RSC Advances

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

View Article Online

PAPER View Journal | View Issue

MesoDyn prediction of a pharmaceutical

Cite this: RSC Adv., 2017, 7, 20293
microemulsion self-assembly consistent with
experimental measurements
Open Access Article. Published on 07 April 2017. Downloaded on 11/04/2017 19:50:25.

a
Xin Zhao, Zhengwu Wang,b Shiling Yuan,c Juan Lua and Zhongni Wang*a

In this work, for the first time a coarse-grained model in a mesoscopic dynamic (MesoDyn) simulation
for a pharmaceutical nonionic surfactant Brij97 was presented. By using this coarse-grained model,
the morphology transition from oil in water (O/W) to a bicontinuous microemulsion formed in the
Brij97/isopropanol/isoamyl acetate/H2O quaternary system was predicted. Furthermore, the effect of
Received 7th February 2017
Accepted 23rd March 2017
emulsifier/isoamyl acetate mass ratio, Brij97/alcohol mass ratio and alcohol structure on the droplets
size and kinetic formation process of the O/W microemulsion were studied. These MesoDyn predictions
DOI: 10.1039/c7ra01541k
were consistent with experimental measurements. This makes it possible to further predict the formation
rsc.li/rsc-advances mechanism of complex drug-loaded aggregates of Brij97 by using this coarse-grained model.

in the presence of SDS. The kinetic process of micelles forma-

1 Introduction tion underwent transitions from three to two stages with the
Recently, pharmaceutical surfactant self-assemblies have increase of L64 concentration, which was hard to be observed
aroused extensive interests due to their excellent abilities of from the experiments. The necessity of the shear and the weak
sustained and controlled release for bioactive drugs.1 Polyoxy- charge to the formation of hexagonal templates in the P123
ethylene oleyl ether Brij97, as a typical non-ionic surfactant, has copolymer were conrmed by Yuan, S. et al. through MesoDyn
its advantages in drug delivery on account of the special ability simulations.12
to enhance the permeation of a drug.2 In our previous studies, Microemulsions, usually consisting of surfactant, co-
Brij97 could form various aggregates including micellar,3 surfactant, oil and water, are dened as optically isotropic
microemulsion,4 lamellar,5 hexagonal6,7 and cubic8,9 liquid transparent and thermodynamically stable systems.14 Micro-
crystals. Although the formation conditions of various Brij97- emulsions have been extensive applied in various elds, such as
based aggregates have been shown clearly, the detailed infor- separation,15 chemical reactions,16 nanomaterial preparation,17
mation of the self-assembled process of Brij97 in the oil–water and drug delivery.18 Microemulsions application are largely
interface at the nanometer scale10 is difficult to obtain from dependent on their different structures, namely, oil-in-water (O/
experiments. W), bicontinuous (B.C.), and water-in-oil (W/O) structures.
Aiming to address the limitation of the experiments, However, as far as we know, the effect of components on the
computer simulations have been proven to be valuable tools for aggregation process of microemulsion from the perspective of
providing additional information at molecular11 or mesoscopic phase behaviour and nanostructure19 was rarely reported.
scale.12 In particular, mesoscopic dynamics (MesoDyn) simula- In present work, the coarse-grained model of Brij97 in
tion have become a powerful analytical approach for exploring MesoDyn simulation was presented for the rst time. The
the phase behaviour and dynamic evolution process of polymer morphologies of Brij97/isopropanol/isoamyl acetate/H2O were
or surfactant in solvents at real conditions (pressure and predicted by using the Brij97 model. The effect of water content,
temperature). Li, Y. et al.13 used MesoDyn method to simulate emulsier/isoamyl acetate mass ratio, Brij97/isopropanol mass
the aggregation behaviour of pluronic copolymer L64 solutions ratio and alcohol structure on the morphology and size of
microemulsion were further studied by a combination of Mes-
oDyn prediction and experimental measurements.
a
College of Chemistry, Chemical Engineering and Materials Science, Collaborative
Innovation Center of Functionalized Probes for Chemical Imaging in Universities of
Shandong, Shandong Normal University, Jinan 250014, P. R. China. E-mail:
2 Materials and methods
zhongniw@hotmail.com 2.1 Materials
b
Department of Food Science Technology, Shanghai Jiaotong University, Shanghai
200240, P. R. China Polyoxyethylene-10-oleyl ether (Brij97, purity $ 99%) was
c
Key Laboratory of Colloid and Interface Chemistry, Institute of Theoretical Chemistry, purchased from Sigma Chemical Company. Isopropanol,
Shandong University, Jinan 250100, P. R. China ethanol, 1,2-propanediol and isoamyl acetate were provided by

