Functionalization of titanium surfaces with polymer brushes prepared from a biomimetic RAFT agent

Article abstract of DOI:10.1021/ma200853w

Well-defined dopamine end-functionalized polymers were synthesized by employing the reversible addition-fragmentation chain transfer (RAFT) polymerization technique. tert-Butyl acrylate, N-isopropylacrylamide, and styrene monomers were polymerized in the

Full text of DOI:10.1021/ma200853w

ARTICLE
pubs.acs.org/Macromolecules
Functionalization of Titanium Surfaces with Polymer Brushes
Prepared from a Biomimetic RAFT Agent
Cꢀedric Zobrist,†,‡,§ Jonathan Sobocinski,†,‡,§ Jo€el Lyskawa,†,‡,§ David Fournier,†,‡
Valꢀerie Miri,†,‡ Michel Traisnel,†,‡,§,^ Maude Jimenez,†,‡,§,^ and Patrice Woisel†,‡,§,^,
*
†Universitꢀe Lille Nord de France, F-59000 Lille, France
‡USTL, Unitꢀe des Matꢀeriaux Et Transformations (UMET, UMR 8207), Equipe Ingꢀenierie des Systꢁemes polymꢁeres (ISP), F-59655
Villeneuve d’Ascq Cedex, France
§Fꢀedꢀeration Biomatꢀeriaux et Dispositifs Mꢀedicaux Fonctionnalisꢀes (BDMF/FED 4123), F-59655 Villeneuve d’Ascq, France
^ENSCL, F-59655 Villeneuve d’Ascq, France
S Supporting Information
b
ABSTRACT: Well-defined dopamine end-functionalized poly-
mers were synthesized by employing the reversible additionÀ
fragmentation chain transfer (RAFT) polymerization tech-
nique. tert-Butyl acrylate, N-isopropylacrylamide, and styrene
monomers were polymerized in the presence of azobis
(isobutyronitrile) and a new catechol-based biomimetic RAFT
agent incorporating a trithiocarbonate unit. All RAFT polymer-
izations exhibited pseudofirst-order kinetics, a linear increase of
the number-average molar mass (Mn SEC) with conversion and
narrow molar mass distributions (polydispersity <1.2). The resulting homopolymers exhibited the electroactive catechol and the
ω-trithiocarbonyl end groups. Subsequent immobilization of dopamine end-functionalized polymers on titanium surfaces was
monitored by using a surface plasmon resonance (SPR) sensor, and the resulting films were characterized by contact angle, infrared
ATR spectroscopy, atomic force microscopy (AFM), and X-ray photoelectron spectroscopy (XPS).
’ INTRODUCTION
are the relatively thermal and solvolytic instabilities due to the
noncovalent grafting. Alternatively, several techniques for cova-
lently tethering well-defined polymer brushes onto surfaces have
been developed, including the covalent attachment of end-
functionalized polymers incorporating an appropriate anchor
(“grafting to”) and the in situ polymerization initiated from the
surface (“grafting from”).17 While the “grafting from” strategy
generally leads to higher grafting surface densities, the “grafting
to” approach, thanks to the known structures and the easy
processability of the linear brush precursors, offers the unique
opportunity to easily tune the surface properties.
Typically, the attachment of well-defined polymer brushes
to a titanium surface was achieved by using either the “graft from”
or the “graft to” approaches, and by mostly applying living
polymerization techniques including opening metathesis18 and
living radical polymerizations (LRP).6,19,20 Of these living radical
polymerization techniques, atom transfer radical polymeri-
zation (ATRP)21 has been the most widely employed.19,20,22
Nevertheless, reversible additionÀfragmentation chain transfer
(RAFT) polymerization presents some advantages over ATRP
polymerization. Both techniques allow the polymerization of a
Titanium and its alloys are attractive materials due to their
unique biocompatible,1 mechanical,2 thermal,3 and electrical4
properties. These properties make them useful in a wide range of
structural, chemical, nanotechnological, and biomaterial applica-
tions. Controlling the functionality of exposed titanium surfaces
is of great importance for improving their performance (and thus
also for their practical applications). Titanium-based orthopedic
implants and nanoparticules, for example, often require chemical
modification of the surface for controlling specific cell adhesion5
and to avoid the aggregation of the TiO2 nanoparticles,6,7
respectively. Hence, over the last two decades, considerable
efforts have been devoted to engineering the properties of the
titanium surface through the development of efficient methods
allowing the manipulation of the surface’s chemical composition
and topology.8À10
Tethering of polymer brushes to a surface11 has emerged
as promising tool to tailor surface properties such as wetta-
bility, corrosion resistance and adhesion for applications includ-
ing microelectronics,4 biomedical devices,12,13 nanopatterning,14
and thermoresponsive adhesives.15,16 Polymer brushes can be
immobilized on appropriate surfaces using either a physisorp-
tion (i.e., chain attachment mainly through van der Waals
interactions) or a covalent (i.e., anchoring by chemical bonds)
strategy. The main disadvantages of the physisorption of polymers
Received: April 13, 2011
Revised:
July 1, 2011
Published: July 14, 2011
r
2011 American Chemical Society
5883
dx.doi.org/10.1021/ma200853w Macromolecules 2011, 44, 5883–5892
|

Macromolecules
ARTICLE
large array of monomers, including (bio)functional monomers,23
under a variety of reaction conditions; however, the RAFT
process does not require the use of a toxic metal (i.e., copper)
catalyst. Moreover, recent cytotoxic results regarding polymers
prepared by a RAFT procedure have shown that polymers
elaborated from a trithiocarbonate RAFT agent did not affect
either the morphology or the viability of different adherent cell
lines (CHO-K1, RAW264,7, NIH3T3).24 Surprisingly, few re-
search works were reported in the literature to tether brush
polymers onto a titanium surface by using a RAFT procedure.
These mostly concern the grafting of well-defined PMMA
polymers onto Nano-TiO2 using a carboxylic acid group as anchor
in a view to overcome the self-aggregation of nanoparticules.25
However, carboxylate-type anchors usually exhibited weak and
unstable bindings with a titanium surface, in particular in aqueous
environments that notably limit their exploitation for the develop-
ment of functionalized materials in contact with biological
media. Therefore, a grafting strategy allowing the immobilization,
through a more robust binding mode, of well-defined polymer
brushes prepared by RAFT procedure would seem to be largely
superior.
