Dual-functional ROMP-based betaines: Effect of hydrophilicity and backbone structure on nonfouling properties

Article abstract of DOI:10.1021/la203683u

Foundational materials for nonfouling coatings were designed and synthesized from a series of novel dual-functional zwitterionic polymers, Poly[NRZI], which were easily obtained via ring-opening metathesis polymerization (ROMP) followed by a single step transformation of the cationic precursor, Poly[NR(+)], to the zwitterion, Poly[NRZI]. The resulting unique dual-functional structure contained the anion and the cation within the same repeat unit but on separate side chains, enabling the hydrophilicity of the system to be tuned at the repeat unit level. These dual-functional zwitterionic polymers were specifically designed to investigate the impact of structural changes, including the backbone, hydrophilicity, and charge, on the overall nonfouling properties. To evaluate the importance of backbone structure, and as a direct comparison to previously studied methacrylate-based betaines, norbornene-based carbo- and sulfobetaines (Poly[NCarboZI] and Poly[NSulfoZI]) as well as a methacrylate-based sulfobetaine (Poly[MASulfoZI]) were synthesized. These structures contain the anion-cation pairs on the same side chain. Nonfouling coatings were prepared from copolymers, composed of the zwitterionic/cationic precursor monomer and an ethoxysilane-containing monomer. The coatings were evaluated by using protein adsorption studies, which clearly indicated that the overall hydrophilicity has a major influence on the nonfouling character of the materials. The most hydrophilic coating, from the oligoethylene glycol (OEG)-containing dual-functional betaine, Poly[NOEGZI-co-NSi], showed the best resistance to nonspecific protein adsorption (Λ_FIB = 0.039 ng/mm²). Both norbornene-based polymers systems, Poly[NSulfoZI] and Poly[NCarboZI], were more hydrophilic and thus more resistant to protein adsorption than the methacrylate-based Poly[MASulfoZI]. Comparing the protein resistance of the dual-functional zwitterionic coatings, Poly[NRZI-co-NSi], to that of their cationic counterparts, Poly[NR(+)-co-NSi], revealed the importance of screening electrostatic interactions. The adsorption of negatively charged proteins on zwitterionic coatings was significantly less, despite the fact that both coatings had similar wetting properties. These results demonstrate that the unique, tunable dual-functional zwitterionic polymers reported here can be used to make coatings that are highly efficient at resisting protein adsorption.

Full text of DOI:10.1021/la203683u

ARTICLE
pubs.acs.org/Langmuir
Dual-Functional ROMP-Based Betaines: Effect of Hydrophilicity and
Backbone Structure on Nonfouling Properties
Semra Colak and Gregory N. Tew*
Department of Polymer Science and Engineering, University of Massachusetts—Amherst, Conte Research Center for Polymers, 120
Governor’s Drive, Amherst, Massachusetts 01003, United States
S Supporting Information
b
ABSTRACT: Foundational materials for nonfouling coatings were
designed and synthesized from a series of novel dual-functional
zwitterionic polymers, Poly[NRZI], which were easily obtained via
ring-opening metathesis polymerization (ROMP) followed by a single
step transformation of the cationic precursor, Poly[NR(+)], to the
zwitterion, Poly[NRZI]. The resulting unique dual-functional structure
contained the anion and the cation within the same repeat unit but on
separate side chains, enabling the hydrophilicity of the system to be
tuned at the repeat unit level. These dual-functional zwitterionic
polymers were specifically designed to investigate the impact of
structural changes, including the backbone, hydrophilicity, and charge,
on the overall nonfouling properties. To evaluate the importance of
backbone structure, and as a direct comparison to previously studied
methacrylate-based betaines, norbornene-based carbo- and sulfobetaines (Poly[NCarboZI] and Poly[NSulfoZI]) as well as a
methacrylate-based sulfobetaine (Poly[MASulfoZI]) were synthesized. These structures contain the anionÀcation pairs on the
same side chain. Nonfouling coatings were prepared from copolymers, composed of the zwitterionic/cationic precursor monomer
and an ethoxysilane-containing monomer. The coatings were evaluated by using protein adsorption studies, which clearly indicated
that the overall hydrophilicity has a major influence on the nonfouling character of the materials. The most hydrophilic coating, from
the oligoethylene glycol (OEG)-containing dual-functional betaine, Poly[NOEGZI-co-NSi], showed the best resistance to
nonspecific protein adsorption (ΓFIB = 0.039 ng/mm2). Both norbornene-based polymers systems, Poly[NSulfoZI] and
Poly[NCarboZI], were more hydrophilic and thus more resistant to protein adsorption than the methacrylate-based Poly-
[MASulfoZI]. Comparing the protein resistance of the dual-functional zwitterionic coatings, Poly[NRZI-co-NSi], to that of their
cationic counterparts, Poly[NR(+)-co-NSi], revealed the importance of screening electrostatic interactions. The adsorption of
negatively charged proteins on zwitterionic coatings was significantly less, despite the fact that both coatings had similar wetting
properties. These results demonstrate that the unique, tunable dual-functional zwitterionic polymers reported here can be used to
make coatings that are highly efficient at resisting protein adsorption.
’ INTRODUCTION
in a 40% increase in fuel consumption to counteract the added
drag. Increased fuel costs combined with the cost of additional
maintenance translates to roughly an extra $500 million annually.6
Therefore, development of new technologies and materials that
reduce or completely prevent any undesired deposition of micro-
organisms is of great importance. However, tailoring an efficient
nonfouling material requires understanding the interactions in-
volved, which is hindered by the complexity of the process, making
it challenging to provide solutions to the problem.
