Facile synthesis of dopa-functional polycarbonates via thiol-Ene-coupling chemistry towards self-healing gels

Article abstract of DOI:10.1002/pola.28111

Since extraction of the naturally occurring mussel-foot proteins is expensive and time-consuming, routes towards synthetic analogues are continuously being explored. Often, these methods involve several protection and deprotection steps, making the synthe

Full text of DOI:10.1002/pola.28111

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Facile Synthesis of Dopa-Functional Polycarbonates via
Thiol-Ene-Coupling Chemistry Towards Self-Healing Gels
Kristina Olofsson, Michael Malkoch, Anders Hult
Department of Fibre and Polymer Technology, KTH Royal Institute of Technology, Teknikringen 56-58, Stockholm, SE 100 44,
Sweden
Correspondence to: A. Hult (E-mail: andult@kth.se)
Received 22 January 2016; accepted 7 March 2016; published online 24 March 2016
DOI: 10.1002/pola.28111
ABSTRACT: Since extraction of the naturally occurring mussel-
foot proteins is expensive and time-consuming, routes towards
synthetic analogues are continuously being explored. Often,
these methods involve several protection and deprotection
steps, making the synthesis of synthetic analogues time-
consuming and expensive as well. Herein, we show that UV-
initiated thiol-ene coupling between a thiol-functional dopa-
mine derivative and an allyl-functional aliphatic polycarbonate
can be used as a fast and facile route to dopa-functional mate-
rials. Different thiol-to-allyl ratios and irradiation protocols
were used and it was found that nearly 50% of the allyl groups
could be functionalized with dopa within short reaction times,
without the need of protecting the catechol. It is also demon-
strated herein that the dopa-functional polymers can be used
31
to form self-healing gels through complexation with Fe ions
2016 Wiley Periodicals, Inc. J. Polym. Sci.,
at increased pH. VC
Part A: Polym. Chem. 2016, 54, 2370–2378
KEYWORDS: dopa; photochemistry; post-polymerization func-
tionalization; ring-opening-polymerization; thiol-ene coupling;
synthetic methods
INTRODUCTION In the field of materials science, a lot of
inspiration towards synthesis of new advanced materials can
be found in nature. The naturally occurring amino acid 3,4-
dihydroxyphenylalanin (DOPA) has been shown to have inter-
esting adhesive properties and is in nature derived from post-
translational modification of tyrosine. Although DOPA is rare
in naturally occurring proteins, it has been found in large
quantities in the adhesive proteins of marine mussels, such as
have e.g. been synthesized via synthesis and/or modification
1
3–15
of peptides.
These synthetic routes are generally time-
consuming and costly, as they often suffer from low yields
and involve multiple reaction steps, as peptide synthesis nor-
mally requires several protection and deprotection steps,
1
6,17
both of the amine and the catechol.
As an alternative,
vinyl catechols have been synthesized by decarboxylation
1
8,19
reactions,
but the subsequent polymerization of these
1
20
the Mytilus edulis. The DOPA-rich adhesive mussel-foot pro-
compounds normally require protection of the catechol.
teins (mfps) of marine mussels have been found to provide
strong adhesion in wet environments and are additionally able
to form cross-linked networks in response to pH changes in
Instead, we envision the functionalization of preformed poly-
mers with the aim of producing materials with similar prop-
erties as more complex dopa-containing biomimetic
materials, but via more simplistic synthetic routes. This
approach is foreseen as a versatile way of tailoring dopa-
functional materials for specific applications, enabling pro-
duction of polymers with additional functionalities and
advanced architectures. Functionalization of preformed poly-
mers with dopa-groups has previously been demonstrated
by e.g. Menyo et al. in 2013, who used carboxylic-acid-
functional 4-arm poly(ethylene glycol) (PEG) and a surplus
of catechol to produce end-functionalized dopa-functional
2
–8
the surroundings or via coordination to metal ions.
Mfps
have therefore attracted a lot of attention in a wide range of
applications in materials science, especially for biomedical
9
applications. Extraction of natural mfps on an industrial scale
is difficult due to low yields, requiring approximately 10,000
1
0
mussels to produce 1 g of the most easily extracted mfps. As
an alternative, recombinant variants of mfps have been
1
1,12
expressed in bacteria such as Escherichia coli.