Recently, the catechol unit has emerged as a very promising
ligand for the functionalization of a wide range of substrates
including polymers, semiconductors, and more importantly,
metals and metal oxides.26À30 In particular, biomimetic dopa-
mine has sparked great interest as an anchor for the functiona-
lization of metal oxide surfaces due to the stability and strength
of the resultant five-membered metallocycle chelate and the
ability of this anchor to be chemically modified through amide
bonds.31À33 Hence, the ability of dopamine derivatives to
strongly bind titanium oxide surfaces has been exploited to
immobilize brush polymers onto titanium surfaces from end-
functionalized dopamine PEG28,34,35 or by using living poly-
merization techniques such as ring-opening metathesis poly-
merization (ROMP)18 or ATRP.22,26,29,36
Analytical Techniques. 1H NMR spectra were recorded at 25 ꢀC,
with a Bruker Avance 300 spectrometer. The number-average molar
mass (Mn), the weight-average molar mass (Mw), and the molar mass
distributions (Mw/Mn) were determined by size exclusion chromatog-
raphy (SEC) in tetrahydrofuran at 40 ꢀC with a flow rate of 1 mL.minÀ1
.
The number-average molar masses (Mn), the weight-average molar
masses (Mw), and the polydispersity indices (PDI = Mw/Mn), were
derived from the refractive index (RI) signal by a calibration curve based
on polystyrene (PS) standards from Polymer Standards Service. In all
plots showing the evolution of Mn with monomer conversion, the
straight line corresponds to the expected evolution of the theoretical
number-average molar mass, Mn,th defined as Mn,th = ([Monomer]/
[CTA]) Â conversion + MCTA
.
A Varian Cary 50 Scan UV/vis spectrophotometer was used for UV
studies. All electrochemical experiments (cyclic voltammetry) were
performed using an Autolab PGSTAT 30 workstation. The electrolyte
solution (0.05 M) was prepared from recrystallized tetrabutylammo-
nium hexafluorophosphate (Bu4NPF6) and dry acetonitrile or dry
dichloromethane. A three-electrode configuration was used with a
platinum disk (2 mm diameter) as working electrode, with an Ag/AgCl
reference electrode and a platinum wire as the counter electrode. All
measurements were recorded at 20 ꢀC under a nitrogen atmosphere.
The solution was purged with nitrogen prior to electrochemical analyses.
Surface Plasmon Resonance (SPR) measurements were carried out
with a mono channel, AutoLab Springle instrument (Eco Chemie, The
Netherlands). Briefly, polarized laser light (l = 670 nm) is directed to the
bottom side of the titanium disk (diameter =2.54 cm) sensor via
a hemispheric lens placed on a prism (BK7 having a refractive index
of n = 1.52) and the reflected light is detected using a photodiode. An
autosampler (Eco Chemie, The Netherlands) is used to inject or remove
the tested solutions. All SPR measurements were done in nonflowing
liquid conditions, i.e. with the circulating pump paused, and at 20 ꢀC.
The noise level of the SPR angle is ∼1 m degree. After each addition, the
cell was thoroughly washed with the used solvent. All measurements
were conducted at 20 ꢀC. The substrates were allowed to equilibrate
until a steady baseline was observed.
Here, we report the use of the RAFT procedure to produce
various well-defined dopamine end-functionalized polymers and
their subsequent and direct immobilization on a titanium surface.
To achieve this goal, we have designed and synthesized a new
trithiocarbonate type RAFT agent incorporating an electro-
active dopamine moiety, which was used to mediate RAFT
polymerizations of common monomers such as tert-butyl
acrylate, N-isopropylacrylamide, and styrene under conditions
that provide kinetic and molar mass control. Subsequent im-
mobilization of polymers on the titanium surface was monitored
by using a surface plasmon resonance (SPR) sensor, and the
resulting films were characterized by contact angle, atomic force
microscopy (AFM) and X-ray photoelectron spectroscopy
(XPS) investigations.
XPS analyses were performed on a Kratos Axis Ultra DLD system
(Kratos Analytical) using a monochromatic Al KR X-ray source (hν =
1486.6 eV). The emission voltage and the current of this source were set
to 15 kV and 10 mA, respectively. The pressure in the analyzing chamber
was maintained at 10À9 mbar or lower during analysis, and the size of
the analyzed spot is 500 μm. Survey (0À1200 eV) and high resolution
(C 1s) spectra were recorded at a pass energy of 160 and 40 eV, respectively.
XPS analyses were performed with a takeoff angle of 90ꢀ relative to the
sample surface. The core level spectra were referenced with the Ti 2p
binding energy at 458.7 eV. Data treatment and peak-fitting procedures
were performed using Casa XPS software.
IR-ATR spectra were recorded on a Spectrum One spectrometer
from Perkin-Elmer coupled with a Zn/Se ATR crystal collecting 64
sample scans.
Contact angle measurements were evaluated with a Digidrop contact
Angle Meter from GBX Scientific Instruments at room temperature. A
water drop was used to measure contact angle values (θ). The
measurement was repeated five times to obtain an average value for
the surface.
Atomic force microscopy (AFM) experiments were carried out in air,
at room temperature, in a Nanoscope III multimode microscope from
Digital Instruments, operating in tapping mode. The experimental set up
consists of integrated silicon tips with a radius of curvature of about
2 nm, and cantilevers (model SSS) with a free oscillation frequency f0 of
ca. 160 kHz. For all experiments, driving oscillation frequency was 250
Hz lower than f0. Images were recorded by maintaining a set point
ratio rsp. (ratio of the magnitude of the engaged amplitude to the free air
’ EXPERIMENTAL SECTION
Materials. All reagents were purchased from Sigma-Aldrich and
used without further purification unless otherwise noted. t-Butyl acrylate
(>99%, Acros Organics) and styrene (99%, Aldrich) were purified by
vacuum distillation under reduced pressure before use. N-Isopropyla-
crylamide (NIPAM, 99% Acros Organics) was recrystallized from
hexane. The primary radical source used in all polymerizations was
azobis (isobutyronitrile) (AIBN, >98%, Fluka) and was used as received.
