Organic-inorganic hybrid materials designed by controlled radical polymerization and mediated using commercial dual functional organophosphorous coupling agents

Article abstract of DOI:10.1039/c4nj01129e

Hybrid materials that are composed of a polymeric material interfaced with an inorganic, metal oxide material have been a rapidly expanding research area in the last few decades. However, interfacial regions remain an important area of focus, and as such hybrid materials do not always possess a very robust or stable interface. Tailor-made interfacial molecules have been successfully reported but material scientists wishing to develop composite interfacial materials would favorably use commercial solutions. Our study shows how we can leverage a commercially available organophosphonic acid group that is coupled with a 2-bromo isobutyrate initiator for surface initiated atom transfer radical polymerization (SI-ATRP) for use as a strong interfacial molecule. We illustrate this mechanism with both nanoparticles of titania and flat titania substrates used as the grafting support and polymerization anchoring points. We demonstrate that the size of the organophosphonic acid initiator, specifically the carbon spacer between the reactive groups, controls the stability of the molecule. The actual covalent linkage between the phosphonic acid group and the titania surface while also leaving the ATRP initiating group able to start the polymerization, is confirmed via³¹P solid state NMR spectroscopy, liquid ¹H NMR spectroscopy XPS, DLS and SEM.

Full text of DOI:10.1039/c4nj01129e

NJC
View Article Online
View Journal | View Issue
PAPER
Organic–inorganic hybrid materials designed by
controlled radical polymerization and mediated
using commercial dual functional
Cite this: New J. Chem., 2014,
38, 6081
organophosphorous coupling agents†
Edgar Cao,ab Eric Prouzet*b and Valerie Heroguez*a
´
´
Hybrid materials that are composed of a polymeric material interfaced with an inorganic, metal oxide
material have been a rapidly expanding research area in the last few decades. However, interfacial
regions remain an important area of focus, and as such hybrid materials do not always possess a very
robust or stable interface. Tailor-made interfacial molecules have been successfully reported but
material scientists wishing to develop composite interfacial materials would favorably use commercial
solutions. Our study shows how we can leverage a commercially available organophosphonic acid
group that is coupled with a 2-bromo isobutyrate initiator for surface initiated atom transfer radical
polymerization (SI-ATRP) for use as a strong interfacial molecule. We illustrate this mechanism with both
nanoparticles of titania and flat titania substrates used as the grafting support and polymerization
anchoring points. We demonstrate that the size of the organophosphonic acid initiator, specifically the
carbon spacer between the reactive groups, controls the stability of the molecule. The actual covalent
linkage between the phosphonic acid group and the titania surface while also leaving the ATRP initiating
Received (in Montpellier, France)
7th July 2014,
Accepted 15th September 2014
DOI: 10.1039/c4nj01129e
1
group able to start the polymerization, is confirmed via 31P solid state NMR spectroscopy, liquid H NMR
www.rsc.org/njc
spectroscopy XPS, DLS and SEM.
bonds from catechols,7 silanes,8 phosphonates,9 or thiols.10,11
Of course, it is very important to understand the nature of the
Introduction
The use of many surface modification techniques is explored in substrate material when considering what type of functional
order to help integrate one object with another through various group can be used to bond to any specific surface.
