Synthesis of magnetic polystyrene nanoparticles using amphiphilic ionic liquid stabilized RAFT mediated miniemulsion polymerization

Article abstract of DOI:10.1021/ma5008013

Imidazole based amphiphilic ionic liquids (ILs) were used as surfactants in miniemulsion polymerization (MEP) of styrene using a free radical process as well as reversible addition-fragmentation chain transfer (RAFT). Monodisperse polystyrene (PS) nanoparticles were obtained, demonstrating the efficiency of the amphiphilic IL as surfactant in MEP. IL stabilized miniemulsion was furthermore used to prepare polystyrene based magnetic nanoparticles (MNP). A large increase of the possible MNP content associated with very good colloidal stability was achieved using IL stabilized RAFT mediated MEP where a carboxyl functionalized chain transfer agent (CTA) was applied, allowing interaction with the MNP surface. The molecular weight and dispersity index of polystyrene, the content of MNP, and the morphologies of the hybrid nanoparticles were controlled by proper optimization of the concentration of initiator and CTA. The materials have been analyzed by NMR, GPC, DLS, SEM, TEM, and TGA. Finally, the magnetic properties of the materials were determined by vibrating sample magnetometer (VSM) analysis.

Full text of DOI:10.1021/ma5008013

Article
pubs.acs.org/Macromolecules
Synthesis of Magnetic Polystyrene Nanoparticles Using Amphiphilic
Ionic Liquid Stabilized RAFT Mediated Miniemulsion Polymerization
Sourav Chakraborty,† Klaus Jah
,‡
nichen, Hartmut Komber, Ahmed A. Basfar, and Brigitte Voit*
†
†
§
,†,‡
̈
†
Leibniz-Institut fu
̈
r Polymerforschung Dresden e.V., Hohe Straße 6, 01069 Dresden, Germany
Organic Chemistry of Polymers, Technische Universitat Dresden, 01062 Dresden, Germany
King Abdulaziz City for Science and Technology, P.O. Box 6086, Riyadh 11442, Saudi Arabia
S Supporting Information
‡
̈
§
*
ABSTRACT: Imidazole based amphiphilic ionic liquids (ILs)
were used as surfactants in miniemulsion polymerization
(
MEP) of styrene using a free radical process as well as
reversible addition−fragmentation chain transfer (RAFT).
Monodisperse polystyrene (PS) nanoparticles were obtained,
demonstrating the efficiency of the amphiphilic IL as surfactant
in MEP. IL stabilized miniemulsion was furthermore used to
prepare polystyrene based magnetic nanoparticles (MNP). A
large increase of the possible MNP content associated with
very good colloidal stability was achieved using IL stabilized
RAFT mediated MEP where a carboxyl functionalized chain
transfer agent (CTA) was applied, allowing interaction with
the MNP surface. The molecular weight and dispersity index
of polystyrene, the content of MNP, and the morphologies of the hybrid nanoparticles were controlled by proper optimization of
the concentration of initiator and CTA. The materials have been analyzed by NMR, GPC, DLS, SEM, TEM, and TGA. Finally,
the magnetic properties of the materials were determined by vibrating sample magnetometer (VSM) analysis.
INTRODUCTION
more than 20 wt % of MNP with respect to the polymer could
15
■
be encapsulated. Landfester et al. reported a strategy of
multistep miniemulsion to encapsulate up to 40% MNP within
the PS particles. Irrespective of the number of steps used for a
successful encapsulation using MEP, several other parameters
like nature of initiator, concentration of surfactant, and surface
modifier of MNP are also quite important for deciding the
characteristics of PMC nanoparticles. Surface modification of
hydrophilic MNP (e.g., Fe O ) is one of the essential
Polymer magnetic composite (PMC) nanoparticles have raised
a great deal of interest due to their potential use in severa1l
biomedical applications like magnetic resonance imaging,
2
3
nucleic acid purification, enzyme immobilization, drug
4
delivery, etc. The magnetic nanomaterials in the colloidal
5
range have been successfully utilized in both therapeutic and
6
diagnostic applications. The major challenge so far was the
3
4
selection of suitable materials based on their various properties
like particle size distribution, colloidal stability, content of
MNP, presence of functionality, and most importantly toxicity.
In the recent years, focus has also been turned on the synthesis
of asymmetric polymer superparamagnetic composites for use
as potential materials for applications in multimodal probes and
requirements in this process. So different kinds of modi-
16−20
fiers
have been used to modify the hydrophilic surface of
Fe O into a hydrophobized one, with oleic acid (OA) being
3
4
the most frequently used one. Besides the surface modifiers,
initiator used for the polymerization also plays an important
role to decide the location of MNP in the polymer particles.
7
,8
biosensors. Besides the biomedical applications, there are
21
Kawaguchi et al. analyzed the effect of initiator concentration
as well as its solubility on the morphological characteristics of
PMC nanoparticles. A complete encapsulation of MNP was
observed when water-soluble initiator was used while surface
localization of MNP on the polymer particles was observed in
several demands for PMC nanoparticles in the production of
9
toner materials and also for the support and easy separation of
10
catalyst in chemical reactions. The above facts necessitate a
fruitful technique to produce PMC nanoparticles in dispersion
or as solid materials. Several methods have been developed
22
11
the case of water-insoluble initiator. Recently, Elaissari et al.
which include microfluidic-based synthesis, selective surface
1
2
prepared anisotropic Janus magnetic polymer nanoparticles
using styrene and acrylic acid as monomer and AIBN as sole
modification, and different kinds of heterogeneous polymer-
1
3,14
ization.
Among all these methods, miniemulsion polymer-
ization (MEP) has been proven to be one of the most effective
methods to prepare such materials. Regarding the synthesis of
PMC nanoparticles using MEP, most of the work in the
literature has proposed a single-step MEP process although no
Received: April 16, 2014
Revised: June 12, 2014
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
initiator. The MNP were found to be accumulated on one side
of the PS surface, thereby producing the Janus morphology.
The influence of the concentration of surfactant has also been a
matter of investigation in the encapsulation of MNP using
ionic one has been reported to have better control over the
29
stability of RAFT mediated MEP. However, a successful
RAFT mediated MEP was reported recently by Zetterlund et al.
using dioctyl sodium sulfosuccinate as an anionic surfactant
which displayed good control and livingness together with good
20
MEP. Focarda et al. reported the effect of surfactant
concentration on the morphology of PMC nanoparticles. The
distribution of MNP among the PS droplets was more
homogeneous with reduction of the concentration of surfactant.
In the absence of surfactant a core−shell morphology, with
MNP as core and PS as shell material, was obtained with
significant reduction in the formation of pure PS particles. The
concentration of surfactant is indeed a very important
parameter to avoid the formation of pure polymer nanoparticles
resulting from homogeneous and micellar nucleation. A
surfactant-free MEP of styrene using sodium p-styrenesulfonate
30
colloidal stability.
The objective of the present study is first to investigate the
influence of amphiphilic IL as surfactant in free radical MEP as
well as in RAFT mediated MEP and thereby to establish a
fruitful environment for the synthesis of PMC nanoparticles.
Finally, a detailed investigation of the characteristics of PMC
nanoparticles along with the influence of RAFT mediated
polymerization on their colloidal stability, morphology,
composition, molecular weight distribution, and magnetic
property will be performed.
2
3
(
NaSS) as self-stabilizing comonomer showed complete
absence of secondary particles in the encapsulation of MNP
by PS. So far most of the work was emphasized on using SDS as
surfactant or on the surfactant-free MEP using NaSS, acrylic
acid, or dextran as emulsion stabilizer.
EXPERIMENTAL SECTION
■
Materials and Characterization. 1-Methylimidazole, 1-bromo-
hexadecane, 1-bromododecane, iron(II) chloride tetrahydrate, iron-
III) chloride hexahydrate, oleic acid, 28−30% aqueous ammonia,
carbon disulfide, and tetrabutylammonium hydrogen sulfate were all
purchased from Aldrich and used as received. Sodium hydroxide
NaOH) pellets (Aldrich) were used to prepare 50% NaOH solution.