The 2-(1-isobutyl) sulfanylthiocarbonylsulfanyl-2-methylpropionic acid
(CTA) was synthesized as previously described.37 All polymerizations
were conducted in a nitrogen atmosphere.
5884
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892

Macromolecules
ARTICLE
Scheme 1. Synthesis of Dopa-CTA and the Resulting Dopamine End-Functionalized Polymers
amplitude of the oscillating silicon tip) of 0.90, using a scanning
frequency of 0.3 Hz.
Anal. Calcd (%) for C17H25NO3S3: C, 52.68; H, 6.51; N, 3.61.
Found: C, 52.86; H, 6.51; N, 3.71.
Synthesis of the Phenyl-Based Chain Transfer Agent (Phi-CTA).
A solution of benzyl alcohol (1.57 g, 14.4 mmol) in dichloromethane
(50 mL) was added slowly, at room temperature and under a nitrogen
atmosphere, to a solution containing 2-(1-isobutyl) sulfanylthiocarbo-
nylsulfanyl-2-methylpropionic acid (CTA, 3.0 g, 12 mmol), DCC (3.0 g,
14.5 mmol) and DMAP (1.5 g, 12 mmol) in dichloromethane (200 mL).
The reaction was stirred overnight. After filtration, the solution was
washed with distilled water (3 Â 50 mL) and dried over MgSO4. The
product was purified by column chromatography (SiO2: dichloro-
methane/petroleum ether 1: 1). The product was obtained as an orange
oil in 60% yield.
Synthesis of the Succinimide-Based Chain Transfer Agent
(Suc-CTA). A suspension of N-hydroxysuccinimide (NHS, 3.3 g, 29.7
mmol) in dry dichloromethane (100 mL) was added dropwise, at
À10 ꢀC and in a nitrogen atmosphere, to a solution containing 2-(1-
isobutyl)sulfanylthiocarbonylsulfanyl-2-methylpropionic acid (CTA,
5.0 g, 19.8 mmol) and dicyclohexylcarbodiimide (DCC, 6.5 g, 29.7
mmol) in dichloromethane (150 mL). The mixture was allowed to stir
overnight at room temperature. The solvent was evaporated, leaving a
crude product which was then subjected to column chromatography
(SiO2: ethyl acetate/petroleum ether 3:1). The product was obtained as
a bright yellow solid in a quantitative yield.
1H NMR (300 MHz, CDCl3), δ (ppm from TMS): 0.92 (d, J = 6.7
Hz, 6H, CHÀ(CH3)2), 1.64 (s, 6H C(CH3)2), 1.88 (m, 1H,
CH2ÀCHÀ(CH3)2), 3.10 (d, J = 6.9 Hz, 2H, SÀCH2ÀCH), 5.05
(s, 2H, C6H5ÀCH2ÀO), 7.26 (m, 5H, C6H5ÀCH2).
1H NMR (300 MHz, CDCl3), δ (ppm from TMS): 0.94 (d, J = 6.7
Hz, 6H, CH-(CH3)2), 1.80 (s, 6H, C(CH3)2), 1.92 (m, 1H, CH2ÀCH-
(CH3)2), 2.75 (s, 4H, (CdO)ÀCH2ÀCH2À(CdO)), 3.15 (d, J = 6.9
Hz, 2H, SÀCH2ÀCH).
13C NMR (75 MHz, CDCl3): 22.1 (C(CH3)2), 25.4 (CH(CH3)2),
27.8 (CH(CH3)2), 45.2 (SÀCH2ÀCH), 56.0 (C(CH3)2), 67.7
(C6H5ÀCH2ÀO), 128.1, 128.5 (CarylÀH),), 135.6 (CIVarylÀCH2),
172.9 (CdO), 221.5 (CdS).
Synthesis of the Dopamine-Based Chain Transfer Agent
(Dopa-CTA). Suc-CTA (5 g, 14.3 mmol) and dopamine hydrochloride
(3 g, 15,8 mmol) were stirred in methanol (150 mL) with triethylamine
(2.4 mL, 17.2 mmol) at room temperature under a nitrogen atmo-
sphere, in the dark, for 48 h. The solvent was evaporated and diethyl
ether (50 mL) was added. The organic phase was washed with water
(3 Â 100 mL) and dried over MgSO4. The solvent was evaporated and
the product was precipitated into hexane to create a light orange solid
in 65% yield.
FT-IR (cmÀ1, KBr): 3037 (dCÀH), 2959 and 2869 (CÀH), 1725
(CdO), 1498 and 1463 (CdC), 1258 (CÀO), 1061 (CdS).
Typical Procedure for RAFT-Mediated Polymerizations. In a Schlenk
tube were added monomer (styrene, tBA, or NIPAM), AIBN, and the
chain transfer agent (CTA) in toluene (3.5 g) or DMF (0.5 g). The
mixture was deoxygenated by nitrogen bubbling for 30 min. The mixture
was then heated to the desired temperature (80 ꢀC for styrene and tBA
and 75 ꢀC for NIPAM). Samples were periodically withdrawn to
measure both the monomer conversion by 1H NMR and polymer
characteristics by size exclusion chromatography. The final polymer was
recovered by precipitation of the mixture in ethanol (Polystyrene and
PtBA) or diethyl ether (PNIPAM). After filtration, the product was
dried under vacuum until achieving a constant weight.
1H NMR (300 MHz, CDCl3), δ (ppm from TMS): 0.94 (d, J = 6.7
Hz, 6H, CH-(CH3)2), 1.58 (s, 6H, C(CH3)2), 1.89 (m, 1H,
CH2ÀCH-(CH3)2), 2.6 (t, J = 7.0 Hz, 2H, CH2ÀCH2ÀAryl), 3.10
(d, J = 6.9 Hz, 2H, SÀCH2ÀCH), 3.37 (q, 2H, NHÀCH2ÀCH2), 6.1
(br, 1H, ArylÀOH), 6.47 (dd, 1H ArylÀHa), 6.62 (t, 1H,
COÀNHÀCH2), 6.73 (d, 1H, ArylÀHb), 6.65 (s, 1H, ArylÀHc), 7.15
(br, 1H, ArylÀOH).
