PCCP

View Article Online

PAPER View Journal | View Issue

Understanding the role of co-surfactants in

microemulsions on the growth of copper oxalate
Cite this: Phys. Chem. Chem. Phys.,
2019, 21, 336 using SAXS
Sunaina,ab Vaishali Sethi,c Surinder K. Mehta, b
Ashok K. Ganguli *c and
Sonalika Vaidya *a
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

This study is an effort to understand the mechanism of the effect of the chain length of co-surfactants
on the growth of copper oxalate inside the core of reverse micelles using small angle X-ray scattering
(SAXS). In this study, we have used two different kinds of co-surfactants viz. 1-butanol (C4) and
1-octanol (C8) for the formation of the microemulsions. Time-dependent SAXS studies were carried out
for these two systems. The data were analyzed using both the model-independent approach and
model-dependent approach. For microemulsions containing only water inside the core of reverse
micelles (no ions), the shape of the reverse micelles was observed to be ellipsoid and spherical in nature
for 1-butanol and 1-octanol respectively. For a system containing copper oxalate nanostructures, the
fitting was carried out using the ellipsoidal core–shell model for reverse micelles and spheres, ellipsoids
Received 5th September 2018, and cylinders for copper oxalate nanostructures with 1-butanol as the co-surfactant. With 1-octanol as
Accepted 27th November 2018 the co-surfactant, the two contributions that were used were the spherical core–shell model for reverse
DOI: 10.1039/c8cp05622f micelles and spheres for copper oxalate nanostructures. Based on the analysis of SAXS data, a growth
mechanism has been proposed. The study discussed here could open the field of understanding the
rsc.li/pccp growth mechanism of complex nanostructures formed using the microemulsion route.

Introduction on the size and morphology of the material. The spectroscopic

and microscopic techniques that are used for studying the
Microemulsions (ME) have gained interest in recent years due growth behavior have disadvantages over SAXS. In the former,
to the efficient control of size, shape, and composition of information about the size could be extracted indirectly from
nanoparticles synthesized by using them. Various parameters changes of the optical properties whereas; in the latter, inter-
of ME that control the shape and size of the nanostructured mediate sample preparation is required. The primary barrier
material are the nature of the surfactant, the co-surfactant, the with these techniques for obtaining information about the
oil phase and the water to surfactant ratio (Wo). The use of formation of nanostructures is that ‘‘seeing through’’ the
these microemulsions for the synthesis of nanostructured liquids for getting an insight into the chemical and physical
materials and the effect of various parameters on the size and events in the solvent during nanocrystal growth is difficult.
shape of these nanostructured materials have been discussed This also leads to a challenge in understanding the relation-
in detail by Pileni,1 Eastoe,2 Lopez-Quintela,3 Capek,4 ship between structure and functionality during material
Holmberg,5 Uskokovic,6 Lombardo,7 and Ganguli.8 Although applications.9 For such studies, SAXS is an ideal technique
there has been a lot of development in this field for the because it can deliver information on shape and size changes
synthesis of nanoparticles, there is still a lack of basic under- with the course of the reaction. Monitoring the growth of
standing of the effect of these parameters on the size and shape nanostructures using SAXS in liquid media becomes easier as
of nanostructured materials. This lack of knowledge affects the X-rays can efficiently penetrate the liquid.10,11 We cite here a
final application of nanostructured materials as it depends few reports from the literature wherein SAXS has been used to
study the nucleation and growth behavior of nanostructured
a
Institute of Nano-Science and Technology, Habitat Centre, Phase – 10, Sector – 64, materials. The kinetics of formation of nanoparticles of Pd
Mohali-160062, India. E-mail: svaidya@inst.ac.in
b
and Rh using the photoreduction method was studied by
Department of Chemistry and Centre for Advanced Studies in Chemistry,
Panjab University, Chandigarh-160014, India
Harada et al.12 using SAXS. They found that the nucleation
c
Department of Chemistry, Indian Institute of Technology, Hauz Khas, process followed an autocatalytic reduction process followed by
New Delhi-110016, India. E-mail: ashokganguliiitd@gmail.com growth of nanoparticles. They also observed a difference in the

336 | Phys. Chem. Chem. Phys., 2019, 21, 336--348 This journal is © the Owner Societies 2019
 View Article Online

Paper PCCP

rate of formation of the two kinds of nanoparticles. Formation nanoparticles using microemulsions, a clear understanding of
of oligomers of silica from tetraethylorthosilicate (TEOS) and how co-surfactants (having different chain lengths) affect the size
aminopropyltrimethoxysilane (APTS) in a reverse micro- and shape of nanoparticles using microemulsions is missing. The
emulsion was studied using in situ SAXS by Lin et al.13 The study reported here focuses on the growth of copper oxalate
authors identified four stages of the evolution of particle nanostructures synthesized using a microemulsion method.
formation, which were referred to as induction, nucleation, Metal oxalates with divalent metal ions, MC2O4 (M = Mn, Co,
diffusion growth and slow down. The nucleation and growth Zn, Ni, and Cu)21–26 are interesting to study as their morphology
mechanism of SnO2 nanoparticles was studied by using in situ can be easily controlled by variation of the parameters affecting
time-resolved SAXS and combined Raman/extended X-ray absorp- the microemulsion. With divalent metal ions, the nature of the
tion fine structure (EXAFS) measurements by Caeteno et al.14 bond between the metal and the ligand results in a linear
Using this combination they proposed a five-step mechanism arrangement, which was reported to be the reason for the
corresponding to the prenucleation of very small nucleates formation of a rod like morphology for these oxalates. Changes
followed by growth of monomers of tin-oxo clusters and in the parameters associated with microemulsions viz. the
cluster–cluster growth which leads to the formation of double surfactant, co-surfactant, and solvent have been shown to affect
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