means. In most cases, these particles or surfaces are modified
Phosphonate groups are promising grafting groups because,
to change their material properties or structure to integrate a unlike silanes, they do not stand homopolymerization. How-
new material into a system, and subsequently create a new ever, it has been previously shown that the Si–O–P bond is very
composite/hybrid material. The exploration of organic–inorganic sensitive to hydrolysis.12,13 Mutin et al. showed that when a
hybrid systems is of particular interest. A lot of work has been grafting reaction of phosphonic acids onto silica is performed
placed into the development of such systems for the sole purpose in aqueous media, very few Si–O–P species could be detected. In
of achieving improved properties or structural characteristics for a contrast, this reaction can be observed in higher yields when
vast spectrum of applications.1–3
using organic media. In contrast, this trend is not followed
Various methods of creating these hybrid organic–inorganic when observed with the Ti–O–P bond. The binding and anchoring
composites involve many surface interactions ranging from of the phosphonate group to the titania surface can be observed
weak interactions such as physical adsorption,4–6 to covalent with the use of aqueous media, yielding a good method of
controlled targeted coating on different surfaces. This raises the
notion that phosphonate anchoring groups can be used to
a
`
´
Laboratoire de Chimie des Polymeres Organiques – Ecole Nationale Superieure de
effectively pattern non-siliceous metal oxides like titania when
using an aqueous media, ensuring that no other surfaces, such as
glass, would be affected.
While anchoring groups serve a purpose of providing a good
coating to a specific surface, they can also be designed to be a
good interfacial region between two different matrices, especially
if one matrix can grow from the other. In the case of surface
´
Chimie, de Biologie et de Physique, UMR CNRS 5629, Universite de Bordeaux,
16 avenue Pey-Berland, F-33607 Pessac Cedex, France. E-mail: heroguez@enscbp.fr
b Department of Chemical Engineering and the Waterloo Institute for Nanotechnology
(WIN), University of Waterloo, Chemistry, 200 University Av. W,
Waterloo N2L 3G1 (ON), Canada. E-mail: eprouzet@uwaterloo.ca
† Electronic supplementary information (ESI) available: The 1H/31P NMR spectro-
scopy spectrum, table of XPS results, overall reaction mechanism and XPS
spectrum are included. See DOI: 10.1039/c4nj01129e
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 6081--6087 | 6081

View Article Online
Paper
NJC
received. 1,1,4,7,10,10-Hexamethyl tetraethyltriamine (HMTETA,
Sigma Aldrich) was also obtained and used as received. Propyl
phosphonic acid-2-bromo-2-methyl propanoate (Br-C(CH3)2(CO)-
O(C3H6)PO3H2) from Specific Polymers (France) was used as a
combined grafting molecule and polymerization initiator (also
referred to as C3-initiator). A slightly modified version of the
initiator, methyl phosphonic acid-2-bromo-2-methyl propanoate
(Br-C(CH3)2(CO)O(CH2)PO3H2, which will be referred to as C1-
initiator) was also provided by Specific Polymers. Polyethylene
glycol methyl ether methacrylate (PEGMEMA-475) (Sigma) was
chosen as the macromonomer and used as received and stored
at 4 1C until use. 18 MO ultrapure DI water was used. Ethanol
was obtained and used after ultrafiltration using a 0.45 mm filter.
Fig. 1 Structure of organophosphorous initiator molecules used. C1 and
C3 names are derived from the hydrocarbon spacer between the ester and
phosphonic acid groups.
initiated atom transfer radical polymerization (or SI-ATRP), it is
possible to grow a uniform layer of polymer starting from a flat
surface substrate. This special controlled radical polymerization
(or CRP) technique has been widely used to form coatings and other
protective polymer shells for various applications.14–18 Similar
initiator molecules were used, both with a nitroxide mediated
radical polymerization,19–21 or with ATRP initiators.21–23 Specifically,
titania particles have been shown to be coated with amine
terminated silane based anchoring groups,24–26 which can be
modified afterwards with typical ATRP initiators. However, this is
a rather complex process, and a direct method with both anchoring
group and initiator is often more desirable commercially.
Other groups have developed such grafting molecules with a
phosphorus-based anchoring group combined with a 2-bromo
isobutyrate ATRP initiator group.22,27 However, such options
refer to ‘‘in-house’’ syntheses and result in a rather low yield of
B30%, which would restrict these studies to the laboratory
demonstration scale. Commercially available products, such as
those shown in Fig. 1 prove to be a viable alternative to the
tedious synthesis method, and as such, the missing link to the
actual development of hybrid or nanocomposite materials.