Chloroform (Fischer Chemicals), acetone (Merck), concentrated
hydrochloric acid (Merck), petroleum ether with boiling point 40−60
°C (Merck), hexadecane (Fluka), and AIBN (Fluka, 98%) were all
used as received. Styrene was twice extracted with 0.1 N NaOH,
followed by washing with water and brine in order to remove the
stabilizer. Prior to use, it was heated over calcium hydride (Fluka) to
(
Through the past two decades ILs have attracted a great deal
of interest among the researchers due to its amazing properties
like wide liquid range, excellent thermal stability, high ionic
conductivity, low volatility, and ease of modification of its
chemical structure. Among the different structural composition
of ILs, amphiphilic ILs have become quite popular in the field
(
2
4
́
of heterogeneous polymerization system. Biczok et al.
reported the ability of micelle formation of the alkylimidazole
based ILs in water. The critical micelle concentration (CMC)
could also be varied by changing the type of alkyl chain or its
length. This kind of surfactant-like behavior of alkylimidazolium
salts opened up the possibility of using such materials as
stabilizer in different heterogeneous polymerization systems.
8
0 °C under vacuum until the evolution of hydrogen disappeared and
distilled under vacuum. It was stored under nitrogen in a refrigerator.
The particle size measurements were performed with a Zetasizer
NANO S (Malvern Instruments, UK) at a fixed scattering angle of
173°. The dynamic light scattering (DLS) measurements were
performed in 0.01 N sodium chloride solution. 250 mg of the
dispersion was weighed, and ca. 20 g of 0.01 N NaCl solution was
added. The given values are the zaverage (intensity based). The error of
the measurements is about 5%.
The zeta potential of the as-synthesized PS latex was determined by
means of electrophoretic mobility measurement, performed using a
Zetasizer nano ZS (Malvern Instruments, UK). 50 mg of the as-
synthesized PS latex was mixed with 20 g of 0.01 N NaCl solution. For
the purpose of the present study to have information about surface
charge of PS particles as well as their stability, only one sample was
investigated at the original pH of the dispersion.
25
Guerrero-Sanchez et al. reported the suspension polymer-
ization of styrene using ILs as stabilizer. Microemulsion
polymerization was also conducted successfully using 1-n-
dodecyl-3-methylimidazolium bromide as surfactant as reported
26
by Yan and Texter. In recent past, microwave-assisted
emulsion polymerization has been performed using 1-n-
2
7
dodecyl-3-methylimidazolium chloride as surfactant. To the
best of our knowledge, ILs have not been used as surfactant in
MEP so far. In the present study, the influence of imidazole
based amphiphilic ILs as surfactant in MEP and for the
synthesis of PMC nanoparticles was investigated.
Gel permeation chromatography (GPC) measurements were
performed with an apparatus of the Agilent Series 1100 (RI detection,
So far the major challenge in the preparation of PMC
nanoparticles was to prepare high magnetic content materials
using one-step MEP and also to avoid the formation of pure
polymer particles. The loss of MNP during MEP due to
insufficient encapsulation can only be reduced if the polymer
droplets somehow attract the MNP through chemical or
physical interaction. Inspired from the structural features of
different surface modifiers used for MNP herein, it was aimed
to synthesize a carboxyl functionalized polymer which was
expected to keep the MNP attached to the polymer droplets. In
the context of functional polymer synthesis, RAFT polymer-
ization has become quite popular as different end functionality
was possible to be achieved through the proper choice of CTA
and its chemical conversion. In the field of MEP, RAFT
polymerization was performed with some difficulties in
colloidal stability and molecular weight distribution which
were supposed to arise due to the superswelling effect caused
by the presence of large amount of oligomers in the initial stage
1
PL_MIXED-B-LS-column [7.5 × 300 mm] and 10 mm PS gel
Agilent column, chloroform 1.0 mL/min). PS was used as standard.
This was the standard method for all of the samples. The samples
containing MNP were filtrated to remove the MNP before the
analysis.
Scanning electron microscopy (SEM) investigations were per-
formed with an Ultra 55 plus (Zeiss). 250 mg of the latex was mixed
with 20 g of water to prepare the samples. One drop was placed on a
C-pad or a wafer. After air drying the samples were sputtered with 3
nm Pt.
Thermogravimetric analysis (TGA) was performed using a TGA
Q5000 (TA Instruments). After 5 min isothermal at room temperature
the temperature was increased up to 800 °C with 10 K/min under
nitrogen.
Transmission electron microscope (TEM) analysis was performed
by mixing ca. 50 mg of the dispersion with ca. 20 g of water. The
sample specimen was prepared by taking 2 μL of the prepared solution
on carbon coated TEM copper grid. After air drying the sample was
investigated by TEM LIBRA 200 (Carl-Zeiss SMT, Oberkochen,
Germany) working with acceleration voltage of 200 kV.
2
8
of polymerization. Using nonionic surfactant instead of the
B
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
3
NMR spectra were recorded on an Advance III 500 NMR
spectrometer (Bruker Biospin, Germany) operating at 500.13 MHz
1.2 (26H, H , H , H , and (CH ) ), 0.84 ppm (t, 3H, H ); δ ( C)
7 9 10 2 10 11
137.62 (C ), 123.41 (C ), 121.67 (C ), 50.17 (C ), 36.74 (C ), 31.84
(C ), 30.24 (C ), 29.7−28.9 (CH ) ), 26.71 (C ), 22.60 (C ), 14.03
ppm (C ). IL2: δ( H) 10.31 (t, 1H, H ), 7.52 (t, 1H, H ), 7.37 (t, 1H,
1
2
3
5
4
1
13
for H and 125.77 MHz for C. CDCl was used as solvent, lock, and
3
9
6
2
10
7
10
1
13
1
standard (δ( H) = 7.26 ppm; δ( C) = 77.0 ppm).
11 1 2
The magnetic properties of the materials were investigated using
vibrating sample magnetometer (VSM) analysis. A Physical Property
Measurement System (PPMS) (Quantum Design) was used which
supplies 9 T field through a superconducting coil and operates with a
vibrating sample magnetometer (Quantum Design) in the temperature
range from 2 to 400 K. The dried sample was fixed in a closed
container using a quick fix adhesive so that it cannot move inside the
container. Then it was attached with the sample holder via an adaptor
for the VSM measurement.
H ), 4.29 (t, 2H, H ), 4.10 (s, 3H, H ), 1.88 (m, 2H, H ), 1.4−1.2
13
3
5
4
6
(18H, H , H , H , and (CH ) ), 0.84 ppm (t, 3H, H ); δ( C)
7
9
10
2
6
11
137.48 (C ), 123.49 (C ), 121.73 (C ), 50.13 (C ), 36.72 (C ), 31.80
1
2
3
5
4
(
C ), 30.22 (C ), 29.5−28.8 (CH ) ), 26.19 (C ), 22.57 (C ), 14.00
9
6
2
6
7
10
ppm (C ).
11
The H and 13C NMR spectra of both ILs are shown in Figure SI2.
Synthesis of Oleic Acid Coated MNP Using the Coprecipitation
Method. OA coated MNP were synthesized according to the
1
2
0
coprecipitation method followed by Focarda et al. with a little
Experiments. Synthesis of 2,2′-[Carbonothiobis(thio)]bis(2-
modification. Briefly, 10.8 g of FeCl ·6H O and 3.9 g of FeCl ·4H O
3
2
2
2
methylpropionic acid) (CTA). Synthesis of CTA was performed
were dissolved in 200 mL of deionized water. Then 3.5 g of OA was
dissolved in 120 mL of acetone, and the above two solutions were
mixed under stirring using a mechanical stirrer at 500 rpm under an
3
1
following the literature. Briefly, carbon disulfide (3.046 g),
chloroform (11.938 g), acetone (5.808 g), tetrabutylammonium
hydrogen sulfate (0.268 g), and 15 mL of petroleum benzene were
mixed together in a double jacketed reactor which was cooled with tap
water under an argon atmosphere. A thermometer was inserted into
the reactor to observe the temperature of the reaction mixture. 50%
aqueous NaOH (22.397 g) was added dropwise to the reaction
mixture for 30 min (care has to be taken so that the temperature does
not exceed 25 °C). After complete addition of sodium hydroxide
solution, the reaction was carried out for 12 h. A yellow solid product
was observed which was dissolved by the addition of 100 mL of water.