This journal is © The Royal Society of Chemistry 2017 RSC Adv., 2017, 7, 20293–20299 | 20293
 View Article Online

RSC Advances Paper

Sinopharm Chemical Reagent Co., LTD (Shanghai, China). cA–W and cE–A calculated from the eqn (2) were incorrect. The
Water was deionized and then doubly distilled. All chemicals possible explanation to the trouble is that the hydrogen
were used without further purication. bonding between beads and solvent was neglected.23 Hence, the
effective Flory–Huggins parameters were dened in Baulin and
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

2.2 Simulation method and parameters Halperin's work,24 c  ¼ ceff + f(F). In general, the miscibility
between polymer and solvent are dependent on the interaction
Before employing the MesoDyn simulation, the components of
parameter. Consequently, cE–W used in present study was
our researched system were represented by coarse-grained
dened as 0.3 on account of the hydrophilic properties of
models. In the present MesoDyn simulation, the coarse-
surfactant EO headgroup, and cC–W was estimated to be 5.0 due
grained model of Brij97 as E3C2 was designed and proposed
to the hydrophobic properties of surfactant alkyl chain tail. The
for the rst time, where the E beads represented the hydrophilic
effective Flory–Huggins interaction parameters between various
EO chain and the C beads represented hydrophobic alkyl chain.
Open Access Article. Published on 07 April 2017. Downloaded on 11/04/2017 19:50:25.

beads used in this work were listed in Table 1.

The ratio of beads E to C was dependent on a simple relation-
In MesoDyn simulation, the dimensionless parameters were
ship for atomistic and Gaussian chains12,20 with a few modi-
chosen as follows: the size of cubic grid 32  32  32 nm, the
cations. It is believed that this modications at the mesoscale
bond length 1.1543, the bead diffusion coefficient 107 cm2 s1,
level does not result in a large deviation.21 The molecular
the noise parameter 100.0, the compressibility parameter 10 kT,
structures and coarse-grained models of all components are
the simulated temperature 298 K, the time step 0.5 ns. For each
shown in Fig. 1. As shown, the bead W represented H2O, the
system, the total number steps of 20 000 were carried out to
bead A represented alcohol, and the bead O represented iso-
reach a kinetic equilibrium. All simulations were carried out
amyl acetate in the MesoDyn simulation.
with MesoDyn module in a commercial soware Materials
The interaction among various beads were represented by
Studio 4.4 from Accelrys, Inc,25 successful used to studied
interaction parameters. The solvent–polymer interaction energy
several surfactant22 or polymer26–28 solution systems.
lij are related to the Flory–Huggins interaction parameter cij
through the following equation:22

lij ¼ cijRT (1)

2.3 Construction of phase diagram
The Flory–Huggins interaction parameter cij can be obtained The pseudo-ternary phase diagram was constructed by the
by the expression:21 previous method described by Wang et al.4 Firstly, the surfac-

2 tant (Brij97) and co-surfactant (isopropanol) were premixed at
VS di  dj
cij ¼ þ cS (2) the weight ratio of 1 : 1 (Km) as the emulsier (EM). Secondly,
RT
the mixtures of the EM and isoamyl acetate with the weight
where di and dj are solubility parameters of components i and j, ratios (K) varying from 0 : 10 to 10 : 0 were stirred to be
respectively. VS is the molar volume of solvent. cS is the entropy homogenized at about 50  C. Finally, individual drops of water
contribution to the mixing energy, which is neglected in this phase was added to each mixture under proper magnetic stir-
study because it is a relatively small correction in comparison to ring at 25  C. The microemulsion region was identied by
other terms.21 a clear and transparent appearance of the solution on visual
The Flory–Huggins interaction parameter between the observation. Samples near the phase boundary line were
hydrophobic tail of Brij97 and isoamyl acetate cC–O was esti- equilibrated at 25  C in a thermostatic water bath for 24 hours.
mated to be 0.1 calculated from the eqn (2), consistent with The microstructural transition from W/O, B.C. to O/W in the
similar dissolve mutually theory. However, the values of cE–W, microemulsion region was characterized by electric conduc-
tivity measurement using a model DDSJ-308A digital conduc-
tivity meter equipped with a DJS-1C platinum bright electrode
(Shanghai INESA Scientic Instrument Co., Ltd., Shanghai,
China).