However, these molecules have not been fully tested against
reaction yield and stability, which is an important first step for
their future use.
Grafting of titania particles
Both P21 and E40 nanoparticles were subjected to a similar
process. In a typical reaction with P21, 0.1 g of P21 was dispersed
using a Powergen 125 homogenizer (Fischer Scientific) into 15 mL
of ethanol containing 0.245 g of the C3-initiator grafting molecule
and subsequently were dispersed in. The suspension was homo-
genized for 3 minutes, before being heated at 85 1C for 2 hours in a
thermostatted bath. The mixture was then centrifuged at 1000 rpm
for 5 minutes and the supernatant was removed. Grafting was
achieved by heating the collected powder in an oven for 4 hours
at 155 1C. The particles were washed from ungrafted molecules by
re-dispersing them by vortex stirring in 40 mL water. The mixture
was left for several minutes, and then centrifuged at 5000 rpm for
10 minutes. The subsequent supernatant was also removed and the
remaining powder was dried overnight.
Grafting of titania surfaces
The titania flat substrates were prepared first via a mixture of
90% w/w P21 and 10% w/w PEG4000 compressed at 200 psi for
10 minutes. The flat substrate (1 cm diameter) was then
sintered at 800 1C for 1 hour, with a ramp-up at 5 1C minÀ1
from room temperature. After the titania supports cooled
down, they were grafted in a similar manner to the titania
particles. In a typical experiment, the C3-initiator (0.24 g) was
mixed with ethanol (15 mL) and sonicated for 5 minutes or
until all the initiator particles were dispersed. The flat sub-
strates were then heated with the C3-initiator–ethanol mixture
at 85 1C for 2 hours. The titania substrate was then removed, rinsed
with DI water quickly, and then heated at 155 1C for 4 hours before
cooling down. A final rinse with water and ethanol was performed
to remove any unbound initiators at this point.
In our study, we aim to discriminate between the commercially
available dual functional organophosphoric acid, displayed in
Fig. 1, which one can actually achieve a well-anchored bond to
titania interfaces, while still promoting the formation of a polymer
layer via atom transfer radical polymerization. We describe
parameters that are mandatory for the molecular stability of
these molecules along the reaction, with the aim of paving the
path to future studies using this method in material science,
especially in membrane science.
Experimental
Materials
Two types of titanium oxide particles were used, including a
nanopowder (average size 21 nm) from Sigma Aldrich (499.7%
Preparation of the macromonomer solution
anatase, CAS 1317-70-0), and a 40 nm aqueous slurry of anatase A 0.3 M aqueous solution of the macromonomer stock solution
(Aerodisp W740X) kindly provided by Evonik. In the following, was prepared by gently dispersing 7.15 g PEGMEMA into 45 mL
TiO2 particles from Sigma were named P21, and those from DI water. The resulting mixture was slowly rocked back and
Evonik, E40. Polyethylene glycol (PEG4000, Sigma Aldrich) was forth for one minute to mix, and the resulting mixture was
used for the synthesis of the flat titania substrates. CuCl then stored at 4 1C until required for use, at which point
(490% purity, Sigma Aldrich), CuBr2 (Sigma Aldrich), and was subjected to a 3-cycle freeze–pump–thaw degassing cycle
N,N-dimethyl formamide (DMF, Sigma Aldrich) were used as before use.
6082 | New J. Chem., 2014, 38, 6081--6087
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
NJC
Paper
Preparation of the catalyst
loaded into the AVANTAGE software provided by ThermoFisher
Scientific.
Following the work prepared by Grajales et al.,28 30 mL of N,N-
dimethylformamide (DMF) was degassed using a three-cycle
freeze–pump–thaw cycle. Copper I chloride (0.06 g) was added
into a vial inside a glovebox, along with copper II bromide (0.04 g)
and the degassed DMF was added into the vial and stirred for
3 hours. The degassed HMTETA ligand (490 mL) was then added to
the vial and was left stirring for another 4 hours. The remaining
mixture was then stored in the glove box until use.