or triple chain structures for SnO2 nanoparticles. These struc- the size and shape of these oxalates.25,26 Thus, the study of this
tures were shown to further interconnect to form SnO2 nano- system (divalent metal oxalate) whose morphology changes
particles. SAXS and electron microscopy were used by drastically with changes in the co-surfactant, we felt, would be
Zhang et al.10 to study the formation of 3D assemblies of Pd interesting to understand the effect of the co-surfactant on the
nanocubes by a solvent evaporation method. They studied growth of nanostructures formed using the microemulsion
the relationship between the nanoparticle shape, the ligand route. Another reason for choosing copper oxalate as a case
shell, and the emergent phase behavior of nanocube assemblies. study to understand the nucleation and growth behavior of
P. Calandra et al.15 carried out studies on the formation of nanostructures in microemulsion using SAXS was the fact that
nanostructures of AgX nanoparticles using dry dispersions of copper oxalate is important from both fundamental as well as
AgNO3 and KX (X = Br, Cl) nanoparticles in AOT/n-heptane commercial aspects. It is known that stacking of ribbons of
solutions. Analysis of AgX/AOT composites by WAXS indicated a Cu(C2O4)–Cu(C2O4) results in the formation of the orthorhombic
shift in the peak position of the (100) plane of the composite structure of copper oxalate.27 Apart from this, copper oxalate
when compared with pure AOT. The authors also suggest exhibits an unusual anti-ferromagnetic property owing to its
structural disorder as observed by a lowering of the intensity polymer-like structure which arises because of stacking of
of the peak. Polte et al.16 studied the growth of gold nano- Cu(C2O4)–Cu(C2O4) ribbons.28 Copper oxalate could be used as
particles (AuNP) by using in situ SAXS. The studies were carried an inorganic blue pigment in ceramics and other commercial
out on a homogeneous mixture formed by the continuous flow powders such as baby blue powder28 and has the potential to be
of an aqueous solution of tetrachloroauric acid and sodium used as a single source precursor for the formation of Cu,29
borohydride. By using SAXS studies they concluded that the Cu2O,29 and CuO.30 To the best of our knowledge, variation of
first step is rapid conversion of the ionic gold precursor into the co-surfactant and study of its effect on the growth of
metallic gold nuclei, followed by particle growth via coales- nanostructures (copper oxalate in this case) using SAXS has
cence of smaller entities. A combination of SAXS and WAXS on not been reported so far. In the present study, the two different
time scales of 100 ms and 2.5 ms with a time resolution of 10 ms co-surfactants that were chosen are 1-butanol and 1-octanol. The
was used by Schiener et al.17 for study of the nucleation and main rationale behind the choice of these two co-surfactants is
growth of CdS nanoparticles. The study gave direct evidence of based on our previous observation26 wherein nanorods of copper
a two-step mechanism involving the formation of prenucleation oxalate were formed in a microemulsion containing CTAB as a
clusters followed by nanoparticle nucleation from coalescence surfactant, 1-butanol as a co-surfactant and isooctane as the oil
of the precursors. Uniform distributions of entrapped cyanamide phase. The morphology of copper oxalate changed to a cube
molecules within reverse micelles of AOT/CCl4 and lecithin/CCl4 when the co-surfactant was changed to 1-octanol. This study will
were observed by Calandra et al.18 using combined results of SAXS, further add to the field of microemulsions and give insight into
FT-IR, and 1H NMR spectroscopy. Ganguli et al. have used SAXS how the chain length of the co-surfactant influences the growth
for studying the growth mechanism of zinc oxalate nanorods of particles and finally their shape.
synthesized using microemulsions19 wherein they observed that
nanostructures with varied size and shape were formed within half
an hour after the start of the reaction. They also observed an Results and discussion
asymmetry in the I(q) vs. q curve concluding that the reverse
micelles had an ellipsoidal shape. Another study by the group The studies carried out in this report are an attempt to get
showed the effect of a hydrotrope on the size and shape of the an insight into the effect of variation in the chain length of the
microemulsion droplets.20 They observed that the size of the co-surfactant on the shape and size of copper oxalate nano-
ellipsoid along the equatorial axis increased from 7 to 40 nm with particles during the reaction. The co-surfactants that were investi-
an increase in the concentration of the hydrotrope. Though there gated are 1-butanol, bearing 4 carbon atoms, and 1-octanol,
have been studies carried out on understanding the formation of bearing 8 carbon atoms. For this, firstly scattering data was

This journal is © the Owner Societies 2019 Phys. Chem. Chem. Phys., 2019, 21, 336--348 | 337
 View Article Online

PCCP Paper
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

Fig. 1 Cross-sectional electron density plots for microemulsions

containing (a) CTAB/1-butanol/iso-octane/water and (b) CTAB/1-octanol/
iso-octane/water. (c) Schematic showing a reverse micelle with regions
A and B.

obtained for a microemulsion containing water inside the core

of the reverse micelles (no ions). In the microemulsion system,
the scattering component is the reverse micelles (consisting
of aqueous cores, surfactant/co-surfactant head groups and
surfactant/co-surfactant tails) as predicted from the contrast
between adjacent regions with different electron densities.
Fig. 1a and b show the cross-sectional electron density profile
of reverse micelles formed using 1-butanol and 1-octanol as the
co-surfactant, respectively. The cross-sectional electron density
profile may be represented by two layers where the interior
hydrophilic core is composed of an aqueous pool, the polar
head of the surfactant and the co-surfactant (depicted by A in
the electron density plot and the schematic, Fig. 1a–c) while the
exterior layer (depicted by B in the electron density plot and the Fig. 2 (a) Double logarithm plot (inset shows the slope obtained),
schematic, Fig. 1a–c) is composed of the hydrophobic chain of (b) Porod plot (inset shows negative deviation in the Porod region) and
the surfactant and the co-surfactant. The center of the hydro- (c) model fitted SAXS data (I vs. q plot) for blank microemulsion (CTAB/
co-surfactant/iso-octane/water) systems.
philic core is defined as r = 0. This core extends to the value of
‘r’ where the sign of r(r) changes. r(r) o 0 corresponds to
the hydrophobic chain of the surfactant and co-surfactant.
Calandra et al.31 have also shown that the contrast between system, the slope was proportional to ‘1’ suggesting that the
the adjacent regions having different electron densities gives reverse micelle has a cylindrical or ellipsoidal shape whereas, in
rise to the scattering and accounts for their choice of model case of 1-octanol, the slope was proportional to ‘0’ which means
used for fitting the data. Based on our interpretation of the that the reverse micelle has a spherical shape. The Porod plot
electron density plots (Fig. 1a and b), it was apparent that shows a negative deviation from Porod’s law as shown in Fig. 2b
the choice of model would be based on a core–shell type of which indicates that the electron density between the two
structure. phases does not change abruptly and varies gradually over a
Fig. 2a shows a double logarithm plot i.e. log(I) vs. log(q) certain range of distance between them, resulting in the
plot. It has been reported that the slope of this curve (eqn (3) formation of a diffuse interfacial layer.33 This diffuse layer
and (4)) gives an indication about the shape of scattering corresponds to CTAB molecules penetrating into iso-octane.
components.32 Hence, the slope of the plot was used to assess The average thickness of the diffuse layer which is obtained
the shape of the reverse micelles. In the case of the 1-butanol from the modified Porod’s law (eqn (6) and (7)) is 1.2 nm and

338 | Phys. Chem. Chem. Phys., 2019, 21, 336--348 This journal is © the Owner Societies 2019
 View Article Online

