Efficient production of block-copolymer-coated ceramic nanoparticles by sequential reversible addition fragmentation chain-transfer polymerizations with particles (SqRAFTwP)

Article abstract of DOI:10.1246/cl.130272

Ceramic nanoparticles (NPs) were surface modified and dispersed by in situ synthesized block copolymers. Living radical polymerization with sequential addition of monomers enables the effective coating of ceramic NPs by block copolymers. The method enable

Full text of DOI:10.1246/cl.130272

doi:10.1246/cl.130272
Published on the web May 31, 2013
801
Efficient Production of Block-copolymer-coated Ceramic Nanoparticles
by Sequential Reversible Addition-Fragmentation Chain-transfer Polymerizations
with Particles (SqRAFTwP)
Toshihiko Arita
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University,
2-1-1 Katahira, Aoba-ku, Sendai, Miyagi 980-8577
(Received March 29, 2013; CL-130272; E-mail: tarita@tagen.tohoku.ac.jp)
Ceramic nanoparticles (NPs) were surface modified and
dispersed by in situ synthesized block copolymers. Living radical
polymerization with sequential addition of monomers enables the
effective coating of ceramic NPs by block copolymers. The
method enables us to add stable polymer layers on any kind of
NPs; therefore, it can help the mass production of polymer-coated
inorganic NPs and drastically reduce the cost, without sacrificing
their performance.
applications of high-performance polymer-grafted NPs, mass
production without sacrificing their performances must be achiev-
ed.
In order to solve this problem, a new method of preparing
polymer-coated ceramic NPs has been developed. The method
proposed here needs no reagent to connect inorganic surfaces and
polymers, reduces the number of chemical processes, and enables
densely polymer-coated ceramic NPs to be produced on a very
large scale (more than 10 g of ceramic NPs can be coated at once in
a 300-mL flask; see Figure S1 in the Supporting Information).8
The polymer-coated NPs produced have similar structures to
polymer-grafted NPs and block-copolymer-micelle-protected NPs,
so their properties and performances are similar.9
Composite materials, especially composites of inorganic
materials and polymers, are very attractive because they can
exhibit promising new properties that have been believed to be
trade-off functions.1 In order to solve the trade-off, stable
dispersions of filler nanoparticles (NPs) in a polymer matrix has
to be achieved. Polymer grafting to the surface of inorganic
materials is promising.2-6 Because polymeric materials are in a
metastable state in most cases, composite materials of just blended
polymers and inorganic materials are not stable, i.e., they will
segregate from each other. When the inorganic NPs have grafted
polymer chains, segregation will be inhibited because the same
polymer material can be miscible.
Many studies of grafting polymer chains on NPs have been
reported. Some of the studies proved the advantage of polymer
grafting on dispersions of inorganic NPs in polymer matrixes.
Polymer grafting is effective, but all of the polymer grafting
methods need anchoring reagents to connect inorganic surfaces
and polymers.3-7 Recently, I proposed a method for the direct
production of polymer-grafted ceramic NPs without using anchor
and/or linker reagents.2 This technique drastically increases the
combinations of ceramics and polymers that can be grafted and
reduces the number of steps needed to prepare polymer-grafted
ceramic NPs; however, the method was not sufficient for mass
production of polymer-grafted ceramic NPs. For industrial
The method is facile, as shown in Scheme 1. As
a
representative example, coating of several ceramic NPs (silica,
¡-Al2O3, £-Fe2O3, BaTiO3, and TiN) by poly(2-hydroxyethyl
acrylate)-b-polystyrene (PHEA-b-PS) was performed as follows:
Ceramic NPs (1 g) were weighed, loaded in a 20-mL test tube with
a screw cap, and transferred to an oxygen- and moisture-free glove
box. Then, 2 mL of hydrophobic solvent (1:5 mixture of CH2Cl2
and CHCl3) was added. The amount of solvent depends on the
volume of the NPs; in the case of NPs with small BET surface
areas, e.g., less than 50 m2 g¹1, 2 mL or less of solvent is enough to
immerse the particles, but in the case of very fine NPs with large
BET surface areas, 5 mL of solvent is necessary to completely
immerse the NPs. Dispersion of the ceramic NPs was enhanced by
using an ultrasonicator (if necessary and/or possible, a homoge-
nizer would be more effective). A hydrophilic monomer, i.e.,
2-hydroxyethyl acrylate (HEA; 0.1 g), 1 mg of a catalyst for
reversible addition-fragmentation chain-transfer (RAFT) polymer-
ization (benzyl hexyl trithiocarbonate), and 0.25 mg of a radical
initiator [2,2¤-azobis(4-methoxy-2,4-dimethylvaleronitrile), V-70]
were added to the test tube. Then, the test tube was sealed and the
first polymerization was carried out, with vortex stirring, at room
Scheme 1. Illustration of encapsulation of ceramics nanoparticles by sequential living radical polymerization with particles (SqLRPwP).