13C NMR (75 MHz, CDCl3): 22.1 (C(CH3)2), 25.8 (CH(CH3)2),
27.8 (CH(CH3)2), 34.5 (CH2ÀCH2ÀNH), 41.8 (CH2ÀCH2ÀNH),
45.5 (SÀCH2ÀCH), 57.0 (C(CH3)2), 115.2 (CarylÀH(o-OH), 115.4
(CarylÀH(o-OH)), 120.7(CarylÀH(p-OH)), 130.6 (CIVarylÀCH2),
142.9 (CIVarylÀOH), 144.2 (CIVarylÀOH), 173.5 (CdO), 220.1 (CdS).
FT-IR (cmÀ1, KBr): 3360 and 3475 (NÀH), 3037 (dCÀH), 2952
(ÀCÀH), 1651 (CdO), 1611 (NÀH), 1444 (CÀN), 1057 (CdS).
Preparation of Functionalized Titanium Surfaces (“Graft to” Method).
Titanium plates (o.d. = 1.5 cm) were first treated with an acidic oxidiz-
ing solution of concentrated sulfuric acid and hydrogen peroxide
H2SO4/H2O2 (1:1) for 2 min to generate the corresponding hydro-
xylated titanium dioxide surface. Titanium plates were thoroughly rinsed
with water, acetone, and ethanol, and dried under nitrogen before
functionalization. The pretreated titanium surfaces were then soaked
5885
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892

Macromolecules
ARTICLE
Figure 1. 1H NMR spectrum of Dopa-CTA (300 MHz, CDCl3).
Table 1. Results for the Polymerizations of Styrene, tBA, and NIPAM Performed in the Presence of Dopa-CTA and
Azobis(isobutyronitrile) (AIBN) as Initiatora
monomer
Styrene
[M]0/[CTA]0/[AIBN]0
150/1/0.3
time (min)
convnb (%)
Mn,th (g molÀ1
)
Mn,SECc (g molÀ1
)
PDId
60
180
240
300
60
4.8
14.2
16.5
19.7
5.3
890
1860
2110
2430
1900
4400
6350
7470
2960
6570
8690
9310
1900
2900
2920
3130
1700
4800
6600
7700
2800
5100
6400
7500
1.03
1.05
1.05
1.05
1.13
1.11
1.11
1.12
1.07
1.09
1.11
1.11
tBA
100/1/0.05
100/1/0.1
120
180
240
120
180
270
330
31.4
46.6
55.4
22.8
54.7
73.4
78.9
NIPAM
a Reaction conditions: T = 75 ꢀC for NIPAM and tBA, 80 ꢀC for styrene; solvent, DMF for NIPAM and toluene for tBA and styrene. b Determined by
1H NMR. c Number-average molar mass, Mn,SEC determined by size exclusion chromatography using PS standards for the synthesized polystyrene,
P(tBA) and for PNIPAM. d Polydispersity index determined by SEC.
in a solution containing 0.5 mM of polymer in water for Dopa-PNIPAM,
acetonitrile for Dopa-PtBA, and THF/water mixture (9:1) for Dopa-PS.
Samples were soaked overnight, washed with the solvent used for the
grafting and, finally, dried under nitrogen to obtain the functionalized
titanium surface.
FTIR spectroscopy. 1H NMR spectrum of Dopa-CTA recorded
in CDCl3 at 25 ꢀC revealed the presence of the characteristic
signals of the catechol unit (6 to 7 ppm) and those bearing by
the isobutylsulfanylthiocarbonylsulfanyl [(CH3)2CHÀCH2ÀSÀ
(CdS)ÀSÀ] moiety (Figure 1). Furthermore, the 13C spectrum
(see Supporting Information) clearly displayed chemical shifts
at 173.4 and 220.0 ppm ascribed to the amide carbonyl group and
the thiocarbonyl fragment of Dopa-CTA, respectively. Finally,
the structure of Dopa-CTA was further confirmed by UVÀvis
(10À4 mM in MeOH, see Supporting Information, Figure SI2)
and IR (see Supporting Information, Figure SI3) investigations,
as a strong UV absorption at around 310 nm and an IR charac-
teristic peak at 1057 cmÀ1 were observed, which can be assigned
to the CdS bond.38
’ RESULTS AND DISCUSSION
Synthesis of the Dopamine-Functionalized Chain Transfer
Agent. A new RAFT agent was specially designed to entail the
catechol fragment required for the titanium grafting (Scheme 1).
The RAFT agent Dopa-CTA was conveniently prepared from
the coupling reaction of the N-hydroxysuccinimide (NHS)
activated ester of the trithiocarbonate 2-(1-isobutyl)sulfanylthio-
carbonylsulfanyl-2-methyl propionic acid (CTA) and the com-
mercially available dopamine hydrochloride. The structure of the
Dopa-CTA was confirmed by 1H NMR, 13C NMR, UVÀvis and
RAFT Polymerizations using Dopa-CTA. First, the abi-
lity of Dopa-CTA to promote RAFT polymerizations of dif-
ferent monomers (NIPAM, tBA, Styrene) was tested. A series of
5886
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892

Macromolecules
ARTICLE
Figure 2. Semilogarithmic kinetic curves for RAFT polymerizations: (A) NIPAM (Δ) and tBA (O); (B) styrene (]). For reaction conditions, refer
to Table 1.
Figure 3. Dependence of the molar masses (filled symbols) and the PDI (open symbols) of obtained polymers on monomer conversion during the
RAFT polymerizations of NIPAM (A, Δ2), tBA(A, Ob) and styrene (B, ][). The straight lines correspond to the theoretically expected Mn.
polymerizations initiated by AIBN in the presence of Dopa-CTA
were conducted in DMF at 75 ꢀC for N-isopropylacrylamide
(NIPAM), in toluene at 75 ꢀC for tert-butyl acrylate (tBA) and in
toluene at 80 ꢀC for styrene.
Information). Both kinetic plots and SEC results proved that
Dopa-CTA allowed for an excellent control of the polymeriza-
tion of different kinds of monomers.