Biofilm formation generally starts within seconds following
implantation of a given material (e.g., medical implant) in body
fluids, such as blood. The first step of the process is the adsorption
of proteins on the substrate surface, which is followed by a cascade
Biofouling, or biofilm formation, remains a challenging pro-
blem for numerous fields ranging from biomedical applications1
and marine coating technologies2 to water purification, transport,
and storage systems.3 Undesirable consequences of this ubiqui-
tous problem include, but are not limited to, reduction in the
efficacy/sensitivity of devices, operational losses, thrombosis,
and microbial infections.4 In the U.S. alone, about half of the
2 million cases of hospital acquired infections are associated with
bacterial biofilm formation on biomedical devices such as
catheters, surgical implants, and prosthetics. In addition to
causing various severe or fatal health complications, such infec-
tions place a significant economic burden on society. Estimated
direct costs associated with these hospital-acquired infections
exceed $3 billion annually.5 Similarly, biofouling on ship hulls,
caused by the buildup of marine microorganisms, plants, algae,
and crustaceans, can reduce vessel speed by up to 10%, resulting
Received: September 20, 2011
Revised:
November 22, 2011
Published: November 29, 2011
r
2011 American Chemical Society
666
dx.doi.org/10.1021/la203683u Langmuir 2012, 28, 666–675
|

Langmuir
ARTICLE
of larger, more complex species that include microorganisms such
as bacteria, fungi, and algae.7 Thus, one can argue that the larger
microorganisms rarely interact with the clean surface but rather
with the proteins adsorbed on it.8 At the interface of materials
science and biology, several different approaches have been em-
ployed to prevent the biofouling process. One approach is to
incorporate antimicrobially active ingredients such as antibiotics,
biocidal molecules, and silver particles in the material, by blending
them with the plastic of the device, covalently attaching them, or
coating/painting them onto the material.5,9 However, for this
approach, once a certain amount of cell debris has formed, the
surface activity decreases as the active material can no longer reach
the pathogen. An alternative approach is to prevent biofouling by
impeding the attachment of a broad range of species rather than
killing them. Since the first step of biofouling is protein adsorption,
efforts focused on preventing their initial adhesion to the substrate,
which would significantly reduce the subsequent adhesion of other
species, would appear to provide a longer-standing solution to the
problem.10 The challenge of this approach is that even 0.1 ng/mm2
of protein adsorption on a surface can induce full scale biofilm
formation, resulting in loss of function of the implantable device.11
Despite the amount of research done in the area, there are few
materials that can effectively fulfill this requirement and resist
biofilm formation. To date, the most widely employed protein-
resistant materials are based on poly(ethylene glycol) (PEG)
or oligo(ethylene glycol) (OEG).12 While the exact mecha-
nism of action of these materials is still debated, the promis-
ing results led to extensive efforts in developing new methods
to immobilize PEG on surfaces, including physisorption,13,14
chemisorption,12,15 and covalent grafting.16À18 Even though
OEG self-assembled monolayers (SAMs) exhibited low protein
adsorption values (0.3À3 ng/mm2),19 their propensity for
defects and limited use on metal surfaces led to the development
of OEG/PEG coatings via surface-initiated polymerization.20,21
This technique produced thicker, more densely packed, robust
coatings composed of PEG brushes which reduced protein
adsorption values to as low as 0.1 ng/mm2.16 PEG shows
effective nonfouling character; however, it is unstable in the
presence of oxygen and transition metal ions, resulting in loss of
function in most biological media.22 As a result, there is still an
ongoing search for alternative, more robust materials that
successfully resist protein adsorption. Whitesides and co-
workers10,23,24 designed and analyzed a series of SAMs contain-
ing different functional groups leading to the important observa-
tion that the best functional groups for inhibiting protein
adsorption shared several common features: the presence of
hydrophilic groups, hydrogen-bond donors, and electrical
neutrality.
monomers.10,33À35 The first example of sulfobetaine-based low-
fouling coatings was reported by Lowe et al.36 The biofouling
resistance of these coatings was tested against bacteria, macro-
phage, and fibroblast cells, and the bioadhesion was found to be
considerably lower than that of the poly(methyl methacrylate)
control coatings. Similar surfaces were prepared by Jiang and co-
workers through surface-initiated atom transfer radical polymer-
ization (ATRP) and had more controlled and densely packed
chains.37,38 These surfaces, composed of zwitterionic brushes,
had a higher packing density and showed better resistance to
protein adsorption (0.003À0.01 ng/mm2) than zwitterionic
SAMs (0.01À0.1 ng/mm2).39 Even though these studies indicate
that zwitterionic systems show comparable, or in some cases,
even better protein resistance than PEG-based materials, they are
still prone to failure in long-term usage.29,40 Methacrylate-based
MPC, sulfo-, and carboxybetaines are all composed of ester
linkages, making them hydrolyzable under physiological condi-
tions. In addition, they suffer from complex or labor-intensive
coating preparation conditions, including requirements for air-
and moisture-free media during the preparation process. Regard-
less of the intended application, there is still a need for more
stable and effective structures as well as efficient ways of incor-
porating them into coatings.
In this report, we introduce novel zwitterionic molecules
based on ring-opening metathesis polymerization (ROMP) as
foundational materials for nonfouling coatings. Two polymer
systems, shown in Figure 1a,1b, were designed and synthesized.
The first system is composed of a series of polymers that carry
dual functionality at the repeat unit level (Figure 1a), consisting
of a zwitterionic functionality coupled with an alkyl moiety that is
systematically tuned to adjust the relative hydrophilicity/phobi-
city of the overall system, using oligoethylene, methyl, and octyl
chains. To the best of our knowledge, this is the first example of
such a system, and it is easily obtained using the ROMP
chemistry platform. The second system, shown in Figure 1b, is
established as a direct comparison to the previously well-studied
methacrylate (MA)-based betaines (Figure 1c) to provide insight
into whether the methacrylate backbone, in addition to the
betaine functionality, is critical for achieving the reported non-
fouling properties. Using these zwitterionic polymers as the base
for nonfouling coatings, we investigated how structural changes
of the zwitterionic material, including the hydrophilicity/pho-
bicity and backbone, impacted their overall nonfouling proper-
ties against several proteins including albumin, fibrinogen, and
lysozyme.
’ RESULTS AND DISCUSSION
This insight has been used in the design strategy of future
nonfouling materials. Among them, zwitterionic polymers, hav-
ing a highly hygroscopic nature similar to PEG, as well as lipid-
like biomimetic features (e.g., 2-methacryloyloxyethylphosphor-
ylcholine (MPC)-based macromolecules),25À30 were shown to
be promising protein-resistant materials. MPC-based coatings
gave protein adsorption amounts varying from 0.03 ng/mm2
(brushes prepared via grafting from)31 to 3 ng/mm2 (bulk
coatings).32 These materials showed suppression of clot forma-
tion and platelet adhesion, even when in contact with human
whole blood without anticoagulants.30 In addition to the MPC-
based materials, other zwitterionic structures that have been
investigated include carboxybetaines, sulfobetaines, and surfaces
prepared from a 1:1 mixture of cationic and anionic SAMs or
Polymer Synthesis and Characterization. A summary of all
the monomers and the strategies used to synthesize them are
shown in Scheme 1. Even though ring-opening metathesis
polymerization (ROMP) can be successfully utilized to synthe-
size well-defined ionic polymers,41À45 it is also well-known that
the reaction kinetics can be significantly retarded or stopped in
the presence of particular functionalities, such as amine, carbox-
ylate, and hydroxyl groups.46À49 Therefore, to prevent the
carboxylate group of the zwitterionic monomers from diminish-
ing the polymerization kinetics, monomers 2aÀ2d were synthe-
sized as quaternary ammonium (cationic) precursors. The
monomer 2e, NSulfoZI (for nomenclature see Figure 1 caption),
was obtained in its zwitterionic form, as the sulfonate function-
ality has been shown not to interfere with ROMP kinetics.43,45
667
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675

Langmuir
ARTICLE
Figure 1. Summary of synthesized zwitterionic polymers and their design principles. (a) ROMP-based, dual-functional betaines with tunable
hydrophilicity/phobicity, (b) ROMP-based carboxy- and sulfobetaines, and (c) MA-based sulfobetaine. N = norbornene, ZI = zwitterion, MA =
methacrylate, OEG = oligoethylene glycol, C1 = methyl, and C8 = octyl.
Scheme 1. Monomer Synthesisa
a Conditions and reagents: (i) 2a: THF, 50 °C, 48 h; 2b: THF, 0 °C, 48 h; 2c: THF, 50 °C, 48 h; 2d: THF, 50 °C, 48 h. (ii) 1,3-Dinitrobenzene (1 wt %),
acetonitrile, 50 °C, 48 h.
The quaternization of the tertiary amine precursor, 1, was done
through its alkylation with an OEG, a methyl, octyl, or tert-butÀyl
or BrÀ) (see Table in Scheme 1 and see Supporting Information
for detailed synthetic procedures). For 2aÀ2c, the alkyl chains
were selected to systematically increase the hydrophobicity of the
targeted zwitterionic functionality.