Paper PCCP

0.6 nm for 1-butanol and 1-octanol systems respectively. It has

been reported that alcohols with longer hydrocarbon chains
favor the formation of reverse micelles compared to alcohols
with shorter hydrocarbon chains. Alcohols with a number of
carbon atoms ranging from 2 to 6 are adsorbed at the interface
resulting in pulling apart the surfactant.34 The interaction
between the alcohol and the surfactant molecule and also with
the oil is negligible. Beyond C6, the alcohols essentially behave
like oils and are present in a lesser amount at the interface.
The interaction with the oil and the surfactant tail increases.34
The smaller value of the average thickness for the diffuse layer
for the microemulsion formed using 1-octanol could thus
be attributed to the increased interaction of the hydrocarbon
tail of 1-octanol with iso-octane and the surfactant tail when
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

compared to that of 1-butanol. Thus, the model-free approach

(electron density curves, the double logarithm plot and Porod’s
plot) helped us to obtain a prior idea of shape and size after
which the SAXS data were fitted with ellipsoidal core–shell
and spherical core–shell models for 1-butanol and 1-octanol
systems respectively (Fig. 2c).
The parameters obtained by fitting the SAXS data for the
reverse micelle containing only water and no ions (blank) and
the proposed schematic for both systems are shown in Table 1
and Fig. 3 respectively. It should be noted that the thickness of
the shell, as obtained after fitting the SAXS data, was found to
be 1.1 nm for the microemulsion formed using 1-butanol and
0.7 nm for the microemulsion formed using 1-octanol. The
values obtained after fitting the SAXS data were found to be in
agreement with the values obtained for the average thickness of
the diffuse layer using Porod’s law (1.2 nm and 0.6 nm
for microemulsions formed using 1-butanol and 1-octanol
respectively). Thus, corroboration of values for the thickness
obtained using two approaches viz. a model-independent Fig. 3 Schematic diagram showing the structure of reverse micelles of (a)
approach and model-dependent approach also reinforces and CTAB/1-butanol/iso-octane/water and (b) CTAB/1-octanol/iso-octane/
water showing various parameters obtained after fitting the SAXS data.
suggests that the model used for analysis of the SAXS data was
A molecule of co-surfactant is represented by the blue color and a
correct. Using these models and the knowledge about the size molecule of surfactant is represented by the red color.
and shape of the reverse micelles containing water, the growth
of copper oxalate in reverse micelles was further investigated.
The growth of copper oxalate in the CTAB/1-butanol/ formation of a well-defined and sharp phase boundary which was
iso-octane/water system was first analyzed by using a double attributed to the formation of copper oxalate nanostructures.33
logarithm plot (Fig. 4a). The slope of the double logarithm plot Based on these preliminary investigations, different models
was around 0.7–0.8 suggesting the ellipsoidal/cylindrical shape of (Fig. 4c) were used for fitting the SAXS data curves.
reverse micelles as well as copper oxalate nanostructures. Positive The fitting parameters are listed in Table 2. Ellipsoid core–
deviation in the Porod plot (Fig. 4b) was observed, indicating the shell and sphere models were used for fitting the scattering

Table 1 Parameters obtained after fitting for the CTAB/co-surfactant/iso-octane/water system

Form factor Structure factor: macro-ion

System Model Dimensions (nm) RHSmic. (nm) Ionic stength wred2
CTAB/1-butanol/iso-octane/water Ellipsoidal core–shell a = 3.8 4.65 0.33 2.2
b = 2.1
t = 1.1

CTAB/1-octanol/iso-octane/water Spherical core–shell R1 = 5.5 5.14 0.39 3.6

R2 = 4.8
RHS: radius of a hard sphere.

This journal is © the Owner Societies 2019 Phys. Chem. Chem. Phys., 2019, 21, 336--348 | 339
 View Article Online

PCCP Paper

indicating coalescence of the micelle and the presence of a

fused micellar structure. Beyond 1.5 h, the size decreases and
remains constant up to the end of the reaction (Fig. 5a). The
variation in the dimension of copper oxalate nanostructures is
quite interesting to observe. Initially (0–20 min), the data could
be fitted using a spherical model indicating that just after
mixing the reacting microemulsions, copper oxalate structures
started forming. The size of these spheres ranged from
1–1.5 nm. From 40 min to 1.5 h, the data could be fitted using
the ellipsoid model. From Fig. 5b, it could be observed that the
increase in the equatorial-semi axis was gradual as compared to
a drastic increase in the size of the semi-principle axis. Beyond
1.5 h, the cylindrical model was used to fit the SAXS curves
wherein a slight increase in the cross-sectional radius of the
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

cylinder and a drastic increase in the length of the cylinder were

observed. From the above analysis, it is quite interesting to
observe that the growth of copper oxalate from spheres to
cylinders occurs only along one dimension i.e. the length (the
semi-principle axis in ellipsoids). The other dimension almost
remains unaltered. Thus, our studies confirm our findings
reported previously26 where it was predicted that the surfactant
molecules are not associated with the end of rods, which are
water-rich domains. This results in an increase in the exchange
of reactants with other micelles, favoring the growth along
this axis and thereby resulting in the formation of rods. Our
findings from current studies using SAXS confirm the predic-
tion reported previously.26
Analysis of SAXS curves for the CTAB/1-octanol/iso-octane/
copper oxalate system obtained at different time intervals was
first carried out on the basis of the preliminary investigation.
From the double logarithm plot (Fig. 6a), it was observed that
the value of the slope varies from 0–0.3 indicating the spherical
shape of the scattering components, reverse micelles and
copper oxalate nanostructures, in this study. From the Porod
plot (Fig. 6b), it was observed that just after mixing the micro-
emulsions to 2 h of the reaction, the Porod plot shows a change
in the slope from positive to negative deviation, indicating
micelle–micelle interaction which results in the formation of
Fig. 4 (a) Double logarithm plot (inset shows the slope obtained), a diffuse phase boundary.33 Beyond 2 h, with the growth of
(b) Porod plot (inset shows positive deviation in the Porod region) and
copper oxalate nanostructures, positive deviation in the Porod
(c) model fitted SAXS data (I vs. q plot) for particle contained inside the
microemulsion (CTAB/1-butanol/iso-octane/copper oxalate) systems. region was observed. This is because copper oxalate in the
microemulsion behaves like a sharply defined phase.33 Hence,
based on the analysis using the model-free approach, SAXS data
curves obtained at 0 and 20 minutes. It should be noted that curves were fitted by selecting an appropriate model using
the ellipsoid core–shell model was used for reverse micelles, as the results of the model-free analysis (Fig. 6c). The fitting
evident from Fig. 2b wherein reverse micelles formed with parameters are listed in Table 3.
1-butanol as the co-surfactant were ellipsoidal in nature. Three Two different models were used for obtaining the fit viz. the
different models viz. spherical, ellipsoid and cylinder were used spherical core–shell model for reverse micelles and spheres for
to fit the scattering contribution from copper oxalate nano- copper oxalate nanostructures. Initially, for the first 2 h, SAXS
structures in SAXS data. From 40 minutes to 2 h, the curves data could be fitted using only the spherical core–shell model
could be fitted with ellipsoid core–shell and ellipsoid models. indicating that only reverse micelles are present as a scattering
Beyond 2 h, ellipsoid core–shell and cylindrical models were component in the system. This also corroborates with the
used to fit SAXS data. From the dimensions obtained for the inference observed from the Porod plot wherein negative
reverse micelles (Fig. 5a) and copper oxalate nanostructures deviation was observed indicating the formation of a diffuse
with time (Fig. 5b), it could be envisaged that the size of phase boundary (interfacial layer) corresponding to the CTAB
ellipsoidal reverse micelles increased with time (up to 1.5 h) molecule penetrating into iso-octane. Thereafter, data were