Table 1 The Flory–Huggins interaction parameters between various

beadsa

Bead E C A O W

E 0 1.6 0.8 4.0 0.3

C 1.6 0 1.0 0.1 5.0
A 0.8 1.0 0 0.6 1.2
O 4.0 0.1 0.6 0 5.0
W 0.3 5.0 1.2 5.0 0
Fig. 1 The coarse-grained models of Brij97, alcohols, isoamyl acetate
a
and H2O in the MesoDyn simulation. The molecular structure of Brij97 E, C, A, O and W represent hydrophilic group of Brij97, hydrophobic
is divided into two types of beads: C and E. group of Brij97, alcohol, isoamyl acetate, and H2O, respectively.

20294 | RSC Adv., 2017, 7, 20293–20299 This journal is © The Royal Society of Chemistry 2017
 View Article Online

Paper RSC Advances

2.4 Diameter distributions of microemulsion formulations inspect the distribution of all compositions in the spherical
droplets, a representative droplet marked by red cubic grid was
The diameter distributions of different microemulsion formu-
segregated and analysed the radial density distribution. As
lations were determined using dynamic light scattering (DLS)
shown in Fig. 2(b), the density distribution of O, C and W beads
with a Zetasizer Nano ZS90 particle size analyzer (Malvern
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

presented a single peak, broad peak and valley respectively,

Instruments Ltd., United Kingdom). The data of size distribu-
indicating the isoamyl acetate molecules and hydrophobic tails
tion was collected at the 90 scattering angle and the tempera-
of Brij97 composed the core of spherical droplets without free
ture was kept at 25  0.1  C.
water molecules. For hydrophilic beads E of Brij97, the density
distribution presented two low peaks far from the centre
of sphere, indicating a hydrophilic shell. This is a typical
3 Results and discussion morphology of O/W microemulsion. While for S2 with 60 wt%
Open Access Article. Published on 07 April 2017. Downloaded on 11/04/2017 19:50:25.

3.1 The morphologies of Brij97/isopropanol/isoamyl acetate/ H2O, the E beads isosurface formed several cylindrical chan-
H2O nels. It is similar to a classical bicontinuous structure with
As we all know, the hydrophilic EO chain of Brij97 tends to water tank and oil pipe. The simulation results presented that
contact with water, and the hydrophobic alkyl chain and oil the microemulsion underwent a transition from B.C. to O/W
phase should come to contact each other. Hence, the with water content increasing from 60 to 80 wt%. In order to
morphology of Brij97 self-assemblies could be simulated by the further conrm the prediction obtained by MesoDyn simula-
isosurface of E and C beads, consistent with mostly previous tion, the phase diagram of Brij97/isopropanol/isoamyl acetate/
reports.13,21,29 Here for convenience, the content of each H2O (Km ¼ 1 : 1) system at 25  C was constructed.
component in system was the weight percentage, consistent
with the experiments. While the concentrations used in the
3.2 Phase behaviour of Brij97/isopropanol/isoamyl acetate/
MesoDyn simulation were transformed to the corresponding
water system
volume fraction.22
Fig. 2(a) showed the dependence of simulated morphology of The pseudo-ternary phase diagram of Brij97/isopropanol/
Brij97 assemblies on water content. The components of sample isoamyl acetate/H2O (Km ¼ 1 : 1) system at 25  C was con-
S1 and S2 were listed in Table 2. For S1 with 80 wt% H2O, the E structed and shown in Fig. 3. The phase boundary was easy to
beads isosurface formed several spherical droplets. In order to determine by visual observation. As shown, a transparent one-
phase microemulsion region is marked “1F” in Fig. 3. The
rest of the phase region marked “2F” is a turbid two-phase
region which is a microemulsion in equilibrium with an
excess water or oil phase. The similar phase behaviour was
observed in Brij97/isopropanol/ethyl acetate/H2O (Km ¼ 1 : 1)
system at 25  C, whereas the microemulsion area was enlarged
by replacing isoamyl acetate with ethyl acetate.4 This indicated
that the microemulsion area was closely related to the chain
length of oil molecule, consistent with the previous result.30 It
was obviously found that both the sample S1 and S2 locate in the
microemulsion region marked “1F”.
In order to conrm the prediction of microstructural tran-
sition obtained from MesoDyn simulation, further experi-
mental measurement was necessary. It is well known, by
measuring the variations of electrical conductivity (k) of
microemulsion as a function of the water content, the micro-
emulsion region was divided into three parts: W/O, B.C. and O/
W (see in Fig. 3). A representative electric conductivity curve was
shown in the insert of Fig. 3. It was obviously found that three
different types of microstructure: W/O (below 54.9 wt%),
bicontinuous (B.C., 54.9–71.7 wt%) and O/W (above 71.7 wt%)
occurred in the present system along the water dilution line
a (marked in Fig. 3). Comparing with the experimental phase
diagram and electrical conductivity results, simulation results
can reproduce corresponding microemulsion structure in
certain water content, which indicates that the selection of
Fig. 2 (a) The beads E isosurface representations for microemulsion
coarse-grained models and interaction parameters is effective
formed by Brij97/isopropanol/isoamyl acetate/H2O system with
different mass ratio of water: S1 (80%) and S2 (60%). (b) The radial and reasonable.13
density distributions of all compositions in the spherical droplet In order to further explore the aggregation behaviour of
marked by a red cubic grid in S1. Brij97-based microemulsion, several samples were prepared