Ionic polymers are intrinsically hydrophilic, making them soluble
in aqueous environments. Therefore, to form stable coatings, the
polymers must be covalently attached to the substrate to prevent
their dissolution into the surrounding media, which would even-
tually result in loss of surface activity. Although attaching these
ROMP-based polymers to solid surfaces is nontrivial, it is essential to
their development as coating materials. Attachment of previously
studied nonfouling polymers required complex and highly sophis-
ticated immobilization methods or grafting brushes to/from sur-
faces which are not suitable for larger-scale applications.16,37,50 In
order to prepare coatings that can be easily applied on a large scale at
ambient conditions, all of the monomers were copolymerized in a
random fashion with an ethoxysilane-containing monomer, 5-bicy-
cloheptenyl triethoxysilane (NSi) (Scheme 2). The ethoxysilane
units on the resulting copolymer chains were used to both covalently
attach them to the silica/glass surface, through the formation of
strong SiÀOÀSi bonds, and to improve the stability of the coatings
through interchain cross-linking.50,51
butyrate chain bearing a good leaving group (TsOÀ, CH3SO4
,
All the monomers were copolymerized in a mixture of
dichloromethane (DCM) and 2,2,2-trifluoroethanol (TFE)
to provide a homogeneous medium for the polymerization,
where DCM was selected as a good solvent for the catalyst and
5-bicycloheptenyltriethoxysilane, and TFE for the ionic
monomers.45,52 The third-generation Grubbs’ catalyst was
used as the initiator for all of the polymerizations and they
were terminated using ethyl vinyl ether as the quencher
1
(Scheme 2). H NMR analysis was used to investigate the
extent of polymerization, where the complete disappearance
of the peaks at 6.50 and 5.15 ppm, corresponding to the
double bond protons of the norbornene ring, and the appear-
ance of broad peaks around 6.15 and 5.90 ppm, corresponding
to the unsaturated protons of the backbone, indicated com-
plete conversion (see Supporting Information). The copoly-
mer compositions were adjusted to contain 15À20 mol % of
the ethoxysilane-containing monomer so that a sufficient
668
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675

Langmuir
ARTICLE
Scheme 2. Copolymer Synthesisa
a Nomenclature: Poly[NR(+)-co-NSi], N = norbornene backbone, R = alkyl group, and NSi = 5-bicycloheptenyl triethoxysilane. All the copolymers were
synthesized with n = 30 and m = 6 as the degree of polymerization values of the corresponding monomers to yield 16.6% Si per polymer chain.
quantity of anchoring groups was present for strong adhesion,
without compromising the overall hydrophilic character of the
coatings.
Table 1. Molecular Weight Characterization of the Copoly-
mers Synthesized
theoreticala
experimentalb
Along with the copolymers of 2aÀ2e and 5-bicycloheptenyl-
triethoxysilane, two additional copolymers, Poly[NCH3-co-NSi]
and Poly[MASulfoZI-co-MASi], were synthesized using the same
design principles (see Supporting Information for detailed
synthetic procedures). These two copolymers were designed to
produce control coatings. Poly[NCH3-co-NSi] was synthesized
as the neutral hydrophobic control (structure shown in Support-
ing Information). Since this polymer contains the identical
backbone of the zwitterionic polymers, its performance would
provide information about the importance of the zwitterionic
functionality in nonfouling behavior. Poly[MASulfoZI-co-MASi], a
methacrylate sulfobetaine monomer copolymerized with an
ethoxysilane monomer, was prepared using ATRP as a control
coating.53,54 It is well-known that the nonfouling character of a
coating is greatly influenced by the surface composition (i.e.,
packing density, thickness, architecture) of the active material.4
In order to compare the protein resistance of these novel
zwitterionic polymers to the well-studied MASulfoZI, it was
critical to prepare “identical”, or as similar as possible, surfaces.
Therefore, Poly[MASulfoZI-co-MASi] was synthesized in a
similar manner to the norbornene-based copolymers to eliminate
any factors that could arise due to differences in coating methods
and enable a direct comparison to the methacrylate sulfobetaines
reported in the literature. Comparing the coatings made from
zwitterions with methacrylate and norbornene backbones, Poly-
[MASulfoZI-co-MASi] and Poly[NSulfoZI-co-NSi], allowed in-
sight as to whether the polymer backbone structure significantly
affects protein adsorption.
polymer
Mn [kDa]
Mn [kDa]
nc
mc %Sie
Poly[NOEG(+)-co-NSi]
Poly[NC1(+)-co-NSi]
Poly[NC8(+)-co-NSi]
Poly[NCarbo(+)-co-NSi]
Poly[NSulfoZI-co-NSi]
Poly[NCH3-co-NSi]
16.9
12.4
14.3
15.3
12.7
6.9
19.7
9.5
34.6 8.0 18.8
22.5 5.0 18.1
30.0 7.5 20.0
21.0 5.6 21.0
32.3d 6.0d 15.7
29.4 7.5 20.3
27.9 5.7 16.9
14.8
11.1
n.d.d
7.2
Poly[MASulfoZI-co-MASi]
10.1
9.5
a Theoretical: n = 30, m = 6, and %Si = 16.6. b Mn, n, m, and %Si
determined by end-group analysis of 1H NMR. c See Scheme 2 for n
and m. d Determined from the ratio of n/m since the end-group was
not seen on the 1H NMR spectra taken in D2O. e %Si = mole percent
of 5-bicycloheptenyl triethoxysilane present in the copolymer com-
position.
shown in Table 1. (For the 1H NMR spectra, see Supporting
Information Figures S4ÀS10.) The molecular weights and
monomer compositions calculated using the degree of polym-
erization of the corresponding monomers, as determined by
1H NMR analysis, were in strong agreement with the the-
oretical ones.
Coating Preparation and Characterization. The copolym-
erization of the ionic monomers with the ethoxysilane function-
ality allowed for a very simple coating preparation procedure,
which is illustrated in Figure S11. The resulting coatings were
durable and resistant to acid and base solutions, which allowed
for the postcure functionalization of the cationic coatings to yield
their zwitterionic counterparts. Among the cationic precursor
coatings, Poly[NCarbo(+)-co-NSi] was subjected to postfunc-
tionalization using acid (4 M HCl/dioxane) to remove the tert-
butyl group and yield Poly[NCarboZI-co-NSi] (see Scheme in
Figure S12), while the remaining ones were treated with dilute
base (0.1 M NaOH) to ring open the cyclic maleimide and obtain
their zwitterionic counterparts (see Scheme in Figure S13).
Poly[NSulfoZI-co-NSi], Poly[MASulfoZI-co-MASi], and Poly-
[NCH3-co-NSi] required no further treatment as they were
polymerized in their zwitterionic or final functional forms.
Surface compositions for coatings requiring postfunctionaliza-
tion were analyzed, before and after treatment, using Fourier
transform infraredÀattenuated total reflectance spectroscopy
(FTIR-ATR) and X-ray photoelectron spectroscopy (XPS).