Scanning electron microscopy images were taken (Leo FESEM
1530 operating at 5 kV) to complete the study. Two series of
concentrations were used in the study – both 0 M (control) and
0.27 M. In both cases, samples of the flat titania substrates were
polymerized for 5 minutes, 15 minutes and 30 minutes. The
subsequent samples were then prepared for imaging by vacuum
drying overnight and gold coated via sputtering before imaging.
Polymerization
Results and discussion
Method
The polymerization mixture was prepared by taking a 9 : 1 volume
ratio between the degassed macromonomer solution to the oxygen
free catalyst (final volume of 10 mL, with a macromonomer
concentration of 0.27 M). This was achieved using a glass neck
that was able to mix both volumes together. The mixture was
stirred for several minutes before use.
As previously mentioned, there were several different initiators
used for this reaction. In each case, the reaction was attempted
using the P21, E40 or flat titania substrates that were grafted with
the attached C3-initiator.
The grafting/initiation dual function of the molecules was tested by
probing the effective binding of the phosphonic acid onto the
different titania surfaces, and the ability of the initiator group to
induce the polymerization of the macromonomer. Besides analyses
using local probes like XPS or NMR spectroscopy, the reaction
was validated using both particles and flat surface structures, by
confirming by DLS or SEM that the polymer was growing from
the titania surface.
Stability of organophosphorous coupling agents
Procedure and equipment
Liquid 1H NMR spectroscopy was used to validate the structure
and characterize the stability of the initiators. The 1H NMR
spectrum was monitored for the C1-initiator (Fig. S1 in the
ESI†) and C3-initiator (Fig. S2 in the ESI†) over a period of time.
It appeared that the storage mode, as initially recommended by
the provider, did not prevent the C1-initiator from being
destabilized, as shown in Fig. 2 and in Table 1 (1H NMR spectra
integration values for the C1-initiator, as received and after a
three-month storage at room temperature).
The structure of the initiator molecules before reaction was checked
using 1H NMR spectroscopy (Bruker AVANCE, 400 MHz) and
compared to the reference spectra provided by Specific Polymers.
In each case, about 10 mg of the initiator was mixed with 450 mL of
CDCL3 (or deuterated chloroform) to be measured.
Solid-state 31P NMR spectra were also recorded (Bruker
AVANCE), at a 10 000 Hz frequency, to confirm that the phos-
phonate groups are actually bound to the surface of the titania.
Samples were prepared through taking portions of the C3-initiator
grafted titania (P21, E40 or titania flat supports) and performing
their analysis in the powdered form.
Dynamic light scattering (DLS) (Vasco Particle Size Analyzer,
Cordouan Technologies, France) helped us to follow the growth
of polymer-titania hybrid particles over polymerization time by
extracting the hydrodynamic diameter calculated from the
measured diffusion coefficient. Each timed measurement sample
was taken from the reaction flask with the polymerization reaction in
progress. A 0.2 mL volume of the reacting nanoparticle suspension
with the growing polymer was taken at varying times and diluted/
dispersed immediately into 15 mL of filtered ethanol. Each sample
was then sonicated for 30 minutes and subsequently measured
using the DLS system, with 6 runs of 2–3 minutes each, to ensure the
statistical accuracy and reproducibility of the data.
1
These H NMR spectra demonstrate that the stability of the
C1-initiator is fairly poor (Fig. 2 & Table 1), with its degradation
resulting in the appearance of two new peaks formed by the
hydrolysis of the ester group (note the peak formation ‘d’ and ‘e’).
