The influence of fluorocarbon chain and phosphorylcholine on the improvement of hemocompatibility: A comparative study in polyurethanes

Article abstract of DOI:10.1039/c3tb21473g

To study the influence of fluorinated surfaces and biomimetic surfaces on the improvement of the blood compatibility of polymers, three monomers containing a fluorinated tail and/or phosphorylcholine groups were designed and synthesized, and were then introduced into polyurethanes based on 4,4′-diphenylmethane diisocyanate (MDI), poly(tetramethylene glycol) (PTMG) and 1,4-butanediol (BDO) via end-capping. The bulk and surface characterization of the polyurethanes was carried out by dynamic mechanical analysis (DMA), Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopic analysis (XPS), atomic force microscope (AFM), and water contact angle measurements. The results indicate that the fluorocarbon chains can drive the phosphorylcholine groups to aggregate at the surface of polyurethane, and the two components show spontaneous arrangement to adapt to the environment when in contact with water. The preliminary evaluation of hemocompatibility was carried out via fibrinogen adsorption and platelet adhesion. The fluorocarbon chains and phosphorylcholine groups showed a synergistic effect on the improvement of hemocompatibility.

Full text of DOI:10.1039/c3tb21473g

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
The influence of fluorocarbon chain and
phosphorylcholine on the improvement of
hemocompatibility: a comparative study in
polyurethanes†
Cite this: J. Mater. Chem. B, 2014, 2,
1344
*
*
Dongsheng Tan, Zhen Li, Xuelin Yao, Chunlan Xiang, Hong Tan and Qiang Fu
To study the influence of fluorinated surfaces and biomimetic surfaces on the improvement of the blood
compatibility of polymers, three monomers containing a fluorinated tail and/or phosphorylcholine
groups were designed and synthesized, and were then introduced into polyurethanes based on 4,40-
diphenylmethane diisocyanate (MDI), poly(tetramethylene glycol) (PTMG) and 1,4-butanediol (BDO) via
end-capping. The bulk and surface characterization of the polyurethanes was carried out by dynamic
mechanical analysis (DMA), Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron
spectroscopic analysis (XPS), atomic force microscope (AFM), and water contact angle measurements.
The results indicate that the fluorocarbon chains can drive the phosphorylcholine groups to aggregate at
the surface of polyurethane, and the two components show spontaneous arrangement to adapt to the
environment when in contact with water. The preliminary evaluation of hemocompatibility was carried
out via fibrinogen adsorption and platelet adhesion. The fluorocarbon chains and phosphorylcholine
groups showed a synergistic effect on the improvement of hemocompatibility.
Received 19th October 2013
Accepted 2nd December 2013
DOI: 10.1039/c3tb21473g
www.rsc.org/MaterialsB
end-cap poly(carbonate urethane)s, and the uorinated poly-
(carbonate urethane)s revealed a unique bilayered surface
1. Introduction
structure which effectively improved the biostability and anti-
platelet adhesion performance of this kind of polyurethane.9
Fluorocarbon chains have also been introduced into the chain
extender (in the hard segment) or the so segment of poly-
urethanes.10,11 The uorocarbon chains could usually enhance
the phase separation and migrate to the top surface, resulting in
polyurethane samples exhibiting low surface energies and good
blood compatibility. In fact, an in vitro platelet test has shown
that the platelet compatibility of polyurethanes could be
improved when the uorocarbon oligomers were directly graf-
ted onto the surface of polyurethanes.12,13
Introducing phospholipid analogues to form biomembrane
mimicking polymers is another effective way to improve hemo-
compatibility by suppressing protein adsorption, platelet adhe-
sion, and platelet activation on the surface of materials.14–17 It is
known that surface characteristics always play a decisive role in
the blood compatibility of biomaterials, and the free and bound
water on the surface of biomimetic polymers have been proven
to play a crucial role in the outstanding biocompatibility.18,19
Therefore, the phospholipid moieties should be designed to
appear at the interface of polyurethane surfaces in order to
function. For example, Iwasaki et al. prepared semi-inter-
penetrating polymer networks (IPNs) by the radical polymeriza-
tion of diffused 2-methacryloyloxyethyl phosphorylcholine
(MPC) in segmented polyurethane membranes;20 Lee et al. used
Various polymers used in the biomaterials eld need to be
modied to improve their biocompatibility, mechanical prop-
erties, hydrolysis resistance properties, or other functionalities.
In blood contacting materials, most research strategies consist
of modulating the interface by modifying the synthetic gra
surface properties, with the aim of improving their hemo-
compatibility.1–6 For instance, polyurethanes (PU) are now
widely used in biomedical apparatus and instruments because
of their relatively good antithrombogenicity, excellent compli-
ance, and benecial physical/mechanical properties. However,
the compatibility of PU with blood is not sufficient for making
small-diameter vascular prostheses for in vivo applications,3,7
and among the various modication methods, surface uori-
nation and surface biomimetic design are two efficient ways for
improving its hemocompatibility.
It is well known that uorocarbon chains exhibit unique
characteristics, such as low surface energy, oxidative stability,
and good water and oil repulsion.4,8 As a result, a large number
of uorinated polymers have been found to be relatively
hemocompatible. Xie et al. applied a uorinated alcohol to
College of Polymer Science and Engineering, State Key Laboratory of Polymer Materials
Engineering, Sichuan University, Chengdu 610065, China. E-mail: qiangfu@scu.edu.
cn; hongtan@scu.edu.cn; Fax: +86-28-85405402; Tel: +86-28-85460961
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c3tb21473g
1344 | J. Mater. Chem. B, 2014, 2, 1344–1353
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
a poly(MPC-co-n-butyl methacrylate) solution to coat poly- with a tapping mode of 0.4–0.7 amplitude at room temperature.
urethane surfaces to form a phospholipid interface, which Contact angles of the samples were measured using the sessile
showed a qualitative improvement in hemocompatibility during drop method with a digital optical contact angle meter DSA100
implantation.5
(KRUSS GmbH, Germany) with 3 mL of water at 25 ꢀC and
Both the uorocarbon chain and phosphorylcholine have relative humidity of 80%. Six lms were tested for each sample.
