectra of C12-NO in SDS/Al(NO ) /water/urea
(2) rodlike micelle (prolate ellipsoid), if the lateral diffusion is
mixtures with increasing urea content; samples are denominated as in
Table 1, where the corresponding compositions are given: (a) SA; (b)
SAU2; (c) SAU4; (d) SAU6.
by some reason too slow; (3) spherical micelle with R > 20 Å.
Unfortunately, none of these possibilities can be completely
discarded. Urea, included in the solvation sphere, may increase
the aggregate hydrodynamic radius to some extent or increase
the lateral diffusion coefficient so as to slow the lateral diffusion
rate below the threshold of ESR time-window.
TABLE 2: Results of H NMR Measurementsa
1
composition (mol‚kg-1
)
11
1
0 Dobs
W
(Hz)
b
(m2‚s-1)
sample SDS Al(NO
3
)
3
H
2
O
urea
φ
agg
p
The neutral spin probe, C12-NO, which is included in SDS
micelles with the nitroxide group close to the micelle interface,
has an isotropic motion in both S and SA samples (Figure 3a
and b). In the SAU series (see Table 1), in a similar way with
the 5-DSA probe, the motion becomes more and more aniso-
tropic (for this probe the anisotropy is reflected in the decreasing
height of the three lines from low to high field ); i.e., only the
local rotation of the probe around the long axis remains rapid,
and there is no contribution from micelle tumbling and lateral
diffusion (Figure 3c,d).
The ESR spectrum of 16-DSA (not shown) is isotropic in all
samples. The high τc value, given in Table 1, reflects the local
viscosity increase within the aggregate core, where the nitroxide
moiety of this probe is supposed to reside, with aluminum nitrate
addition sample (SA), which continues with urea addition, in
the SAU series. It shows that structural modifications in the
aggregates also take place, along with dimensional and geo-
metrical changes.
1
2
3
4
5
0.5
0.5
0.42
0.37
0.44
0
69.0
69.0
0
0
0.0498
0.0498
4.45
0.26
0.28
1.46
2.50
-
13.7
0.56
0.89
0.82
0.97
28.5 50.8
21.0 41.8
13.5 22.0
5.8 16.3
69.0 4.57 0.0717
69.0 8.57 0.0560
69.0 15.62 0.0558
a
D
obs, surfactant self-diffusion coefficient; W, the line width; p, the
29
b
calculated axes ratio. Volume fraction of the aggregates.
diffusion of the micelle itself. Addition of Al3 in large
concentration (Figure 1, region III) to an aqueous SDS solution
results in a very strong decrease of Dobs (Table 2). This reflects
the micelle growth induced by the excess of Al salt, leading to
+
1
8
long, wormlike aggregates. Subsequent addition of urea to
these wormlike micellar solutions produces a significant increase
of Dobs (which is also reflected in the dramatic macroscopic
viscosity decrease). The enhanced mobility of the aggregates
as the system takes up urea, may partly be due to a dilution
effect, but the order of magnitude increase in Dobs can only be
explained by a significant reduction of aggregate sizes. This
fact can be roughly quantified by estimating the axial ratio, p,
of the elongated micelles, following a numerical procedure
described in detail in ref 30. As expected, the calculated p values
(shown in Table 2) decrease as the urea concentration increases.
By this procedure, the micelles were found to be still elongated
even at the highest urea/SDS molar ratio (35:1) as the assump-
ii. NMR Surfactant Self-Diffusion. The experimental self-
diffusion coefficients of the surfactant in different SDS/Al-
(NO3)3/urea/H2O solutions are presented in Table 2. As the
overall concentration of the surfactant is much higher than the
critical micellar concentration (cmc), the contribution of the free
monomer to the observed self-diffusion coefficient can be
neglected, and consequently, the measured diffusion reflects the

Synthesis of Mesoporous Alumina on SDS
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7739
urea is present. The binding of urea is, however, calculated with
high uncertainty due to the low fraction of “bound” urea.
Nevertheless, one can estimate that about one to three urea
molecules are bound to each surfactant headgroup. This finding
is in line with ESR observation of local viscosity increase
with urea addition and with the proposed mechanism for
urea action, suggesting that urea replaces some of the water
molecules in the solvation sphere of the polar headgroup of the
3
3-35
amphiphile.