340 | Phys. Chem. Chem. Phys., 2019, 21, 336--348 This journal is © the Owner Societies 2019
 View Article Online

Paper PCCP

Table 2 Parameters obtained after fitting for the CTAB/1-butanol/iso-octane/copper oxalate system

Contribution 1: for reverse micelles Contribution 2: for copper oxalate nanostructures

Form factor: Structure factor:
ellipsoidal core–shell macro-ion Form factor Structure factor
Time RHSmic. Ionic Zparticle Rpart. b/Lpart.
(min) a (nm) b (nm) t (nm) (nm) strength Model (cm1) (nm) (nm) Model R (nm) L (nm) wred2
10
0 3.57 1.50 1.01 4.51 1.10 Sphere 1.8  10 1.0 — Hard sphere 0.9 — 6.3
20 3.61 1.61 0.79 4.26 1.04 Sphere 2.1  1010 1.2 — Hard sphere 1.1 — 4.7
40 3.64 1.70 0.68 4.01 0.96 Ellipsoid 2.6  1010 1.1 2.2 Hard sphere 2.1 — 3.9
60 3.80 1.79 1.12 4.08 0.99 Ellipsoid 3.6  1010 0.6 4.1 Hard sphere 4.0 — 8.9
90 3.56 1.62 0.85 4.61 0.91 Ellipsoid 4.2  1010 1.5 5.3 Hard sphere 5.2 — 7.9
120 3.53 1.47 0.90 4.32 0.93 Cylinder 5.7  1010 7.0 18.8 Cylinder prism 7.0 18.4 10.9
240 3.46 1.43 0.91 4.37 0.87 Cylinder 7.1  1010 8.3 19.9 Cylinder prism 8.1 19.8 3.7
360 3.32 1.35 0.83 3.95 0.85 Cylinder 8.9  1010 8.5 50.2 Cylinder prism 8.5 49.5 11.7
480 3.29 1.32 0.74 4.02 0.82 Cylinder 9.1  1010 9.1 53.1 Cylinder prism 9.2 52.9 6.6
540 3.26 1.26 0.87 4.21 0.80 Cylinder 10.2  1010 9.5 63.0 Cylinder prism 9.3 62.9 5.1
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

600 3.28 1.29 0.93 4.05 0.79 Cylinder 12.7  1010 10.8 70.4 Cylinder prism 10.5 70.5 6.1
1200 3.32 1.22 1.12 4.21 0.72 Cylinder 14.8  1010 14.4 74.5 Cylinder prism 14.4 74.3 5.3
RHS: radius of a hard sphere.

fitted with spherical core–shell and sphere models for 7–20 h. the separation of nanostructures from reverse micelles could be
The fitted SAXS data are shown in Fig. 6c. We found that the attributed to the slow rate of the reaction within the reverse
size obtained using a spherical core–shell model decreased micelles due to an increase in the film rigidity as compared to
after 7 h whereas the size obtained for spheres increased after the microemulsion formed with 1-butanol. The change in the film
7 h indicating growth in the particle formation as shown in rigidity with an increase in the chain length of the co-surfactant
Fig. 7a and b. The fit of the averaged SAXS profile yields has been discussed previously.8,26 In the case of the 1-butanol
a spherical particle whose volume corresponds to the volume system, due to film flexibility, the micelles interact with each other
of a cube.10 in no time and the growth of copper oxalate starts immediately
In view of the above observations, a schematic for the after mixing of the two microemulsions. But in the case with
formation of copper oxalate using the microemulsion method 1-octanol as the co-surfactant, the surfactant film is more rigid
with a variation of the co-surfactant is proposed which is shown resulting in a reduced coalescence rate and hence leading to a
in Fig. 8. It is thus proposed that for the microemulsion system slow nucleation and growth process.26 Another interesting obser-
formed using 1-butanol as the co-surfactant, nucleation of vation that we have come across while analyzing SAXS data and
copper oxalate nanostructures started immediately after mixing correlating with the formation of copper oxalate is that apart from
of microemulsions. Up to one hour after the start of the the change in the rigidity of the surfactant film with the chain
reaction, we propose that a series of events had occurred which length of the co-surfactant, it is the shape of the reverse micelles
included formation of nuclei of copper oxalate inside the core that changes when the chain length was changed from C4
of reverse micelles (after coalescence and decoalescence of (1-butanol) to C8 (1-octanol). This change in the shape of the
reverse micelles containing the reactants), growth of nano- reverse micelles along with the rigidity of the surfactant film
structures inside the core of reverse micelles and separation affects the morphology of the final product, copper oxalate in
of these nanostructures from the core of reverse micelles. The this case.
proposal was based on the fact that the positive deviation from In order to corroborate the results obtained using SAXS,
Porod’s law was observed from the start of the reaction and also TEM studies were carried out on the product obtained after
the fact that initial fitting of the SAXS data could be carried out centrifugation (after different times of the reaction). For
using the spherical model for copper oxalate nanostructures. 1-butanol, the studies were carried out on sample isolated at
Beyond one hour, these separated copper oxalate nanostructures 20 minutes, 2 h, 7 h and 20 h of the reaction. From Fig. 9a–d,
capped by the surfactant interacted with each other resulting in the growth of nanostructures of copper oxalate with 1-butanol
the formation of ellipsoids and finally nanorods (cylinders). The as the co-surfactant could be observed. At 20 minutes, spherical
reason for the formation of rod shapes has been attributed to the particles could be observed in the TEM image (Fig. 9a) which
association of surfactant molecules on the lateral side of the rods. corroborates well with our SAXS analysis. Small rod-like struc-
The ends of the rods, being water-rich domains, are not associated tures assembling with each other could be clearly observed in
with surfactant molecules. This results in an increase in the the inset of Fig. 9b. Discrete rod-like structures could be
exchange of reactants with other micelles, favoring the growth observed at 7 h (Fig. 9c) and 20 h (Fig. 9d) of the reaction.
along this axis and thereby resulting in the formation of rods. In the case of copper oxalate nanostructures synthesized using
In the case of the reaction occurring in the microemulsion formed 1-octanol as the co-surfactant, spherical particles could be
with 1-octanol as the co-surfactant, separation of copper oxalate observed in the TEM image (Fig. 10a) for the sample isolated
nanostructures occurred after 7 hours of reaction. This delay in after 7 h of the reaction. These particles assembled to form