been demonstrated to effectively improve the blood compati-
bility of polymers by being introduced on their surfaces. It is
unclear, however, which is more effective in the improvement of
2.3. Synthesis
hemocompatibility, or if a synergistic effect can occur between
2.3.1. Synthesis of tert-butyl 2-(2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-
them when they both appear in the same polymer surface? This pentadecauorooctanamido) ethylcarbamate. A solution of
motivated us to study the inuence of uorocarbon chains and 2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-pentadecauorooctanoicacid (a, 10
phosphorylcholine on the improvement of blood compatibility g, 24.2 mmol) and N-methylmorpholine (2.92 mL, 26.6 mmol) in
in polyurethanes. In the present work, four kinds of poly- anhydrous tetrahydrofuran (100 mL) was cooled to ꢁ15 ꢀC, then
urethanes with different chain ends (uorocarbon chain and/or isobutyl chlorocarbonate (3.17 mL, 5.0 mmol) was added. Aer
phosphorylcholine) were synthesized and characterized to the mixture was stirred for 10 min, N-boc-ethylenediamine
explore the mechanism of the anticoagulation effect through (BDA) (b, 3.87 g, 24.2 mmol) in 10 mL anhydrous tetrahydro-
comparison.
furan was added dropwise to the reaction mixture. The reaction
mixture was stirred for 2 h at room temperature and the solvent
was removed using a rotary evaporator. Then the residue was
dissolved in ethyl acetate. The solution was washed with dilute
HCl, brine and distilled water, dried over MgSO4, ltered under
vacuum and evaporated in a vacuum to give crude c. The crude c
was puried by silica gel column chromatography using a
gradient of petroleum ether–ethyl acetate to yield 6.8 g of c
(50.5%, Rf ¼ 0.61 in 60% petroleum ether–ethyl acetate).
APCI MS (positive) m/z: theoretical 556 g molꢁ1, observed
2. Experimental
2.1. Materials
2-Chloro-1,3,2-dioxaphospholane-2-oxide (COP, 95%) was
purchased from Acros Organics. 2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-
Pentadecauorooctanoicacid (95%), 5-amino-1-pentanol (95%),
and dibutyltin dilaurate (DBTDL, 95%) were purchased from
Aldrich. 4,40-Diphenylmethane diisocyanate (MDI, Aladdin) and
1,4-butanediol (BDO, Aladdin) were used as received. Poly-
(tetramethylene glycol) (PTMG, Mw ¼ 1000, obtained from
Aldrich) was dehydrated at 100 ꢀC for 4 h under vacuum. Other
reagents and solvents were all obtained from Kelong Co.
(Chengdu, China) in AR grade and were used as received.
579 g molꢁ1
.
1H NMR (CDCl3, 400 MHz) d ppm: 1.44 (s, 9H, –C(CH3)3),
3.36–3.41 (m, 2H, –CH2–), 3.45–3.50 (m, 2H, –CH2–), 4.92 (s, 1H,
–NH–), 7.95 (s, 1H, –NH–).
2.3.2. Synthesis of N-(2-aminoethyl)-2,2,3,3,4,4,5,5,6,7,7,8,8,8-
tetradecauorooctanamide. Compound c (5.0 g, 9.0 mmol) was
dissolved in 50 mL of ethyl acetate, 20 mL of saturated HCl–ethyl
acetate was added, and the mixture was stirred at ambient
2.2. Instrumentation
1H NMR spectra were recorded on a Varian Unity INOVA (400 temperature for 1 h. Aer evaporation of the solvent, ethyl ether
MHz) spectrometer, using tetramethylsilane (TMS) as an (3 ꢂ 50 mL) was added to the residue and the precipitated white
internal standard, and CDCl3 or DMSO-d6 as solvents. Mass crude product was collected by ltration. Then, the crude
spectra (MS) were obtained on a TSQ Quantum Ultra LC/MS/MS product was dissolved in methanol, and adjusted to pH 10 by
system (San Jose, CA, USA), and the electrospray ionization the addition of 1 N NaHCO3. Aer the removal of the methanol
source was run in positive ion mode for all experiments. Gel under vacuum, the N-(2-aminoethyl)-2,2,3,3,4,4,5,5,6,7,7,8,8,8-tet-
permeation chromatography (GPC) was performed with a radecauorooctanamide (FA) was extracted with ethyl acetate and
Waters-1515 apparatus using N,N-dimethylformamide (DMF)– further washed with brine and distilled water, dehydrated over
LiBr as eluent. The molecular weights are relative to polymethyl anhydrous MgSO4, ltered under vacuum and evaporated in a
methacrylate (PMMA) standards. The sample concentration was vacuum to yield 3.5 g of FA (85.3%).
2–3 mg mLꢁ1, and the ow rate was 1.000 mL minꢁ1 at 40 C.
APCI MS (positive) m/z: theoretical 456 g molꢁ1, observed
ꢀ
XPS was carried out on an XSAM-800 electron spectrometer 457 g molꢁ1
equipped with a Mg Ka achromatic X-ray source (20 kV, 10 mA).
1H NMR (DMSO, 400 MHz) d ppm: 2.89 (t, 2H, –CH2N–), 3.43
The contents of phosphorus (P) and uorine (F) were detected (t, 2H, –NCH2–), 5.32 (t, 1H, –NH–), 7.85 (br, 2H, –NH2).
by an inductive coupled plasma emission spectrometer (ICP- 2.3.3. Synthesis of tert-butyl-5-hydroxypentylcarbamate. To
.
AES, IRIS 1000, Thermo Fisher Scientic) and an ion chro- a solution of 5-amino-1-pentanol (d, 10.3 g, 0.1 mol) in
matograph (ICS-90, DIONEX). Three samples were tested. The dichloromethane (150 mL) was added dropwise a solution of di-
thermal transitions of the PUs were measured by dynamic tert-butyl dicarbonate (e, 21.8 g, 0.1 mol) in dichloromethane
mechanical analysis (Q800, TA Instruments-Waters LLC). (50 mL) with stirring for 30 min, and then the mixture stirred for
Fourier transform infrared (FTIR) spectroscopy analysis was 24 h at room temperature. The volatile components were
performed on a Nicolet-6700 spectrophotometer equipped with evaporated, and the residue was dissolved in ethyl acetate and
a Ge prism at an incident angle of 45ꢀ. AFM measurements were further washed twice with 0.5 M citric acid and once with H2O.
performed on an SPA400/SPI3800N (Seiko Instruments Inc.) Aer the mixture was dried (MgSO4) and the solvent evaporated,
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 1344–1353 | 1345

View Article Online
Journal of Materials Chemistry B
Paper
the residue was evaporated in a vacuum to yield an oil (f) (17.5 g, for 1 h. Following this, APPC dissolved in DMF was added to the
86.2%).
mixture and stirred at 80 ꢀC for 1 h. The solution was then
APCI MS (positive) m/z: theoretical 204 g molꢁ1, observed cooled to room temperature. The polymer was precipitated in a
205 g molꢁ1
.
methanol–distilled water mixed solvent (v/v ¼ 4 : 1) to remove
1H NMR (CDCl3, 400 MHz) d ppm: 1.25 (t, 2H, –CH2OH), 1.44 any residual APPC and the low molecular weight fraction, then
(s, 9H, –C(CH3)3), 1.51 (m, 2H, –CH2–), 1.59 (m, 2H, –CH2–), 1.65 dried in an oven for 24 h, followed by drying at 60 ꢀC under
(s, 1H, –OH), 3.13 (q, 2H, –CH2NH–), 3.65 (m, 2H, –CH2–), 4.56 vacuum for 48 h.