V. Dynamic Fluorescence Quenching. The quenched fluo-
rescence decay of pyrene in the SDS/Al(NO3)3/urea/water
solutions, was analyzed within two models, depending on how
the mean length of the aggregates, Lmic, relates to the diffusion
length Ldiff. Here
Ldiff
)
πD/k0
(4)
x
Figure 4. 1H NMR spectra of dodecyl sulfate in SDS/Al(NO
) /water/
3 3
where D is the mutual diffusion coefficient of the probe and
quencher and k0 represents the deactivation rate constant of the
excited probe in the absence of quenchers.
urea mixtures with increasing urea content. Samples are numbered as
in Table 2: (a) 2; (b) 3; (c) 4; (d) 5. The spectra were recorded by
pulsed FT technique at 200 MHz.
Model I describes the fluorescence decay for the excited
probe-quencher pair solubilized in aggregates when Lmic .
tion of spherical aggregates resulted in an un-physically large
hydrodynamic radius.
3
6,37
2Ldiff:
1
iii. NMR Relaxation of Surfactant. The H NMR spectra of
ln F(t)/F(0) )
-k t - B c {exp(B t) erf(B t) - 1 + 2(B t/π) } (5)
the surfactant in elongated aggregates present in various SDS/
Al(NO3)3/H2O/urea mixtures are displayed in Figure 4. The
corresponding line widths, defined as the width at the half-height
of the main NMR alkyl chain signal of the surfactant, are
presented in Table 2. Without urea the spectra exhibit a non-
Lorentzian band shape, with the alkyl and terminal methyl NMR
proton bands highly overlapped (Figure 4a). This type of
spectrum represents the typical relaxation observed in wormlike
aggregates.3 As a result of urea addition, the two bands become
narrower and their shape changes toward a Lorentzian shape
1
/2
1/2
0
0
0
1
1
1
The two parameters, B and B , are given by
0
1
2
2
kqR
3D
kqR
1
B0 )
B1 )
(6)
(
)
(
)
3
D
1
where k is the first-order quenching rate constant in the reaction
q
volume of length 4R/3 surrounding the excited probe and c is
0
(see Figure 4d). The large, 50 Hz, line width in the absence of
the quencher concentration expressed as molecules per unit
urea, which corresponds to long cylinder-like aggregates,
decreases gradually with the urea content, reaching a value of
length.
When Lmic < 2Ldiff, the fluorescence decay data can be
analyzed, within the framework of Infelta-Tachiya-type equa-
tion (model II)
16.3 Hz at the highest urea content. This value is still higher
3
8,39
than the one recorded for spherical micelles in the binary
surfactant/water system (∼14 Hz). Although the line width does
not have a direct and simple physical meaning, the observed
line narrowing clearly points to the decrease of the mean
F(t) ) F(0) exp(-A t + A (exp(-A t) - 1))
(7)
(8)
2
3
4
3
1
aggregate length.
where Ai parameters are explicitly
iV. NMR SolVent Self-Diffusion. The observed self-diffusion
coefficients of water and urea are presented in Table 3. The
addition of the aluminum salt to water/urea mixtures leads to a
marked decrease in urea self-diffusion, probably due to forma-
2
2
A
) k
A
) n
A
) 4Γ D/L
1
2
0
3
4
n is the average number of quenchers per micelle, and Γ1
depends on the quenching rate constant, k , and the encounter
radius of excited probe and quencher.
In a recent study of the SDS/Al(NO3)3/water system with
excess Al salt (Figure 1, region III), large wormlike aggregates
have been identified with a number of experimental techniques.18
The fluorescence quenching data carried out in the SDS/Al-
tion of a complex involving urea and Al3 or NO3 . Formation
of such complexes would influence not only the electrostatic
interactions between the charged species but also the presence
of urea molecules near to the polar-apolar interface. The
characterization of these complexes is, however, beyond the aim
of the present paper.