Then the aqueous layer was acidified by 20 mL of concentrated HCl
with stirring for 30 min under an argon atmosphere. The solid product
thus obtained was filtered and washed with distilled water and then
dried until constant weight. Figure 1 shows the chemical structure of
argon atmosphere. After 30 min, 25 mL of 28−30 wt % NH OH
4
solution was added over a period of 10−15 min. The resulting
suspension was stirred for 1 h at room temperature and then heated at
8
5 °C for 1 h more. Temperature was raised to 110 °C and kept for 1
h more to remove excess ammonia. Then it was cooled to room
temperature. The black particles were separated using a magnet from
outside, and the supernatant liquid was decanted. The particles were
washed several times with distilled water until the washed liquid comes
to the pH range ∼7−8. Then the particles were washed with methanol
to remove excess OA, which is not adsorbed onto the surface of MNP.
Finally the particles were dried in vacuum oven at 50 °C until constant
weight.
Miniemulsion Polymerization Using IL as Surfactant. The
aqueous phase was prepared first by dissolving IL in water. Then
the organic phase was prepared by mixing purified styrene,
hexadecane, AIBN, and RAFT agent (in the case of RAFT mediated
MEP) until the solid particle dissolves. Both the phases were mixed,
and the mixture was degassed 5 times using the Ar/vacuum cycle.
Then the mixture was stirred at 600 rpm using a glass stirrer under an
argon atmosphere for 1 h to prepare the pre-emulsion. Then the
miniemulsion was prepared by sonication of the pre-emulsion for 10
min (duty cycle 90%) with an ultrasonic disintegrator Branson 450 W
using a 1/2 in. minitip under inert atmosphere. During the sonication
a cooling of the reaction vessel by ice water was performed in order to
avoid a heating of the mixture. Finally, the prepared miniemulsion was
polymerized at 70 °C under an Ar atmosphere using a mechanical
stirring at 400 rpm. The detailed recipe is shown in Table 1.
Synthesis of PMC Nanoparticles Using IL Stabilized Miniemul-
sion. The organic phase was prepared by mixing styrene, hexadecane,
OA coated MNP, and CTA (in the case of RAFT mediated MEP)
followed by sonication for 1 min in an ultrasound bath at room
temperature. AIBN was added after sonication. The aqueous phase was
prepared by mixing IL with water and mixed with organic phase. After
mixing, the procedure described previously was repeated using the
recipe described in Table 1.
Removal of Coagulum (If Any). After the polymerization, the
formed dispersion was poured through a mesh (pore size 20 μm) and
then used for the analytical investigations. Finally, the rest in the mesh
and the rests from the stirrer and the vessel were transferred into a frit
using water. The coagulum was washed with water and dried in a
vacuum at room temperature in order to determine the quantity of
coagulum.
Isolation of Solid Particles for Characterization. About 2 g of the
final dispersion was weighed in a Petri dish and kept overnight at room
temperature. The air-dried products were further dried in a vacuum at
room temperature until the weight was constant, and then the solid
Figure 1. Structures of carboxyl functionalized CTA and surfactant ILs
used in this study.
1
13
CTA. The H and C spectra of CTA are shown in Figure SI1. NMR
1
13
(
CDCl ): δ( H) 1.71 ppm (s, H ); δ( C) 217.2 (C ), 180.0 (C ),
3 3 4 1
5
5.8 (C ), 25.2 ppm (C ); mp 172−175 °C.
2
3
Synthesis of Amphiphilic ILs. Synthesis of ILs was performed
32
according to the method reported by Yao et al. Briefly, N-
methylimidazole (0.125 mol) and 1-bromoalkane (0.15 mol) were
added into a 250 mL three-necked round-bottomed flask. Isopropanol
(
2
20 mL) was added to reduce the viscosity. The mixture was stirred for
0 h under reflux at 70 °C. After removing isopropanol, the product
was dissolved in 50 mL of water, and the excess materials were
extracted 3 times with 25 mL of ethyl acetate each. The water was
evaporated at 40 °C using a rotational evaporator. The product was
then dried in a vacuum oven at 40 °C until constant weight. The
structure of ILs is shown in Figure 1.
content was calculated. P O10 was used as drying agent in the vacuum
4
oven. The dried product was used for characterization.
1
-n-Hexadecyl-3-methylimidazolium bromide (IL 1); mp 62−65
Calculation of Monomer Conversion. The conversion of monomer
was determined by the gravimetric method considering the solid
content of the dispersion. In the case of PMC nanoparticles,
considering the contribution of MNP in solid content of the
°
C; 1-n-dodecyl-3-methylimidazolium bromide (IL 2); mp 43−46 °C.
NMR (CDCl ): IL1: δ ( H) 10.32 (t, 1H, H ), 7.55 (t, 1H, H ), 7.38
t, 1H, H ), 4.28 (t, 2H, H ), 4.10 (s, 3H, H ), 1.88 (m, 2H, H ), 1.4−
1
3
1
2
(
3
5
4
6
C
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
a
Table 1. Experimental Details for the Synthesis of PS and
PMC Nanoparticles
MNP by OA was performed using the coprecipitation
20
method. The content of OA on the MNP surface was
investigated by TGA analysis, as shown in Figure 2a.
Considering the residue in the TGA curve as the weight
fraction of the inorganic substance (here MNP), the content of
OA was estimated to be about 22 wt %. The weight loss curve
of pure OA showed a Tmax of 272 °C. In the TGA curve of OA
modified MNP two ranges of weight loss were found. The small
weight loss in the first range between 166 and 266 °C might
reflect the liberation of residual free OA with a Tmax1 at 211 °C,
whereas a significant weight loss in the second range between
b
MNP
AIBN
(g)
CTA
(g)
IL1
(g)
IL2
(g)
react time
(h)
sample
(wt %)
c
PS 1
0.130
0.130
0.022
0.022
0.022
0.130
0.130
0.130
0.130
0.022
0.022
0.022
0.130
0.130
0.10
6
6
c
PS 2
0.09
0.09
PS 3
0.076
0.113
0.076
0.10
0.10
14
14
14
8
PS 4
PS 5
PMC 1
PMC 2
PMC 3
PMC 4
PMC 5
4
8
0.10
0.10
0.10
8
2
90 and 440 °C with a Tmax at 333 °C was observed which is
12
8
8
possibly due to the liberation/decomposition of OA attached to
the surface of MNP. The magnetization curve of OA coated
MNP is shown in Figure 2b.The saturation magnetization was
observed to be 53 emu/g.
0.09
8
8
0.038
0.076
0.150
0.076
0.076
0.10
0.10
0.10
0.10
14
14
14
20
20
d
PMC 6
PMC 7
PMC 8
PMC 9
8
8
The hydrophobization of the surface was confirmed by a
comparative MNP dispersion analysis between water and
toluene as medium (Figure SI 3a). The MNP were easily
dispersed in toluene just by manual shaking whereas in water it
was not possible to disperse even with the help of sonication.
The morphology of the MNP was analyzed by TEM using the
MNP dispersed in toluene. The particle size was mostly around
10−20 nm. A relatively broad distribution of particle size was
observed, which has to be expected in the case of the
coprecipitation method. The MNP were observed to appear as
aggregates of dark spheres up to 28−42 nm surrounded by the
attached OA layer appearing gray. The TEM image of OA
coated MNP is shown in (Figure SI 3b).