It is hypothesized that the single carbon spacer between the ester
group and the phosphonic acid group results in an easily hydro-
lyzable bond. We concluded from this study that the C3-initiator,
tested in parallel without noticeable evolution after a three month
XPS (ThermoFisher Scientific K-ALPHA) was used for surface
analysis using a monochromatized AlKa source (hn = 1486.6 eV)
and a 200 micron spot size. A pressure of 10À7 Pa was maintained
in the chamber during analysis. The full spectra (0–1350 eV) were
obtained with a constant pass energy of 200 eV and high resolution
spectra at a constant pass energy of 40 eV. Charge neutralization
was required for all insulating powders pressed onto indium foil.
High-resolution spectra were quantified using the Scofield library
Fig. 2 Organophosphorous initiator: C1-initiator, degraded C1-initiator
and C3-initiator with their associated 1H NMR spectrum labeling.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 6081--6087 | 6083

View Article Online
Paper
NJC
Table 1 Integration values for C1-initiator taken from 1H NMR spectra
provided from Specific Polymers and after 3 months storage
Integration value
(original)
Integration value
(3 months)
Peak
a (1.93 ppm)
d (1.95 ppm)
e (3.8 ppm)
b (4.5 ppm)
c (10–11 ppm)
6
4.8
1.1
0.3
1.4
2.2
—
—
2
2.3
Table 2 Integration values for C3-initiator taken from 1H NMR spectrum –
measurements upon receiving C3-initiator and post 3 months
Integration value
(original)
Integration value
(3 months)
Peak
f (1.93 ppm)
g (4.23 ppm)
h (8–10 ppm)
9.8
2.0
2.4
10.2
2.0
3.1
Fig. 4 XPS results for the presence of bromine and phosphorous on:
(a) Sigma P21 particles, (b) Evonik E40 particles, and (c) flat titania support.
the use of both Evonik and Sigma particles yielded similar
results, but with the exception that the binding ability of the
phosphonic acid increased substantially with particles, as the
signal intensity was much greater than that performed using
the solid flat surface. 31P NMR spectrum analyses also revealed
that this grafting mechanism takes place at room temperature,
an important asset for future grafting (see ESI,† Fig. S3 & S4).
XPS allowed us to further characterize the nature of the surface
on the powders and flat supports, with a special emphasis on the
stability of the bromine group located on the other side of the
initiator molecule. Both bromine and phosphorous atoms are clearly
identified on the XPS spectra, Fig. 4, which confirm the actual
grafting and the stability of the initiator function upon grafting.
These resulting signals only exist for samples that were
bound to the organophosphorous/C3-initiator, and not to the
C1-initiator. We quantified the proportion of each component
as shown in Table 3. The actual phosphorus to bromine ratio
was not the expected 1 : 1 ratio, which we assigned to the
potential degradation under the exposure to sunlight during
Fig. 3 31P Solid State NMR spectroscopy for (A) C3-initiator, (B) C3-
initiator bound to the flat titania substrate (unwashed), (C) the C3-
initiator bound to the flat titania substrate (after washing), (D) C3-initiator
bound to Evonik titania particles after washing and (E) C3-initiator bound
to Sigma titania particles after washing.
storage, was the only organophosphorous coupling agent that
remained stable (Table 2), which could be used in this study.
Organophosphorous coupling agent anchoring onto titania
The actual grafting of the C3-initiator was followed by 31P NMR
spectroscopy performed on the unbound C3-initiator along
with the C3-initiator bound to the titanium oxide based surface
(with and without washing). The results are presented in Fig. 3.
A narrow peak at around B35 ppm is observed for the
unbound initiator (Fig. 3A), which is assigned to the free
phosphonic acid group. This peak remains after grafting but
before washing (Fig. 3B), along with a broader peak at 18 ppm.