+
-
q
(NO3)3/water system with excess Al salt (Figure 1, region III)
The reduced diffusion coefficients of the solvent, D/D0,
defined as the ratio of the self-diffusion coefficients in the
presence and in the absence of the aggregates, are also shown
in Table 3. The fraction f of solvent molecules bound to the
2
have shown that the statistical parameter ø can be reliably used
as a criterion to discriminate between the two models and
implicitly to evaluate the mean length of the aggregates as
compared with the diffusion length, Ldiff
.
Both models have been applied to fit the fluorescence
quenching curves of pyrene measured in selected SDS/Al/urea/
1
9
32
charged headgroups was estimated from
D ) fDmic + A(1 - f)D0
(3)
-
water mixtures (urea were added in DS wormlike micelles
Here Dmic is the diffusion coefficient of the aggregates and A
represents the obstruction factor. Thus, the hydration number
solutions) (Table 4). In the solution with a moderate quantity
-
1
of urea (2.68 mol kg ), both models give approximately the
2
(i.e., the number of water molecules bound per surfactant) was
same value of ø , implying a decrease in the mean aggregate
found to decrease from 12 in the absence of urea to 7 when
length to 2Ldiff. By further addition of urea, model II gives more

7740 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Caragheorgheopol et al.
TABLE 3: Reduced Self-Diffusion Coefficient of Urea and Water, D/D
composition (mol‚kg-1)
0
, and the Obstruction Factor, A
09Dobs(H
-1
9
D/D
0
1
2
O)
)
10 Dobs(urea)
a
2
2
-1
SDS
Al(NO
3
)
3
H
2
O
urea
φ
agg
(m ‚s
(m ‚s
)
2
H O
urea
A
0
0
0
0
0
0
0
0
0
0
69.0
69.0
69.0
69.0
69.0
69.0
69.0
69.0
6.81
3.62
8.33
14.18
4.60
4.57
1.67
1.77
1.14
0.96
1.19
1.08
1.04
0.86
0.92
1.00
0.21
0.17
0.22
0.18
0.20
0.19
0.79
0.93
0.89
0.89
0.82
0.97
.42
.37
.44
0.072
0.056
0.056
0.91
0.93
0.92
0.67
0.83
0.95
0.955
0.964
0.964
8.57
15.62
a
Volume fraction of the aggregates.
TABLE 4: TRFQ Dataa
composition (molar ratio)
03[SDS]
(M/kg)
3
model I
model II
1
10 [DPCl]
(M/kg)
b
(10 s-1)
8
D (10 m s-1)
b
-10
2
ø2 c
ø2 c
SDS
Al(NO
3
)
3
H
2
O
urea
k
q
N
0
.4
1.4
.8
1.2 ( 0.1
1.2 ( 0.1
1.3 ( 0.1
0.72 ( 0.2
3.9 ( 0.2
4.3 ( 0.2
4.5 ( 0.2
0.64 ( 0.2
0
.25
0.57
0
69.0
183
1.20
3.27
1
0.35
0.41
0.42
1.08
0.89
0.97
2.68
4.56
14.71
69.0
69.0
69.0
211
237
183
0.42
0.87
0.7
1.06
1.30
1.96
1.09
1.10
1.49
220
218
a
Quenching rate constant, k
q
, and mutual diffusion coefficient, D, from the one-dimensional quenching model and the apparent mean aggregation
b
40
c
2
numbers, N, from the Infelta-Tachiya type equation. A hydrophobic radius of the cylinder of 17 Å was considered.
Reduced ø .