However,
due to limitations in the production processes, other routes
towards synthetic analogues are continuously being explored.
2
1
PEG; and by Xu et al. in 2012, who functionalized poly
(pentafluorophenyl acrylate) with dopamine to modify tita-
Biomimicking materials expressing the catechol of DOPA,
2
2
3,4-dihydroxyphenyl (herein referred to as a dopa-group),
nium surfaces. In 2015, Mattson et al. furthermore showed
Additional Supporting Information may be found in the online version of this article.
2016 Wiley Periodicals, Inc.
V
C
2
370
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
that a co-polymer of ethylene oxide and allyl glycidyl ether
Size exclusion chromatography (SEC) was conducted using a
TOSOH EcoSEC HLC-8320GPC system from PSS GmbH
equipped with an EcoSEC RI detector and three columns: PSS
PFG 5 mm; Microguard, 100 Å; and Microguard 300 Å (MW
(
P(EO-co-AGE)) could successfully be functionalized with
2
3
dopa-groups using thiol-ene coupling (TEC) chemistry.
Though effective and highly controlled, this approach
required protection of the catechol before functionalization,
resulting in a multistep synthesis. The versatility and specific
2
1
resolving range: 100–300,000 g mol ). DMF with 0.01 M LiBr
was used as the mobile phase at 50 8C and a flow rate of 0.2
2
4
21
character of TEC, however, promote TEC as a promising
chemistry for the synthesis of various dopa-functional mate-
rials. TEC is tolerant toward other functional groups and can
be carried out in a variety of solvents, making TEC suitable
for the functionalization of a wide range of polymers with
mL min . A conventional calibration method was created
using poly(methyl methacrylate) (PMMA) standards and tolu-
ene was used as an internal standard to correct for flow-rate
fluctuations. Obtained data was processed using PSS WinGPC
Unity software (version 7.2).
2
5–32
unsaturated bonds.
Differential scanning calorimetry (DSC) was carried out on a
Mettler Toledo DSC-1 equipped with Gas Controller GC100.
Analyses were run at a temperature interval of 270 8C to
Herein, we present a facile alternative route to dopa-
functional polymers, where allyl-functional linear polycar-
bonate (PC) is functionalized with a thiol-functionalized
derivative of dopamine via UV-initiated TEC. Linear PC has
shown great potential in the biomedical field and can fur-
thermore be synthesized with a variety of functional reactive
groups, creating a suitable platform for further functionaliza-
21
150 8C under nitrogen atmosphere (30 mL min ) with a
21
heating/cooling rate of 10 8C min
and 2 min isotherms.
Two heating cycles were conducted in order to remove influ-
ence of thermal history of the samples and data was
obtained from the second heating cycle. Evaluation of
3
3
e
tion. Linear PC has previously been successfully postfunc-
acquired data was done using STAR Excellence Software.
3
4–36
tionalized using thiol-functional compounds,
but little
UV-irradiation was carried out using a Blak-Ray B-100AP
High Intensity UV lamp equipped with a 100 W long-wave
spot lamp (365 nm, 100 W) in combination with subsequent
irradiation using a UV conveyor belt (Fusion Systems, USA,
work has so far been conducted using dopa-functional thiols
to functionalize PCs.
This study focuses on the synthesis and characterization of
linear dopa-functional PCs, introducing a route that does not
require protection of the catechol. By limiting the number of
reaction steps, a faster and easier route to synthetic dopa-
functional materials is presented than the ones previously
shown in the literature. It is additionally demonstrated that
the herein synthesized materials can be used to form auton-
21
model MC6R) with a line speed of 5 m min . The MC6R
conveyor belt was equipped with Fusion electrodeless bulbs
2
2
(standard type BF9 (320–360 nm, 3 W cm )).
Rheological measurements were carried out at 25 8C on a
DHR-2 Rheometer from TA Instruments using parallel-plate
geometry (8 mm in diameter). Frequency-sweep measure-
ments (0.01–10 Hz at 0.1% strain) and strain-sweep meas-
urements (0.01–100% strain at 1 Hz) were conducted.
Additionally, a time-sweep measurement was conducted,
where a sample was exposed to 0.1% strain alternated with
3
1
omously self-healing gels in the presence of Fe ions.