This journal is © The Royal Society of Chemistry 2017 RSC Adv., 2017, 7, 20293–20299 | 20295
 View Article Online

RSC Advances Paper

Table 2 The components, size and polydispersity index (PDI) of all microemulsions testeda

Sample K Km Alcohol Water (%) Size (nm) PDI Type

S1 9:1 1:1 Isopropanol 80 14.81 0.150 O/W

This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

S2 9:1 1:1 Isopropanol 60 — — B.C.

S3 8:2 1:1 Isopropanol 80 17.33 0.255 O/W
S4 9:1 3:1 Isopropanol 80 12.30 0.307 O/W
S5 9:1 1:1 Ethanol 80 15.52 0.189 O/W
S6 9:1 1:1 1,2-Propanediol 80 16.02 0.216 O/W
a
K and Km represent the ratio of emulsier/isoamyl acetate and Brij97/alcohol, respectively. The values of size and PDI were obtained by DLS
measurement.
Open Access Article. Published on 07 April 2017. Downloaded on 11/04/2017 19:50:25.

Fig. 3 Pseudo-ternary phase diagram of EM/isoamyl acetate/H2O

system at 25  C. The insert is a representative curve of electrical
conductivity (k) vs. H2O (wt%) along dilution line a.

and investigated by using a combination of MesoDyn prediction

and experiment methods in the following research. The
components and measurement results of all samples tested
were listed in Table 2.

3.3 Effect of emulsier/isoamyl acetate mass ratio (K)

To obtain a more clear understanding on the aggregation
behaviour of Brij97-based microemulsion, sample S3 with
a certain water content of 80 wt% was simulated by changing
the mass ratio of emulsier/isoamyl acetate (K) from 9 : 1 to
8 : 2. The MesoDyn simulation could provide the density Fig. 4 (a) The density slice of beads E, C and O for microemulsion S1
distribution of different beads in microemulsion droplets, and S3. (b) The droplet size distribution of S1 and S3.
which help us to explore the aggregation behaviour. In the cubic
grid, the density slices of beads E, C and O along the x–z plane
were created and divided as depicted in Fig. 4. As shown, high dynamic light scattering. As shown in Fig. 4 and Table 2, the
density was represented by red colour, and the blue colour DLS results showed that the value of both average size and
represents low density. As shown in the Fig. 4, the beads E was polydispersity index (PDI) increased with the K value
distributed in the outer layer of the droplets in contact with the decreased. This is consistent with the MesoDyn prediction
water phase. While the beads C was distributed in the inner that beads E, C and O dispersed over a wider area. Moreover, it
layer of the droplets in contact with the oil phase. These were was obviously found that the droplet size presented in simu-
typical O/W microemulsion structure. Comparing with the lated cubic grid (32  32  32 nm) was slightly smaller than
density slices of S1, it was easy to nd that the density of beads O the DLS result. The similar result was found that the size of
of S2 dispersed over a wider area. This induced beads E and C to microemulsion droplets in electron microscopy (TEM) images
disperse loosely. was also smaller than the DLS results.31 This was due to the
To conrm the MesoDyn prediction, the droplet size size obtained by DLS was hydrodynamic diameter rather than
distribution of microemulsion S1 and S2 were measured by real diameter.