This journal is © the Owner Societies 2019 Phys. Chem. Chem. Phys., 2019, 21, 336--348 | 341
 View Article Online

PCCP Paper
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

Fig. 5 Variation in (a) inner and outer radii of ellipsoid core–shells (model
used for reverse micelles) and (b) dimensions of spheres, ellipsoids and
cylinders (models used for copper oxalate) obtained after fitting SAXS data
curves for the CTAB/1-butanol/iso-octane/copper oxalate system.

nanocubes after the reaction was carried out for 20 h (Fig. 10b).
The size of the nanostructures observed from TEM was larger Fig. 6 (a) Double logarithm plot (inset shows the slope obtained),
than that observed by fitting the SAXS data. This anomaly (b) Porod plot (inset shows both positive and negative deviation in the
could be attributed to the fact that SAXS studies were carried Porod region) and (c) model fitted SAXS data (I vs. q plot) for particle
out during the progress of the reaction whereas TEM studies contained inside the microemulsion (CTAB/1-octanol/iso-octane/copper
oxalate) systems.
were carried out on the product obtained after isolation
by centrifugation. During isolation of the product, self-
aggregation of the oxalate nanostructures could have resulted
in large size. This self-aggregation could have been the result of monohydrate in the case of 1-butanol as well as 1-octanol as the
the interaction between the hydrophobic chains of the surfac- co-surfactant. All the reflection planes in the pattern could
tant that surrounds the particles. Such interaction has been be indexed using the orthorhombic unit cell of CuC2O4H2O
reported in the literature which has resulted in assembled (JCPDS-00-021-0297).
nanostructures of BaCrO4.35 The average crystallite size of both the samples was calcu-
PXRD was done to check the purity of the final product lated using Scherrer’s equation and has been reported in
formed after the reaction. The final product was extracted Table 4. Refined lattice parameters a, b and c were calculated
from the microemulsion system by centrifugation followed by by using UnitCell software (based on a non-linear least square
washing with a 1 : 1 chloroform/methanol mixture to remove cell refinement program). The cell parameters calculated for
surfactant and a non-polar solvent. The PXRD pattern (Fig. 11) copper oxalate are given in Table 4. Not much change in the
of the product obtained via the reverse micelle method (free lattice parameters for the two systems was obtained when
from surfactant) shows the formation of pure copper oxalate compared to the standard.

342 | Phys. Chem. Chem. Phys., 2019, 21, 336--348 This journal is © the Owner Societies 2019
 View Article Online

Paper PCCP

Table 3 Parameters obtained after fitting the CTAB/1-octanol/iso-octane/copper oxalate system

Contribution 1: for reverse micelles Contribution 2: for copper oxalate nanostructures

Form factor: spherical core–shell Structure factor: macro-ion Form factor: sphere Structure factor: hard-sphere
1
Time (min) R1 (nm) R2 (nm) RHSmic. (nm) Ionic strength Zpart (cm ) Rpart. (nm) RHSpart. (nm) wred2
0 2.0 1.7 1.85 1.09 — — — 9.3
20 2.1 1.7 2.02 0.99 — — — 3.6
40 2.2 1.7 2.09 0.97 — — — 6.3
60 2.2 1.8 2.21 0.98 — — — 9.1
80 2.3 1.9 2.23 0.94 — — — 4.9
100 2.4 1.9 2.35 0.91 — — — 5.1
120 2.6 2.0 2.59 0.93 — — — 8.9
420 1.2 0.9 1.15 0.86 7.4  1010 0.9 1.0 10.9
480 1.6 0.2 1.52 0.84 9.0  1010 1.0 1.2 11.7
540 2.6 1.4 2.58 0.81 10.4  1010 2.9 3.0 8.6
600 3.0 1.3 2.86 0.77 12.8  1010 6.4 6.5 6.6
1200 3.0 1.7 2.92 0.71 14.4  1010 20.1 20.2 6.2
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

RHS: radius of a hard sphere.

(Merck, purity Z99%) and ammonium oxalate (Merck, purity

Z99%) were used. Distilled water was used for preparing the
aqueous phase for the reverse micellar system. All chemicals were
used without further purification.

(ii) Synthesis of copper oxalate nanostructures

Copper oxalate nanostructures were synthesized using a micro-
emulsion method as reported earlier.24 Briefly, the micro-
emulsions were formed by using CTAB as the surfactant,
1-butanol/1-octanol as the co-surfactant, iso-octane as the non-
polar phase, and a 0.1 M aqueous solution of Cu2+ and C2O42 as
the aqueous phase The weight fraction of these constituents in
the microemulsion was 16.76% of CTAB, 13.9% of 1-butanol/
1-octanol, 59.29% of iso-octane and 10.05% of the aqueous
phase. Two microemulsions were prepared, one containing
a 0.1 M aqueous solution of copper nitrate while the other
containing a 0.1 M aqueous solution of ammonium oxalate.
The two microemulsions were slowly mixed and stirred for 20 h
using a magnetic stirrer. Aliquots were taken out at different
time intervals for carrying out SAXS studies. The details of the
experiment are discussed in the next section. The product
obtained at the end of the reaction was separated from the
surfactant and non-polar phase by centrifugation and washed
with a 1 : 1 chloroform/methanol mixture to remove the surfac-
tant and dried at room temperature.