TABLE 5: ESR Parameters of the Spin Probes in the Initial Solution (Molar Ratio SDS/Al(NO
during Reaction and in the As-Synthesized Alumina
)
3 3
/Water/Urea ) 0.42/1/69/21),
treatment
1
0
state
temp/time
pH final
spin probe
5-DSA
10 τ
C
(s)
S
2Azz (G)
observations
initial sol
precipitate
precipitate
precipitate
0.60
80 °C/6 h
80 °C/15 h
50 °C/3 days
6.1
7.3
60.0
64.8
6 4.8
initial sol
precipitate
precipitate
precipitate
16-DSA
CAT 16
C12-NO
16.2
80 °C/6 h
80 °C/15 h
50 °C/3 days
6.1
7.3
∼25
outside “fast motion” regime
idem
idem
∼25
∼25
initial sol
precipitate
precipitate
precipitate
80 °C/6 h
80 °C/12 h
50 °C/3 days
6.1
7.3
42.2
50.0
anisotropic features
anisotropic features
initial sol
precipitate
precipitate
precipitate
8.6
h
h
h
h
-1 > h
-1 > h
0
0
> h+1
> h+1
80 °C/6 h
80 °C/12 h
50 °C/3 days
6.1
7.3
10.0
15.0
15.5
0
> h-1 > h+1
0
> h-1 > h+1
reliable statistics than model I. At the highest urea content (14.71
2. ESR of Spin Probes during Alumina Synthesis. The EPR
studies aimed at characterizing the aggregates during the
formation of mesostructured alumina are reported. The measure-
ments were performed at different stages of synthesis (for details
see the Experimental Section), and the relevant parameters of
the ESR spectra are collected in Table 5. In the following, only
the spectra of the probes in the sample Y2, with the molar ratios
Al(NO3)3/SDS/urea/water ) 1/0.42/21/69 will be discussed in
a detail, the values for sample Y1 being almost identical.
-
1
mol kg ), the residuals obtained from model I clearly deviate
2
from an even distribution and the corresponding ø is higher
than the value found for model II. These facts imply that the
mean length of the wormlike micelles decreases to values lower
than Ldiff, so the quenching process is no longer dominated by
diffusion.
An average number of quenchers per micelle, n (and
consequently a mean aggregation number, N), can be obtained
only for the shortest micelles found at the highest urea content
The ESR spectra of the starting solution show that the spin
probes are included in surfactant aggregates. The shape and the
parameters of the ESR spectrum of 5-DSA report on a high
degree of ordering of the surfactant chains in the interface region
and on absence of rapid aggregate tumbling and/or lateral
diffusion (see the discussion in the section devoted to precursor
systems (Figure 5a)). Heating for 4 h to 80 °C does not change
significantly the spectrum. We have also measured carefully a
sample heated at 60 °C over 40 h, to evidence any possible
changes in the aggregate geometry as the solution composition
changes during urea hydrolysis, before precipitation. At pH ∼
(see Table 4) when model II applies. However, we consider
that the result is uncertain, owing to the expected polydispersity
in aggregate sizes and the given aggregation number should
only be seen as a minimum estimated value. By fitting this last
curve with the classical Infelta-Tachiya model for probe-
quencher pairs solubilized exclusively in monodisperse spherical
5
-1
micelles, the quencher exit rate found, k-, of 6.9 × 10 s
-
was much higher than expected for DS spherical micelles
5
-1 19
(
(0.3-0.7) × 10 s ), implying that the micelles are still
elongated.

Synthesis of Mesoporous Alumina on SDS
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7741
Figure 5. ESR spectra of 5-DSA during alumina synthesis: (a) initial
solution, Y2, molar ratio SDS/Al(NO
3 3
) /water/urea ) 0.42/1/69/21; (b)
precipitate after 6 h at 80 °C, pH ) 6.1; (c) precipitate after 15 h at 80
°
C, pH ) 7.3; (d) precipitate after aging (c) at 50° C for 3 days. The
Figure 6. ESR spectra of CAT 16 during alumina synthesis: (a) initial
trace enhancing the small peak at high field was recorded with a higher
3 3
solution, molar ratios SDS/Al(NO ) /water/urea ) 0.42/1/69/21; (b)
modulation amplitude (2-3 G) as compared to the complete spectrum
precipitate after 6 h at 80 °C, pH ) 6.1; (c) precipitate after 12 h at 80
°C. The traces enhancing the small peak at high field were recorded
with a higher modulation amplitude (2-3 G) as compared to the
complete spectra (1 G).
(1 G).
5
the spectrum of the bulk showed immobilization of the radical.
Filtration showed that the signal was associated with very fine
solid particles. Thus no significant structural changes take place
with the surfactant aggregates in solution before the onset of
precipitation.