EXPERIMENTAL
Materials
1
6
00% strain at a frequency of 1 Hz with the duration of
0 s per step. The two steps were alternated and repeated
2,2-Bis(methylol) propionic acid (Bis-MPA) was supplied by
Perstorp Holding AB. Dimethylformamide (DMF) and 1,4-
dioxane were purchased from Lab-Scan Analytical Science.
three times each, with a total measurement time of 6 min.
Dried dichloromethane (DCM) (max. 0.004%
H
2
O,
UV-Vis absorption measurements were acquired on a Tecan
Infinite M200 Pro plate reader at 230 to 800 nm.
SeccosolvVR ) and tetrahydrofuran (THF) (EMSUREVR ACS,
Reag. Ph Eur) were purchased from Merck. The UV-initiator
Irgacure 184 was purchased from Ciba Speciality Chemicals,
Inc. and pH buffer solutions with pH 4, 7, and 10 were pur-
Synthesis of 5-Methyl-5-Allyloxycarbonyl-1,3-Dioxan-
2-One (MAC)
The monomer, MAC, was synthesized in two steps according
chased from Scharlau. Deuturated chloroform (CDCl ) and
3
3
7
to a previously published procedure as follows.
deuturated methanol (MeOD) for NMR analyses were pur-
chased from Cambridge Isotope Laboratories, Inc. All other
materials and reactants were purchased from Sigma-Aldrich
and used as received, unless otherwise stated.
Synthesis of Allyl 3-Hydroxy-2-(Hydroxymethyl)-2-
Methylpropanoate
Bis-MPA (54.0 g, 403 mmol) and KOH (25.8 g, 459 mmol)
were dissolved in 120 mL DMF in a flask equipped with a
cooler. The mixture was heated under stirring to 100 8C for
1 h, after which allyl bromide (41.8 mL, 484 mmol) was
added. The reaction was thereafter allowed to proceed at
100 8C overnight, whereafter formed KBr salt was filtered
off and washed with DMF. After removal of the solvent under
reduced pressure, further purification was carried out using
Instrumentation
Nuclear magnetic resonance (NMR) analyses were conducted
1
on a Bruker Avance 400 MHz instrument. H-NMR spectra
1
3
were acquired at 400 MHz and
C-NMR spectra were
acquired at 101 MHz. Deuturated solvents were used and
the residual solvent peak was used as internal standard.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378
2371

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Conditions and Amounts of Reactants used in Polymerizations of MAC
DP target
MAC (g)
1-Pyrene Butanol (mg)
TU (mg)
DBU (lL)
Reaction time (min)
Yield (%)
2
5
7
5
0
5
5.00
5.00
5.00
274
137
91.4
444
222
148
180
30
30
50
88.6
63.1
82.1
89.7
60.0
3
0.0 mL DCM were used in all polymerizations and all were terminated by addition of 0.5 mL acetic acid.
flash chromatography in a silica column (gradient from hep-
tane to ethyl acetate), to give 39.1 g of product (55.7%
yield).
whereafter the catalyst DBU (180 lL, 1.20 mmol) was added
to start the reaction. Samples for H-NMR analyses were taken
1
every 5 min for kinetics and the reaction was terminated by
addition of 0.5 mL acetic acid after 30 min. Purification of the
polymer was achieved through precipitation in cold methanol.
The methanol was carefully decanted off after 1 h in a freezer
at 218 8C. Precipitation was conducted twice to give pMAC25
as a high viscous liquid (4.68 g, 88.6% yield).
1
H NMR (CDCl , 400 MHz) d 1.06 (s, 3H, -CH ), 3.33 (s, 2H, -
3
3
OH), 3.76 (dd, 4H, 1CH -O-), 4.62 (d, 2H, 5-CH -O-), 5.21
2
2
(
1
dd, 1H, 5CHH (cis)), 5.30 (dd, 1H, 5CHH (trans)), 5.88 (m,
1
3
H, 5CH-); C NMR (CDCl , 101 MHz) d 17.24, 49.37,
3
65.52, 67.32, 118.37, 131.85, 175.57.