8
8
0.09
a
7
.98 g of styrene, 36.0 g of water, and 0.575 g of hexadecane were
used in all cases. The sonication time and polymerization temperature
were fixed to 10 min and 70 °C, respectively (exceptions PS 1 and PS
). With respect to monomer. A series of investigations were
performed using this recipe to investigate the influence of the
concentration of surfactant and sonication time duration on particle
size of PS. Polymerization was continued up to 40 h to investigate the
influence of monomer conversion on the final MNP content of the
b
c
2
d
material.
dispersion, we have calculated the conversion of monomer using the
equation
Miniemulsion Polymerization Using Amphiphilic ILs
as Surfactant. The surfactant behavior of amphiphilic ILs has
not been explored in the field of miniemulsion so far. So we
aimed to establish an IL stabilized miniemulsion process which
can produce monodisperse polymer nanoparticles based on
polystyrene. From the chemical structure of amphiphilic ILs
used in this study, 1-n-hexadecyl-3-methylimidazolium bromide
(IL 1) and 1-n-dodecyl-3-methylimidazolium bromide (IL 2)
(Figure 1), they were expected to behave like a cationic
surfactant, and thus the PS particles should have a positive
charge on their surface. Zeta potential determination of PS latex
at the original pH (6.4) of the latex confirmed the presence of
positive charge on the surface of PS particles. On the other
S − M − V
%
conversion =
× 100%
L × W
where S represents the weight of dried sample, M is the theoretical
weight of MNP in the dried sample, V represents the theoretical
weight of the nonvolatile, chemically nonreactive component of the
recipe in the dried sample, L is the weight of the latex taken for
determination of monomer conversion, and W is the weight fraction of
monomer in the recipe.
RESULTS AND DISCUSSION
Synthesis of Oleic Acid Coated MNP. In order to
disperse MNP in styrene, hydrophobization of the surface of
■
Figure 2. Thermal and magnetic properties of OA modified MNP: (a) TGA curve of pure OA and OA modified MNP and (b) magnetization curve
of OA modified MNP.
D
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 3. Dependence of particle size on the recipe and the sonication conditions: (a) influence of surfactant to monomer molar ratio on average
particle size and (b) influence of sonication time during miniemulsion on average particle size. In both cases (a) and (b), recipe was followed as given
for PS 1 and PS 2 in Table 1.
Figure 4. SEM image of PS particles synthesized by (a) IL 1 (PS 1) and (b) IL 2 (PS 2) stabilized MEP. Sonication time was performed for 10 min
in both cases.
hand, the value of zeta potential (54 ± 0.8 mV) also indicated a
high stability of the PS latex. The influence of IL concentration
on the particle size (DLS) was observed to follow a similar kind
of behavior to that of a conventional anionic surfactant like SDS
as well as of cetyltrimethylammonium bromide (CTAB) as an
example for a cationic surfactant used in MEP. In the
Supporting Information, experimental details and results of
IL, SDS, and CTBA stabilized MEP under identical conditions
are presented, proving that ILs are as effective as surfactants in
MEP as SDS and even more effective in stabilizing and
controlling the size of polymer particles in MEP than CTAB
performed and is shown in Figure 3a. The pattern of the curves
did not show any dissimilarity between two ILs; moreover, the
average particle size and dispersion index value were quite close
in both cases. It indicates that the occurrence of micellar
nucleation in the case of IL 1 stabilized MEP was almost
negligible; otherwise, it would produce a much broader
distribution in the particle size analysis. In the miniemulsion
system the creation of droplets as well as their size can be
influenced on changing the sonication time, and thereby we
investigated the influence of sonication time on particle size
distribution of the PS latex. A gradual decrease in the average
particle size was observed in both systems with increasing
sonication time, as shown in Figure 3b. Antonietti et al.
(see Supporting Information, Figures SI 8 and SI 9). The
particles size shows in all cases a tendency to decrease with
increasing the amount of surfactant. In an ideal scenario for
MEP, the concentration of surfactant should be below the
CMC. In the present system, the surfactant IL 1 has a very low
33
reported a similar kind of behavior with other surfactants.
This also indicates the predominant droplet nucleation in both
the systems. The particle size became nearly constant after 10
min of US for both the cases which is attributed to the lowest
particle size that was possible to achieve using the mentioned
recipe and conditions.
2
4
CMC (0.61 mM), and therefore, it had to be used in MEP at
a concentration higher than its CMC, since it was impossible to
produce a stable miniemulsion at a concentration below its
CMC value. In order to reduce the chance for micellar
nucleation as much as possible, we used another imidazole
based IL surfactant with a dodecyl hydrocarbon chain, 1-
dodecyl-3-methyl imidazolium bromide (IL 2), which has a
The morphology of PS particles was investigated using SEM
(
Figure 4). Monodisperse particles were observed in both
systems with an average diameter of 120 nm down to 80 nm,
depending on the surfactant concentration and sonication time.
In some cases, very few particles in the order of 300−400 nm
were observed. But considering the low fraction of such big
particles, it is almost negligible compared to the dominating
fraction of smaller PS particles.
24
CMC value of 9.8 mM, much higher than that of IL 1. Thus, a
stable MEP was possible to conduct at a much lower
concentration of IL2 compared to its CMC. A comparative
study between two surfactants regarding the influence of
surfactant concentration on particle size of PS latex was
E
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 5. Stability of latexes after dilution (a) immediately and (b) 3 days after polymerization.
Synthesis of PMC Nanoparticles Using IL Stabilized
MEP. For the preparation of PMC nanoparticles, at first, three
experiments were performed with IL 1 as surfactant using
different amounts of feed MNP content. The colloidal stability
was observed by keeping the diluted latex for long duration.
The dilution was made at the same concentration as for the
DLS study (without addition of electrolyte). Figure 5 indicates
a gradual decrease in the stability of dispersion with increasing
amount of feed MNP. The experiment with 12 wt % feed MNP
the immediate precipitation of PMC nanoparticles in the case
of sample PMC 3, DLS could not be performed.
In order to check the dispersion scenario using IL 2 as
surfactant, an experiment (PMC 4) with 8 wt % feed MNP
content was performed which showed similar latex stability as
that of PMC 2. The DLS analysis (Table 2) also displayed very
similar results regarding the particle size distribution. Thus, one
can assume again that there is no high influence from the
micellar nucleation in the case of IL 1 stabilized MEP;
otherwise that would result in broader particle size distribution.
Apart from the instability in the dispersion of PMC 3, the
polymerization eventually ended up producing a high amount
of coagulum (∼35%). The final MNP content of PMC
nanoparticles was investigated using TGA. The results are
summarized in Table 2.
(
PMC 3) showed very poor stability that was indicated by an
immediate precipitation of the PMC nanoparticles from the
dispersion (magnetism of the particles was confirmed by
attraction toward an external magnet, as shown in Figure 6a).
In the case of PMC 1, the final MNP content fitted
reasonably well with the feed MNP content. Considering the
monomer conversion of 95% along with 5 wt % coagulum the
reduction of ca. 0.8 wt % MNP content in the final composite
material was satisfactory. But with increasing the feed MNP to
8
wt % (PMC 2), the reduction of MNP content in the
particles became higher up to 2.5 wt %, which was quite a
significant loss. The loss of MNP became even more
pronounced with increasing the amount of feed MNP to 12
wt % (PMC 3). This can be correlated with an increased
amount of coagulum with increase in feed MNP content as
reported in Table 2. In order to know the nature of coagulum,
we performed a TGA investigation of the coagulum from PMC
3 which suggested that a hybrid material (containing both
organic and inorganic materials) made up the coagulum. The
coagulum showed attraction toward an external magnet which
confirmed the presence of MNP within it (Figure 6b). Altering
the surfactant from IL1 to IL2 keeping the molar surfactant
concentration constant, thereby using IL2 below its CMC did
not improve the stability of the dispersion as well as the final
content of MNP (compare PMC 2 to PCM 4, Table 2). We
assume that inefficient interactions between the PS matrix and
OA coated MNP or the inability to hold the MNP inside or on
the surface of PS spheres due to phase segregation is
Figure 6. Magnetism of the nanoparticles from PMC 3 proved by the
attraction toward external magnet (a) and presence of MNP in the
coagulum from PMC 3 proved by the attraction toward external
magnet (b).