SCHEME 1: Diagram Representing the Presumed
Positions of the Nitroxide Moieties of Different Spin
Probes vs the Organic/Inorganic Interface
a
After 6 h at 80 °C, when pH 6.1 is attained and precipitation
has started, the spin probe 5-DSA appears partly immobilized
in the precipitate (Figure 5b). After 15 h of reaction, when pH
7
.3 is attained and the precipitation is complete, the spectrum
of 5-DSA in precipitate indicates complete immobilization of
the probe (Figure 5c). Changes during aging were not evidenced
(Figure 5d).
The spectrum (not shown) of a probe with the radical moiety
located in the core region of the micelles (16-DSA) indicates a
strong increase in local viscosity during alumina precipitation,
but the probe maintains a fairly high degree of mobility in the
as-synthesized alumina (Table 5). This is considered as strong
evidence that the measured data refer to probes in surfactant
aggregates in the mesoporous frame.
a
Changes of radical geometry during alumina precipitation are
suggested.
For the CAT 16 probe, the peculiar line shapesresembling
the 5-DSA spectrasobserved in the precipitate first formed (6
h reaction at 80 °C) (Figure 6b) and which is maintained in the
final stage (Figure 6c), indicates a tilt angle of the radical moiety
in CAT 16 the radical moiety is positioned in front of the
charged quaternary ammonium group, which will be anchored
at the water interface) Such a relative radial positioning was
well established in aggregates of nonionic block-copolymers
(Pluronics). In support of this statement, the probe maintains
a fairly high mobility in the final solid. Therefore the evolution
observed is an indication of structural changes continuing during
the precipitation of alumina and affecting deeper regions of the
template.
3. Influence of Composition on the Structure of As-
Synthesized Alumina. A number of syntheses have been carried
out, starting from different ratios of the components, to evidence
the effect of the aggregate size and shape in the precursor system
on the structure of the product. In all compositions the water/
urea molar ratio was maintained constant and the Al salt content
was higher than 5 wt %, ensuring the stability of the solution
as a single micellar phase (i.e., Figure 1, region III).
All these syntheses led to ordered materials, whose typical
X-ray diffractograms are presented in Figure 8. The XRD
patterns of as-synthesized alumina exhibited two or three Bragg
peaks. Their relative positions on the scattering vector axis, q,
4
2
(piperidine cycle) vs the long molecular axis. This way the
nitrogen pπ orbital (of the free electron) becomes approximately
parallel with the long molecular axis, which is the preferred
rotational axis, as in doxyl stearic acid probes. This probe is
anchored at the interface with the quaternary nitrogen atom and
exposes the piperidine cycle outside the organic phase. The
observed tilt of the cycle indicates certain constraints on the
probe geometry at the organic/inorganic interface, when the
alumina precipitate is formed (Scheme 1).
A small change in the tilt angle between the long molecular
axis and the nitroxide moiety, causing a change in the relative
amplitudes of the three derivative line (see ref 41 for a similar
change of spectrum, simulated by a small change of the tilt
angle) (Figure 7c,d) is evidenced also by the C12NO probe in
the precipitate, but only after 12 h of reaction.
This neutral probe is supposed to be included deeper in the
SDS aggregate than the charged CAT 16 (especially because

7742 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Caragheorgheopol et al.
TABLE 6: Position of the First Diffraction Peak, d100, and
the Unit Cell Dimension, a, of As-Synthesized Alumina
composition (molar ratio)
no. SDS Al(NO
)
3 3
H
2
O
urea pH
d
100 (Å) aa (Å)
1
2
3
Y2
4
5
6
7
8
0.24
0.26
0.26
0.42
0.54
0.61
1.23
1.23
1.88
2
0.69
1.09
1.09
1
69
69
69
69
69
69
69
69
69
69
31
31
31
21
31
31
31
31
31
30
7.8
6.4
7.3
7.2
7.2
7.5
7.3
6.4
6.1
7.2
33.5
38.4
35.6
34.8
32.7
35.3
35.0
36.4
38.2
33.5
38.6
44.3
41.1
40.1
37.7
40.7
40.4
42.0
44.1
38.6
2.1
1.69
0.83
0.83
2.12
1
Y1
a
The standard deviation for a is 1 Å.
temperature (350 °C) was used. A similar result, the collapse
of the hexagonal framework, was also obtained in the case of
surfactant removal by ionic exchange.