Synthesis of 5-Methyl-5-Allyloxycarbonyl-
,3-Dioxan-2-One (MAC)
Synthesis of N-(3,4-Dihydroxyphenetyl)24-
Mercaptobutanamide (Dopa-Thiol)
1
Carbonate formation was achieved using ethyl chloroformate.
Synthesized allyl 3-hydroxy-2-(hydroxymethyl)22-methylpro-
panoate (30.0 g 172 mmol) was dissolved in 150 mL THF,
after which ethyl chloroformate (82.0 mL, 861 mmol) was
added. TEA (143 mL, 1030 mmol) was dissolved in 143 mL
THF, whereafter the solution was slowly added to the reac-
tion flask. Formed HCl salt was filtered off and the remaining
solution was concentrated using reduced pressure. The prod-
uct was precipitated in cold diethyl ether two times and
then recrystallized in diethyl ether from THF to give the
product, MAC, as white crystals (21.8 g, 63% yield).
Dopamine hydrochloride (10.0 g, 52.7 mmol) and c-
thiobutyrolactone (4.80 ml, 55.4 mmol) were dissolved in a
100 mL aqueous solution of NaHCO3 (9.30 g, 111 mmol).
The reaction mixture was heated to 95 8C and allowed to
react under reflux for 2 h; 100 mL of brine was added and
the product was extracted two times with 500 mL THF. The
organic phase was dried over magnesium sulfate and fil-
tered, after which the solvent was evaporated under reduced
pressure. Further purification was carried out using flash
chromatography, giving 9.40 g (36.8 mmol) of the product as
a slightly yellow solid. 70% yield.
1
H NMR (CDCl , 400 MHz) d 1.36 (s, 3H, -CH ), 4.21 (d, 2H,
1
3
3
H NMR (MeOD, 400 MHz) d 1.85 (p, 2H, -CH -CH -SH), 2.28
t, 2H, -CH -CO-), 2.47 (t, 2H, -CH -SH), 2.65 (t, 2H, -CH -Ar),
2 2 2
2
2
5
-CH -O-), 4.70 (m, 4H, 1CH -O-), 5.31 (dd, 1H, 5CHH
2
2
(
1
3
(cis)), 5.35 (dd, 1H, 5CHH (trans)), 5.91 (m, 1H, 5CH-);
C
3
6
3
1
.35 (t, 2H, -CH -NH-), 6.53 (dd, 1H, Ar), 6.66 (d, 1H, Ar),
.70 (d, 1H, Ar); C NMR (MeOD, 101 MHz) d 24.42, 31.27,
5.53, 35.85, 42.14, 116.32, 116.83, 121.03, 131.96, 144.70,
46.18, 175.25.
2
13
NMR (CDCl , 101 MHz) d 17.9, 40.4, 66.9, 73.2, 119.8, 131.1,
3
147.6, 170.9.
Synthesis of Poly(5-methyl-5-allyloxycarbonyl-
,3-dioxan-2-one) (pMAC)
1
Ring-opening of MAC from 1-pyrene butanol was conducted
in dry DCM under inert atmosphere at 20 8C using the cata-
lyst/co-catalyst system of 1,8-diazabicycloundec-7-ene (DBU)
and a thiourea co-catalyst (TU) (see Supporting Information
for synthesis of TU). Three polymers with different length
Synthesis of pMAC(Allyl/Dopa)
Dopa functionalization of pMAC was achieved through UV-
initiated TEC between the allyl-groups of pMAC and dopa-
thiol as follows.
(
degree of polymerization (DP) of 25, 46, and 70) were pro-
Synthesized pMAC (400 mg), dopa-thiol (see Table 2), and 1
wt % Irgacure 184 were dissolved in 4.00 mL 1,4-dioxane,
duced within short reaction times according to the following
general procedure. Amounts and conditions for all polymer-
izations are listed in Table 1.
after which each mixture was bubbled through with N2(g)
.
Each sample was then subjected to UV-irradiation according
to the following protocol:
General Polymerization Procedure
MAC (5.00 g, 25.0 mmol), initiator 1-pyrene butanol
1. Two times 30 min irradiation in the Blak-Ray UV-Lamp
(365 nm, 100 W, 10 cm distance from UV source), with an
extra addition of 3 mg Irgacure 184 after 30 min.