The sample with 8 wt % feed MNP (PMC 2) showed a
reasonable stability where the precipitation started only after
keeping the dispersion standing for nearly 3 days. The least
MNP content dispersion (PMC 1 with 4 wt % feed MNP) was
observed to be a very stable throughout an observation time of
even 3 weeks. The particle size distribution was investigated
using DLS analysis (Table 2) which shows an increase in the
average particle size as well as the dispersion index with
increasing the feed MNP content from 4 to 8 wt %. Because of
Table 2. Characterization Results for PMC Nanoparticles Synthesized by IL Stabilized MEP
a
feed MNP
(wt %)
av particle size (nm)
[dispersion index]
coagulum
(wt %)
solid content
(wt %)
conversion of monomer
(%)
MNP content from TGA
(wt %)
sample
PMC 1
PMC 2
PMC 3
4
8
122 [0.03]
136 [0.07]
n.d.
5
8
19
18
17
18
96
90
80
90
3.2
5.5
8.2
5.8
12
8
35
7
b
PMC 4
143 [0.06]
a
b
With respect to monomer. In the case of PMC 4, IL 2 was used as surfactant. In all other cases, IL 1 was used as surfactant.
F
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 7. Schematic representation of the enrichment of styrene phase with OA-coated MNP as a result of phase separation between MNP and
developing PS phase in the AIBN initiated MEP.
Table 3. Characterization Results from IL Stabilized Non-RAFT and RAFT Mediated MEP
AIBN
(g)
CTA
(g)
AIBN:CTA
(mole ratio)
av particle size (nm)
[dispersion index]
dispersity
index Đ
solid content
(wt %)
conv
(%)
sample
M (g/mol)
n
a
2.61 × 106
2.43 × 106
18000
PS 1
0.130
0.130
0.022
0.022
0.022
93 [0.03]
103 [0.05]
125 [0.05]
140 [0.07]
133 [0.04]
2.0
2.2
1.7
1.6
1.8
18
18
16
13
16
100
100
84
b
PS 2
a
PS 3
0.076
0.113
0.076
1:2
1:3
1:2
a
PS 4
12000
70
b
PS 5
19000
82
a
b
IL 1 as surfactant. IL 2 as surfactant.
responsible for the low MNP content in the latex compared to
and the trithiocarbonyl moiety of CTA would enhance the
interaction of the PS chain with MNP; consequently, the
distribution of MNP among the PS droplets would be more
homogeneous. At first, the RAFT mediated MEP was
conducted in the absence of MNP in order to investigate the
morphology of the RAFT polymerized PS particles and the
terminal functionalization of the PS chain. Two experiments
were performed using IL1 as surfactant to observe the influence
of concentration of CTA in the molecular weight development
and dispersion index. In addition to that, one IL2 stabilized
RAFT mediated MEP was also performed. Table 3 represents
the GPC investigation of PS synthesized by different non-
RAFT and RAFT mediated MEP. Irrespective of the nature of
22
the feed. Elaissari et al. also found a similar scenario in the
SDS stabilized MEP for the synthesis of anisotropic magnetic
polymer particles where a final MNP content of only 19 wt %
was possible to achieve using 30 wt % of feed MNP, indicating
a significant loss of MNP through the formation of coagulum.
The amount of feed MNP also had a measurable influence on
the monomer conversion which became lower as the amount of
MNP increased. With increase in the amount of feed MNP
from 4 to 12 wt %, the conversion of monomer goes down
from 95% to 80% (Table 2). The reduction in monomer
conversion in the presence of MNP has also been reported in
15
1
5,22
the literature.
Involvement of OA in the polymerization
might be a reason for such reduction in conversion, leading to
consumption of initiator, although increasing the amount of
AIBN did not increase the monomer conversion. It can be
assumed that the influence of MNP may arise from their
preferred location in the styrene monomer phase within the
droplet due to their immiscibility with the final PS phase as
shown in Figure 7.
Synthesis of PMC Nanoparticles Using RAFT Medi-
ated MEP. In order to overcome the problem of
inhomogeneity in the distribution of MNP among the PS
particles as well as the loss of MNP content in the composite
materials, RAFT mediated MEP was performed to synthesize
the PMC nanoparticles. Considering the affinity of carboxylic
acid group toward the surface of MNP, we choose a carboxylic
acid terminated CTA (Figure 1) expecting the carboxyl group
IL, a large reduction in the molecular weight value M was
observed in the presence of CTA as compared to the non-
n
RAFT MEP. The reduction of M could be controlled by
n
tuning the AIBN to CTA mole ratio and the dispersity index Đ
in all RAFT mediated MEP was with 1.5−1.9 significantly lower
as in the free radical process, both indicating an effective chain
transfer, even though Đ was not as low as for a typical bulk
31
RAFT polymerization. The purpose of the present study was
to successfully attach the terminal carboxylic group to the PS
chain via RAFT mediated MEP. The confirmation of the
presence of terminal carboxyl group in PS chain was not
possible by any spectroscopic analysis due to the still high
molecular weight. Hence, a model reaction via bulk RAFT
polymerization was performed with the same CTA, which
produced low molecular weight PS, to confirm the attachment
G
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
of terminal carboxylic group at the PS chain ends. Figure 8
of the dispersion from RAFT mediated MEP also indicates a
narrow particle size distribution (see Table 3).
1
depicts the H NMR spectrum of this polymer (M
g/mol).
∼ 4400
n,NMR
In order to synthesize the PMC nanoparticles by RAFT
mediated MEP and to compare the results with the non-RAFT
system that was discussed earlier, we used feed MNP
concentration of 8 wt % with respect to monomer. The
influence of initiator to CTA mole ratio on the content of MNP
in the final composites along with the morphology was
investigated in detail. The TGA investigation of the PMC
nanoparticles obtained from RAFT mediated MEP showed an
enhancement in the MNP content with increasing the amount
of CTA in the recipe (Table 4). A high value of 27 wt % of
MNP content (PMC 7) in a stable latex could be achieved from
a feed MNP concentration of 8 wt %. Comparing the TGA
values of the samples prepared by non-RAFT MEP (Table 2),
it was quite amazing to reach such a high value of MNP
content. The colloidal stability was also an important factor for
such particles with high MNP content. An observation of the
diluted latex of PMC 6 confirmed a reasonable stability of the
latex as it remained stable even after 1 week of standing. Figure
9 shows a comparative stability difference between PMC 2 and
Figure 8. 1H NMR spectrum of α,ω-carboxylic acid terminated
polystyrene synthesized using 2,2′-[carbonothiobis(thio)]bis(2-meth-
ylpropionic acid) as CTA (solvent: CDCl ).
3
The signal groups of methine protons neighboring to the
trithiocarbonyl group (4.4−4.9 ppm) and of the methyl
protons of the 2-methylpropionic acid end groups (0.8−1.05
ppm) can be well observed. The chemical shift of the methyl
protons corresponds with a 2-methylpropionic acid group
bonded to the PS chain but not with a trithiocarbonyl group
(
∼1.7 ppm from CTA data). Considering the SCH:CH3
intensity ratio of ∼1:6, a nearly perfect α,ω-termination by 2-
methylpropionic acid groups bonded to PS can be concluded.
This is also confirmed by the C NMR spectrum (Figure SI 4).
1
3
Figure 9. Stability of the diluted latex of PMC nanoparticle dispersion
13
(
8 wt % feed MNP content) synthesized by non-RAFT (PMC 2) and
Based on the C NMR data of the CTA, the carboxyl signal of
a 2-methylpropionic acid group bonded to trithiocarbonyl is
expected at ∼180 ppm. Only a very weak signal at 179.3 ppm
can be identified by 2D NMR (Figure SI 5) whereas an intense
signal at 183.4 ppm represents the 2-methylpropionic acid
group bonded to PS. This model RAFT polymerization
suggests that this CTA results in well-defined PS bearing a
trithiocarbonyl group in the backbone and two carboxylic acid
groups as α,ω-termination. This is in accordance with the
RAFT mediated (PMC 6) MEP. (a) Immediately and (b) 1 week after
polymerization.
PMC 6. The DLS analysis of all PMC nanoparticles prepared in
this study also suggests a reasonably good distribution of the
particle size (Table 4).