(s, 1H, –NH–).
The reaction scheme for the synthesis of the polyurethanes is
2.3.4. Synthesis of 5-aminopentyl-2-(trimethylammonio) shown in Scheme 2. The characteristics of the polymer (PU-PC)
ethyl phosphate. Compound f (10.0 g, 49.0 mmol) and are summarized here:
triethylamine (7.5 mL, 53.9 mmol) were dissolved in anhydrous
FTIR (neat, cmꢁ1): 3297 (NH); 2940, 2856, 2799 (CH2); 1728,
tetrahydrofuran (80 mL). Aer cooling the solution down to ꢁ15 1706 (carbonyl of NHCOO); 1651 (carbonyl of NHCONH); 1221,
ꢀC, COP (7.0 g, 49.0 mmol) dissolved in anhydrous tetrahydro- 1103 (C–O–C).
furan (10 mL) was added dropwise to the stirred solution under
1H NMR (DMSO, 400 MHz) d ppm: 1.24 (s, –CH2–), 1.49 (m,
nitrogen. The temperature of the reaction mixture was main- –CH2–), 3.13 (s, –N+(CH3)3), 3.24 (br, –CH2–), 3.34 (m,
tained at ꢁ15 ꢀC for 1 h and then allowed to slowly warm up to –CH2OCH2–), 3.78 (s, Ar-CH2-Ar), 4.06 (t, –COOCH2), 7.08, 7.34
room temperature. The triethylammonium chloride precipitate (m, Ar-H), 9.50 (s, –NHC]O).
was ltered off and washed with anhydrous tetrahydrofuran.
2.3.6. Sample preparation. Films used for testing were
The ltrate was evaporated under vacuum to give the residue as coated onto a clean glass disk from 10 wt% polymer solutions in
ꢀ
a colorless oil. The residue was dissolved in 60 mL dry aceto- DMF. The solvent was evaporated in an oven at 60 C for 24 h,
nitrile and transferred to a 100 mL glass pressure bottle. Aer and the lms were further dried in a vacuum oven at 60 ꢀC for 48
the pressure bottle was cooled to ꢁ18 ꢀC, excess trimethylamine h. The lms were approximately 0.3 mm thick.
was rapidly added to the solution. The pressure bottle was then
2.3.7. Protein adsorption assay. The adsorption of human
sealed and maintained at 60 ꢀC for 24 h. Subsequently the brinogen onto the lms was evaluated using the enzyme-linked
solution was evaporated under vacuum to produce a viscous immunosorbent assay (ELISA),21 and the procedure followed that
liquid, and then dissolved in methanol (20 mL) followed by the described previously.22,23 Polyurethane lms with 1.0 ꢂ 1.0 cm2
addition of saturated HCl–ethyl acetate (10 mL). The mixture surface area were incubated with 500 mL of Phosphate Buffered
was stirred for 2 h, concentrated under reduced pressure, Saline (PBS) at 37 ꢀC for 60 min, and were then incubated
washed with diethyl ether several times, dissolved in methanol sequentially with 500 mL of 0.5 mg mLꢁ1 human brinogen (Fg,
(20 mL), adjusted to pH 9 by the addition of saturated sodium F3879, 55–75% protein, purchased from Sigma, Shanghai) in PBS
bicarbonate solution, and ltered. The ltrate was concentrated solution at 37 ꢀC for 120 min, 500 mL of bovine serum albumin
under reduced pressure and the crude product was puried (BSA, 2 mg mLꢁ1) in PBS solution at 37 ꢀC for 90 min, and 500 mL
by ash column chromatography (reverse phase silica gel, of 5.5 mg mLꢁ1 horseradish peroxidase (HRP) conjugated goat
water : methanol ¼ 10 : 1) to yield APPC as a white solid (10.2 g, anti-human brinogen (RCe-0363G, Y-Y Chemical Reagents,
38.1 mmol, 78%).
APCI MS (positive) m/z: theoretical 268 g molꢁ1, observed lms were rinsed 5 times with 500 mL of washing buffer and
Shanghai) in PBS solution at 37 ꢀC for 30 min. The polyurethane
269 g molꢁ1
.
transferred into clean wells, and the enzyme-induced color
1H NMR (DMSO, 400 MHz) d ppm: 1.38 (m, 2H, –CH2–), 1.51 change reaction was then carried out by adding 500 mL of 0.1 M
(m, 2H, –CH2–), 1.60 (m, 2H, –CH2–), 2.74 (t, 2H, –CH2NH2), 3.15 citrate-phosphate buffer (pH ¼ 5.0) containing 1 mg mLꢁ1
(s, 9H, –N+(CH3)3), 3.55 (t, 2H, –CH2N+), 3.68 (m, 2H, –CH2O–), chromogen of o-phenylenediamine (OPD) and 0.03% hydrogen
4.05 (s, 2H, –OCH2–), 8.20 (s, 2H, –NH2).
peroxide, and nally the absorbance of light at 490 nm was
13C NMR (D2O, 200 MHz) d ppm: 19.5 (–CH2–), 23.9 (–CH2–), determined by a microplate reader (Model 550, Bio-Rad, USA).
26.7 (–CH2–), 36.9 (–CH2NH2), 51.4 (–N+(CH3)3), 56.8 (–CH2N+), The absorbance from the control poly(ether urethane) lms was
56.9 (–CH2O–), 63.5 (–OCH2–).
equivalent to 100% for calculating relative adsorption values.