20296 | RSC Adv., 2017, 7, 20293–20299 This journal is © The Royal Society of Chemistry 2017
 View Article Online

Paper RSC Advances

Fig. 5(a) showed the variation of the order parameters of energy density decreased asymptotically to approach a stable
beads W and O as a function of time steps for microemulsion S1 value during the simulations when the system reaches dynamic
and S3, respectively. As we all known, the formation process of equilibrium.32 This indicated that the aggregation process of
microemulsion is a self-assembling evolution from disorder to microemulsion is spontaneous. It is worth mentioning that the
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

order. Hence, the order parameter of beads in system increased comparison of the free energy density values between S1 and S3
along with the aggregation process of microemulsion. It was is totally meaningless because it is not routinely calculated for
clear seen from Fig. 5(a), the formation process of the O/W real systems.33 While in this case, the value of free energy
microemulsion can be divided into three stages. In the rst change (|DG|) before and aer the aggregation could reect the
stage, all of the order parameter of beads are close to zero, stability of microemulsion to some extent. As shown in Fig. 5(b),
indicating that each component in the system is highly the |DG| value of S3 was higher than that of S1, indicating that
dispersed. In the second stage, the order parameter of each the stability of microemulsion increased with the K value
Open Access Article. Published on 07 April 2017. Downloaded on 11/04/2017 19:50:25.

component increased dramatically, which indicated that the decreasing.

ordered aggregates began to form. The third stage is the
formation and stabilization of microemulsion. In this stage, the
3.4 Effect of Brij97/isopropanol mass ratio (Km)
value of order parameter is not constant but waved up and down
in a small range. It is not surprising because the microemulsion In general, the formation and stability of microemulsion are
is a dynamic equilibrium system. closely related to the ratio of surfactant to co-surfactant (Km).30 It
Comparing with the order parameters curves of bead E, C will be meaningful to predict the effect of Km value on the
and A (not presented), the changes of beads W and O were morphology and aggregation behaviour of microemulsion by
obvious. As shown in Fig. 5(a), the time of rst stage for sample MesoDyn simulation. The beads E isosurface representations
S3 was shorter than that of sample S1. This indicated that the of sample S4 was shown in Fig. 6(a). Comparing with the
formation rate of microemulsion accelerated with the K value morphology of microemulsion S1 shown in Fig. 2(a), it is clearly
decreasing. The result may be due to the enhancement of seen that the droplets size of S4 is smaller than that of S1. The
hydrophobic interaction. Besides, the order parameter of iso-
amyl acetate and water increased signicantly. The former may
be result from the fact that more isoamyl acetate molecules
embed into the hydrophobic chains of surfactant to arrange
orderly. The latter may be due to the result that larger droplets
permitted more water molecules to form hydrogen bond with
EO chains of surfactant.
Apart from the size and formation rate of microemulsion,
the practical application of microemulsion is also largely
dependent on the thermodynamically stability. As we all know,
the free energy change of system can reect the stability. The
MesoDyn simulation can provide the free energy change in the
aggregation process of microemulsion, while it is hard to obtain
from experiment. Fig. 5(b) showed the free energy density plots Fig. 6 (a) The beads E isosurface representations for microemulsion
with time step for microemulsions S1 and S3. As shown, the free S4; (b) droplet size distribution of microemulsion S4.

Fig. 5 (a) Order parameter plot of beads W and O with increasing time steps for microemulsion S1 and S3. (b) Free energy density plot with time
step for microemulsions S1 and S3.

This journal is © The Royal Society of Chemistry 2017 RSC Adv., 2017, 7, 20293–20299 | 20297
 View Article Online

RSC Advances Paper

DLS resulted showed that the hydrodynamic diameter was also

decreased from 14.81 nm to 12.30 nm when the Km value
increased from 1 : 1 to 3 : 1. It also can be seen from Fig. 6(a),
several droplets contacted with each other in the simulated
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

morphology of microemulsion S4, resulting in the nonunifor-

mity of microemulsion. This is consistent with the experimental
results that the PDI value of microemulsion S4 is higher than
that of S1.
Fig. 7(a) showed the variation of the order parameters of
beads W, E and C as a function of time steps for microemulsion
S1 and S4, respectively. From the gure we can see, the forma-
Open Access Article. Published on 07 April 2017. Downloaded on 11/04/2017 19:50:25.