Characterizations of Polymers. The structure of synthesized
polymers was elucidated by 1H NMR. For instance, as shown in
Figure 4 for dopamine-terminated PNIPAM, the presence of the
catechol unit in polymers was indicated by its 1H NMR typical
resonances between 6.5 and 6.75 ppm (for Hf, Hg and Hh, see
enlarged part, Figure 4), and at 2.5 ppm for the benzylic protons
Hi. Moreover, 1H NMR spectra also displayed resonances
around 3.2 ppm belonging to the methylene protons Hk of the
isobutylsulfanylthiocarbonyl sulfanyl moiety, providing further
evidence of the ability of Dopa-CTA to act as the CTA in RAFT
polymerizations. By taking into account the integration of one
proton from the Dopa-CTA (Hg, Hh or Hf, Figure 4) and one
proton from the repeating unit NIPAM (Hc, Figure 4), the
The polymerization conditions and results of RAFT polymer-
izations of NIPAM, tBA and styrene are summarized in Table 1.
The ln([M]0/[M]) vs time plots (Figure 2) for RAFT poly-
merizations of various monomers is reported. After a typical
induction period,39 the first-order kinetic plot was linear in all
cases indicating that the concentration of active species was
constant, which is essential in controlled radical polymerizations.
Using this trithiocarbonate as the CTA agent, it appears that the
polymerization rate was higher for NIPAM polymerization than
tBA polymerization, the polymerization rate for styrene polym-
erization being the lowest, consistent with our previous study.40
Figure 3 shows the evolution of the molecular masses and
polydispersity indices of the resulting polymers as a function of
monomer conversion. In all cases, molecular weights of the
PNIPAM, P(tBA) and PS increased linearly and the polydisper-
sity indices remained relatively low as the monomer conversion
progressed. Moreover, all of the synthesized polymers showed
rather symmetrical and narrow SEC traces (see Supporting
molecular weight of the polymer could be calculated (Mn,NMR
=
12 500 g molÀ1).
3
The UVÀvis spectroscopy confirmed the presence of the
terminal trithiocarbonate group on the dopamine end-function-
alized polymers. Indeed, the UVÀvis spectra of dopamine end-
functionalized polymers (Figure 5) exhibited a wide characteristic
5887
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892

Macromolecules
ARTICLE
Figure 4. 1H NMR spectrum of dopamine-terminated PNIPAM (300 MHz, D2O). Peaks marked with asterix correspond to remaining solvent.
Figure 5. UVÀvis spectrum of dopamine end-functionalized PtBA
(À), PNIPAM (red line), PS (blue line), Dopa-CTA (green line)
(ε=13460 L.molÀ1.cmÀ1, λmax =308nm) andPhi-PNIPAM(violetline).
absorption band around 310 nm as observed for the Dopa-CTA
(green line) and Phi-PNIPAM (violet line) (which does not
incorporate a catechol unit), corresponding to the chromophoric
CdS bond of the CTA group.
Finally, the presence of catechol fragments in the polymer
chains was further evidenced by recording cyclic voltamograms
(CV) in acetonitrile (or acetonitrile/dichloromethane 1:1 in the
case of Dopa-PS) using Bu4NPF6 (0.05 M) as the supporting
electrolyte (Figure 6). Indeed, for all dopamine end-functional-
ized polymers and as observed for the Dopa-CTA derivative, CV
gave rise to the irreversible two-electron oxidation wave around
1.1 V vs Ag/AgCl, corresponding to the two-step oxidation of the
catechol unit.31 Dopa-based polymers showed no significant
shifts in the position of the oxidation wave, suggesting that the
polymer chain does not affect the redox properties of the
Dopamine unit. An interesting feature of the CV of the Dopa-
PNIPAM derivative is the difference in the shape of the peak
corresponding to the formation of oxidized species of the
dopamine fragment compared to other polymer systems, indi-
cating that the PNIPAM backbone may interact with the
electroactive unit.40
Figure 6. Cyclic voltamograms of (a) Dopa-CTA, Dopa-PtBA, and Phi-
PNIPAM in acetonitrile and Dopa-PS in acetonitrile/dichloromethane
(1:1, v:v) and (b) Dopa-CTA and Dopa-PNIPAM in acetonitrile. CV
were recorded at C = 10À3 M in the presence of Bu4NPF6 (0.05 M).
Platinum working electrode; Ag/AgCl reference electrode. Scan rate of
50 mV sÀ1
.
Immobilization of Dopamine End-Functionalized Polymers
on a Titanium Surface. Prior to the immobilization, the titanium
surface was pretreated with an acidic oxidizing solution of
concentrated sulfuric acid and hydrogen peroxide (H2SO4/H2O2
1:1) to generate the corresponding hydroxylated titanium
5888
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892

Macromolecules
ARTICLE
Scheme 2. Illustration of the Functionalization of Titanium
Surfaces with Dopamine End-Functionalized Brushes Poly-
mers Prepared from the Dopa-CTA RAFT Agent
Table 2. XPS Atomic Ratios and Contact Angle Measure-
ments of Polymers Grafted onto Titanium Surfaces
Figure 8. IR-ATR Spectra of titanium surfaces functionalized with (a)
Dopa-PS, (b) Dopa-PNIPAM, and (c) Dopa-PtBA. The 1900À2300 cmÀ1
region corresponding to the ZnÀSe crystal absorption band is omitted
for more clearness.
XPS atomic concentration (%)
static water contact
samples
C 1s
O 1s
Ti 2p N 1s
angle (deg)
Ti
30.0
37.6
28.9
21.2
26.32
19.3
10.4
0.9
23 ( 3
74 ( 2
90 ( 2
110 ( 2
Dopa-PNIPAM 46.9
4.14
were also detected, which are likely due to contamination arising
from the polishing and cleaning of the titanium samples.1,41,42
After polymer grafting, an increase in the carbon signal and a
decrease in the titanium and oxygen ratios were observed in the
XPS survey (Figure 7), which are consistent with the immobi-
lization of polymers on the surface. Furthermore, in the case of
titanium surfaces modified with Dopa-PNIPAM, an increase of
the nitrogen ratio was observed as evidenced in the XPS survey
spectra (Figure 7). Finally, a film thickness of 3À5 nm could be
estimated from the attenuation of the Ti signal.43
Dopa-PtBA
64.6
54.6
6.27 0.6
8.40 1.03
Dopa-PS
Contact Angle Measurements. The impact of polymer
grafting on the titanium surface wettability was investigated with
static contact angle measurements (Table 2). The unmodified
titanium surface exhibited a low contact angle of 23 ( 3ꢀ
consistent with the presence of hydrophilic hydroxyl TiÀOH
groups on the titanium surface subsequent to the oxidation step.