This broadening and peak shift is consistent with previous
results found with phosphonic acid group binding.29,30 Upon
washing (Fig. 3C), the narrow peak disappears and only the
broad peak assigned to the phosphonic group bound to the
titania surface remains. This is observed for flat titania sub-
strates (Fig. 3C) and particles (Fig. 3D & E). Confirmation with
Table 3 XPS atomic percentages obtained for the particles P21 (sigma),
E40 (Evonik) and flat titania surfaces
P21- P21 + C3- E40- E40 + C3- Flat titania- Flat titania +
blanka initiatora blanka initiatora blankb
C3-initiator b
(%)
Element (%)
(%)
(%)
(%)
(%)
Br3p3
P2p
C1s
N1s
Ti2p
O1s
—
—
0.46
3.73
—
—
0.41
3.59
—
0.13
2.86
22.3
1.8
20.1
52.2
0.22
17.6
0.19
23.3
53.8
30.44 28.03
0.38 0.38
19.34 17.25
47.38 48.79
21.39 27.37
1.04 0.45
21.64 17.40
53.40 49.04
a
Data was obtained via averaging of 4 different points on the surface of
b
the particles. Data was obtained via one point measurement.
6084 | New J. Chem., 2014, 38, 6081--6087
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
NJC
Paper
reaction, as it was previously shown to potentially break the P–C
bond.30 However, comparative tests with grafted samples
exposed to UV light (see ESI,† Table S1) preliminary appeared
to indicate this was not the main cause of bromine loss in our
case. We hypothesize that the heating temperature could be a
factor in bromine loss, which makes further grafting at room
temperature important, but will be subject to future studies.
Confirming polymeric growth
After this first step of local analysis regarding the occurrence of
grafting and molecule stability, dynamic light scattering was used
as a means to confirm that the grafted initiator would allow
polymer growth on the particle surface. We applied the loading
procedure with the C3-initiator to each set of titania particles (both
P21 and E40), which were further exposed to a solution of 0.27 M
macromonomer (PEGMEMA) to form the polymeric hybrid core–
shell structure. The resulting data were collected at varying time
intervals. The graphs of the auto-correlation functions are reported
in Fig. 5a & c, as per the corresponding hydrodynamic diameters
(Fig. 5b & d) calculated from these curves.
Fig. 6 SEM images obtained for flat titania substrates with bound initiators
exposed to different concentrations of macromonomer and time. The
substrates exposed to 0 M macromonomer were in A–C for 5 min, 15 min
and 30 min, respectively. Images D–F were exposed to 0.27 M macro-
All autocorrelation functions could be fitted with a single
size cumulant fit, as the proof of a monodisperse suspension of
particles all along the polymerization step. The decay of these
autocorrelation functions, as a function of time, translates in a monomer for 5 min, 15 min and 30 min respectively as well.
Fig. 5 Dynamic light scattering of growing core–shell hybrid particles of titania taken at different reaction times. The samples shown are (a) log-scale
graph of P21, (b) subsequent size of the P21 particles, (c) log-scale graph of E40, and (d) subsequent size of the E40 particles over time.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 6081--6087 | 6085

View Article Online
Paper
NJC
concomitant increasing hydrodynamic diameter, which is quantified left to polymerize in either 0 M (used as a reference), or 0.27 M
in the size evolution as a function of time. The fast reaction kinetics macromonomer solution (with the catalyst). Comparison of the
noted from our results are in line with the rates reported in the 0.27 M macromonomer sample with the reference (0 M macro-
literature.31,32 Thus, there is clear evidence that the grafted initiator monomer) yields striking differences with the 15 minute and
particles are able to initiate a homogeneous polymerization on a 30 minute samples, and shows the formation of a veil that coats
curved surface to create an integrated hybrid core–shell structure.
the ceramic surface and reduces the surface roughness. This
This polymerization with the C3-initiator was further confirmed helps to further confirm and solidify that these commercially
on the flat titania substrates by direct observation using SEM (Fig. 6). available organophosphorous initiator molecules are able to
The flat titania substrates were loaded with the C3-initiator and were bond effectively and initiate an ATRP reaction from a metal
oxide surface, regardless of the surface topology.