The cationic coatings of the dual-functional system, Poly[NR-
(+)-co-NSi], were treated with 0.1 M NaOH solution to yield
Gel permeation chromatography (GPC) of charged polymers
is challenging due to the likelihood of large aggregate formation
or interactions with the stationary phase of the chromatographic
column.55 It requires specialized columns and an aqueous mobile
phase. Additionally, the copolymers synthesized here contain
ethoxysilane units that are readily hydrolyzable in aqueous media,
facilitating cross-linking reactions and the formation of insoluble
material. Thus, the molecular weight characterization of the
synthesized polymers was done by end-group analysis using 1H
NMR spectroscopy. Previously, we demonstrated the living
ROMP of Poly[NCarboZI] and Poly[NSulfoZI] as well as a
1
direct comparison between H NMR and absolute molecular
weight, which showed high correlation.45 The molecular weights
as well as the molar compositions of the copolymers were
determined by comparing the integration values of the phenyl
end-group protons to the protons specific to each comonomer.
Detailed molecular weight characterization of all the polymers is
669
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675

Langmuir
ARTICLE
Table 2. Hydrophilicity Measurements of the Monomers and Zwitterionic Coatings Prepared
water contact angle (deg)
polymer
Rt (min)a
θstatic
θadvancing
θreceding
roughness (nm)b
Poly[NOEGZI-co-NSi]
Poly[NC1ZI-co-NSi]
5.9
7.3
32 ( 1
36 ( 4
70 ( 3
43 ( 2
49 ( 2
63 ( 4
32 ( 2
38 ( 5
54 ( 1
42 ( 3
50 ( 4
70 ( 1
18 ( 1
23 ( 3
31 ( 1
25 ( 3
26 ( 2
41 ( 2
0.5
0.5
1.3
0.4
0.4
0.3
Poly[NC8ZI-co-NSi]
30.9
6.5
Poly[NSulfoZI-co-NSi]
Poly[NCarboZI-co-NSi]
Poly[MASulfoZI-co-MASi]
9.8
7.2
a Retention times (Rt) of the corresponding monomers as measured by HPLC using a C8 column with a gradient of 1% CH3CN/min starting with
100% water. b Rms roughness was calculated with the AFM manufacturer’s provided software on images of 1 μm  1 μm scan size.
their zwitterionic forms, Poly[NRZI-co-NSi]. The cyclic imide
functionality of the norbornene monomer, provided that it
contains a neighboring quaternary ammonium group, can be
easily ring-opened under basic conditions to yield amide and
carboxylic acid moieties (see scheme in Figure S13).45 This
single step transformation, easily obtainable using our ROMP
platform, was the key point in the design and synthesis of the
dual-functional systems introduced here (Figure 1a). As a result,
novel structures that carry a hydrophilic/phobic functionality
coupled with the zwitterionic one, in the same repeat unit, were
obtained. In comparison to the traditional zwitterionic polymers
that have the cation and the anion on the same side chain, these
dual-functional zwitterions contained the cationic and the anio-
nic functionalities on separate side chains, however, still within
the same repeat unit. Figure S13 shows the FTIR-ATR spectra of
a representative coating before and after postfunctionalization, in
which the reduction of the peak at 1706 cmÀ1, corresponding to
the imide carbonyl stretching and appearance of new peaks at
1660 and 1587 cmÀ1, corresponding to amide carbonyl stretch-
ing along with a broad peak at 3300 cmÀ1 of NÀH stretching,
indicated extensive conversion. A small peak remaining at
1706 cmÀ1 was an indication that the bottom of the coating
was not fully functionalized. However, we further confirmed that
the unreacted material was only at the bottom of the coatings, as
full conversion on the coating surface was observed by XPS,
where a shift in the nitrogen 1s binding energy from 400 to 398
eV was observed upon conversion from imide to amide (see
Supporting Information Figure S14).
The deprotection of the carboxylate of Poly[NCarbo(+)-co-
NSi] was performed via acid treatment to yield Poly[NCarboZI-
co-NSi] (scheme in Figure S12). Figure S12 shows the FTIR
spectra obtained before and after the deprotection step. A
decrease in the intensity of the peak that corresponds to both
the ester and the imide carbonyl stretching at 1703 cmÀ1 was
observed due to the conversion of the ester to an acid, while the
imide stayed intact. The appearance of two new peaks at 1665
and 1578 cmÀ1, both corresponding to the carboxylate carbonyl
stretching, indicated a successful transformation. This ring-
opening reaction and the removal of the tert-butyl group were
previously reported for similar structures in solution.45 Both 1H
NMR and FTIR indicated that those reactions proceeded
quantitatively to generate the desired zwitterions.45
single wavelength ellipsometry and ranged from 20 to 35 nm.
These results were in agreement with the measurements ob-
tained from atomic force microscopy (AFM) using section
analysis (Figure S17). The surface morphology of these coatings,
analyzed by AFM, was found to be relatively smooth with no
phase separation and roughness ranging from 0.5 to 1.3 nm (see
Table 2 and Figure S18). AFM also revealed only minor swelling
(2À3 nm) of the films upon immersion in water, indicating
highly cross-linked network formation.
Contact angle measurements using the sessile drop method
were performed to analyze the hydrophilicity of the coatings. The
results are summarized in Table 2. Poly[NOEGZI-co-NSi] had
the lowest contact angle, θstatic = 32° ( 1° (θA/θR = 32°/18°),
indicating that this was the most hydrophilic coating obtained.
Within the dual-functional series (Figure 1a), increasing contact
angle values were observed with increasing hydrophobicity of the
alkyl chain, with Poly[NC8ZI-co-NSi] being the most hydro-
phobic coating (θA/θR = 54°/31°). These results were in
agreement with HPLC retention times (Rt) of the corresponding
zwitterionic monomers (Table 2). The contact angle values for
Poly[NCarboZI-co-NSi] and Poly[NSulfoZI-co-NSi] were with-
in the same range, while Poly[MASulfoZI-co-MASi] demon-
strated a relatively higher contact angle. HPLC results also
tracked with contact angles for these monomers; however,
HPLC and contact angle values did not track across zwitterionic
type. For example, Poly[NC1ZI-co-NSi] has θA/θR = 38°/23°
and a 7.3 min Rt of its corresponding monomer (NC1ZI);
however, while MASulfoZI has a very similar Rt of 7.2 min, its
corresponding coating, Poly[MASulfoZI-co-MASi], has higher
contact angles (θA/θR = 70°/41°) (Table 2). Overall, it can be
clearly seen that within the dual system the wettability of the
coatings is strongly influenced by the hydrophilicity of the side
chains, which was in accordance with expectations. The contact
angle values of the cationic precursors showed similar wetting
properties to their zwitterionic counterparts (Table S1). The
comparison of the cationic precursors to their zwitterionic
forms, having the same molecular composition and thus similar
wetting properties, but different ionic characters, allowed the
effect of charge on the nonfouling character of a coating to be
examined.