Apart from this, the MNP content of the final composite
could also be controlled from a constant feed amount just by
varying the monomer conversion, as shown in Table 4 in the
case of PMC 6. We continued the polymerization of PMC 6 up
to 40 h to achieve maximum possible conversion in the given
recipe. The conversion of monomer increased with the time of
polymerization, as expected for a typical RAFT polymerization,
34
polymerization of n-butyl acrylate using the same CTA.
The morphology of the functionalized PS particles prepared
by RAFT mediated MEP was investigated by SEM (Figure SI
6
). Nice spherical particles were observed associated with
apparently narrow particle size distribution. The DLS analysis
Table 4. Characterization Results for PMC Nanoparticles Synthesized by RAFT Mediated IL Stabilized MEP
polymerization
time (h)
AIBN:CTA
(mole ratio)
av particle size (nm)
[dispersion index]
coagulum
(wt %)
SC
monomer conv MNP content from TGA
a
sample
PMC 5
(wt %)
(%)
(wt %)
14
14
22
40
14
20
20
1:1
1:2
1:2
1:2
1:4
3:1
3:1
182 [0.09]
224 [0.12]
222 [0.17]
216 [0.22]
242 [0.29]
206 [0.10]
210 [0.12]
3
0
2
7
0
4
6
13
11
13
15
7
60
52
64
74
34
75
74
13.4
19.2
11.1
8.9
PMC 6 (14)
PMC 6 (22)
PMC 6 (40)
PMC 7
27.3
6.8
PMC 8
15
15
b
PMC 9
7.0
a
b
In all experiments, feed MNP concentration was 8 wt % with respect to styrene, and IL 1 was used as surfactant (exception PMC 9). IL 2 was used
as surfactant.
H
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
which finally led to a decrease in the final content of MNP since
it approached the feed value. Thus, the MNP content in the
resulting PMC, starting from 8 wt % in the feed, varied from
19.2 to 8.9 wt % depending on the monomer conversion. To
investigate the influence of AIBN to CTA mole ratio on the
final MNP content of PMC as well as on the outcome of the
polymerization, a series of experiments were performed by
changing AIBN to CTA mole ratio but keeping the MNP
content in the feed at 8 wt %. Sample PMC 6 and PMC 7 were
absolutely free of coagulum whereas the others ended up
producing low amount of coagulum. With increase in the
amount of AIBN, or better lowering the CTA content in
comparison to AIBN, the coagulum started to appear, reaching
a large amount associated with the reduction of MNP content
in the final PMC nanoparticles (PMC 8). A similar trend was
also observed in case of IL2 stabilized RAFT mediated MEP
(PMC 9).
In order to explain the production of PMC with a higher final
MNP content than the initial feed amount, we took in account
the solid content (SC) of each experiment and the monomer
conversion calculated from SC. The value of SC and
corresponding monomer conversion value is shown in Table
4. With the increase in concentration of CTA, the SC as well as
the conversion was reduced. This was in accordance with the
expectation as the CTA is able to control the rate of
polymerization and thus, with increasing the amount of CTA
while keeping the amount of AIBN constant at fixed
polymerization conditions, one should obtain a lower
conversion of monomer. The residual monomer that remained
in the droplets evaporated out during the vacuum drying
leaving the organic content of the material much lower
compared to the theoretical one. Hence, the TGA investigation
of the PMC nanoparticles prepared by RAFT mediated MEP
always revealed a higher content of MNP compared to the
theoretically predicted one. In case of non-RAFT MEP (Table
Figure 10. TEM image of the IL stabilized MEP showing
inhomogeneous distribution of MNP. (a) and (b) represent the
morphology from IL 1 stabilized MEP (PMC 2) at two different
magnifications, whereas (c) and (d) represent the same from IL 2
stabilized MEP (PMC 4).
from being distributed all over the PS particles. Also, within the
individual composite particles, MNP were not delocalized over
the entire particle; rather they were accumulated on some
specific areas near the surface of the polymer particle (Figure
2
2
10b,d), forming a patchy morphology. Elaissari et al.
2
4
), the monomer conversion was above 90% (PMC 2 and PMC
) accompanied by a loss of MNP via the formation of
described the reasons for the surface localization and
agglomeration of OA coated MNP in the PS particles
synthesized by MEP. It was explained on the basis of the
mechanism of AIBN initiated MEP where the polymerization
propagates toward the surface of monomer droplets and as a
consequence the MNP were pushed toward the surface of the
droplet due to the poor compatibility between PS matrix and
coated OA layer on MNP surface, as indicated in Figure 7. As a
result, an anisotropic Janus like morphology was developed.
Beside the above morphology, some other areas of the
specimen were also investigated in order to explain the
relatively broad particle size distribution as well as the
formation of coagulum. Eventually, it was observed that in
some areas of the specimen with MNP patches the particles
showed a tendency toward agglomeration due to MNP
attraction which might lead to inefficient stabilization and the
formation of coagulum. Furthermore, some particles were
found to stick together forming aggregates which may result in
a broader particle size distribution (Figure SI 7).
TEM investigations of the samples prepared by RAFT
mediated MEP were also performed to explain the large MNP
content of the final PMC nanoparticles combined with still a
high stability of the latex and to gain information about the
distribution of MNP inside the PS particles. From Table 4, it
can be observed that the gradual enhancement in the MNP
content of PMC nanoparticles is associated with reduction in
monomer conversion with increase in the AIBN to CTA mole
ratio. Figure 11 depicts TEM image of PMC 6 and PMC 7 with
coagulum. Thus, the final content of MNP could never reach
even near the theoretical value. As discussed before, in case of
PMC 6, different MNP content was possible to achieve in a
single experiment just by controlling the time of polymer-
ization. This is quite in accordance with the fact that with
increasing the polymerization time, following the typical RAFT
kinetics, the monomer conversion also increases, consequently
the amount of residual styrene that evaporates during the
drying process is reduced and as a result of that the organic and
inorganic content of the material approaches the feed
composition.
Investigation of Morphology Using TEM Analysis. In
order to make a correlation between the above results and the
morphological characteristics of the PMC nanoparticles, TEM
investigations were performed with the samples containing
similar feed MNP content of 8 wt %. It is clear from the images
of the samples PMC 2 and PMC 4 that in both cases the MNP
were not homogeneously distributed over all PS particles. A
large number of pure PS nanoparticles were observed beside
the PMC nanoparticles (Figure 10a,c). The reason for the
existence of pure PS particles was not, or not only, the micellar
nucleation that might happen in the case of IL 1 stabilized
MEP; otherwise, we would not observe the presence of such
particles in the IL 2 stabilized MEP where the concentration of
IL 2 was far below its CMC. It can be speculated that the strong
magnetic attraction between the individual MNP restricts them
I
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Besides enhancing the MNP content of the PMC nanoparticles,
it could control the morphology as well, depending on the mole
ratio of initiator to CTA. The different morphologies that were
possible to achieve are schematically outlined in Figure 13.
The morphology I of PMC nanoparticles was observed in
several AIBN initiated MEPs and was well explained by Elaissari
22
et al. in the SDS stabilized MEP to synthesize anisotropic
Janus magnetic polymer particle. But the occurrence of pure
polymer particles could not be avoided. On the other hand, a
real homogeneous distribution of MNP associated with a nice
dispersion of MNP inside the PS particles was observed in
morphology II, in which the molar concentration of CTA was
higher than that of AIBN. The reason for such morphology was
possibly due to the affinity of the surface of MNP toward the
carboxylic acid group of CTA attached to the polymer chain
Figure 11. TEM image of the PMC nanoparticles synthesized by
RAFT mediated MEP, showing homogeneous distribution of MNP
among the PS particles. (a) PMC 6 (14), AIBN to CTA = 1:2; (b)
PMC 7, AIBN to CTA = 1:4.