(
274 mg, 0.999 mmol), and TU (444 mg, 1.20 mmol) were
transferred to a thoroughly dried round-bottom flask (2 h,
50 8C), after which three cycles of vacuum/N2(g) were
1
2. Addition of 3 mg Irgacure 184, followed by irradiation:
0.552 s on the Fusion UV conveyor belt (320–360 nm,
applied in order to ensure an inert atmosphere. 30.0 mL dry
2
2
DCM (<0.004% H 0) was added to dissolve all reactants,
1.656 J cm
)
2
2
372
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 2 Specified Amounts and Ratios of Dopa-Thiol used in
the Functionalization of pMAC, as well as the Radiation Proto-
col used for each Sample
ring-opening polymerization (ROP) using an organometallic
catalyst complex consisting of DBU and a TU co-catalyst
(
Supporting Information), according to Scheme 1. Three dif-
ferent DPs were targeted and the kinetics of the polymeriza-
tions were investigated using H NMR and SEC in DMF with
PMMA standards. The polymers will be referred to as short
DP
Dopa-thiol
(mg)
Irradiation
protocol
1
Sample
(pMAC)
Thiol:allyl
(
DPtarget 25), medium (DPtarget 50), and long (DPtarget 75),
1
2
3
4
5
6
7
8
9
25
25
25
46
46
46
70
70
70
1:2
1:1
3:2
1:2
1:1
3:2
1:2
1:1
3:2
242
484
725
248
495
743
250
500
751
1 1 2
unless the structure is otherwise specified.
1 1 2
1 1 2 1 3
1 1 2
During the polymerizations, samples for NMR and SEC were
taken every 5 min to enable kinetic evaluation. However, due
to high polymerization rates, the polymerizations had to be
terminated before analysis of these samples could be con-
ducted and it was decided that the polymerizations were to
be terminated after 30 min. Monomer conversion was deter-
mined from H-NMR spectra by comparing integrals of peaks
from the monomer and the polymer with each other (Sup-
porting Information Fig. S5). H-NMR spectra (Supporting
1 1 2
1 1 2 1 3
1 1 2
1 1 2
1
1 1 2 1 3
1
4
00 mg of pMAC were used in all reactions.
Information Fig. S6) were also used to calculate the final DP
of the polymers after integration of the peaks belonging to
the initiator, 1-pyrene butanol. Final DP was found to be 25,
3
. Addition of 10 mg Irgacure 184, followed by irradiation:
0
.92 s on the Fusion UV conveyor belt (320–360 nm, 2.76
46, and 70 for short, medium, and long pMAC, respectively.
2
2
J cm
)
Specified amounts of dopa-thiol used for all samples can be
seen in Table 2, along with details regarding the irradiation
protocol used for each sample. Repeated irradiation was uti-
lized in an attempt to increase the functionalization degree.
Evaluation of samples from the kinetic study of the polymer-
izations of pMAC-short and pMAC-medium (Fig. 1) suggested
that the maximum monomer conversion for pMAC-short was
reached after only 20 min and also showed a clear decrease
in polymerization rate when comparing pMAC-short and
pMAC-medium. Aiming for a monomer conversion of approx-
imately 80%, the results from pMAC-short and pMAC-
medium thus suggested that pMAC-long would require a lon-
ger reaction time than 30 min and the polymerization of
pMAC-long was therefore terminated after 50 min.
31
Gel Formation Using Fe
0.0 mg of pMAC(allyl29/dopa17) (2.89 lmol, 49.2 lmol
dopa groups) was dissolved in 54.7 lL of FeCl ꢀ6H O solu-
4
3
2
3
1
tion (0.3 M FeCl ꢀ6H O in methanol, 16.4 lmol Fe ). 25.3
3
2
lL of 2 M NaOH(aq) was thereafter added dropwise during
stirring in order to initiate gel formation.