Treatment of this surface with Dopa-PtBA, Dopa-PS, and Dopa-
PNIPAM induced, in both cases, a dramatic increase of the
contact angle. The measured contact angles were 90 ( 2ꢀ, 110 (
2ꢀ, and 74 ( 2ꢀ with Dopa-PtBA, Dopa-PS, and Dopa-PNIPAM,
respectively, which agreed with the reported values.44À46
IR-ATR. The grafting of dopamine end-functionalized poly-
mers onto the titanium surface was further characterized by
infrared spectroscopy using the attenuated total reflectance
(ATR) technique. The IR-ATR spectrum of the titanium sample
functionalized with Dopa-PS (Figure 8a) exhibited the charac-
teristic absorption bands at 3025 cmÀ1 and at 1450, 1490, and
1600 cmÀ1 assigned to CÀH aromatic and CdC aromatic
stretching of the polystyrene backbone, respectively.47
Figure 7. Partial XPS survey spectra of titanium surface before and after
grafting of the dopamine end-functionalized PNIPAM.
dioxide surface31 which allows the bidentate catechol ligand to
strongly adhere to the TiO2 surface. The pretreated titanium
surface was then soaked in a solution containing 0.5 mM of
dopamine end-functionalized polymer overnight, resulting in
(after washing) the polymer functionalized titanium surface
(“grafting to” method, Scheme 2).
XPS Investigations. Successful immobilization of the poly-
mers on the titanium surface was first investigated by XPS.
Chemical compositions of end-functionalized grafted polymer
and unmodified titanium surfaces are summarized in Table 2. As
expected, the XPS-determined elemental composition of the
unmodified titanium surface showed the presence of titanium
and oxygen components. Carbon, nitrogen, and cationic species
The grafting of the Dopa-PNIPAM polymer onto the titanium
surface was revealed in the IR spectrum (Figure 8b) by the
appearance of NÀH and CdO stretching bands at 3350 and
1650 cmÀ1, respectively. An additional peak at 1546 cmÀ1
assigned to the NÀH bending further indicated the presence
of the amide group in the polymer layer.
Finally, the functionalization of the titanium surface with
Dopa-PtBA was evidenced by IR-ATR by the appearance of
the characteristic absorption band of the carbonyl group at
1725 cmÀ1
.
5889
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892

Macromolecules
ARTICLE
Table 3. Estimation of Surface Coverage Γ from SPR
Investigations
SPR Surface Coverage
Mn (g molÀ1 PDI C (mM) Γ (molecules/cm2)
polymers
)
Dopa-PNIPAM
12500
9000
1.10
1.12
0.25
1
4.4 Â 1012
Dopa-tBA
12.4 Â 1012
Figure 9. SPR sensograms recorded on the titanium surface during the
injection of (a) Dopa-PNIPAM (Mn = 12 500 g molÀ1, PDI = 1.10) and
3
(b) Phi-PNIPAM (Mn = 9500 g molÀ1, PDI = 1.10); [polymer] =
0.25 mM in water; buffer = water, T = 20 ꢀC.
3
Figure 11. AFM image of the titanium surface modified with dopamine
end-functionalized PNIPAM. The AFM image is a topographical image
with an image size of 2 Â 2 μm.
evidential impact on the resonance angle, implying that a stable
polymer brush was formed.
Figure 10. SPR sensograms recorded on the titanium surface during
injection of (a) Dopa-PtBA (Mn = 9000 g molÀ1, PDI = 1.12) and (b)
3
Phi-PtBA (Mn = 10 000 g molÀ1, PDI = 1.14); [polymer] = 1 mM in
acetonitrile; buffer = acetonitrile.
SPR experiments were also performed on the Dopa-PtBA
3
(Mn = 9000 g molÀ1, PDI = 1.12) (Figure 10) in acetonitrile and
3
showed trends similar to those observed for Dopa-PNIPAM,
indicating that dopamine end-functionalized polymers can also
bind with the titanium surface in the nonaqueous phase.
In order to investigate the role played by the catechol anchor,
Surface Plasmon Resonance. The ability of dopamine end-
functionalized PNIPAM and tert-butyl polymers to coordinate
with a titanium surface was also investigated by surface plasmon
resonance (SPR) spectroscopy in water and acetonitrile,
respectively.48 SPR is a label-free optical technique which is
sensitive to the changes in refractive index and thickness in the
close vicinity of the metal layer.49,50 Here, we have exploited this
technique to (i) study the binding events between polymers
(Dopa-PNIPAM and Dopa-PtBA) and the titanium surface, (ii)
investigate the role played by the catechol anchor in the
immobilization process, and (iii) get quantitative information
such as surface coverage.
we have studied the ability of Phi-PNIPAM (Mn = 9500 g molÀ1
3
PDI = 1.12) and Phi-PtBA (Mn = 10000 g molÀ1 PDI = 1.14),
3
which do not bear a catechol fragment, to interact with the
titanium surface by SPR in aqueous media and acetonitrile,
respectively. As shown in Figures 9b and 10b, in both cases
rinsing almost restored the SPR signal to the baseline, suggesting
that no coordination occurred and thereby indicating the im-
portant role of the catechol unit in the formation of stable brush
polymers.
Finally, the amount of bonded Dopa-PNIPAM and Dopa-
PtBA on the Ti sensor surface was calculated from the difference
between SPR signal of samples after the immobilization proce-
dure was finished and the SPR signal of the solvent (Table 3).
By using the De Feijter equation51 (see Supporting In-
formation), the surface coverage Γ of the grafted polymers on
the sensor surface was calculated to be 4.4 Â 1012 and 12.4 Â
1012 molecules/cm2 for Dopa-PNIPAM and Dopa-PtBA, re-
spectively. These grafting densities are lower than those expected
for self-assembled monolayers (1014 molecules/cm2).52 Thus,
taking also into account the thicknesses (3À4 nm) estimated by
As illustrated in Figure 9a, sensorgrams obtained for the
dopamine end-functionalized PNIPAM (Mn,NMR = 12500 g
3
molÀ1, PDI = 1.10), recorded below its LCST (≈ 32 ꢀC, see
Supporting Information), successively exhibited a dramatic shift
of the SPR angle after injection and then a progressive increase of
reflectance, which is indicative of adsorption and the formation of
an organic layer on the titanium surface.