2.3.5. Synthesis of polyurethanes. The control poly- Four replicate samples were measured for each concentration in
urethane based on a 3 : 2 : 1 molar ratio of MDI : PTMG : BDO, a given experiment. Experiments were done three times.
and end-capped polyurethanes based on a 3 : 2 : 0.5 : 1 molar
2.3.8. Platelet adhesion assay. Platelet adhesion measure-
ratio of MDI : PTMG : BDO : end-capping reagent, were ments were carried out according to the standard protocol as
prepared by solution polymerization in a suitable solvent with described briey below. Fresh blood from healthy, drug-free
DBTDL catalyst. All polyurethanes were synthesized by similar donors was collected using vacuum tubes, containing sodium
methods. A representative synthesis of PU-PC (PTMG-MDI-BDO- citrate as an anticoagulant (anticoagulant and blood with the
PC) is described as follows.
volume ratio 1 : 9), and platelet-rich plasma (PRP) was obtained
In the rst reaction step, MDI and 0.5% DBDTL were added from blood centrifuged at 190g for 10 min. The polyurethane
to the stirred toluene solution of PTMG under a dry nitrogen lms (10 mm ꢂ 10 mm) were placed into a 24-well tissue culture
atmosphere at 60 ꢀC. The reaction was maintained at 60 ꢀC for 2 plate and were immersed in 1.0 mL PBS (pH 7.4) and equili-
h. In the second step, the chain extender (BDO) was added ꢀto brated at 37 ꢀC for 1 h. Aer the PBS was removed, 1.0 mL PRP
the reaction solution, while the temperature was kept at 70 C was poured into each well, and the lms were soaked in the PRP
1346 | J. Mater. Chem. B, 2014, 2, 1344–1353
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
ꢀ
for 2 h at 37 C. Aer the PRP solution was removed, the lms
were rinsed 3 times with PBS (3 ꢂ 1.0 mL), and then treated with
2.5% glutaraldehyde for 1 day at 4 ꢀC. Aer rinsing with PBS
solution, the lms were dehydrated with a series of composition
gradients of aqueous ethanol solutions (ethanol composition:
0%, 25%, 50%, 75% and 100%). Critical point drying of the
specimens was done with liquid CO2. The dried lms were gold
coated and then examined with a JEOL Superprobe 733 (JEOL,
Japan) scanning electron microscope at 20 kV. Quantitative
measurements were made for 5 representative areas of 12
different specimens by image analysis.
2.4. Statistical analysis
All data are expressed as the mean ꢃ standard deviation. One-
way ANOVA with Tukey post hoc tests were used for hypothesis
testing, with p < 0.05 as the measure for statistical signicance.
The number of independent tests has been listed for each
experiment.
Scheme 2 The synthesis of polyurethanes end-capped with monomers.
biomimetic functional group. Monomer III (FASPC) was
synthesized according to our previous study.22,23 In this mole-
cule, the PC head group and long uorinated tail are both linked
to L-serine, and the amine in L-serine is retained as an active
group. These three monomers decorated with active amines
could provide the reactivity with the isocyanato group of MDI,
and thus endow the polyurethanes with distinctive features.
It is possible to achieve a broad range of properties by
varying the hard and so segments of polyurethanes.25,26 In our
previous work, FASPC was introduced into various poly-
urethanes, and the hemocompatibility of all samples were
obviously improved without specicity.23 MDI, PTMG and BDO
were chosen to synthesize the control polyurethane, which are
also the components of some commercial polyurethanes (e.g.,
pellethane). The stoichiometries of the reaction are shown in
Table 1, and the detailed descriptions of the syntheses are
presented in the Experimental section. The control poly-
urethane and FA, APPC, FASPC end-capped polyurethanes were
labeled as PU, PU-F, PU-PC and PU-FPC, respectively.
3. Results and discussion
3.1. Synthesis of end-capping reagents and polyurethanes
To explore the inuence of the uorocarbon chain and phos-
phorylcholine on the improvement of blood compatibility in
polyurethanes, three monomers with characteristic structures
were designed and synthesized, and then were utilized to end-
cap polyurethanes. The synthetic routes are shown in Schemes 1
and 2.
As shown in Scheme 1, each of the three end-capping
reagents contains an amine (–NH2) as the active functional
group to block polyurethane chains, while each of them also has
its own distinctive structure. Monomer I (“FA” for short) is based
on peruorooctanoic acid, and the main structure of FA was
conrmed through MS and 1H NMR results, which are listed in
the Experimental section. The synthesis of monomer II (APPC)
was based on 5-amino-1-pentanol, 2-choloro-1,3,2-dioxaphos-
pholane-2-oxide (COP) and trimethylamine according to the
Ishihara method.24 The phosphorylcholine polar head (PC) is the
GPC, 1H NMR, XPS, and element determination were used to
conrm the structure and composition of the polyurethanes.
The weight-average molecular weight (Mw) and number-average
molecular weight (Mn) of the polyurethanes were determined by
GPC (as shown in Table 1), and calibrated based on narrow
molecular weight PMMA standards. It is shown that all their
average molecular weights are on the order of 104 (between
2.69 ꢂ 104 and 5.67 ꢂ 104 in Mn). Due to the leaching of
unreacted end-cappers and the low molecular weight fraction in
methanol–distilled water mixed solvent and the nature of
solution polymerization, the polydispersities of the poly-
urethanes are all less than 2.0.
The results of 1H NMR are shown in the Experimental
section. The characteristic peak of phosphorylcholine
(–N+(CH3)3) at 3.13 ppm appears clearly in both PU-PC and PU-
FPC. The characteristic peak of FA does not appear obviously in
PU-F due to the small fraction of characteristic hydrogen atoms
(–CF2–CH2–). However, the successful end-capping can be
veried by the results of XPS and F/P element determinations,
the results of which are listed in Table 2. The appearance of F
Scheme 1 The synthesis routes and structures of the monomers.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 1344–1353 | 1347

View Article Online
Journal of Materials Chemistry B
Paper
Table 1 Stoichiometry, molecular weights and yields of the polyurethanes
Molecular weighta
Stoichiometry
MDI : PTMG : BDO : end-capper
Sample
Mn ꢂ 10ꢁ4
Mw ꢂ 10ꢁ4
Mw/Mn
Yield (%)
PU
PU-F
PU-PC
PU-FPC
3 : 2 : 1 : 0
5.67
2.75
2.69
3.50
8.53
4.87
4.64
5.68
1.50
1.77
1.72
1.62
78
71
82
75
3 : 2 : 0.5 : 1
3 : 2 : 0.5 : 1
3 : 2 : 0.5 : 1
a
Gel permeation chromatography (GPC) was performed with a Waters-1515 using N,N-dimethylformamide (DMF)–LiBr as eluent. The molecular
weights are relative to polymethyl methacrylate (PMMA) standards.