Discussion
The effect of urea on the aggregate structure in the redisso-
lution area of the pseudoternary SDS/Al(NO3)3/H2O system
Figure 7. ESR spectra of C12-NO during alumina synthesis: (a) initial
solution, molar ratios SDS/Al(NO /water/urea ) 0.42/1/69/21; (b)
precipitate after 6 h at 80 °C, pH ) 6.1; (c) precipitate after 12 h at 80
C; (d) precipitate after aging (c) at 50 °C for 3 days.
(Figure 1, region III) was investigated by several spectroscopic
3 3
)
methods and all results agree that addition of urea produces a
very significant reduction of the length of the wormlike
aggregates.
°
During synthesis (heating) no changes in the aggregate shapes
are observed in solution, before the formation of the precipitate.
Thus, we do not see any intermediate structure between the
starting aggregates and the precipitate. This result tends to
support the mechanism of formation of the mesostructured
hexagonal alumina proposed by Sicard et al. on the basis of
1
7
fluorescence measurements.
In the precipitate, 5-DSA is strongly immobilized by elec-
trostatic interaction with positively charged aluminum oligo-
meric species. The immobilization degree (described by 2Azz
value) increases from the precipitate first formed (∼6 h) to the
final precipitate (pH ∼ 7.3 and ∼15 h).This interaction reflects
-
the DS binding to alumina precursors. It should be remarked
3
+
that in the hexagonal DS-Al crystalline phase (Figure 1,
region II) 5-DSA maintains rapid lateral diffusion, presumably
because its counterion is also free to move. The cationic probe
CAT 16, appears to have the piperidine ring tilted vs the long
molecular axis (Scheme 1) in the precipitate, as soon as it is
formed, as a result of steric or specific interaction with aluminum
species. A similar, but smaller change in tilt angle is found with
the neutral probe C12-NO, not immediately after precipitation,
but after about 6 h. This effect is due to the formation and the
development of the layer of the alumina on the surface of the
surfactant template, during reaction. 16-DSA, located in the
hydrocarbon core of SDS retains isotropic rotation mobility,
but with reduced rate, in connection with the viscosity increase
in the core of the template.
Figure 8. XRD powder diffraction patterns of as-synthesized alumina
obtained from precursors with the SDS/Al(NO /water/urea molar
ratios: (a) 0.26/1.1/69/31, final pH ) 7.3; (b) 2/1/69/30, final pH )
.2. The aluminum-based solids were allowed to age 3 days at 50 °C.
3 3
)
7
followed in all cases the ratios 1: 3:2, implying that hexago-
x
nal structures were formed. The absence of the third diffraction
peak, corresponding to a (200) reflection was noticed in a few
syntheses of concentrated surfactant systems. It can be inferred
that the absence of the third peak indicated the formation of a
less ordered material.
Regarding the aging effect, the 3 days of aging at 50 °C had
a rather limited effect as the dimension of the unit cell remained
unchanged and only a slight enhancement of the Bragg peaks
was observed in case of less ordered materials.
However, the question of how the complex between the
inorganic polymers and the surfactant micelles evolve toward
a hexagonal structure is not elucidated. Earlier results concerning
18
the pseudoternary SDS/Al(NO3)3/water system may offer some
clues for understanding the formation of hexagonal alumina.
The dimensions of the unit cells, a, calculated from the
position of (100) reflection are shown in Table 6. Its value is
approximately 40 Å, with a slight dependence on the composi-
tion of the precursor system. The hexagonal framework always
collapsed during calcination. The disappearance of the morphol-
ogy took place even in those cases when, to prevent the removal
of the SO42 group from the inorganic frame, a low calcination
3
+
Thus, it was observed that concentrations of Al lower than
those used in the syntheses do not stabilize the SDS micellar
solutions and lead to the precipitation of the surfactant (Figure
1, region II). Intriguingly, the precipitate has a hexagonal
structure with the dimension of the unit cell of 44 Å,18 which
is only slightly higher than those of the as-synthesized hexagonal
-

Synthesis of Mesoporous Alumina on SDS
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7743
References and Notes
aluminas (∼40 Å). These observations suggest that the arrange-
ment of the surfactant/inorganic species complex into the
hexagonal pattern may occur in the same way as the formation
(1) Kresge, C. T.; Leonovicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck,
J. S. Nature 1992, 359, 710.
-
3+
of the insoluble hexagonal DS -Al complex in the SDA/
Al(NO3)3/H2O system.