Next, rinsing the SPR cell with water caused a small decrease
of the resonance angle, suggesting that no significant physisorp-
tion occurred. Moreover, subsequent extensive rinsing had no
5890
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892

Macromolecules
ARTICLE
XPS, these SPR results suggest that polymers are highly canted
on the surface or are in a mushroom formation.53,54
(5) Brunette, D.; Wieland, M.; Tosatti, S.; Textor, M. Use of
Molecular Assembly Techniques for Tailoring Chemical Properties of
Smooth and Rough Titanium Surfaces. In Bio-Implant Interface; CRC
Press: Boca Raton, FL, 2003.
(6) Xiong, L.; Liang, H.; Wang, R.; Chen, L. J. Polym. Res. 2010, 1–5.
(7) Meuer, S.; Fischer, K.; Mey, I.; Janshoff, A.; Schmidt, M.; Zentel,
R. Macromolecules 2008, 41, 7946–7952.
(8) Zoulalian, V.; Monge, S.; Z€urcher, S.; Textor, M.; Robin, J. J.;
Tosatti, S. J. Phys. Chem. B 2006, 110, 25603–25605.
(9) Zoulalian, V.; Z€urcher, S.; Tosatti, S.; Textor, M.; Monge, S.;
Robin, J.-J. Langmuir 2009, 26, 74–82.
(10) Portꢀe-Durrieu, M. C.; Guillemot, F.; Pallu, S.; Labrugꢁere, C.;
Brouillaud, B.; Bareille, R.; Amꢀedꢀee, J.; Barthe, N.; Dard, M.; Baquey, C.
Biomaterials 2004, 25, 4837–4846.
Atomic Force Microscopy. Finally, atomic force microscopy
was used to analyze the surface topography of the titanium film
deposed from Dopa-PNIPAM (Mn = 21000 g molÀ1, PDI = 1.15).
3
For this purpose, a titanium sample was half immersed in a 0.5 mM
solution of Dopa-PNIPAM in water overnight. After rinsing with
water and drying, AFM images were recorded in tapping mode.
The resulting AFM image (Figure 11) clearly indicates the
grafting of the end-functionalized polymer onto the titanium
surface, as a neat step between the modified and unmodified
surface is evidenced. Visual inspection suggests a near-homogeneous
grafted layer having a thickness of around 4À5 nm, in accordance
with XPS results.
(11) Barbey, R. l.; Lavanant, L.; Paripovic, D.; Sch€uwer, N.; Sugnaux,
C.; Tugulu, S.; Klok, H.-A. Chem. Rev. 2009, 109, 5437–5527.
(12) Ma, H.; Wells, M.; Beebe, T. P.; Chilkoti, A. Adv. Funct. Mater.
2006, 16, 640–648.
’ CONCLUSIONS
In conclusion, well-defined dopamine end-functionalized
polymers have been prepared via RAFT polymerization. A new
RAFT agent including a catechol unit fragment was designed and
used to mediate homopolymerizations of tert-butyl acrylate, N-
isopropylacrylamide, and styrene. Characteristic features of con-
trolled polymerizations were observed and a range of polymers
with controlled molar mass and narrow molar mass distributions
were prepared. The grafting of dopamine end-functionalized
polymers was monitored in real time by SPR and confirmed by
XPS, ATR, contact angle, and AFM investigations. Our future
work will focus on the preparation of well-defined dopamine end-
functionalized polymers incorporating long side-chain stimulable
and/or bioactive units and on their subsequent immobilization
on titanium surfaces, with the goal of creating smart biomaterials.
These works will be reported in due course.
(13) Ma, H.; Li, D.; Sheng, X.; Zhao, B.; Chilkoti, A. Langmuir 2006,
22, 3751–3756.
(14) B€unsow, J.; Kelby, T. S.; Huck, W. T. S. Acc. Chem. Res. 2009,
43, 466–474.
(15) Liu, X.; Ye, Q.; Yu, B.; Liang, Y.; Liu, W.; Zhou, F. Langmuir
2010, 26, 12377–12382.
(16) Nagase, K.; Kobayashi, J.; Okano, T. J. R. Soc. Interface 2009,
6, S293–S309.
(17) Minko, S. Grafting on Solid Surfaces: “Grafting to” and “Graft-
ing from” Methods. In Polymer Surfaces and Interfaces, Stamm, M., Ed.;
Springer: Berlin and Heidelberg, Germany, 2008; pp 215À234.
(18) Ye, Q.; Wang, X.; Li, S.; Zhou, F. Macromolecules 2010, 43,
5554–5560.
(19) Raynor, J. E.; Petrie, T. A.; Fears, K. P.; Latour, R. A.; García,
A. s. J.; Collard, D. M. Biomacromolecules 2009, 10, 748–755.
(20) Raynor, J. E.; Petrie, T. A.; García, A. J.; Collard, D. M. Adv.
Mater. 2007, 19, 1724–1728.
(21) Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921–2990.
(22) Gao, C.; Li, G.; Xue, H.; Yang, W.; Zhang, F.; Jiang, S.
Biomaterials 2010, 31, 1486–1492.
(23) Rowe-Konopacki, M. D.; Boyes, S. G. Macromolecules 2007,
40, 879–888.
’ ASSOCIATED CONTENT
S
Supporting Information. Characterization spectra of
b
Dopa-CTA and Phenyl-CTA. and GPC traces of polymers.
This material is available free of charge via the Internet at
http://pubs.acs.org
(24) Pissuwan, D.; Boyer, C.; Gunasekaran, K.; Davis, T. P.; Bulmus,
V. Biomacromolecules 2010, 11, 412–420.
(25) Hojjati, B.; Sui, R.; Charpentier, P. A. Polymer 2007, 48,
5850–5858.
’ AUTHOR INFORMATION
(26) Fan, X.; Lin, L.; Dalsin, J. L.; Messersmith, P. B. J. Am. Chem.
Soc. 2005, 127, 15843–15847.