tion process of the microemulsion with different Km can also be

divided into three stages. While for S4, the rst stage of
microemulsion formation was remarkably shortened. This is
because that the hydrophobic interaction between various
components increased with alcohol content decreasing.
Besides, the order parameter of surfactant and water increased
signicantly. Comparing with the hydrophilic headgroup E, the
Fig. 8 The E beads isosurface representations (S1, S5, S6), density slice
order parameter of hydrophobic tail C increased more signi- of E (E1, E5, E6) and A (A1, A5, A6) for microemulsion with different
cantly. This indicated that isopropanol could weaken the alcohols: isopropanol, ethanol and 1,2-propanediol.
hydrophobic interaction of surfactant by mainly inserting into
the hydrophobic tails of surfactant. Fig. 7(b) showed the free
energy density plots with time step for microemulsions S1 and Fig. 8 S1, S5 and S6 showed the morphologies of micro-
S4. As shown, the |DG| value of S4 was higher than that of S1, emulsion formed by isopropanol, ethanol and 1,2-propanediol,
indicating that the stability of microemulsion increased when respectively. All microemulsion were typical O/W microstruc-
the Km value increased from 1 : 1 to 3 : 1. ture. It was obviously found that the droplets size increased by
the order of S1, S5 and S6. The MesoDyn predictions were
consistent with the DLS results listed in Table 2. This indicated
that the droplets size were dependent on the polarity of co-
3.5 Effect of alcohol structure surfactant alcohol. In order to understand clearly the mecha-
Alcohol as a co-surfactant have been widely applied in micro- nism of alcohol as a co-surfactant, the density slices of beads E
emulsion. Generally, alcohol can contribute to the formation of and C of microemulsion with different alcohols were obtained.
micelle by the means of changing the hydrophile–lipophile As we can see from A1, A5, and A6 in Fig. 8, the distribution of
balance of surfactant and adjusting the polarity of water and oil alcohol molecules in microemulsion were obviously different.
phase. Usually, the chain length and polarity of alcohol are For microemulsion S1, isopropanol mainly dispersed in oil to
responsible for the hydrophile–lipophile ability and surface improve the polarity of oil. Ethanol mainly dispersed into
activity. Here, MesoDyn simulation was used to predict the surfactant molecules to adjust the hydrophile–lipophile balance
morphologies and distribution of microemulsion formed by for microemulsion S5. For microemulsion S6, 1,2-propanediol
different alcohols. mainly dispersed in water phase to reduce the polarity. The

Fig. 7 (a) Order parameter plot of beads W, C and E with increasing time steps for microemulsion S1 and S4. (b) Free energy density plot with time
step for microemulsions S1 and S4.

20298 | RSC Adv., 2017, 7, 20293–20299 This journal is © The Royal Society of Chemistry 2017
 View Article Online

Paper RSC Advances

different distribution of alcohol molecules may be the reason 11 B. Meng and H. S. Ashbaugh, Langmuir, 2015, 31, 3318–3325.
why the droplets size of microemulsion changes. 12 S.-L. Yuan, X.-Q. Zhang and K.-Y. Chan, Langmuir, 2009, 25,
2034–2045.
4 Conclusions 13 Y. Li, G. Xu, Y. Zhu, Y. Wang and H. Gong, Colloids Surf., A,
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.

2009, 334, 124–130.

In present work, the phase behaviour and mesoscopic structure 14 S. P. Moulik and B. K. Paul, Adv. Colloid Interface Sci., 1998,
of Brij97/isopropanol/isoamyl acetate/water were studied by using 78, 99–195.
a combination of experiment and MesoDyn simulation methods. 15 Z. Lou, C. Guo, X. Feng, S. Zhang, Z. Xing, W. Shan and
The structural transition from B.C. to O/W with water content Y. Xiong, Hydrometallurgy, 2015, 157, 199–206.
increasing was conrmed by electric conductivity measurement 16 I. Capek, Adv. Colloid Interface Sci., 1999, 80, 85–149.
and MesoDyn simulation. At a certain water content of 80 wt%, 17 S. N. Khadzhiev, K. M. Kadiev, G. P. Yampolskaya and
Open Access Article. Published on 07 April 2017. Downloaded on 11/04/2017 19:50:25.