and/or P elements in PU-F, PU-PC, and PU-FPC is direct adapt to the values in the bulk. The wt% of F on the surfaces of
evidence for the successful end-capping synthesis. The F PU-F and PU-FPC are 48.71% and 37.72% (ignoring the
content in the bulk of PU-FPC is lower than that in PU-F, and the hydrogen fraction), which are 6.3 times and 37.03 times to that
same trend is observed in the comparison of P content between in the bulk; the wt% of P in PU-PC and PU-FPC on the surface
PU-FPC and PU-PC. It is assumed that the FASPC monomer has are 2.5 times and 5.8 times that in the bulk, respectively. The
lower reactivity than FA and APPC due to steric hindrance of the results indicate that the chain ends of polyurethane possess
amine.
high mobility and tend to migrate to the surface. However, the
end-capping of the PC group on polyurethane chains has
limited effect on migration, compared to the easier migration of
the FA and FASPC groups.
3.2. Structure analysis
As potential biomaterials, both the surface and bulk properties
play important roles in the performance of polyurethane
materials. Therefore, the surface chemical constitution and the
phase separation of polyurethanes were examined aer the
introduction of end-cappers.
Microphase separation is the main concern for poly-
urethanes, which plays an important role in determining the
surface and bulk properties.28,29 In this work, dynamic
mechanical analysis (DMA) was employed to study the thermal
transitions of the polyurethanes, and the changes in glass
transition temperature (Tg) of the so segment phase deter-
mined from the peaks of tan d curves were used as an indicator
of the degree of phase separation.29 The DMA data are shown in
Fig. 2A, and the Tg data are listed in Table 3. The Tg values of the
end-capped polyurethanes are all lower than that of the control
sample, which suggests a trend of enhanced microphase sepa-
ration in the bulk.
The XPS data could give direct and atomic scale chemical
information about the surface groups. The atomic concentra-
tion of carbon (C 1s), oxygen (O 1s), nitrogen (N 1s), phosphorus
(P 2p), and uorine (F 1s) in the 5–10 nm depth range from the
surface were detected.23,27 The C 1s peaks in the four kinds of
polyurethanes were tted to seven component peaks. Among
them, the CF2 and CF3 peaks belong to the uorinated tail,
which appear in the C 1s spectra of PU-F and PU-FPC. The C–O–
P peak belonging to PC group appears in the C 1s spectra of
both PU-PC and PU-FPC. However, the C–O–P peak area of PU-
FPC is much larger than that of PU-PC, as shown in Fig. 1a and
c. For composition, the weight contents of F and P in the bulk
were also detected by the element determination method, and
the results are also shown in Table 2. The atomic percentages of
F and P on the surface were calculated as a weight percentage to
Fluorinated groups are considered as microphase segre-
gating groups, where the peruoro moiety prefers to be sepa-
rated from the aliphatic groups and tends to assemble at the
Table 2 Element contents of polyurethanes on the surface and in the
bulk
Element contents on the surfacea Element contents in the
(at%)
bulkb (wt%)
Sample
C
O
N
P
F
P
F
PU
PU-F
PU-PC
76.27 21.90 1.83
48.81 7.65 4.79
77.48 19.73 2.50 0.29
0
0
0
—
0
—
38.75
0
7.67 ꢃ 0.26
0.28 ꢃ 0.02
0
PU-FPC 51.94 12.07 6.16 0.53 29.30 0.19 ꢃ 0.06 1.02 ꢃ 0.01
a
b
Fig. 1 C 1s XPS spectra of PU-FPC (a), PU-F (b), control PU (c), and PU-
Values detected by XPS. Values detected by element determination;
“—” not detected.
PC (d).
1348 | J. Mater. Chem. B, 2014, 2, 1344–1353
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 2 (A) Dynamic mechanical spectra of polyurethanes; (B) C]O stretching band in the FTIR spectra of polyurethanes; (C) AFM phase images
of control PU (a), PU-F (b), PU-PC (c), and PU-FPC (d).
Table 3 Glass transition temperatures detected by DMA, and C]O stretching band information of the polyurethanes
“Free” C]O
H-bonded C]O
Tg of
Stretching band
Area percentage
(%)
Stretching band
Area percentage
(%)
Sample
so segment (ꢀC)
n (cmꢁ1
)
n (cmꢁ1
)
PU
PU-F
PU-PC
PU-FPC
ꢁ17.05
ꢁ18.88
ꢁ24.31
ꢁ18.33
1729
1729
1730
1730
37.8
39.7
40.4
38.7
1707
1706
1704
1702
62.2
60.3
59.6
61.3
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 1344–1353 | 1349

View Article Online
Journal of Materials Chemistry B
Paper
interface,30–32 therefore the enhanced microphase separation polyurethanes drastically. The uorocarbon chains of PU-F,
results in a lower Tg of the so segment of PU-F than that of the which are decorated on the hard segment, drive the hard
control PU. Zwitterions in block polymers usually play the role segment to the surface due to the lower surface energy, and the
of a “binder” between two phases and cause strong microphase hard domains almost cover the surface of PU-F.41 This result is
separation.33,34 Moreover, the steric hindrance of the phos- in agreement with the XPS data. In the phase image of PU-PC, a
phorylcholine group with the hard segment and so segment decrease in the content of hard domains is observed, while the
further strengthens the microphase separation of PU-PC, and size of them increases obviously and the boundary between
therefore the Tg of the so segment of PU-PC is lowest of the hard domains and so domains becomes sharper than that in
four kinds of polyurethanes. As for PU-FPC, since the uoro- the PU phase image. The phase inversion is complete in the
carbon chain and the PC group are linked, the weak intermo- PU-FPC phase image. Driven by the uorinated tail, the hard
lecular forces acting on the uorocarbon chains can partly domains aggregate on the surface and become a continuous
shield the polarizable environment offered by the PC group,35 phase, and the so domains become the dispersion phase with
and this shield effect inhibits the improvement of microphase “island” shape, as shown in Fig. 2C and D. The comparision of
separation and results in a higher Tg compared with that of surface phase morphology between PU and the end-capped
PU-PC.
polyurethanes in the AFM data conrms the similar trend of
In the infrared spectra of polyurethanes, the amide I band, enhanced microphase separation behavior for the end-capped
which is principally due to carbonyl stretching vibration modes, polyurethanes as that observed in the DMA and FTIR analysis.