(2) Corma, A. Chem. ReV. 1997, 97, 2373.
(
3) Ying, J. Y.; Mehnert, Ch. P.; Wong. M. S Angew. Chem., Int. Ed.
1
999, 38, 56.
We therefore propose the following model for the formation
of hexagonal meso-structured alumina: the starting SDS/Al-
(
(
4) Schuth, F. Chem. Mater. 2001, 13, 3184.
5) Cˇ ejka, J. Appl. Catal. A 2003, 54, 327 and references therein.
(
NO3)3/water/urea mixture is stabilized by the high content of
(6) Vaudry, F.; Khodabandeh, S.; Davis, M. E. Chem. Matter 1996, 8,
1451.
3+
Al . Then, whereas the pH increases as a result of urea
decomposition, the content of the charged hydroxoaluminum
(
(
7) Zhang, W.; Pinnavaia, T. J. Chem. Commun. 1998, 1185.
8) Cabrera, S.; Mascouri, J. E.; Alamos, J.; Beltran, A.; Beltran, D.;
3+
species becomes significant at the expense of Al . The decrease
Mendioroz, S.; Marcos, M. D.; Amoros, P. AdV. Matter. 1999, 11, 379.
(9) Yang, P.; Zhao, D.; Margolese, D. I.; Chmelka, B. F.; Stucky, G.
D. Chem. Mater. 1999, 11, 2813.
in Al3 content leads to destabilization of the micellar phase,
+
-
and DS starts to precipitate, forming, together with Al species,
(
(
10) Yada, M.; Machida, M.; Kijima, T. Chem. Commun. 1996, 769.
11) Yada, M.; Hiyoshi, H.; Ohe, K.; Machida, M.; Kijima, T. Inorg.
an ordered hexagonal structure. Subsequent increase of pH
would have as the only result the condensation of the polyions
of aluminum contained in the precipitate, with the preservation
of the hexagonal arrangement. This mechanism does not exclude
the formation of alumina without a hexagonal framework as a
result of the electrostatic interaction between the micelles in
solution and the inorganic polymeric species.
Chem. 1997, 36, 5565.
(12) Sicard, L.; Llewellyn, P. L.; Patarin, J.; Kolenda, F. Micropor.
Mesopor. Mater. 2001, 44, 195.
(
13) Yada, M.; Kuroki, S.; Kuroki, M.; Ohe, K.; Kijima, T. Langmuir
002, 18, 8714.
14) Firouzi, A.; Kumar, D.; Bull, L. M.; Besier, T.; Sieger, P.; Huo,
2
(
Q.; Walker, S. A.; Zasadzinski, J. A.; Glinka, C.; Nicol, J.; Margolese, D.
I.; Stuky, G. D.; Chmelka, B. F. Science 1995, 267, 1138.
(
15) Zana, R.; Frasch, J.; Soulard, M.; Lebeau, B.; Patarin, J. Langmuir
999, 15, 2603.
16) Frasch, J.; Lebeau, B.; Soulard, M.; Patarin, J.; Zana, R. Langmuir
000, 16, 9049.
17) Sicard, L.; Lebeau, B.; Patarin, J.; Zana, R. Langmuir 2002, 18,
Concluding Remarks
1
2
7
(
The present study has shown, in the first part, the progressive
decrease of the mean length of the wormlike micelles in the
ternary SDS/Al(NO3)3/H2O system under the influence of added
urea.
The use of the nitroxide spin probes located in different
regions of the micelles allowed the investigation of the complex
formed by the growing inorganic polymer and dodecyl sulfate
aggregates during synthesis. No structural changes were ob-
served for the aggregates in solution before the onset of
precipitation. In the precipitate, the presence of the surfactant
template is demonstrated as well as its interaction with the
aluminum species at the interface. An evolution of this interac-
tion in time is also observed.