Among the possible uses for such hybrid structures are the
applications linked with liquid or gas separation. Gas separation
is a challenge since dense polymers offer high gas selectivity but
low permeability and difficulty to scale-up very large surfaces of
thin films (several hundreds of square meters at least). On the
other hand, porous ceramic membranes are excellent regarding
durability and permeability but display low gas selectivity.
Combining both high selective polymer thin film and high
permeability robust ceramic support is a promising design if
the polymer film can be homogeneously coated onto a large
surface with no point defect and regular thickness. As growing
the polymer film directly from the ceramic support would be the
best solution, we conducted exploratory studies with the current
system to assess its potential in membrane development. SEM
observations (Fig. 7) confirm that a thick polymer carpet can be
generated from the ceramic support using our method.
Fig. 7a shows how the polymer is actually growing from the
ceramic support, not only coated onto it. The reliefs observed in
Fig. 7b and others result from the high vacuum drying required
for SEM observation that reduces the polymer film thickness.
Finally, Fig. 7c illustrates how a continuous relatively uniform
polymer film can be created on the ceramic support with no
interfacial defect. These preliminary tests are promising for the
development of our hybrid materials synthesized with the help
of commercially available dual functional organophosphorous
agents.
Conclusions
From the information presented, we have found that initiator
groups with a phosphonate-anchoring group are highly effec-
tive for covering different morphologies of surfaces on titania
surfaces. Besides, the function of each component of these
organophosphorous initiator molecules is independent of the
other, and enables us to potentially use the molecule on any
terrain/morphology that yields a metal oxide surface. It is also
important to note that the phosphonate group must be sepa-
rated from the ester group (or any other reactive group) by more
than one hydrocarbon spacer to ensure stability of the mole-
cule, and that reactions in the dark might be required for a
quantitative reaction.
Polymerization is not affected whether the initiator molecule is
bound to the surface or free in solution. This provides us proof that
Fig. 7 SEM images obtained with a flat titania substrate grafted and exposed to
a 0.27 M solution of macromonomer for 24 hours: (a) ceramic/polymer inter-
the molecule functions properly without fear of interference from
other initiator molecule functional groups or from functional groups
face, (b) polymer carpet topography and (c) side view of ceramic/polymer
interface.
6086 | New J. Chem., 2014, 38, 6081--6087
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
NJC
Paper
14 K. Matyjaszewski, P. J. Miller, N. Shukla, B. Immaraporn,
A. Gelman, B. B. Luokala, T. M. Siclovan, G. Kickelbick,
T. Vallant, H. Hoffmann and T. Pakula, Macromolecules,
1999, 32, 8716–8724.
15 R. R. Shah, D. Merreceyes, M. Husemann, I. Rees,
N. L. Abbott, C. J. Hawker and J. L. Hedrick, Macromolecules,
2000, 33, 597–605.
that exist within the organophosphorous molecule itself. This
undoubtedly will help provide an excellent interface towards
metal-oxide/polymer materials that have the potential, through the
synthesis in a less toxic environment, to offer innovative solutions in
many application domains such as membrane-based gas separation.
Acknowledgements
16 J. T. Park, J. H. Koh, J. K. Koh and J. H. Kim, Appl. Surf. Sci.,
2008, 255, 373.
17 J. T. Park, J. H. Koh, J. A. Seo, Y. S. Cho and J. H. Kim, Appl.
Surf. Sci., 2011, 257, 8301–8306.
18 R. Barbey, L. Lavanant, D. Paripovic, N. Schu¨wer,
C. Sugnaux, S. Tugulu and H.-A. Klok, Chem. Rev., 2009,
109, 5437–5527.
´
We would like to thank the Conseil regional d’Aquitaine along-
side the European program IDS-FunMat (Erasmus Mundus) and
the European Multifunctional Material Institute (EMMI) for
funding and supporting this project. Discussions with Christine
`
Labrugere were instrumental in achieving the XPS results.