has a relatively well-resolved adsorption doublet. The higher The surface hard segment content of PU-PC is less than that of
wavenumber peak (at about 1730 cmꢁ1) is associated with “free” PU, while the surface hard segment contents of PU-F and
(non-bonded) C]O stretching adsorption, and the lower wave- PU-FPC are much greater than that of PU, which is the reverse of
number peak (at about 1710 cmꢁ1) mainly represents the the trend detected in the FTIR spectra. It can be deduced that
adsorption of hydrogen-bonded C]O (with –NH– groups) in the surface morphologies of the end-capped polyurethanes,
hard segments.36,37 Therefore, the area and wavenumber of the especially PU-F and PU-FPC, are different from the morphol-
H-bonded carbonyl peak can reect the morphology of the hard ogies in the bulk, which corresponds to the chemical variation
segment in polyurethanes. The analysis of the FTIR spectra from bulk to surface detected by XPS, and the uorocarbon
shown in Fig. 2B makes use of the above general assignments for chains play an important role in this system.
the C]O stretching adsorption. To quantitate the examination of
the carbonyl stretching band, the resolution results of the C]O
peaks are listed in Table 3.38 As observed in the FTIR data, upon
3.3. Surface properties
careful examination of the C]O band, it is evident that the peak The time-related water contact angles on the surfaces of the
area of the H-bonded C]O stretching band decreases aer end- polyurethane lms were determined, and typical images are
capping, which indicates that the content of the hard segment shown in Fig. 3a. It can be observed that the water drops on all
decreases in the bulk of PU-F, PU-PC and PU-FPC. In addition, all the polyurethane lms show relatively stable contact angles: the
the H-bonded C]O bands of end-capped polyurethanes show a change from momentary angles to equilibrium values in 40
similar trend of shiing to lower wavenumbers, which should be seconds is less than 10ꢀ. The high values of the water contact
due to the improvement of microphase separation compared angles on PU-F and PU-FPC lms (above 115ꢀ) are caused by the
with conventional polyurethane.39,40
appearance of uorocarbon chains on the surfaces, which has
According to the analysis of infrared spectroscopy and DMA been veried by XPS data. The angles of PU-PC are slightly lower
data, the enhanced microphase separation between so than that of PU due to the phosphorylcholine groups on the
segments and hard segments is due to the introduction of a surface. The results indicate that the uorocarbon chains tend
uorinated tail and/or phosphorylcholine group. In addition, to cover the top surface to minimize the surface free energy aer
the hard segment contents of all the end-capped polyurethanes the polyurethane lms are formed, while the phosphorylcholine
are decreased in the bulk.
groups do not show this trend.
AFM was used to further investigate the microphase struc-
To investigate the rearrangement ability, the polyurethane
ture of the polyurethane surface in the form of visualization, as lms were soaked in distilled water at room temperature for one
shown in Fig. 2C. Working at tapping mode, the phase image week, and then the water contact angles were determined again
provides information about surface stiffness variation related to aer drying under nitrogen, the results of which are shown in
changes in Young's modulus, and the microphase-separated Fig. 3b. The water contact angle pictures of PU-FPC are also
structures can be seen in the phase images.29,41,42 The light color shown for comparison. Aer the water treatment, all the end-
regions generally represent the hard segment due to its higher capped polyurethane lms have lower water contact angles than
modulus.
the control PU lm. This result could be explained by the
It should be noted that the contrast in phase images is based restructuring of the surface structure:43–45 aer water penetra-
on the same level of tapping force (amplitudes with range 0.4– tion into the surface and consequent reconstruction of the
0.7). A typical microphase separation pattern is displayed in the atoms and molecules at the surface of the substrate, some of the
phase image of the control PU, and the hard segment phase size strongly hydrophobic uorocarbon chains migrate to the sub-
is uniformly dozens of nanometers. However, the introducing of surface, and the hydrophilic structures, mainly phosphoryl-
end-capping groups changes the surface microdomains of the choline groups in this system, tend to migrate to the top
1350 | J. Mater. Chem. B, 2014, 2, 1344–1353
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 3 The water contact angles of polyurethanes before (a) and after (b) water treatment for one week. The water contact angle pictures are of
the PU-FPC sample.
surface. The strong mobility of uorocarbon chains is demon- adsorption. The PU-FPC sample shows the lowest amount of Fg
strated in this process. However, since the uorocarbon chain adsorption, which can be mainly attributed to the greater
are linked with the phosphorylcholine group on the molecular number of phosphorylcholine groups. In this preliminary
scale, it should have an effect on the improvement of the evaluation, Fg adsorption was measured as a single reference to
hydrophility of the whole FASPC moiety for PU-FPC, and it could evaluate the protein adsorption. However, it is important to
be understood that the PU-FPC lm with more phosphor- note that other plasma proteins (such as von Willebrand factor,
ylcholine groups on the surface has a higher water contact angle plasminogen or bronectin), which also play important roles in
than the PU-PC lm.
the coagulation process,50,51 were not tested in this study.
Fig. 5 shows typical SEM photographs of platelet adhesion to
the polyurethane lms. In this study, the blood platelets from
multiple donors (three donors) were taken for the test, with the
aim of avoiding potential donor-to-donor variations in platelets
and their reactivity/adhesion to surfaces.50,52 Statistically
analyzed by images of parallel samples, the platelet counts on
3.4. Hemocompatibility evaluation
Platelet adhesion and activation on the surface of blood-con-
tacting devices will trigger the coagulation of blood leading to
thrombus formation, and protein adsorption is the rst step in
a cascade of surface processes that induce platelet deposition
and thrombus formation. Thus, protein adsorption and platelet
adhesion are usually investigated as a preliminary evaluation of
the hemocompatibility of materials,46–48 which have also been
employed in this work.
the surfaces of PU, PU-F, and PU-PC are (1.7 ꢃ 0.5) ꢂ 106 cmꢁ2
,
(4.1 ꢃ 1.4) ꢂ 104 cmꢁ2, and (2.9 ꢃ 0.6) ꢂ 105 cmꢁ2, respectively.
On the surface of PU-FPC, hardly any platelets can be found
aer soaking for 2 h. There are not only a substantial number of
adhered platelets, but also some deformed platelets on the
Fibrinogen (Fg) is one of the most abundant proteins in
blood and plays a vital role in thrombosis on the surface of
materials, and the adsorption of human brinogen on poly-
urethane lms was detected by the ELISA method in this study.