Finally, the ability of the SDS/Al(NO3)3/water/urea system
to produce mesostructured alumina over a large domain of
concentrations was demonstrated. The structure and the topology
of the final product was fixed and did not depend on the size
of the aggregates in the precursor systems. The thermal stability
of the as-synthesized alumina was, however, poor and the
removal of the surfactant has always led to the collapsing of
the hexagonal framework.
(
4.
(18) Angelescu, D.; Caldararu, H.; Khan, A. Colloids Surf. A, in press.
19) Vasilescu, M.; Angelescu, D.; Caldararu, H.; Almgren, M.; Khan
A. Colloids Surf. A 2004, 235, 57.
(
(20) Angelescu, D.; Linse, P. Langmuir 2003, 19, 9661.
(21) Waggoner, A. S.; Keith, A. D.; Griffith, O. H. J. Phys. Chem. 1968,
72, 7109.
(22) Stone, T. J.; Buckman, T.; Nordio, P. L.; McConell, H. M. Proc.
Natl. Acad. Sci. U.S.A. 1965, 54, 1010.
(23) Seelig, J. J. Am. Chem. Soc. 1970, 92, 3881
(24) Gaffney, B. J. In Spin Labeling I; Berliner, L. J., Ed.; Academic
Press: New York, San Francisco, CA, and London, 1976; p 567.
(25) Stilbs, P. Prog. NMR 1987, 19, 1.
(
26) K e´ kicheff, P.; Grabielle-Madelmont, C.; Ollivon, M. J. Colloid
Interface Sci. 1989, 131, 112 and 133.
27) Lasic, D. D. J. Colloid Interface Sci. 1986, 113, 188.
(28) Wikander, G.; Johansson, L. B. Langmuir 1989, 5, 728.
29) Kuznetsov, A. N. Spin Probe Technique; Nauka: Moscow, 1976;
pp 53 (in Russian).
(
(
(30) Nilsson, F.; S o¨ derman, O. Langmuir 1996, 12, 902.
(31) Ulmius, J.; Wennerstr o¨ m, H. J. Magn. Reson. 1977, 28, 309.
(32) Johannesson, H.; Halle, B. J. Chem. Phys. 1996, 104, 6807.
(33) Tsai, J.; Gerstein, M.; Levitt, M. J. Chem. Phys. 1996, 104, 9417.
(
(
34) Baglioni, P.; Ferroni, E.; Kevan, L. J. Phys. Chem. 1990, 94, 4296.
35) Baglioni, P.; Rivara-Minten, E.; Dei, L.; Ferroni, E. J. Phys. Chem.
A tentative explanation of the precipitation mechanism is
proposed, by analogy with the destabilization of the micellar
solutions in the SDS/Al(NO3)3/water system and formation of
1
990, 94, 8218.
(36) Almgren, M.; Hansson, P.; Mukhtar, E.; Van Stam, J. Langmuir
1992, 8, 2405.
3
+
-
3+
a hexagonal Al -DS precipitate when the Al concentration
(37) Almgren, M.; Alsins, J.; Mukhtar, E.; Van Stam, J. J. Phys. Chem.
drops below a certain value.
1
988, 92, 4479.
(
38) Van der Auweraer, M.; Deschryver, F. C. Chem. Phys. 1987, 111,
05.
(39) Van der Auweraer, M.; Reekmans, S.; Boens, N.; Deschryver, F.
C. Chem. Phys. 1989, 132, 91.
Acknowledgment. The financial support by NATO Science
for Peace (NATO project no. SfP-974217) is gratefully ac-
knowledged. The Romanian coauthors acknowledge support
from the Ministry of Education and Research (Romania) (grant
1
(40) Almgren, M.; Swarup, S. J. Phys. Chem. 1983, 87, 876.
(41) Caragheorgheopol, A.; Schlick, S. Macromolecules 1998, 31, 7736.
(42) Caragheorgheopol, A.; Caldararu, H.; Dragutan, I.; Joela, H.; Brown,
8
/2001). D.A. is indebted to the European Union for a Marie
Curie Fellowship at Physical Chemistry 1, Lund University.
W. Langmuir 1997, 13, 6912.