Protein Adsorption Studies. Preventing, or at least minimiz-
ing, the adsorption of proteins is believed to be the key point in
providing longer-lasting solutions to the biofouling problem.4
Therefore, evaluating the ability of these coatings to resist
nonspecific protein adsorption was studied using five different
proteins (Table S2). These particular proteins were chosen in
order to investigate two factors. First, to test the efficacy of the
Coating thickness was easily tuned by varying the polymer
concentration of the spin-coating solution (see Supporting
Information Figure S16). However, to be consistent throughout
all the studies, coatings were prepared from 1% w/v polymer
solutions. The resulting coating thicknesses were measured using
670
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675

Langmuir
ARTICLE
Table 3. Protein Adsorption Amounts on the Zwitterionic Coatings and the Controls, As Obtained from Ellipsometry
Measurements
protein adsorption amount (ng/mm2)
polymer
BSA
fibrinogen
lysozyme
myoglobin
cytochrome c
Poly[NOEGZI-co-NSi]
Poly[NC1ZI-co-NSi]
Poly[NC8ZI-co-NSi]
Poly[NSulfoZI-co-NSi]
Poly[NCarboZI-co-NSi]
Poly[MASulfoZI-co-MASi]
Poly[NCH3-co-NSi]
silica
0.046 ( 0.01
0.49 ( 0.06
4.79 ( 1.04
0.12 ( 0.08
0.59 ( 0.06
0.83 ( 0.02
1.07 ( 0.09
0.47 ( 0.10
0.04 ( 0.02
0.76 ( 0.22
3.44 ( 0.24
1.03 ( 0.23
0.84 ( 0.26
3.39 ( 0.11
4.67 ( 0.40
4.15 ( 0.02
0.65 ( 0.11
0.70 ( 0.15
3.51 ( 0.58
0.96 ( 0.16
0.56 ( 0.07
1.45 ( 0.27
2.94 ( 0.16
1.29 ( 0.10
0.24 ( 0.06
0.45 ( 0.13
1.63 ( 0.10
0.63 ( 0.04
0.95 ( 0.10
2.01 ( 0.29
1.04 ( 0.10
1.37 ( 0.03
0.41 ( 0.08
0.26 ( 0.07
1.30 ( 0.04
0.93 ( 0.07
0.65 ( 0.17
0.94 ( 0.08
2.62 ( 0.11
2.02 ( 0.10
synthesized coatings to resist the adsorption of common pro-
teins, including bovine serum albumin (BSA), fibrinogen, and
lysozyme. Second, to investigate the role of electrostatics on
protein adsorption,56 while eliminating the effect of protein size
(molecular weight). Although albumin, fibrinogen, and lysozyme
span a range of electrophoretic mobilities (zeta potentials) from
relatively negative (albumin) to positive (lysozyme), they are
quite different in size and shape. Larger proteins are known to
provide more hydrophobic and van der Waals interaction con-
tacts, and this was shown to enhance protein adsorption.57,58
Therefore, to eliminate these effects, two additional proteins
were also studied. These were myoglobin and cytochrome c,
with negative and approximately neutral electrophoretic mo-
bilities (zeta potentials), respectively, but molecular weights
essentially identical to positive lysozyme. Table S2 summarizes
the properties of the proteins, including molecular weight,
isoelectric point, density, and electrophoretic mobility values
(zeta potential) as measured in phosphate buffered saline
(PBS) (pH = 7.4).
Prior to protein adsorption, all the surfaces were extensively
soaked with PBS to extract any unreacted material. Thickness
measurements on these coatings, before and after, showed no
significant change, indicating successful cross-linking and strong
adhesion to the substrate. A literature procedure19,59 was fol-
lowed for the protein adsorption studies, where the coatings were
first hydrated in PBS and then incubated in the protein solutions
for 2 h (see Supporting Information for details). These coatings
were removed and further washed with an excess of PBS followed
by deionized water to remove any loosely adhered proteins, and
dried under vacuum prior to any analysis. Protein adsorption
on the coatings was studied using two different techniques.
Ellipsometry measurements60À62 were performed to quantify the
amount of adsorbed protein and fluorescence microscopy63,64
was employed for qualitative analysis, using fluorescein isothio-
cyanate (FITC) labeled BSA to visually study protein adsorption.
The differences in coating thickness before and after protein
adsorption was measured using ellipsometry to yield the thick-
ness of the adsorbed protein layer. The amount of protein
adsorbed was calculated by multiplying the thickness of the
adsorbed protein layer by its density.57,60 Protein adsorption
results obtained by both techniques (fluorescence microscopy
and ellipsometry) were consistent with each other. In addition to
the zwitterionic coatings, their corresponding cationic precursors
were also analyzed using the same techniques. This direct
comparison between cationic and zwitterionic coatings was
performed to provide further information about the effect of
electrostatic interactions on protein adsorption.
Adsorption Summary. Table 3 provides a complete summary
of the amounts of protein adsorbed on the zwitterionic coatings
and the controls (clean silica, the hydrophobic uncharged Poly-
[NCH3-co-NSi], and Poly[MASulfoZI-co-MASi]). (Protein ad-
sorption on the corresponding cationic precursors can be found
in Supporting Information Table S3.) This table shows that
regardless of the protein used, all the ROMP-based zwitterionic
coatings, except the most hydrophobic one (Poly[NC8ZI-co-
NSi]), performed better in resisting nonspecific protein adsorp-
tion than the controls. Among them, Poly[NSulfoZI-co-NSi]
and Poly[NOEGZI-co-NSi] showed the best resistance, with
adsorption amounts as low as ΓBSA = 0.12 ng/mm2 and ΓFIB
=
0.039 ng/mm2, respectively. A similar trend was observed among
both the dual-functional system (Figure 1a) and the more
traditional zwitterions in which the amount of protein adsorption
increased with increasing hydrophobicity (contact angle). In
general, this was common to all the proteins studied independent
of their physical characteristics. Figure 2 shows the adsorption of
fibrinogen on all of the zwitterionic coatings and controls. This
protein was selected to simplify the discussion as it emphasizes
the trends found among all the zwitterionic coatings summarized
in Table 3. In addition, fibrinogen has been widely studied
allowing stronger comparisons between these coatings and other
surfaces in the literature.21,67À69
Figure 2 shows that, within the dual-functional coating system,
as the length of the alkyl chain, or hydrophobicity, was increased,
the protein adsorption values increased (Figure 2, dual-
functional). While this is the first example of tuning the hydro-
philicity of a zwitterionic system at the repeat unit level, similar
studies were performed for OEG-based SAMs, where the hydro-
philicity of the system was tuned either by modifying the end
groups65 or by varying the length of the SAMs.66,67 Our
observations were consistent with the conclusions drawn from
these previous studies; lower protein adsorption values were
obtained on coatings with higher hydrophilicity. Among the
dual-functional coatings, Poly[NOEGZI-co-NSi] showed the
lowest fibrinogen adsorption, ΓFIB = 0.039 ng/mm2, which was
also below the 0.1 ng/mm2 limit.11 This compares well to the
“best performing” surfaces in the literature including surfaces
composed of OEG-SAMs68 and MASulfoZI-brushes39 that
showed fibrinogen adsorption of 0.18 and <0.02 ng/mm2,
respectively. The Poly[NC1ZI-co-NSi] coating also performed
better than the three controls including the one containing
671
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675

Langmuir
ARTICLE
Figure 2. Fibrinogen adsorption on controls and zwitterionic coatings. Dotted line indicates the 0.1 ng/mm2 protein adsorption limit that can induce
full scale biofilm formation, resulting in loss of function of the implantable device.11 Error bars represent standard error from three independent
measurements. Data for BSA and lysozyme adsorption are shown in Figures S19 and S20, respectively.