This testing method cannot obtain absolute values for the
amount of Fg adsorbed on a lm surface in units (mg cmꢁ2),21,49
so the Fg adsorption to the control PU lm was used as a
reference, and the relative Fg adsorptions on the end-capped
polyurethane lms with respect to that on the PU lm are
shown in Fig. 4. In comparison with the control PU, the amount
of adsorbed human Fg on PU-F, PU-PC, and PU-FPC decreased
by 23.5%, 37.6%, and 78.7%, respectively. Even though the
surface of PU-F is covered by uorocarbon chains, the decrease
of adsorbed Fg on PU-F is still less than that of PU-PC, which
indicates that the phosphorylcholine groups are more effective
Fig. 4 Relative human fibrinogen adsorption on various polyurethanes
determined from ELISA with control PU as a reference. * Statistically
than the uorocarbon chain for the resistance to protein significant difference with respect to control PU (p < 0.05).
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 1344–1353 | 1351

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 5 SEM images of the surfaces of polyurethane films after 2 h of PRP exposure. (a) PU; (b) PU-F; (c) PU-PC; (d) PU-FPC. Actual magnification:
left 1000ꢂ ; right 5000ꢂ.
surface of the control PU, which indicates that activation of the on the improvement of hemocompatibility of the poly-
adhered platelets may occur. Most of the platelets adhered on urethanes, and the advanced properties are mainly due to these
the PU-F and PU-PC surfaces remained a normal shape.
The varying degrees of resistance to protein adsorption and
reasons:
(1) The migration ability of the uorocarbon chains can drive
platelet adhesion by phosphorylcholine groups and uoro- the phosphorylcholine groups to the surface of the poly-
carbon chains in polyurethanes is an interesting phenomenon urethane lms.
in this study. It has been reported that both uorocarbon chains
(2) The spontaneous arrangement of uorocarbon chains
and PC groups on a polymer surface can prevent protein and phosphorylcholine groups results in a balance on the
adsorption and platelet adhesion to improve blood compati- surface to adapt to the water environment.
bility.12,20,53 The experimental data in this study also verify this
(3) Both uorocarbon chains and phosphorylcholine groups
trend, while the degree of improvement in each aspect is in polyurethanes have their own advantages for anti-
different. In the protein adsorption study, the number of PC coagulation, and each of them can play a vital role in preventing
groups present on the surface of PU-PC is much lower than that the progress of a clotting cascade.
of uorocarbon chains at the PU-F surface (from XPS and AFM
Although the synergistic effect of uorocarbon chains and
data), but the amount of brinogens adsorbed on PU-PC is still phosphorylcholine groups on anticoagulation still needs to be
lower than that on PU-F, which obviously indicates the greater further claried in detail and in vivo, the results of this work
efficiency of the PC group in preventing protein adsorption. The indicate that the combination of anticoagulation components
opposite trend is observed in the platelet adhesion experiment. in the same material has real potential to develop biomaterials
The amount of platelets adhered to PU-F is one order of for contact with blood.
magnitude smaller than that adhered to PU-PC, which means
that the uorocarbon chains in PU-F reduce the adhesion of
platelets more efficiently than PC groups in PU-PC. It can be
Acknowledgements
concluded that, in polyurethanes, the uorocarbon chains are
This work is a collaborative research project sponsored by the
more efficient for resisting platelet adhesion, while the PC
Ministry of Science and Technology of China (grant no.
groups prevent brinogen adsorption more effectively. More-
2010DFA31190) and the Government of Quebec, Quebec, Can-
over, when uorocarbon chains and PC groups are simulta-
ada (grant no. PSR-SIIRI-285). We would also like to express our
neously introduced into polyurethane, the hemocompatibility
sincere thanks to the National Natural Science Foundation of
(in both protein adsorption and platelet adhesion experiments)
China for nancial support (51121001).
is improved dramatically in this preliminary study. This syner-
gistic effect triggered by two anticoagulation components in
polyurethane is promising, and is worthy of quantitative
research in detail and in other materials in the future.
Notes and references
1 H. S. Lee, N. Tomczyk, J. Kandel, R. J. Composto and
D. M. Eckmann, J. Mater. Chem. B, 2013, 1, 6382–6391.
2 F. Tong, X. Chen, L. Chen, P. Zhu, J. Luan, C. Mao, J. Bao and
4. Conclusions
In this work, three monomers containing a uorinated tail and/
or phosphorylcholine for end-capping were synthesized and
introduced into polyurethanes. It is found that the uorocarbon
chains and phosphorylcholine groups have a synergistic effect
J. Shen, J. Mater. Chem. B, 2013, 1, 447–453.
3 R. Y. Kannan, H. J. Salacinski, P. E. Butler, G. Hamilton and
A. M. Seifalian, J. Biomed. Mater. Res., Part B, 2005, 74B, 570–
581.
1352 | J. Mater. Chem. B, 2014, 2, 1344–1353
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
4 K. Kashimoto, J. Yoon, B. Y. Hou, C. H. Chen, B. H. Lin, 29 H. Tan, J. H. Li, M. Guo, R. N. Du, X. Y. Xie, Y. P. Zhong and
M. Aratono, T. Takiue and M. L. Schlossman, Phys. Rev.
Lett., 2008, 101, 076102.
Q. Fu, Polymer, 2005, 46, 7230–7239.
30 S. J. Cowling, A. W. Hall and J. W. Goodby, J. Mater. Chem.,
2011, 21, 9031–9042.
31 G. De Luca, A. Liscio, M. Melucci, T. Schnitzler, W. Pisula,
C. G. Clark, Jr, L. M. Scolaro, V. Palermo, K. Muellen and
P. Samori, J. Mater. Chem., 2010, 20, 71–82.
32 Y. Zhang, L. Wang, Z. Zhang, Y. Zhang and X. Tuo, J. Polym.
Sci., Polym. Chem. Ed., 2013, 51, 2161–2170.
5 I. Lee, K. Kobayashi, H. Y. Sun, S. Takatani and L. G. Zhong,
J. Biomed. Mater. Res., Part A, 2007, 82A, 316–322.
6 H. Yin, T. Akasaki, T. L. Sun, T. Nakajima, T. Kurokawa,
T. Nonoyama, T. Taira, Y. Saruwatari and J. P. Gong,
J. Mater. Chem. B, 2013, 1, 3685–3693.
7 A. P. Khandwekar, D. P. Patil, A. A. Hardikar, Y. S. Shouche
and M. Doble, J. Biomed. Mater. Res., Part A, 2010, 95A, 33 E. Marwanta, T. Mizumo and H. Ohno, Solid State Ionics,
413–423. 2007, 178, 227–232.