(iii) Characterization
To get an insight into the morphological changes during
Fig. 7 Variation in (a) inner and outer radii of the spherical core–shell
(model used for reverse micelles) and (b) radii of spheres (model used for
the growth process of the copper oxalate nanoparticles, small
copper oxalate) obtained after fitting SAXS data curves for the CTAB/ amounts were periodically collected from the continuously
1-octanol/iso-octane/copper oxalate system. stirred reacting microemulsion and filled into a glass capillary.
SAXS studies on the CTAB/1-octanol/iso-octane system were
carried out at Indian beamline-18B at the Photon Factory
Experimental synchrotron, Tsukuba, Japan. The data were recorded at
room temperature using a Pilatus detector. The data were
(i) Materials used first normalized to the intensity of the transmitted beam, after
Hexadecyltrimethylammonium bromide (CTAB) (CDH, purity which they were radially averaged to obtain a one-dimensional
Z99%), iso-octane (Merck, purity Z99%), 1-butanol (Merck, intensity (I) versus scattering vector (q) plot. SAXS studies on
purity Z99%), 1-octanol (CDH, 499%), copper nitrate trihydrate the CTAB/1-butanol/iso-octane system were carried out on a

This journal is © the Owner Societies 2019 Phys. Chem. Chem. Phys., 2019, 21, 336--348 | 343
 View Article Online

PCCP Paper
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

Fig. 8 Schematics for the growth mechanism of copper oxalate formation within reverse micelles having 1-butanol and 1-octanol as the co-surfactant.

Fig. 9 TEM images of copper oxalate obtained after (a) 20 minutes,

(b) 2 h, (c) 7 h and (d) 20 h of reaction using CTAB/1-butanol/iso-octane.

SAXSess mc2 instrument (Anton-Paar, Austria) using a CuKa

source with a wavelength of 1.54 Å. Data were acquired through
image plates and read on a PerkinElmer cyclone image plate
reader. The SAXSQuant 2D software was used to reduce the two-
dimensional data to a one-dimensional intensity (I) versus
scattering vector (q) plot. For both the particle dispersion and
the reverse micellar solution, the scattering of the solvent
was subtracted before data analysis. Transmission electron
microscopy (TEM) studies were carried out on a JEOL JEM
2100 operated at 200 kV. Powder samples were dispersed in
ethanol and a drop of the dispersed solution was put on a
copper grid supported by a carbon film and dried. Powder X-ray Fig. 10 TEM images of copper oxalate obtained after (a) 7 h and (b) 20 h
diffraction (PXRD) studies of powdered samples, obtained after of reaction using CTAB/1-octanol/iso-octane.

344 | Phys. Chem. Chem. Phys., 2019, 21, 336--348 This journal is © the Owner Societies 2019
 View Article Online

Paper PCCP

where Ibkg(q) is the scattering intensity of the iso-octane filled

capillary.20,36
Model-independent approach. No assumption on the scat-
tering system is made in the model-free approach.37 It utilizes a
set of standard plots which give an idea about the scattering
particles. The first plot is a power-law variation in I(q) which
is observed in SAXS obtained from differently shaped
nanostructures
I(q) = Io + qm. (3)
The values of the exponent ‘m’ can be determined from the slope
of linear parts of a log I(q) vs. log(q) plot (double logarithm plot)

log(I(q)) = log(Io)  m log(q) (4)

Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

Fig. 11 PXRD pattern of copper oxalate monohydrate obtained via the where Io is defined as the intensity of scattering at zero angle.32
reverse micelle method in the case of (a) 1-butanol and (b) 1-octanol as the In the double logarithm plot, the value of ‘m’ as 0, 1 and 2
co-surfactant. corresponds to spherical, cylindrical and lamellar shapes
respectively. The second set of standard plots is obtained by
applying Porod’s law which gives information about the nature
Table 4 Calculated lattice parameters for copper oxalate (refined with
standard deviation) of the phase boundary. For a system having a sharp boundary
with different electron densities33 which are constant, the
dcrystallite Volume equation followed in Porod’s law is
Sample (nm) a (Å) b (Å) c (Å) (Å3)
1-Butanol 23.8 5.3887(6) 5.5820(4) 2.5533 (1) 76.805(7) ln[q4I(q)] = ln K. (5)
1-Octanol 21.6 5.4383(6) 5.6206(4) 2.5637(1) 78.364(7)
JCPDS-00-021-0297 5.40300 5.57100 2.54600 76.63 When the system has a diffuse phase boundary with an average
thickness E between two phases, then the scattering is reduced
at high angles. This results in a negative deviation from Porod’s
completion of the reaction, were carried out on a Bruker D8 law.38,39 Under such conditions, Porod’s law modifies to the
Advance diffractometer (Ni-filtered Cu Ka, l = 1.5406 Å). following equation
Normal scans were recorded with a step size of 0.021. Raw data
ln[q4I(q)] = ln K  s2q2 (6)
were subjected to background corrections and Ka2-lines were
stripped off. where the average thickness of the interface layer is given by

(iv) Method of data analysis E = (2P)1/2s (7)

SAXS data were analyzed by using two approaches: (a) a model- For a quasi-two-phase system having sharply defined phase
independent approach to get a prior idea about the shape and size boundaries such that there are small fluctuations of electron
of scattering particles by using a power law and Porod’s law and density between the two-phases, additional scattering is observed
(b) direct fitting of SAXS curves by using the appropriate model. which also results in a positive deviation from Porod’s law,40
The total scattering intensity for monodisperse interacting whereby the equation modifies to
particles as a function of their single-particle form factor P(q) ln[q4I(q)] = ln K + bq2 (8)
and structure factor S(q) can be written as
where b is a constant related to the size of the regions with
I(q) = nmDr2V2P(q)S(q) + Ibkg (1) micro-fluctuations of electron density. When the data are
acquired using the line collimation mode, q3 is used instead
where nm represents the number density of scattering particles,
of q4 because of instrument broadening.41
Dr represents the scattering length density difference between
Model-dependent approach. For performing fitting of SAXS
the particles and the solvent, V is the volume of the particles,
curves, the software suite SASfit (version 0.94.8) was used. The
P(q) is the form factor related to the shape and size of scattering
fitting was based on the fact that during the reaction, the
particles, S(q) is the structure factor, representing inter-particle
system consists of two components: (a) the reverse micelles
interactions between the neighboring scattering particles and
and (b) copper oxalate nanostructures bound by the surfactant
Ibkg is the background intensity.
and/or micelles. In order to account for the polydispersity of
The scattering intensity I(q) was measured as a function of
both systems, the Schultz–Zimm distribution was used.
scattering vector q. To obtain more quantitative information,
The models used for studying the growth of copper oxalate
the net scattering intensity of the particles (Ip(q)) is derived
in the CTAB/1-butanol/iso-octane system were an ellipsoidal
from the observed scattering intensity (I(q)), which is given by
core–shell model for reverse micelles and sphere, ellipsoid and
Ip(q) = I(q)  Ibkg(q) (2) cylindrical models for copper oxalate nanostructures. The ellipsoidal

This journal is © the Owner Societies 2019 Phys. Chem. Chem. Phys., 2019, 21, 336--348 | 345
 View Article Online