MASulfoZI, which is known to be highly resistant to protein
adsorption in polymer brushes.39
For the coatings containing the more traditional zwitterionic
groups, protein adsorption also increased with hydrophobicity
(contact angle). As shown in Figure 2, both norbornene-based
betaines, Poly[NSulfoZI-co-NSi] (ΓFIB = 1.026 ng/mm2)
and Poly[NCarboZI-co-NSi] (ΓFIB = 0.84 ng/mm2), performed
better than the three controls including the methacrylate-based
sulfobetine, Poly[MASulfoZI-co-MASi] (ΓFIB = 3.39 ng/mm2).
Both Poly[NSulfoZI-co-NSi] and Poly[MASulfoZI-co-MASi]
contain the same sulfobetaine functional groups and only dif-
fer by one methylene group. In methacrylate-based polycarbo-
betaine systems it has been shown previously that when shorter
spacers (methylenes) are present between the cationic and anionic
group, better resistance to protein adsorption is achieved.69 The
fact that Poly[NSulfoZI-co-NSi] outperforms Poly[MASulfoZI-
co-MASi], given that these coatings are made in a similar fashion
and that Poly[NSulfoZI-co-NSi] contains one extra meth-
ylene,69 indicates that the polymer backbone can positively
influence the ability of zwitterionic coatings to resist protein
adsorption.
Figure 3. Adsoprtion vs contact angle for each zwitterionic coating. The
diamond symbols represent the average protein adsorption of all five
proteins studied. The bars represent the adsorption range observed for
all five different proteins (the highest point indicating the highest protein
adsorption value observed or vice versa). See Supporting Information
Figure S22 for the graph showing all the data points of the corresponding
proteins.
Another important observation also shown in Figure 2 is that
Poly[NC1ZI-co-NSi] and Poly[NCarboZI-co-NSi], composed of
the same functional groups (quaternary ammonium as the cation
and carboxylate as the anion), showed similar wetting properties
and similar protein adsorption amounts (ΓFIB = 0.76 ng/mm2
and ΓFIB = 0.84 ng/mm2, respectively). However, while Poly-
[NCarboZI-co-NSi] represents a traditional betaine structure,
Poly[NC1ZI-co-NSi] is quite different (see Figure 1). It contains
a trimethylammonium cation and a separate carboxylate anion.
The comparison of these two coatings demonstrates that “se-
paration” of the charges in this new architecture does not impair
the nonfouling character of the material. This is an important fact
because it expands the scope of chemistry one can envisage. As
shown here for the first time, this new approach allows the
incorporation of various side chains into the zwitterionic
functionalities. These moieties can be used to tune the final
properties of the zwitterionic materials. The best example of
this is Poly[NOEGZI-co-NSi], which showed the strongest
resistance to all five proteins studied. The zwitterionic func-
tionality inherently increases the hydrophilicity of the polymer
system, which is further improved by the OEG side chain,
creating a synergistic effect.
672
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675

Langmuir
ARTICLE
Figure 4. Adsorption behavior of proteins with different isoelectric points and electrophorectic mobilities on cationic and zwitterionic coatings. Error
bars represent standard error from three independent measurements. The contact angles for the four samples are θA/θR = 32°/18°, 32°/18°, 38°/21°,
and 38°/23° respectively for coatings labeled a, b, c, and d in this figure.
Fluorescence microscopy images of FITC-BSA adsorption on
Poly[NSulfoZI-co-NSi] and Poly[MASulfoZI-co-MASi] are
shown in Figures S21a and S21b, respectively. These results,
which are consistent with the ellipsometry data, indicate that
indeed the backbone structure has an important impact on the
nonfouling character of the coatings. According to the contact
angle measurements (Table 2), it is clear that norbornene-based
Poly[NSulfoZI-co-NSi] and Poly[NCarboZI-co-NSi] generate
more hydrophilic surfaces than Poly[MASulfoZI-co-MASi],
which results in increased wettability of the coatings and thus
improved nonfouling character. Figures S21c and S21d compare
the fluorescence microscopy images of FTIC-BSA adsorption on
the cationic coating Poly[NOEG(+)-co-NSi] and its zwitterionic
counterpart Poly[NOEGZI-co-NSi]. Both coatings have identi-
cal advancing and receding contact angles (θA/θR = 32°/18°) yet
show remarkably different fluorescent intensities. The fluores-
cence microscopy image of the cationic coating is much brighter
green than the one corresponding to the zwitterionic coating
which is again consistent with ellipsometry measurements (also
see ellipsometry data discussed below). This suggests that it may
be possible to use fluorescence microscopy to screen new coat-
ings for protein adsorption before selecting the “best” ones for
quantitative analysis using ellipsometry methods.
Figure 3, which plots protein adsorption as a function of
contact angle for all six zwitterionic coatings, also supports the
conclusion that backbone structure impacts the nonfouling
character of the coating. This graph shows the corresponding
average protein adsorption amounts on the specific zwitter-
ionic coatings (red diamonds) as well as the range of the
maximum and minimum adsorption observed (the black
bars). Figure 3 shows that Poly[NSulfoZI-co-NSi], on average,
had lower protein adsorption than Poly[MASulfoZI-co-MASi]
and that Poly[NOEGZI-co-NSi] had the lowest average pro-
tein adsorption of all the coatings studied. Overall, a clear
trend is seen, where the protein adsorption increases with in-
creasing contact angle values, indicating that the greater the
hydrophilicity of a zwitterionic coating, the more efficient it is
in resisting nonspecific protein adsorption.
Cationic versus Zwitterionic Coatings. Figures 4 and 5 show
protein adsorption for four coatings as a function of protein
studied. Figure 4 compares the three widely studied proteins —
albumin, fibrinogen, and lysozyme—while Figure 5 compares
the three proteins of varying electrophoretic mobility (or zeta
potential) but similar size (molecular weight) and shape. From
these two figures, three trends are apparent: First, zwitterionic
coatings show less protein adsorption than their corresponding
cationic coatings despite the fact that they have essentially
identical wetting properties (see Figure 4 caption). This result
is not necessarily surprising since the importance of electrostatic
attraction between surfaces and protein adsorption has been
documented; however, the systems described here represent very
closely related coatings for making this comparisons.56 Second,
the amount of adsorbed protein on the cationic coatings tracks
with their measured electrophoretic mobilities (zeta potential);
as the eletrophorectic mobility values of the proteins shifted from
more negative to positive, their adsorption decreased. The
protein adsorption trend on Poly[NC1(+)-co-NSi] was ΓBSA
>
ΓFIB > ΓLYS (Figure 4) and ΓMYG > ΓCYT‑C > ΓLYS (Figure 5),
while Poly[NOEG(+)-co-NSi] showed a similar trend, but one
that was slightly shifted to more negative eletrophorectic mobi-
lities, where ΓBSA > ΓFIB ∼ ΓLYS (Figure 4) and ΓMYG > ΓCYT‑C
∼ ΓLYS (Figure 5). This indicates that the cationic OEG contain-
ing coatings can limit the adsorption of less negatively charged
proteins better than the cationic NC1 containing coatings. Third,
smaller proteins (myoglobin, cytochrome c, and lysozyme)
adsorbed more than the larger ones (albumin and fibrinogen)
on the zwitterionic coatings, especially on Poly[NOEGZI-co-
NSi]. One possible explanation for the observed higher adsorp-
tion of the smaller proteins is that these smaller proteins are able
to adsorb more effectively onto small defects on the coating
surface that are inaccessible to larger proteins. At the same time
other explanations, like a higher packing density, yielding higher
673
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675

Langmuir
ARTICLE
Figure 5. Effect of the protein’s electrophorectic mobility (zeta potential) on its adsorption behavior with cationic and zwitterionic coatings. Error bars
represent standard error from three independent measurements. The contact angles for the four samples are θA/θR = 32°/18°, 32°/18°, 38°/21°, and
38°/23° respectively for coatings labeled a, b, c, and d in this figure.