33
end. Lu et al. reported a ligand exchange reaction of OA
coated MNP with a carboxyl functionalized CTA in which a
partial replacement of OA was described by the CTA. The
interaction of CTA with the MNP was mostly through the
terminal carboxyl group and the trithio group. In our system,
the presence of carboxyl terminated CTA in each of the
monomer droplets was proved to be an effective way to
distribute the MNP homogeneously between all the styrene
droplets which was not possible in the case of the non-RAFT
process as shown in Figure 13. During the RAFT polymer-
ization the oligomer PS containing a carboxylic acid and
trithiocarbonyl group at their chain possibly interacted with the
surface of MNP to disperse them inside each PS droplet, and as
a result of that morphology II developed. Because of the
complexity of the system under investigation, FTIR analysis
could not confirm such interaction of MNP with the carboxyl
functionalized PS chain. Another important factor was the low
conversion of monomer in RAFT mediated MEP and the
resulting lower mass of the PS chains which made the mobility
of MNP inside the droplet much easier compared to the non-
RAFT MEP and thereby producing a homogeneous dispersion
within the PS droplet. Increasing the concentration of AIBN
produced a much faster polymerization which led to reasonably
higher conversion of monomer (PMC 8 and PMC 9 in Table
an AIBN to CTA mole ratio of 1:2 and 1:4, respectively. A real
homogeneous distribution of MNP was observed among the PS
composite particles. Almost no pure PS particles were found
which is quite impressive compared to the non-RAFT one
(
Figure 10a,c). Similar to PMC 2, a significant amount of MNP
was found to be attached to the surface of polymer particles
which was quite expected as all of the experiments in our study
use AIBN initiated polymerization.
To analyze the influence of AIBN to CTA mole ratio on the
morphology of PMC nanoparticles, TEM investigation of PMC
8
was performed with a high AIBN content compared to CTA
of 3:1 (Figure 12a). The MNP were observed to distribute
4
) in a similar reaction time as for PMC 6 or PMC 7.
Figure 12. TEM image of (a) PMC 8 and (b) PMC 9 showing
anisotropic distribution of MNP (Janus-like morphology) in all PMC
nanoparticles.
Consequently, the mobility of MNP inside the PS droplet
became very low, and accumulation of the nanoparticles at the
surface of PS particles due to low compatibility was observed
which is represented by morphology III in Figure 13. Attaining
the anisotropic Janus morphology in all PMCs reproducibly has
been rather a tricky one as a proper optimization of the amount
of initiator and CTA needs to be done. Otherwise, a lower
concentration of AIBN will lead to morphology II or a higher
concentration of AIBN will finally end up producing a high
amount of coagulum in the system.
Investigation of Molecular Weight and Dispersity
Index. The molecular weight distribution of the PS particles
was highly influenced by the presence of MNP in the system. In
absence of MNP, IL stabilized free radical MEP led to high
molecular weight of PS (e.g., M of PS1:2.61× 10 g/mol)
which was expected from a heterogeneous polymerization
system, whereas via RAFT MEP lower molecular weights were
obtained (e.g., M of PS3:1.8 × 10 g/mol, Table 3). In the
presence of MNP the M value of the product of free radical
MEP remained high but the distribution became broader. The
dispersity index increased from about 2 to 4−5. The
appearance of broader distribution was possible to overcome
using RAFT mediated MEP. In all the samples prepared by
among all the PS particles as found in case of PMC 6 (Figure
11a). But a striking difference was detected in the location of
MNP in the individual PS particles. The MNP were not
dispersed homogeneously within the PS particles; rather, they
were confined to a specific area on the PS surface nearly similar
to that of PMC 2 or PMC 4. But due to the efficiency of CTA
regarding the distribution of MNP in all polymer particles (as
shown in Figure 11), no existence of pure PS particles was
found. Consequently, the anisotropic (Janus-like) location of
MNP in the polymer was spread over all the PS particles in the
latex. Altering the surfactant to IL 2 (PMC 9) did not influence
the morphology as the MNP were nicely distributed in the PS
particles (Figure 12b). Thus, a similar anisotropic morphology
developed also in this case. One can speculate that with
increasing the AIBN content, less PS chains are functionalized
with COOH end groups, and thus, a lower number of chains
can undergo specific interactions with the MNP.
6
n
4
n
n
Hence, the carboxyl terminated CTA proved to be an
effective component in the synthesis of PMC nanoparticles.
J
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 13. Representation of the different morphologies I−III that are possible to develop through IL stabilized MEP in the absence and presence of
CTA.
Table 5. Molecular Weight Distribution of PS in PMC Nanoparticles
a
sample
PMC 1
feed MNP (wt %)
AIBN (g)
CTA (g)
AIBN:CTA (mole ratio)
conv of monomer (%)
M (g/mol)
dispersity index Đ
n
4
8
0.130
0.130
0.130
0.130
0.022
0.022
0.022
0.022
0.022
0.130
0.130
96
90
80
90
60
52
64
74
34
75
74
2.4 × 105
5.8
5.1
4.1
4.8
1.9
1.5
1.7
1.7
1.5
1.7
1.8
5
PMC 2
PMC 3
PMC 4
PMC 5
2.2 × 10
12
8
2.0 × 105
1.2 × 105
25000
7000
b
8
0.038
0.076
0.076
0.076
0.150
0.076
0.076
1:1
1:2
1:2
1:2
1:4
3:1
c
c
c
PMC 6 (14)
PMC 6 (22)
PMC 6 (40)
PMC 7
8
8
14000
18000
2500
8
8
PMC 8
8
20000
20000
b
PMC 9
8
3:1
a
b
In all experiments, feed MNP concentration was 8 wt % with respect to styrene. In the case of PMC 4 and PMC 9, IL 2 was used as surfactant. In
c
all other cases, IL 1 was used as surfactant. Values in the parentheses represent the time of polymerization in the synthesis of PMC 6.
RAFT MEP, the dispersity index was reduced down to 1.5.
Thus, much better control over the polymerization was possible
to achieve. The M value could also be controlled by a proper
n
tuning of AIBN to CTA mole ratio and by monomer
conversion. In the case of PMC 6, a nearly linear increase in
M value with time of polymerization and consequently with
n
the increase in the conversion of monomer clearly fits with a
controlled radical polymerization system. The slight deviation
from the linearity in the development of M with respect to
n
time of polymerization (or conversion of monomer) may arise
from the influence of MNP within the droplets.
Magnetic Property Investigation. The magnetic proper-
ties of the synthesized PMC nanoparticles were investigated
using VSM analysis. All the materials showed a typical
paramagnetic behavior at the experimental condition (Figure
1
4). The saturation magnetization (M ) of the samples was in
s
accordance with the MNP content value. The morphological
divergence did not show any significant difference in the Ms
values. Among the PMC nanoparticles synthesized in this
Figure 14. Magnetization curve of PMC nanoparticles with different
MNP content which were synthesized by IL 1 stabilized MEP.
study, PMC 7 displayed the highest value of M (16.60 emu/g)
s
as expected from its high MNP content composition. The
control the M value by tuning the MNP content of the PMC
s
magnetization curve clearly indicates that it is possible to
nanoparticles through RAFT mediated MEP. From the similar
K
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
feed MNP content, a different range of M could be achieved.
AUTHOR INFORMATION
s
■
This provides a big advantage in the application purpose of the
magnetic materials where the extent of magnetization holds a
vital key for the outcome of the experiment.
Corresponding Author
E-mail voit@ipfdd.de (B.V.).
*
Notes
CONCLUSION
The authors declare no competing financial interest.
■
In this work we could successfully perform MEP of styrene
using imidazole based ILs as surfactants. Furthermore,
molecular weight and dispersity index of the polymer could
be controlled by employing RAFT mediated IL stabilized MEP.
Because of the availability of CTA with different functional
groups, one can obtain several kinds of internally functionalized
polymer particles through this approach. On the other hand,
the presence of IL at the surface of polymer particles can be
easily utilized through the ion exchange capability of IL that can
offer the possibility of dispersing the same polymer particle in
different solvents ranging from polar to nonpolar in nature.