PCCP Paper

core–shell model which is used to describe the micelles of CTAB/ Table 5 Scattering length densities of all the components of the
1-butanol/iso-octane/copper oxalate can be expressed by the microemulsion
following equations42 Molecular mass Density Scattering length
ð1 Component (g mol1) (g cm3) density (cm2)
 
IECS ðqÞ ¼ F 2 ðq; mÞ ¼ ½F ðq; mÞ2 dm (9) Iso-octane 114.22 0.69 6.77  1010
0
CTAB 364.45 0.50 4.73  1010
Water 18 0.99 9.35  1010
ð 1 2 1-Butanol 72.12 0.81 7.80  1010
½F ðq; mÞ2 ¼ F ðq; mÞdm (10) 1-Octanol 130.23 0.82 7.96  1010
0

   
3jiðxc Þ 3jiðxt Þ with
F ðq; mÞ ¼ ðZc  Zsh ÞVc þ ðZsh  Zsol ÞVt (11)
xc xt
4
K ðq; R; DZÞ ¼ pR3 DZ3ðsin qR  qR cos qRÞ ððqRÞÞ3 (21)
3
sin x  x cos x
jiðxÞ ¼ (12)
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

x2 where R1 is the overall radius of a spherical shell, R2 is the

radius of the core, Z is the scattering length density difference
pffiffiffiffiffiffiffiffiffi
xc ¼ q a2 m2 þ b2 ð1 þ mÞ2 (13) between the shell and the matrix, and m is the scattering length
density difference between the core and the matrix relative to
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi the shell contrast.
xc ¼ q ða þ tÞ2 m2 þ ðb þ tÞ2 ð1 þ mÞ2 (14)
To account for the inter particle interaction between the
scattering components, a structure factor is introduced to fit
4 the SAXS data. It is important to select an appropriate model
Vc ¼ pab2 (15)
3 for the structure factor which can account for intermicellar
interactions. To account for the interactions between the
4
Vt ¼ pða þ tÞðb þ tÞ2 (16) reverse micelles, we have used the macro-ion model with a
3
local monodisperse approximation. The local monodisperse
where Zc is the scattering length density of the core, Zsh is the approximation is generally used for systems with larger poly-
scattering length density of the shell, Zsol is the scattering length dispersities and higher concentrations.43 In our system, large
density of the solvent, a is the semi-principle axis of the elliptical polydispersity arises because of a large number of scattering
core, b is the equatorial semi-axis of the elliptical core, t is the components viz. reverse micelles and copper oxalate nano-
thickness of the shell, Vc is the volume of the core and Vt is structures. Also, the microemulsion system is formed using
the total volume of the core along with the shell. Similarly, the a high concentration of surfactant, which also justifies the
equations42 used to fit data for a sphere, ellipsoid and cylinder application of the local monodisperse approximation. The
are given as follows choice of macro-ion as a model to incorporate the structure
4 sin qR  qR cos qR factor in the analysis was based on the fact that the micelles
Isphere ¼ pR3 DZ3  (17) formed with CTAB as the surfactant consist of charged surfac-
3 qR
tant molecules, associated counter ions and an aqueous
 2 ð p solution of the reactants (copper nitrate and ammonium
4 3 2 2 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Iellipsoid ¼ pR DZ K ðq; RÞ n 2 cos2 y þ sin2 y  sin ydy oxalate).20 To account for the interaction between copper
3 0 oxalate nanostructures, hard spheres and the cylinder model
(18) with the monodisperse approximation were used.
0 The scattering length densities of all the components of the
0  12 1
qL microemulsion were firstly calculated by the SASfit software
B2 Bsin C
B 2 C
C C (Table 5) using their molecular mass, density, and X-ray energy.
Icylinder ¼ DZ2 L2 B Si ðq; LÞ  B
@ qL A C (19)
@qL A The calculated scattering length density was further used for a
2 model approximation for the fitting of SAXS data.

In the case of the CTAB/1-octanol/iso-octane/copper oxalate

system, the reverse micelles were fitted with the spherical Conclusions
core–shell model and copper oxalate nanoparticles during their
SAXS was used to understand the effect of variation in the chain
growth were fitted with the sphere. The spherical core–shell
length of the co-surfactant on the growth of copper oxalate
model which was used to describe the micelles of CTAB/
nanostructures. The inferences obtained by analyzing the data
1-octanol/iso-octane/copper oxalate can be expressed by the
using a model-independent approach were used to select the
following equation42
model used for fitting the SAXS data. We observed that
Ishell(q, R1, R2, Z, m) = [K(q, R1, DZ)  K(q, R2, DZ(1  m))]2 the chain length of the co-surfactant affects the shape of the
(20) micelles. With 1-butanol (4 carbon atoms), the shape of the

346 | Phys. Chem. Chem. Phys., 2019, 21, 336--348 This journal is © the Owner Societies 2019
 View Article Online

Paper PCCP

micelles was observed to be ellipsoidal in nature whereas a 10 Y. Zhang, F. Lu, D. van der Lelie and O. Gang, Phys. Rev.
spherical shape of the micelles was observed with 1-octanol Lett., 2011, 107, 135701.
(8 carbon atoms) as the co-surfactant. Copper oxalate nano- 11 T. M. Stawski and L. G. Benning, Methods Enzymology, ed.
structures were found to form immediately inside the core of J. J. De Yoreo, Academic Press, 2013, vol. 532, pp. 95–127.
the reverse micelles, after mixing of the reactant microemulsions, 12 M. Harada, N. Tamura and M. Takenaka, J. Phys. Chem. C,
when 1-butanol was used as the co-surfactant, indicated by the 2011, 115, 14081–14092.
positive deviation from Porod’s law. Negative deviation followed 13 C.-H. Lin, J.-H. Chang, Y.-Q. Yeh, S.-H. Wu, Y.-H. Liu and
by positive deviation from Porod’s law with time (for the reaction C.-Y. Mou, Nanoscale, 2015, 7, 9614–9626.
with 1-octanol as the co-surfactant) indicated that nanostructures 14 B. L. Caetano, F. Meneau, C. V. Santilli, S. H. Pulcinelli,
were formed (inside the core of reverse micelles) after 2 h of M. Magnani and V. Briois, Chem. Mater., 2014, 26, 6777–6785.
mixing the microemulsions containing the reactants. Finally, 15 P. Calandra, A. Longo, V. Marcianò and V. Turco Liveri,
based on the fitting of SAXS data, a model for the growth J. Phys. Chem. B, 2003, 107, 6724–6729.
mechanism was proposed. The main interesting observation of 16 J. Polte, R. Erler, A. F. Thünemann, S. Sokolov, T. T. Ahner,
the study is that apart from the change in the rigidity of the K. Rademann, F. Emmerling and R. Kraehnert, ACS Nano,
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