19 R. Matsuno, K. Yamamoto, H. Otsuka and A. Takahara,
Macromolecules, 2004, 37, 2203–2209.
20 R. Matsuno, H. Otsuka and A. Takahara, Soft Matter, 2006,
2, 415.
21 M. Kobayashi, R. Matsuno and H. Otsuka, Science and
Technology, 2006.
Notes and references
1 P. J. Hagrman, D. Hagrman and J. Zubieta, Angew. Chem.,
Int. Ed., 1999, 38, 2638–2684.
2 C. R. Kagan, Science, 1999, 286, 945–947.
´
3 C. Sanchez, B. Julian, P. Belleville and M. Popall, J. Mater. 22 K. Babu and R. Dhamodharan, Nanoscale Res. Lett., 2008, 3,
Chem., 2005, 15, 3559. 109–117.
4 N. G. Hoogeveen, M. A. C. Stuart and G. J. Fleer, Faraday 23 S. Devillers, B. Barthelemy, J. Delhalle and Z. Mekhalif, ACS
Discuss., 1994, 98, 161. Appl. Mater. Interfaces, 2011, 3, 4059–4066.
5 N. G. Hoogeveen, M. A. C. Stuart and G. J. Fleer, J. Colloid 24 G. K. Raghuraman, J. Ru¨he and R. Dhamodharan,
Interface Sci., 1996, 182, 133–145. J. Nanopart. Res., 2007, 10, 415–427.
6 N. G. Hoogeveen, M. A. C. Stuart and G. J. Fleer, J. Colloid 25 L. Xiong, H. Liang, R. Wang and L. Chen, J. Polym. Res.,
Interface Sci., 1996, 182, 146–157. 2010, 18, 1017–1021.
7 X. Fan, L. Lin, J. L. Dalsin and P. B. Messersmith, J. Am. 26 H. Liang, X. Li, L. Xiong and S. Huang, Synth. React. Inorg.,
Chem. Soc., 2005, 127, 15843–15847. Met.-Org., Nano-Met. Chem., 2013, 43, 1034–1039.
8 A. Papra, N. Gadegaard and N. B. Larsen, Langmuir, 2001, 27 I. Minet, J. Delhalle, L. Hevesi and Z. Mekhalif, J. Colloid
17, 1457–1460. Interface Sci., 2009, 332, 317–326.
9 L. N. Djafer, A. Ayral, B. Boury and R. M. Laine, J. Colloid 28 S. T. Grajales, X. Dong, Y. Zheng, G. L. Baker and
Interface Sci., 2013, 393, 335–339. M. L. Bruening, Chem. Mater., 2010, 22, 4026–4033.
10 P. E. Laibinis, G. M. Whitesides, D. L. Allara, Y.-T. Tao, A. N. 29 O. Sel, T. Azaıs, M. Marechal, G. Gebel, C. Laberty-Robert
Parikh and R. G. Nuzzo, J. Am. Chem. Soc., 1991, 113, 7152–7167. and C. Sanchez, Chem. – Asian J., 2011, 6, 2992–3000.
11 J. Rundqvist, J. H. Hoh and D. B. Haviland, Langmuir, 2005, 30 G. Guerrero, A. Vioux and P. H. Mutin, Chem. Mater., 2001,
21, 2981–2987. 13, 4367–4373.
12 P. H. Mutin, V. Lafond, A. F. Popa, M. Granier, L. Markey 31 X. S. Wang and S. P. Armes, Macromolecules, 2000, 33,
and A. Dereux, Chem. Mater., 2004, 16, 5670–5675. 6640–6647.
13 P. H. Mutin, G. Guerrero and A. Vioux, J. Mater. Chem., 2005, 32 J. H. Truelsen, J. Kops, W. Batsberg and S. P. Armes, Polym.
15, 3761. Bull., 2002, 49, 235–242.
´ ´
¨
´
´
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 6081--6087 | 6087