Chem. Lett. 2013, 42, 801-803
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

802
temperature. After 14 h for the first polymerization, 0.4 g of a
hydrophobic monomer (styrene) was added to the test tube, and the
second polymerization was carried out. After another 22 h, the test
tube was removed from the glove box, and CHCl3 was added to
terminate the polymerization.
polymerization. In the case of other smaller NPs like silica,
¡-Al2O3, TiN, and £-Fe2O3, the NPs were not coated in a totally
separated state but a few particles aggregated and agglomerated
state. The initial particle size of BaTiO3 was larger than those of
the other NPs; therefore, the dispersion by ultrasonication was
effective, although it was not strong enough for the other NPs, with
initial particle diameters smaller than 100 nm. In addition, the
BaTiO3@PHEA-b-PS could be separated by just evaporating the
solvent, and they formed a BaTiO3 solid sheet, shown in the inset
image in Figure 1, not a powder. The BaTiO3@PHEA-b-PS solid
was able to be repeatedly dispersed in CHCl3; however, the solid
became elastic and difficult to redisperse after each solidification.
The PS chains immobilized on the BaTiO3 surface swelled in
CHCl3 and interpenetrated each other; therefore, redispersion
became difficult. These results show that the PHEA-b-PS chains
were strongly bound to the surface of the ceramic (BaTiO3).
The ceramic surface before and after coating with PHEA-b-PS
was observed by FESEM. As shown in Figures 2a-2d, the ceramic
surfaces after polymer coating were much smoother than those of
the untreated ceramics. The SEM observations also show that the
ceramics were successfully coated by the polymer. In addition, as
shown in Figures 2e-2h, STEM images of PHEA-b-PS-coated
SiO2 and £-Fe2O3 NPs were recorded. Comparing the SEM- and
TEM-mode images, it can be seen that all the SiO2 and £-Fe2O3
NPs are covered with low-electron-density materials, which
implies the formation of polymer layers on their surfaces.
The as-synthesized NPs were washed three times with
dimethyl sulfoxide (DMSO) by a dispersion and centrifugation
method to remove residual monomers and free (unbound)
polymers. In this case, no polymeric fraction was found in the
supernatant, i.e., all the polymers synthesized were in the
precipitate. After the purification processes, the polymer-coated
ceramic NPs were dried overnight at 70 °C. The characterization
of the polymer-coated NPs was carried out by field-emission
scanning electron microscopy (FESEM), scanning transmission
electron microscopy (STEM), Fourier-transform infrared spec-
troscopy (FTIR), thermogravimetric analysis (TGA), and dynamic
light scattering (DLS). The molecular-weight distribution (MWD)
of the polymer was measured using gel permeation chromatog-
raphy (GPC) and NMR spectroscopy.