adsorption amounts, cannot be ruled out. This observation is in
contrast to the expectation that larger proteins can adsorb more
due to the increased number of hydrophobic and van der Waals
interactions with the surface.58
modifying the side chain. Poly[NOEGZI-co-NSi], the most
hydrophilic material, demonstrated the best resistance to non-
specific protein adsorption among all the coatings made. Protein
adsorption values within the range of previously reported metha-
crylate-based sulfo- or carbobetaine SAMs (0.01À0.1 ng/mm2) and
in some cases, significantly lower than the 0.1 ng/mm2 limit
(amount above which full biofilm formation occurs), were
obtained. Poly[NOEGZI-co-NSi], Poly[NC1ZI-co-NSi], Poly-
[NCarboZI-co-NSi], and Poly[NSulfoZI-co-NSi] were shown
to be promising nonfouling zwitterionic materials easily applic-
able to large-scale systems. These materials were specifically
designed without any hydrolyzable units for improved stability
under physiological conditions, making them excellent candi-
dates for applications such as nonfouling coatings, hydrogels, and
biological scaffolds.
’ CONCLUSIONS
Novel, norbornene-based zwitterionic polymers were studied
as promising materials for nonfouling applications. A novel dual
functional system, Poly[NRZI-co-NSi], composed of a hydro-
philic/phobic side chain coupled with the zwitterionic function-
ality, was introduced for the first time. This unique design
enabled tuning of the wetting properties of the zwitterionic
system by varying the side-chain hydrophobicity at the repeat
unit level. In addition, the screening of any electrostatic interac-
tions that might occur between the protein and the coating was
shown to be important by the direct comparison of the cationic
precursors to their zwitterionic forms, where considerably lower
protein adsorption was observed on the zwitterionic coatings.
Poly[NCarboZI-co-NSi] and Poly[NSulfoZI-co-NSi], re-
sembling traditional zwitterionic polymers with the charged
groups on the same pendant side chain, were synthesized as a
direct comparison to the methacrylate-based sulfobetaine,
Poly[MASulfoZI-co-MASi]. This enabled the first direct study
on the effect of the backbone structure. Poly[NOEGZI-co-NSi]
and Poly[NSulfoZI-co-NSi] outperformed Poly[MASulfoZI-
co-MASi] indicating that the methyl methacrylate backbone is
not essential and new macromolecular structures can generate
better nonfouling materials than these classical zwitterions.
The resistance to protein adsorption of these coatings, investi-
gated by measuring the adsorption amounts of proteins having
different physical characteristics, demonstrated the role of improved
wettability on imparting nonfouling character to zwitterionic mate-
rials. A similar trend was observed within both systems; increased
hydrophilicity (lower contact angles) resulted in reduced protein
adsorption. The unique dual-functional system has the added
advantage of expanding the chemical scope of zwitterions by simply
’ ASSOCIATED CONTENT
S
Supporting Information. All experimental procedures,
b
including monomer and polymer synthesis, as well as detailed
information about characterization techniques and protein ad-
sorption studies. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: tew@mail.pse.umass.edu.
’ ACKNOWLEDGMENT
This work was supported by the Office of Naval Research
(N00014-10-1-0348) and the National Science Foundation
(DMR-0805061, DMR-0820506). Mass spectral data were ob-
tained at the University of Massachusetts Mass Spectrometry
Facility, which is supported, in part, by the National Science
Foundation. Dr. Ke Zhang is gratefully acknowledged for his
674
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675

Langmuir
ARTICLE
helpful discussion. Authors also thank Ms. Melissa Lackey and
Dr. Abhigyan Som for their invaluable comments on early drafts.
(32) Xu, Y.; Takai, M.; Konno, T.; Ishihara, K. Lab Chip 2007,
7, 199.
(33) Holmlin, R. E.; Chen, X. X.; Chapman, R. G.; Takayama, S.;
Whitesides, G. M. Langmuir 2001, 17, 2841.
(34) Chen, S. F.; Zheng, J.; Li, L. Y.; Jiang, S. Y. J. Am. Chem. Soc.
2005, 127, 14473.
’ REFERENCES
(1) Werner, C.; Maitz, M. F.; Sperling, C. J. Mater. Chem. 2007, 17,
3376.
(2) Krishnan, S.; Weinman, C. J.; Ober, C. K. J. Mater. Chem. 2008,
18, 3405.
(3) Flemming, H. C.; Griebe, T.; Schaule, G. In Antifouling Strategies
in Technical Systems - A Short Review; 18th Biennial Conference of the
International-Association-on-Water-Quality; Pergamon-Elsevier Science
Ltd.: Dordrecht, 1996; p 517.
(4) Banerjee, I.; Pangule, R. C.; Kane, R. S. Adv. Mater. 2011, 23, 690.
(5) Darouiche, R. O. New Engl. J. Med. 2004, 350, 1422.
(6) http://www.onr.navy.mil/Media-Center/Fact-Sheets/Biofouling-
Prevention.aspx (accessed on May 2011).
(7) Hall-Stoodley, L.; Costerton, J. W.; Stoodley, P. Nat. Rev.
Microbiol. 2004, 2, 95.
(8) Knetsch, M. L.W.; Koole, L. H. Polymer 2011, 3, 340.
(9) Madkour, A. E.; Dabkowski, J. A.; Nusslein, K.; Tew, G. N.
Langmuir 2009, 25, 1060.
(10) Kane, R. S.; Deschatelets, P.; Whitesides, G. M. Langmuir 2003,
19, 2388.
(11) Tsai, W. B.; Grunkemeier, J. M.; Horbett, T. A. J. Biomed. Mater.
Res. 1999, 44, 130.
(35) Chen, S.; Jiang, S. Adv. Mater. 2008, 20, 335.
(36) Lowe,A. B.;Vamvakaki, M.;Wassall, M. A.; Wong, L.;Billingham,
N. C.; Armes, S. P.; Lloyd, A. W. J. Biomed. Mater. Res. 2000, 52, 88.
(37) Zhang, Z.; Chao, T.; Chen, S.; Jiang, S. Langmuir 2006, 22,
10072.
(38) Zhang, Z.; Chen, S. F.; Jiang, S. Y. Biomacromolecules 2006, 7,
3311.
(39) Chang, Y.; Liao, S.; Higuchi, A.; Ruaan, R.; Chu, C.; Chen, W.
Langmuir 2008, 24, 5453.
(40) Wang, D. A.; Williams, C. G.; Li, Q. A.; Sharma, B.; Elisseeff,
J. H. Biomaterials 2003, 24, 3969.
(41) Ilker, M. F.; Schule, H.; Coughlin, E. B. Macromolecules 2004,
37, 694.