the size of microemulsion droplets increased with the value of K M. K. Kadieva, Adv. Colloid Interface Sci., 2013, 197–198,
or Km decreasing. The MesoDyn simulation results showed that 132–145.
the swelling of microemulsion droplets was due to the broad 18 A. W. Du and M. H. Stenzel, Biomacromolecules, 2014, 15,
dispersion of isoamyl acetate molecules in the hydrophobic 1097–1114.
chains of surfactant. Besides, the incensement of alcohol content 19 D. P. Acharya and P. G. Hartley, Curr. Opin. Colloid Interface
could induce the formation of large droplets. The effect of co- Sci., 2012, 17, 274–280.
surfactant on the formation and stability of microemulsion was 20 B. A. C. van Vlimmeren, N. M. Maurits, A. V. Zvelindovsky,
depended on the hydrophile–lipophile ability of alcohol. G. J. A. Sevink and J. G. E. M. Fraaije, Macromolecules,
1999, 32, 646–656.
Acknowledgements 21 S. Yang, S. Yuan, X. Zhang and Y. Yan, Colloids Surf., A, 2008,
322, 87–96.
Support of this work by the National Natural Science Founda- 22 Y.-m. Li, G.-y. Xu, D. Wu and W.-p. Sui, Eur. Polym. J., 2007,
tion of China (31271933, 31071603) is gratefully acknowledged. 43, 2690–2698.
23 S. L. Guo, T. J. Hou and X. J. Xu, J. Phys. Chem. B, 2002, 106,
References 11397–11403.
24 V. A. Baulin and A. Halperin, Macromolecules, 2002, 35,
1 I. Martiel, N. Baumann, J. J. Vallooran, J. Bergfreund, 6432–6438.
L. Sagalowicz and R. Mezzenga, J. Controlled Release, 2015, 25 Materials Studio Release Notes, Release 4.4, Accelrys Soware
204, 78–84. Inc., San Diego, USA, 2008.
2 S.-C. Shin, C.-W. Cho and I.-J. Oh, Int. J. Pharm., 2001, 222, 26 X. Guo, S. Yuan, S. Yang, K. Lv and S. Yuan, Colloids Surf., A,
199–203. 2011, 384, 212–218.
3 Z. Wang, T. Wu, W. Zhou, X. Wei and J. Zhao, J. Surfactants 27 C. Do, W.-R. Chen, K. Hong and G. S. Smith, Phys. B, 2013,
Deterg., 2010, 14, 391–400. 430, 87–94.
4 Z. Wang, F. Guo, J. Lu, L. Wei and X. Liu, Adv. Mater. Res., 28 B. Zhang, R. Liu, J. Zhang, B. Liu and J. He, Comput. Theor.
2014, 924, 10–17. Chem., 2016, 1091, 8–17.
5 Z. Wang and W. Zhou, J. Solution Chem., 2009, 38, 659–668. 29 B. Zhang, R. Liu, J. Zhang, B. Liu and J. He, Comput. Theor.
6 Z. Wang, F. Liu, Y. Gao, W. Zhuang, L. Xu, B. Han, G. Li and Chem., 2016, 1091, 8–17.
G. Zhang, Langmuir, 2005, 21, 4931–4937. 30 R. X. Guo, X. Fu, J. Chen, L. Zhou and G. Chen, J. Agric. Food
7 X. Liu, Z. Wang, L. Wei and X. Wei, J. Dispersion Sci. Technol., Chem., 2016, 64, 6286–6294.
2014, 35, 1699–1708. 31 X. Zhao, Y. Yao, K. Tian, T. Zhou, X. Jia, J. Li and P. Liu, Eur. J.
8 J. Fan, F. Liu and Z. Wang, Int. J. Pharm., 2016, 497, 248–254. Pharm. Biopharm., 2016, 108, 91–99.
9 L. Zhang, Z. Wang, J. Fan, X. Zhao and M. Wang, J. Drug 32 S. S. Jawalkar and T. M. Aminabhavi, Polymer, 2006, 47,
Delivery Sci. Technol., 2016, 32, 49–55. 8061–8071.
10 Y. Wang, D. D. Zhu, J. Zhou, Q. L. Wang, C. Y. Zhang, 33 Y. Fu, L. Liao, Y. Lan, L. Yang, L. Mei, Y. Liu and S. Hu, J. Mol.
Y. J. Liu, Z. M. Wu and X. D. Guo, Colloids Surf., B, 2015, Struct., 2012, 1012, 113–118.
136, 536–544.

This journal is © The Royal Society of Chemistry 2017 RSC Adv., 2017, 7, 20293–20299 | 20299