Figure 1 depicts the DLS results for PHEA-b-PS-coated
BaTiO3 NPs (BaTiO3@PHEA-b-PS). The z-average and peak size
of the BaTiO3@PHEA-b-PS were almost the same size as those of
BaTiO3 NPs observed by SEM (Figure 2a). This implies that the
BaTiO3 NPs were coated separately by the polymers during the
In order to further examine the surface nature of the NPs and
interactions between the ceramic NPs and polymer chains, the
FTIR spectra of polymer-coated NPs, polymer-grafted NPs, PS,
and £-Fe2O3 NPs were analyzed. Figure 3a (red line) shows
the FTIR spectrum of the PHEA-b-PS-coated £-Fe2O3 NPs
(£-Fe2O3@PHEA-b-PS). The bands observed in the 3000-2800
and 3100-3000 cm¹1 regions were attributed to the C-H stretching
modes of alkyl and aromatic groups, respectively. The bands at
¹1
1600 and 1500-1430 cm
correspond to the C-C vibration
¹1
frequency of the PS aromatic group. The band at 1700 cm
corresponds to the C=O vibration frequency of the carboxy group
of PHEA. All these bands were clearly observed in the spectrum of
£-Fe2O3@PHEA-b-PS. In addition, the intensities of the bands
attributed to PS were stronger than those for PS-grafted £-Fe2O3
NPs (£-Fe2O3-g-PS) prepared using a previously developed
method.2 This implies that the amount of immobilized PS was
larger than that of directly PS-grafted £-Fe2O3, even though the
Figure 1. DLS results of untreated BaTiO3 NPs and BaTiO3@PHEA-b-
PS in chloroform. The concentration of BaTiO3 fraction was 0.01 wt % for
both solutions. Inset depicts the image of BaTiO3@PHEA-b-PS in
chloroform and dried BaTiO3@PHEA-b-PS. Very high concentration
showed precipitation (a), on the other hand, when the concentration was
lowered to 0.05 wt %, the dispersion was enough stable to measure DLS
(b). The dried BaTiO3@PHEA-b-PS sheet (c).
Figure 2. FESEM images of BaTiO3 (a), BaTiO3@PHEA-b-PS (b), TiN (c), and TiN@PHEA-b-PS (d); SEM-mode STEM image of SiO2 (e), TEM-mode
STEM image of SiO2@PHEA-b-PS (f), SEM-mode STEM image of £-Fe2O3 (g), and TEM-mode STEM image of £-Fe2O3@PHEA-b-PS (h).
Chem. Lett. 2013, 42, 801-803
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

803
the growing polymers (macroradicals) are precipitated in the
solution. The place at which precipitation preferentially occurs is
the ceramic surface. Covering the ceramic surface with the
polymer is again thermodynamically preferable. According to the
precipitation polymerization technique, the precipitated macro-
radical can still propagate;10 therefore, follow-up propagations
with monomers are possible. Polymers synthesized by follow-up
polymerizations form the outer layer; again, this is thermodynami-
cally stable. Because of the thermodynamically consistent proce-
dure and the particle structure, block-copolymer-coated ceramic
particles produced by SqRAFTwP are extraordinarily stable
against solvents and heat. Ichinose and Kunitake reported that
thin polymer films prepared by polymerization-induced adsorption
had stronger interactions with ceramic surfaces than did those
produced by simple coating;11 this is another reason why the
block-copolymer-coated NPs prepared by SqRAFTwP had stable
polymer layers on their surfaces.
SqRAFTwP is also advantageous in terms of the principle
of radical polymerization. The propagation rate coefficients of
hydrophilic monomers in hydrophobic solvents are often larger
than those in hydrophilic solvents.12,13 This makes SqRAFTwP
more attractive and powerful because the amount of hydrophilic
monomers can be reduced to a negligible level, i.e., the adsorbing
layer will not be so thick.
In conclusion, a new, facile, and effective procedure for
preparing block-copolymer-protected ceramic NPs was developed.
As described in the main text, operation of SqRAFTwP is very
simple, but its theoretical background is strong; therefore, it is
applicable to almost all ceramics. The method is also suitable for
mass production of polymer-coated NPs. The developed method is
expected to become one of the standard methods for producing
polymer-coated NPs.
Figure 3. (a) FTIR spectra of PS (blue), £-Fe2O3@PHEA-b-PS (red),
£-Fe2O3-g-PS (green), and £-Fe2O3 (black). (b) TG curves of £-Fe2O3-g-PS
(black). TG curves of £-Fe2O3@PHEA-b-PS synthesized without any
dispersing operation (blue), as-synthesized with ultrasonication and vortex
mixing (green), and cleaned (red). Weight loss attributed to PHEA-b-PS
was calculated from 300 °C.
background slope caused by scattering was smaller. This implies
that the volume fraction of NPs loaded in the sample for FTIR of
£-Fe2O3@PHEA-b-PS was smaller than that of £-Fe2O3-g-PS. The
FTIR analysis also confirms that PHEA-b-PS was successfully
immobilized on the surfaces of the ceramic NPs.