8 R. Leitsmann, O. Bohm, P. Planitz, C. Radehaus, M. Schaller 34 S. Pispas, G. Floudas and N. Hadjichristidis, Macromolecules,
and M. Schreiber, Surf. Sci., 2010, 604, 1808–1812. 1999, 32, 9074–9077.
9 X. Y. Xie, R. F. Wang, J. H. Li, L. Luo, D. Wen, Y. P. Zhong and 35 X. Wang, P. Long, S. Dong and J. Hao, Langmuir, 2013, 29,
C. S. Zhao, J. Biomed. Mater. Res., Part B, 2009, 89B, 223–241. 14380–14385.
10 T. Hong, X. Xingyi, L. Jiehua, Z. Yingping and F. Qiang, 36 M. M. Coleman, D. J. Skrovanek, J. Hu and P. C. Painter,
Polymer, 2004, 45, 1495–1502. Macromolecules, 1988, 21, 59–68.
11 G. Jiang, X. Tuo, D. Wang and Q. Li, J. Mater. Sci.: Mater. 37 D. Tan, X. Zhang, J. Li, H. Tan and Q. Fu, Appl. Surf. Sci.,
Med., 2012, 23, 1867–1877. 2012, 258, 2697–2706.
12 K. Y. Chen and J. F. Kuo, J. Mater. Sci.: Mater. Med., 2002, 13, 38 D. Tan, X. Zhang, J. Li, H. Tan and Q. Fu, Chin. J. Polym. Sci.,
37–45. 2011, 29, 615–626.
13 J. C. Lin, S. L. Tiong and C. Y. Chen, J. Biomater. Sci., Polym. 39 D. A. Wang, J. Ji and L. X. Feng, Macromolecules, 2000, 33,
Ed., 2000, 11, 701–714. 8472–8478.
14 H. Wang, F. Xu, Y. Wang, X. Liu, Q. Jin and J. Ji, Polym. 40 G. Trovati, E. Ap Sanches, S. C. Neto, Y. P. Mascarenhas
Chem., 2013, 4, 3012–3019.
15 P. Wan, D. Tan, Z. Li, X. Zhang, J. Li and H. Tan, Chin.
J. Polym. Sci., 2012, 30, 190–198.
16 T. Goda, H. Furukawa, J. P. Gong and K. Ishihara, So
Matter, 2013, 9, 2166–2171.
17 M. Tan, Y. Feng, H. Wang, L. Zhang, M. Khan, J. Guo,
Q. Chen and J. Liu, Macromol. Res., 2013, 21, 541–549.
18 T. Hatakeyama, M. Tanaka and H. Hatakeyama, Acta
Biomater., 2010, 6, 2077–2082.
19 K. Ishihara, H. Nomura, T. Mihara, K. Kurita, Y. Iwasaki and
N. Nakabayashi, J. Biomed. Mater. Res., 1998, 39, 323–330.
20 Y. Iwasaki, Y. Aiba, N. Morimoto, N. Nakabayashi and
K. Ishihara, J. Biomed. Mater. Res., 2000, 52, 701–708.
21 Z. Zhang, T. Chao, S. F. Chen and S. Y. Jiang, Langmuir, 2006,
22, 10072–10077.
and G. O. Chierice, J. Appl. Polym. Sci., 2010, 115,
263–268.
41 M. Penoff, W. Schreiner, P. Oyanguren and P. Montemartini,
Macromol. Symp., 2012, 321–322, 186–190.
42 Y. Yanagihara, N. Osaka, S. Murayama and H. Saito, Polymer,
2013, 54, 2183–2189.
43 X. Cui, S. Zhong, Y. Gao and H. Wang, Colloids Surf., A, 2008,
324, 14–21.
44 S. Perutz, J. Wang, E. J. Kramer, C. K. Ober and K. Ellis,
Macromolecules, 1998, 31, 4272–4276.
45 D. L. Schmidt, R. F. Brady, K. Lam, D. C. Schmidt and
M. K. Chaudhury, Langmuir, 2004, 20, 2830–2836.
46 H. W. Chien, C. C. Tsai, W. B. Tsai, M. J. Wang, W. H. Kuo,
T. C. Wei and S. T. Huang, Colloids Surf., B, 2013, 107, 152–
159.
22 D. Tan, X. Zhang, J. Li, H. Tan and Q. Fu, J. Biomed. Mater. 47 J. Tan, W. G. McClung and J. L. Brash, J. Biomater. Sci., Polym.
Res., Part A, 2012, 100A, 380–387. Ed., 2013, 24, 497–506.
23 X. Zhang, D. Tan, J. Li, H. Tan and Q. Fu, Biofouling, 2011, 27, 48 L. Ma, H. Qin, C. Cheng, Y. Xia, C. He, C. Nie, L. Wang and
1175. C. Zhao, J. Mater. Chem. B, 2014, 2, 363.
24 K. Ishihara, T. Uchida and N. Nakabayashi, Polym. J., 1990, 49 J. Li, Y. Zhang, J. Yang, H. Tan, J. Li and Q. Fu, J. Biomed.
22, 355–360. Mater. Res., Part A, 2013, 101A, 1362–1372.
25 Y. W. Tang, R. S. Labow and J. P. Santerre, J. Biomed. Mater. 50 M. Zhang and T. A. Horbett, J. Biomed. Mater. Res., Part A,
Res., 2001, 56, 516–528. 2009, 89A, 791–803.
26 T. Choi, J. Weksler, A. Padsalgikar and J. Runt, Polymer, 51 L. M. Szott and T. A. Horbett, Curr. Opin. Chem. Biol., 2011,
2009, 50, 2320–2327. 15, 677–682.
27 J. Tan and J. L. Brash, J. Appl. Polym. Sci., 2008, 108, 1617– 52 J. M. Grunkemeier, W. B. Tsai and T. A. Horbett, J. Biomed.
1628. Mater. Res., 1998, 41, 657–670.
28 L. C. Xu, J. Runt and C. A. Siedlecki, Acta Biomater., 2010, 6, 53 M. Pagliaro and R. Ciriminna, J. Mater. Chem., 2005, 15,
1938–1947.
4981–4991.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 1344–1353 | 1353