surfactant film, it is the shape of the reverse micelles which 2010, 4, 1076–1082.
affects the shape of the final product i.e. copper oxalate in this 17 A. Schiener, A. Magerl, A. Krach, S. Seifert, H. G. Steinrück,
case. The time-dependent SAXS studies add to the field of J. Zagorac, D. Zahn and R. Weihrich, Nanoscale, 2015, 7,
microemulsions and give an insight into the influence of 11328–11333.
the chain length of the co-surfactant on the growth of nano- 18 P. Calandra, A. Longo, A. Ruggirello and V. Turco Liveri,
structures. This study is an effort to understand the growth J. Phys. Chem. B, 2004, 108, 8260–8268.
mechanism of copper oxalate, though time resolution on an 19 S. Sharma and A. K. Ganguli, J. Phys. Chem. B, 2014, 118,
even shorter time scale could help in increasing the overall 4122–4131.
understanding of the dynamics of the growth process of 20 V. Sethi, J. Mishra, A. Bhattacharyya, D. Sen and A. K.
nanostructured materials using microemulsions. Ganguli, Phys. Chem. Chem. Phys., 2017, 19, 22033–22048.
21 T. Ahmad, K. V. Ramanujachary, S. E. Lofland and
A. K. Ganguli, J. Mater. Chem., 2004, 14, 3406–3410.
Conflicts of interest 22 J. Ahmed, T. Ahmad, K. V. Ramanujachary, S. E. Lofland and
There are no conflicts to declare. A. K. Ganguli, J. Colloid Interface Sci., 2008, 321, 434–441.
23 T. Ahmad, S. Vaidya, N. Sarkar, S. Ghosh and A. K. Ganguli,
Nanotechnology, 2006, 17, 1236.
Acknowledgements 24 T. Ahmad, R. Chopra, K. V. Ramanujachary, S. E. Lofland
and A. K. Ganguli, J. Nanosci. Nanotechnol., 2005, 5,
Sunaina thanks the Council of Scientific and Industrial 1840–1845.
Research (CSIR), Govt. of India for a fellowship. AKG and SV 25 S. Vaidya, P. Rastogi, S. Agarwal, S. K. Gupta, T. Ahmad,
thank the Department of Science and Technology, India and Saha A. M. Antonelli, K. V. Ramanujachary, S. E. Lofland and
Institute of Nuclear Physics, India for facilitating the experiments A. K. Ganguli, J. Phys. Chem. C, 2008, 112, 12610–12615.
at the Indian Beamline, Photon Factory, KEK, Japan. 26 R. Ranjan, S. Vaidya, P. Thaplyal, M. Qamar, J. Ahmed and
A. K. Ganguli, Langmuir, 2009, 25, 6469–6475.
References 27 N. Jongen, P. Bowen, J. Lemaı̂tre, J.-C. Valmalette and
H. Hofmann, J. Colloid Interface Sci., 2000, 226, 189–198.
1 M. P. Pileni, J. Phys. Chem. C, 2007, 111, 9019–9038. 28 M. Rahimi-Nasrabadi, S. M. Pourmortazavi, A. A. Davoudi-
2 J. Eastoe, M. J. Hollamby and L. Hudson, Adv. Colloid Dehaghani, S. S. Hajimirsadeghi and M. M. Zahedi,
Interface Sci., 2006, 128–130, 5–15. CrystEngComm, 2013, 15, 4077–4086.
3 M. A. López-Quintela, C. Tojo, M. C. Blanco, L. Garcı́a Rio 29 B. Kumar, S. Saha, K. Ojha and A. K. Ganguli, Mater. Res.
and J. R. Leis, Curr. Opin. Colloid Interface Sci., 2004, 9, Bull., 2015, 64, 283–287.
264–278. 30 T. Ahmad, R. Chopra, K. V. Ramanujachary, S. E. Lofland
4 I. Capek, Adv. Colloid Interface Sci., 2004, 110, 49–74. and A. K. Ganguli, Solid State Sci., 2005, 7, 891–895.
5 K. Holmberg, J. Colloid Interface Sci., 2004, 274, 355–364. 31 P. Calandra, A. Longo and V. Turco Liveri, J. Phys. Chem. B,
6 V. Uskoković and M. Drofenik, Adv. Colloid Interface Sci., 2003, 107, 25–30.
2007, 133, 23–34. 32 T. Li, A. J. Senesi and B. Lee, Chem. Rev., 2016, 116,
7 D. Lombardo, M. A. Kiselev, S. Magazu and P. Calandra, 11128–11180.
Adv. Condens. Matter Phys., 2015, 2015, 22. 33 Z.-H. Li, Chin. Phys. C, 2013, 37, 108002.
8 A. K. Ganguli, A. Ganguly and S. Vaidya, Chem. Soc. Rev., 34 D. S. Mathew and R.-S. Juang, Sep. Purif. Technol., 2007, 53,
2010, 39, 474–485. 199–215.
9 H.-G. Liao, K. Niu and H. Zheng, Chem. Commun., 2013, 49, 35 M. Li, H. Schnablegger and S. Mann, Nature, 1999, 402, 393.
11720–11727. 36 L. Boldon, F. Laliberte and L. Liu, Nano Rev., 2015, 6, 25661.

This journal is © the Owner Societies 2019 Phys. Chem. Chem. Phys., 2019, 21, 336--348 | 347
 View Article Online

PCCP Paper

37 J. A. Lake, Acta Crystallogr., 1967, 23, 191–194. 41 J. T. Koberstein and R. S. Stein, J. Polym. Sci., Polym. Phys.
38 J. T. Koberstein, B. Morra and R. S. Stein, J. Appl. Crystallogr., Ed., 1983, 21, 2181–2200.
1980, 13, 34–45. 42 J. Kohlbrecher, SASfit: A program for fitting simple structural
39 Z. H. Li, Y. J. Gong, D. Wu, Y. H. Sun, J. Wang, Y. Liu and models to small angle scattering data, https://kur.web.psi.ch/
B. Z. Dong, Microporous Mesoporous Mater., 2001, 46, 75–80. sans1/SANSSoft/sasfit.pdf.
40 N. Cohaut, J. M. Guet and R. Diduszko, Carbon, 1996, 34, 43 I. Bressler, J. Kohlbrecher and A. F. Thunemann, J. Appl.
674–676. Crystallogr., 2015, 48, 1587–1598.
Published on 27 November 2018. Downloaded on 7/16/2019 4:53:14 AM.

348 | Phys. Chem. Chem. Phys., 2019, 21, 336--348 This journal is © the Owner Societies 2019