(27) Statz, A. R.; Meagher, R. J.; Barron, A. E.; Messersmith, P. B.
J. Am. Chem. Soc. 2005, 127, 7972–7973.
Corresponding Author
*E-mail: patrice.woisel@ensc-lille.fr.
(28) Dalsin, J. L.; Hu, B.-H.; Lee, B. P.; Messersmith, P. B. J. Am.
Chem. Soc. 2003, 125, 4253–4258.
(29) Ye, Q.; Wang, X.; Hu, H.; Wang, D.; Li, S.; Zhou, F. J. Phys.
Chem. C 2009, 113, 7677–7683.
(30) Liu, J.; Yang, W.; Zareie, H. M.; Gooding, J. J.; Davis, T. P.
Macromolecules 2009, 42, 2931–2939.
(31) Watson, M. A.; Lyskawa, J.; Zobrist, C.; Fournier, D.; Jimenez,
M.; Traisnel, M.; Gengembre, L.; Woisel, P. Langmuir 2010, 26,
15920–15924.
(32) Goldmann, A. S.; Sch€odel, C.; Walther, A.; Yuan, J.; Loos, K.;
M€uller, A. H. E. Macromol. Rapid Commun. 2010, 31, 1608–1615.
(33) Song, H. M.; Ye, P. D.; Ivanisevic, A. Langmuir 2007, 23, 9472–
9480.
(34) Dalsin, J. L.; Lin, L.; Tosatti, S.; V€or€os, J.; Textor, M.; Messersmith,
P. B. Langmuir 2004, 21, 640–646.
(35) Wach, J.-Y.; Malisova, B.; Bonazzi, S.; Tosatti, S.; Textor, M.;
Z€urcher, S.; Gademann, K. Chem.—Eur. J. 2008, 14, 10579–10584.
(36) Fan, X.; Lin, L.; Messersmith, P. B. Compos. Sci. Technol. 2006,
66, 1198–1204.
’ ACKNOWLEDGMENT
The authors would like to thank CNRS, the Region Nord/Pas-
de-Calais and the European FEDER program for funding to
Surface Plasmon Resonance (SPR) apparatus at UMET. We thank
Aurꢀelie Malfait for the GPC analysis and Catherine MELIET for
her assistance with the elemental analyses. The Region Nord Pas-
de-Calais and PRIM SP are thanked for funding of the ELE-
MENTAR, vario MICRO equipment.
’ REFERENCES
(1) Brunette, D.; Tengvall, P.; Textor, M., Titanium in Medicine:
Material Science, Surface Science, Engineering, Biological Responses and
Medical Applications (Engineering Materials); Springer: Berlin, 2001.
(2) Niinomi, M. Mater. Sci. Eng., A 1998, 243, 231–236.
(3) Thomas, R. Geothermics 2003, 32, 679–687.
(4) Li, L.; Hu, W.; Chi, L.; Fuchs, H. J. Phys. Chem. B 2010,
114, 5315–5319.
5891
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892

Macromolecules
ARTICLE
(37) Qiu, X.-P.; Tanaka, F.; Winnik, F. M. Macromolecules 2007,
40, 7069–7071.
(38) Ran, R.; Yu, Y.; Wan, T. J. Appl. Polym. Sci. 2007, 105, 398–404.
(39) Moad, G.; Barner-Kowollik, C., The Mechanism and Kinetics of
the RAFT Process: Overview, Rates, Stabilities, Side Reactions, Product
Spectrum and Outstanding Challenges; Wiley-VCH Verlag GmbH & Co.
KGaA: Weinheim, Germany, 2008; pp 51À104.
(40) Bigot, J.; Charleux, B.; Cooke, G.; Delattre, F.; Fournier, D.;
Lyskawa, J.; Stoffelbach, F.; Woisel, P. Macromolecules 2010, 43, 82.
(41) Viornery, C.; Chevolot, Y.; Lꢀeonard, D.; Aronsson, B.-O.;
Pꢀechy, P.; Mathieu, H. J.; Descouts, P.; Gr€atzel, M. Langmuir 2002,
18, 2582–2589.
(42) Nanci, A.; Wuest, J. D.; Peru, L.; Brunet, P.; Sharma, V.; Zalzal,
S.; McKee, M. D. J. Biomed. Mater. Res 1998, 40, 324–335.
(43) Laforgue, A.; Addou, T.; Bꢀelanger, D. Langmuir 2005,
21, 6855–6865.
(44) Walters, K. B.; Hirt, D. E. Polymer 2006, 47, 6567–6574.
(45) Li, Y.; Pham, J. Q.; Johnston, K. P.; Green, P. F. Langmuir 2007,
23, 9785–9793.
(46) Gilcreest, V. P.; Carroll, W. M.; Rochev, Y. A.; Blute, I.;
Dawson, K. A.; Gorelov, A. V. Langmuir 2004, 20, 10138–10145.
(47) Pei, Z.; Busenlehner, L.; Worley, S. D.; Tang, Y.; Curtis, C. W.
J. Phys. Chem. B 1997, 101, 10450–10454.
(48) The immobilization of dopamine end-functionalized PS was
not studied by SPR due to its poor solubility in employed solvents
(49) Homola, J.; Yee, S. S.; Gauglitz, G. Sens. Actuators B: Chem.
1999, 54, 3–15.
(50) Chang, Y.; Chu, W.-L.; Chen, W.-Y.; Zheng, J.; Liu, L.; Ruaan,
R.-C.; Higuchi, A. J. Biomed. Mater. Res., Part A 2010, 93A, 400–408.
(51) De Feijter, J. A.; Benjamins, J.; Veer, F. A. Biopolymers 1978,
17, 1759–1772.
(52) Zareie, H. M.; Boyer, C.; Bulmus, V.; Nateghi, E.; Davis, T. P.
ACS Nano 2008, 2, 757–765.
(53) Wu, T.; Efimenko, K.; Genzer, J. J. Am. Chem. Soc. 2002, 124,
9394–9395.
ꢂ
(54) Wu, T.; Efimenko, K.; Vlꢂcek, P.; Subr, V.; Genzer, J. Macro-
molecules 2003, 36, 2448–2453.
5892
dx.doi.org/10.1021/ma200853w |Macromolecules 2011, 44, 5883–5892