(42) Alfred, S. F.; Al-Badri, Z. M.; Madkour, A. E.; Lienkamp, K.;
Tew, G. N. J. Polym. Sci., Polym. Chem. 2008, 46, 2640.
(43) Rankin, D. A.; Lowe, A. B. Macromolecules 2008, 41, 614.
(44) Colak, S.; Nelson, C. F.; Nusslein, K.; Tew, G. N. Biomacro-
molecules 2009, 10, 353.
(45) Colak, S.; Tew, G. N. Macromolecules 2008, 41, 8436.
(46) Slugovc, C. Macromol. Rapid Commun. 2004, 25, 1283.
(47) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18.
(48) Lynn, D. M.; Mohr, B.; Grubbs, R. H. J. Am. Chem. Soc. 1998,
120, 1627.
(12) Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1993,
115, 10714.
(13) Elbert, D. L.; Hubbell, J. A. J. Biomed. Mater. Res. 1998, 42, 55.
(14) Liu, V. A.; Jastromb, W. E.; Bhatia, S. N. J. Biomed. Mater. Res.
2002, 60, 126.
(15) Bearinger, J. P.; Terrettaz, S.; Michel, R.; Tirelli, N.; Vogel, H.;
Textor, M.; Hubbell, J. A. Nature Mater. 2003, 2, 259.
(16) Hucknall, A.; Rangarajan, S.; Chilkoti, A. Adv. Mater. 2009,
21, 2441.
(17) Jo, S.; Park, K. Biomaterials 2000, 21, 605.
(18) Malmsten, M.; Emoto, K.; Van Alstine, J. M. J. Colloid Interface
Sci. 1998, 202, 507.
(49) Lynn, D. M.; Mohr, B.; Grubbs, R. H.; Henling, L. M.; Day, M. W.
J. Am. Chem. Soc. 2000, 122, 6601.
(50) Sambhy, V.; Peterson, B. R.; Sen, A. Langmuir 2008, 24, 7549.
(51) Colak, S.; Kusefoglu, S. H. J. Appl. Polym. Sci. 2007, 104, 2244.
(52) Rankin, D. A.; P’Pool, S. J.; Schanz, H.; Lowe, A. B. J. Polym. Sci.,
Polym. Chem. 2007, 45, 2113.
(53) Dong, H.; Ye, P.; Zhong, M.; Pietrasik, J.; Drumright, R.;
Matyjaszewski, K. Langmuir 2010, 26, 15567.
(54) Kobayashi, M.; Terada, M.; Terayama, Y.; Kikuchi, M.; Takahara,
A. Macromolecules 2010, 43, 8409.
(19) Sharma, S.; Johnson, R. W.; Desai, T. A. Langmuir 2004,
20, 348.
(20) Schonenberger, C.; Sondaghuethorst, J. A. M.; Jorritsma, J.;
Fokkink, L. G. J. Langmuir 1994, 10, 611.
(55) Volet, G.; Lesec, J. J. Liq. Chromatogr. 1994, 17, 559.
(56) Mueller, B.; Zacharias, M.; Rezwan, K. Adv. Eng. Mater. 2010,
12, B53.
(57) Voros, J. Biophys. J. 2004, 87, 553.
(21) Ma, H. W.; Wells, M.; Beebe, T. P.; Chilkoti, A. Adv. Funct.
Mater. 2006, 16, 640.
(58) Jeon, S. I.; Andrade, J. D. J. Colloid Interface Sci. 1991, 142, 159.
(59) Sharma, S.; Popat, K. C.; Desai, T. A. Langmuir 2002, 18, 8728.
(60) Malmsten, M. Abstr. Pap. Am. Chem. Soc. 1994, 207, 71.
(61) Defeijter, J. A.; Benjamins, J.; Veer, F. A. Biopolymers 1978,
17, 1759.
(22) Shen, M. C.; Martinson, L.; Wagner, M. S.; Castner, D. G.;
Ratner, B. D.; Horbett, T. A. J. Biomater. Sci., Polym. Ed. 2002, 13, 367.
(23) Chapman, R. G.; Ostuni, E.; Takayama, S.; Holmlin, R. E.; Yan,
L.; Whitesides, G. M. J. Am. Chem. Soc. 2000, 122, 8303.
(24) Ostuni, E.; Chapman, R. G.; Holmlin, R. E.; Takayama, S.;
Whitesides, G. M. Langmuir 2001, 17, 5605.
(25) Ishihara, K.; Nomura, H.; Mihara, T.; Kurita, K.; Iwasaki, Y.;
Nakabayashi, N. J. Biomed. Mater. Res. 1998, 39, 323–330.
(26) Ishihara, K.; Aragaki, R.; Ueda, T.; Watenabe, A.; Nakabayashi, N.
J. Biomed. Mater. Res. 1990, 24, 1069.
(62) Elwing, H. Biomaterials 1998, 19, 397.
(63) Kim, J.; Somorjai, G. A. J. Am. Chem. Soc. 2003, 125, 3150.
(64) Togashi, D. M.; Ryder, A. G.; Heiss, G. Colloids Surf., B 2009,
72, 219.
(65) Herrwerth, S.; Eck, W.; Reinhardt, S.; Grunze, M. J. Am. Chem.
Soc. 2003, 125, 9359.
(66) Schilp, S.; Rosenhahn, A.; Pettitt, M. E.; Bowen, J.; Callow,
M. E.; Callow, J. A.; Grunze, M. Langmuir 2009, 25, 10077.
(67) Schilp, S.; Kueller, A.; Rosenhahn, A.; Grunze, M.; Pettitt,
M. E.; Callow, M. E.; Callow, J. A. Biointerphases 2007, 2, 143.
(68) Li, L.; Chen, S.; Zheng, J.; Ratner, B. D.; Jiang, S. J. Phys. Chem.
B 2005, 109, 2934.
(27) Ueda, T.; Watanabe, A.; Ishihara, K.; Nakabayashi, N.
J. Biomater. Sci., Polym. Ed. 1991, 3, 185.
(28) Ishihara, K.; Ziats, N. P.; Tierney, B. P.; Nakabayashi, N.;
Anderson, J. M. J. Biomed. Mater. Res. 1991, 25, 1397.
(29) Ueda, T.; Oshida, H.; Kurita, K.; Ishihara, K.; Nakabayashi, N.
Polym. J. 1992, 24, 1259.
(30) Ishihara, K.; Oshida, H.; Endo, Y.; Ueda, T.; Watanabe, A.;
Nakabayashi, N. J. Biomed. Mater. Res. 1992, 26, 1543.
(31) Sibarani, J.; Konno, T.; Takai, M.; Ishihara, K. In Surface
Modification by Grafting with Biocompatible 2-Methacryloyloxyethyl Phos-
phorylcholine for Microfluidic Devices; Kim, Y. H., Cho, C. S., Kang, I. K.,
Kim, S. Y. Kwon, O. H., Eds.; ASBM7: Advanced Biomaterials VII; 2007;
Vol. 342À343, p 789.
(69) Zhang, Z.; Vaisocherova, H.; Cheng, G.; Yang, W.; Xue, H.;
Jiang, S. Biomacromolecules 2009, 10, 2686.
675
dx.doi.org/10.1021/la203683u |Langmuir 2012, 28, 666–675