The amount of block copolymer immobilized was quantified
by TG analysis. Figure 3b shows TG curves of directly PS-grafted
and PHEA-b-PS-coated ¡-Al2O3 NPs. The amount of immobilized
PHEA-b-PS was much higher than that of PS grafted by a direct
grafting technique.2 The amount of immobilized polymer was
dependent on the dispersion method during the first polymerization
step. When ultrasonication was used before polymerization, a very
large amount of polymer was immobilized on the surface (red and
green lines in Figure 3b). However, when no special dispersion
method was used before and during the polymerization (with only
stirring), a smaller amount of polymer was immobilized (blue
line). Because dispersed particles have larger surface areas than
aggregated particles, the dispersion operation is important for
obtaining well-polymer-coated ceramic particles. The MWDs of
polymers grown on the ceramic surfaces were also measured.
Details are presented in the Supporting Information.8 The polymer
synthesized was actually a block copolymer of PHEA and PS, and
the MWD of the PHEA-b-PS was Mn = 47000 Da, with Mw/Mn =
1.7.
The reason why the developed sequential reversible addition-
fragmentation chain-transfer polymerization with particles
(SqRAFTwP) works well is now discussed. When a liquid with
a large surface free energy (surface tension) is penetrated in a
liquid (or gas) with a smaller surface free energy, the penetrated
liquid forms spheres, such as water in air or organic solvents does.
This phenomenon is thermodynamically spontaneous. SqRAFTwP
is based on this simple principle. In addition, polymers with large
surface free energies can be dissolved in hydrophobic organic
solvents in their monomeric form. As the polymerization proceeds,
This work was supported by a Scientific Research Grant from
the Ministry of Education, Culture, Sports, Science and Technol-
ogy of Japan. The author thanks KAKENHI No. 24685034, Eno
Foundation, and the Ogasawara Foundation.
References and Notes
1
Hybrid Materials: Synthesis, Characterization, and Applica-
tions, ed. by G. Kickelbick, WILEY-VCH Verlag GmbH & Co.
KgaA, 2007. doi:10.1002/9783527610495.
2
3
T. Arita, Nanoscale 2010, 2, 2073.
E. Marutani, S. Yamamoto, T. Ninjbadgar, Y. Tsujii, T. Fukuda,
M. Takano, Polymer 2004, 45, 2231.
4
5
6
7
8
R. Matsuno, K. Yamamoto, H. Otsuka, A. Takahara, Chem.
Mater. 2003, 15, 3.
J. Pyun, S. Jia, T. Kowalewski, G. D. Patterson, K.
Matyjaszewski, Macromolecules 2003, 36, 5094.
T. Arita, K.-i. Moriya, K. Minami, T. Naka, T. Adschiri, Chem.
Lett. 2010, 39, 961.
C. Boyer, V. Bulmus, P. Priyanto, W. Y. Teoh, R. Amal, T. P.
Davis, J. Mater. Chem. 2009, 19, 111.
Supporting Information is available electronically on the
CSJ-Journal Web site, http://www.csj.jp/journals/chem-lett/
index.html.
9
R. D. Bennett, A. C. Miller, N. T. Kohen, P. T. Hammond, D. J.
Irvine, R. E. Cohen, Macromolecules 2005, 38, 10728.
10 K. N. N. Jayachandran, P. R. Chatterji, J. Macromol. Sci., Polym.
Rev. 2001, 41, 79.
11 I. Ichinose, T. Kunitake, Adv. Mater. 1999, 11, 413.
12 S. Beuermann, Macromol. Rapid Commun. 2009, 30, 1066.
13 K. Liang, R. A. Hutchinson, Macromol. Rapid Commun. 2011,
32, 1090.
Chem. Lett. 2013, 42, 801-803
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/