Magnetic polymer particles were synthesized using IL
stabilized MEP employing first a free radical process, but high
content of MNP in the composite particles could not be
achieved due to increasing instability in the dispersion with
increasing MNP in the feed associated with the formation of a
high amount of coagulum. This problem was overcome using
RAFT mediated IL stabilized MEP. Different MNP contents
ranging from 8 to 27 wt % could be reached from 8 wt % feed
concentration, depending on the final monomer conversion,
with good to moderate colloidal stability. Considering the
advantages of a single step miniemulsion process, the RAFT
miniemulsion approach proved to be a promising method to
enhance the final MNP content in the composites. On the
other hand, different morphologies could be developed by
tuning the initiator to CTA mole ratio, but fully avoiding the
preparation of polystyrene particles lacking any MNP due to
the improved interactions of the MNP with carboxylic acid end
groups of the polystyrene chains prepared with the RAFT
mechanism. Combining several characterization techniques, the
colloidal stability, morphology, structural elucidation, and
composition of the materials were confirmed. As RAFT
chemistry employed here is quite compatible with the biological
system, a biocompatible monomer can well be utilized
according to this approach to prepare magnetic biocompatible
polymer nanoparticles for suitable biomedical applications.
Beside the above benefits of RAFT mediated MEP approach for
synthesizing PMC nanoparticles, a major drawback lies in the
low conversion of monomer in case of experiments consisting
of high feed MNP concentration which may restrict this
method in several applications, even though it has been shown
that this can be overcome by increasing the polymerization
time. Therefore, at present, we are further improving the
formation of PMC nanoparticles through the RAFT mediated
MEP approach.
ACKNOWLEDGMENTS
■
The authors acknowledge Petr Fomanek and Uta Reuter for
TEM, Maria Auf der Landwehr for SEM, Kerstin Arnhold for
TGA, and Petra Treppe for GPC analysis. We have been
grateful to Dr. Volker Neu of IFW Dresden (Germany) for
providing the VSM instrument for magnetic property measure-
̈
ment. S. Chakraborty thanks Dr. Jurgen Pionteck for important
advice in this work.
REFERENCES
■
(
1) Cheng, F.-Y.; Su, C.-H.; Yang, Y.-S.; Yeh, C.-S.; Tsai, C.-Y.; Wu,
C.-L.; Wu, M.-T.; Shieh, D.-B. Biomaterials 2005, 26, 729−738.
(
2) Elaïssari, A.; Rodrigue, M.; Meunier, F.; Herve, C. J. Magn. Magn.
Mater. 2001, 225, 127−133.
3) Akgol, S.; Kacar, Y.; Denizli, A.; Arıca, M. Y. Food Chem. 2001, 74
(3), 281−288.
(
̈
̧
(4) Gupta, P. K.; Hung, C. T. Life Sci. 1989, 44, 175−186.
(5) Langer, R. Science 1990, 249, 1527−1533.
6) Hancock, J. P.; Kemshead, J. T. J. Immunol. Methods 1993, 164,
(
5
(
2
(
8
(
1−60.
7) Yang, S.-M.; Kim, S.-H.; Lim, J.-M.; Yi, G.-R. J. Mater. Chem.
008, 18, 2177−2190.
8) Yoon, J.; Lee, K. J.; Lahann, J. J. Mater. Chem. 2011, 21, 8502−
510.
9) Ahmadi, A.; Williamson, B. H.; Theis, T. L.; Powers, S. E. J.
Cleaner Prod. 2003, 11, 573−582.
(10) Ding, S.; Xing, Y.; Radosz, M.; Shen, Y. Macromolecules 2006,
39, 6399−6405.
(11) Yuet, K. P.; Hwang, D. K.; Haghgooie, R.; Doyle, P. S. Langmuir
2
009, 26, 4281−4287.
12) Lattuada, M.; Hatton, T. A. J. Am. Chem. Soc. 2007, 129,
2878−12889.
13) Braconnot, S.; Eissa, M. M.; Elaissari, A. Colloid Polym. Sci. 2013,
91, 193−203.
14) Chen, K.; Zhu, Y.; Zhang, Y.; Li, L.; Lu, Y.; Guo, X.
Macromolecules 2011, 44, 632−639.
15) Ramírez, L. P.; Landfester, K. Macromol. Chem. Phys. 2003, 204,
(
1
(
2
(
(
22−31.
(16) Chen, Z.; Peng, K.; Mi, Y. J. Appl. Polym. Sci. 2007, 103, 3660−
3666.
(
17) Luo, Y.-D.; Dai, C.-A.; Chiu, W.-Y. J. Polym. Sci., Part A: Polym.
Chem. 2008, 46, 1014−1024.
18) Mahdavian, A. R.; Ashjari, M.; Mobarakeh, H. S. J. Appl. Polym.
Sci. 2008, 110, 1242−1249.
19) Qian, Z.; Zhang, Z.; Chen, Y. J. Colloid Interface Sci. 2008, 327,
(
(
3
(
54−361.
20) Ramos, J.; Forcada, J. Langmuir 2011, 27, 7222−7230.
21) Mori, Y.; Kawaguchi, H. Colloids Surf., B 2007, 56, 246−254.
22) Kaewsaneha, C.; Tangboriboonrat, P.; Polpanich, D.; Eissa, M.;
(
(
ASSOCIATED CONTENT
Supporting Information
■
*
S
Elaissari, A. J. Polym. Sci., Part A: Polym. Chem. 2013, 51, 4779−4785.
(
(
23) Lu, S.; Ramos, J.; Forcada, J. Langmuir 2007, 23, 12893−12900.
24) Vanyur, R.; Biczok, L.; Miskolczy, Z. Colloids Surf., A 2007, 299,
NMR spectra of ILs, CTA and carboxyl functionalized PS,
comparative dispersion analysis of OA coated MNP in water
and toluene, SEM image of PS 3, TEM image of OA coated
MNP sticky particles of PMC 2, SEM images of PS
nanoparticles prepared by SDS/CTAB stabilized MEP and
comparative study between IL, SDS and CTAB regarding the
influence of surfactant to monomer ratio on average particle
size of polystyrene. This material is available free of charge via
the Internet at http://pubs.acs.org.
́
́
2
56−261.
(
̌
25) Guerrero-Sanchez, C.; Erdmenger, T.; Sereda, P.; Wouters, D.;
Schubert, U. S. Chem.Eur. J. 2006, 12, 9036−9045.
(
(
26) Yan, F.; Texter, J. Chem. Commun. 2006, 2696−2698.
27) Costa, C.; Santos, V. H. S.; Dariva, C.; Santos, A. F.; Fortuny,
M.; Araujo, P. H. H.; Sayer, C. J. Appl. Polym. Sci. 2013, 127, 448−455.
(28) Huang, X.; Sudol, E. D.; Dimonie, V. L.; Anderson, C. D.; El-
Aasser, M. S. Macromolecules 2006, 39, 6944−6950.
L
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(
29) de Brouwer, H.; Tsavalas, J. G.; Schork, F. J. Macromolecules
000, 33, 9239−9246.
30) Dong, S.; Cheng, S.; Zetterlund, P. B. J. Polym. Sci., Part A:
Polym. Chem. 2013, 51, 2104−2109.
31) Lai, J. T.; Filla, D.; Shea, R. Macromolecules 2002, 35, 6754−
756.
32) Zhang, Q.; Yang, F.; Tang, F.; Zeng, K.; Wu, K.; Cai, Q.; Yao, S.
Analyst 2010, 135, 2426−2433.
33) Bechthold, N.; Tiarks, F.; Willert, M.; Landfester, K.; Antonietti,
M. Macromol. Symp. 2000, 151, 549−555.
34) Jiang, X.; Schoenmakers, P. J.; Lou, X.; Lima, V.; van Dongen, J.
L. J.; Brokken-Zijp, J. J. Chromatogr. A 2004, 1055, 123−133.
35) Xiao, Z.-P.; Yang, K.-M.; Liang, H.; Lu, J. J. Polym. Sci., Part A:
Polym. Chem. 2010, 48, 542−550.
2
(
(
6
(
(
(
(
M
dx.doi.org/10.1021/ma5008013 | Macromolecules XXXX, XXX, XXX−XXX