1
In Figure 1(a), the monomer conversion (calculated from H
NMR) has been plotted against reaction time. It can be seen
that the monomer conversion converges toward approxi-
mately 90% and that the rate of polymerization decreases as
the targeted DP is increased, as shown by the differences in
the inclination of the curves. As the monomer concentration
was kept constant for all polymerizations, increased DP was
achieved by decreasing both initiator and catalyst/co-catalyst
RESULTS AND DISCUSSION
Polymerization Kinetics and Characterization
In order to produce a suitable platform for further function-
alization using TEC, a linear PC with allyl functionalities in
the repeating unit was envisioned. MAC was considered a
suitable monomer and the polymer was synthesized through
SCHEME 1 Reaction scheme for the ring-opening polymerization of MAC and UV-initiated TEC postpolymerization functionaliza-
tion with dopa-thiol leading to pMAC(allyl/dopa).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378
2373

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 1 Kinetic study of the ring-opening polymerization of MAC. a) Monomer conversion versus reaction time, and b) 2ln([M]/
1
[
M]
0
) versus reaction time. Monomer conversion was calculated by comparing monomer and polymer integrals from H NMR
(
monomer peaks: 4.17 and 4.20 ppm (2H, d), and polymer peak: 4.28 ppm (4H)). [Color figure can be viewed in the online issue,
which is available at wileyonlinelibrary.com.]
concentration. Due to this, a decrease in the concentration of
active growing chains was expected, which should lead to a
decrease in polymerization rate for the higher DPs.
1,4-dioxane in glass vials (14 mm in diameter). Three differ-
ent thiol-to-allyl ratios were used for the three lengths of
pMAC: 1:2, 1:1, and 3:2, resulting in a total of nine samples
(
see the Experimental for specified amounts of reactants). 1
The rate profiles in Figure 1(b) imply first-order kinetics in
the beginning of the polymerizations for all DPs. In the end
of the polymerizations, the profiles, however, deviate from
linearity, probably because the reactions are approaching
equilibrium monomer concentration. This is clearly the case
for the short pMAC, where final conversion was reached
after approximately 20 min.
wt % of the UV-initiator Irgacure 184 was initially added to
all samples and the samples were thereafter subjected to
UV-light to initiate TEC according to the following protocol:
1
2
3
. Two times 30 min irradiation using a Blak-Ray UV-Lamp
365 nm, 100 W, 10 cm distance from UV source). 3 mg of
(
Irgacure 184 were added after 30 min.
. Addition of 3 mg Irgacure 184, followed by 0.552 s irradi-
ation on a Fusion UV conveyor belt (320–360 nm, 1.656
The molecular weights obtained for the polymers using SEC
(
Table 3) are notably lower than the theoretical values calcu-
2
2
1
J cm
)
lated from H-NMR spectra, probably due to differences in
hydrodynamic volume between the synthesized pMAC and
the PMMA standards used for the calibration of the instru-
ment. These differences between molecular weights calcu-
lated from NMR and molecular weights obtained from SEC
were furthermore more pronounced for the longer polymers,
probably because the solubility of the polymers in DMF var-
ied with chain length, thereby influencing the hydrodynamic
volume. It should also be mentioned that some of the SEC
traces (Supporting Information Figs. S7 and S8) were a bit
asymmetric, with signs of a peak at double molecular weight,
indicative of chain-chain coupling. The low dispersities
. Addition of 10 mg Irgacure 184, followed by 0.92 s irradi-
ation on the Fusion UV conveyor belt (320–360 nm, 2.76
2
2
J cm ) (only samples with a thiol-to-allyl ratio of 3:2)
Additional amounts of initiator were added before each radi-
ation step to ensure that not all initiator had been con-
sumed. (See Supporting Information for details regarding the
molar attenuation coefficient of Irgacure 184 in 1,4-dioxane
and expected absorbance values for each irradiation step.) In
TABLE 3 Molecular Weight of Synthesized pMAC Calculated
from NMR Compared with Molecular Weights Obtained from
SEC
reported in Table 3, as well as the linearity of M as a func-
n
tion of monomer conversion (Supporting Information Fig.
S9), none the less suggest a low level of side reactions.
a
a
b
b
w
Sample
M
NMR
M
n
M
Ð
M
Functionalization of pMAC with Dopa-Thiol
pMAC25
pMAC46
pMAC70
5,300
9,500
5,000
6,700
7,500
5,300
7,400
9,500
1.14
1.25
1.25
After successful synthesis of allyl-functional polycarbonates,
pMAC, subsequent functionalization with a thiol-functional
dopamine derivative, dopa-thiol, was achieved through UV-
initiated TEC. Due to the radical-scavenging properties of the
catechol, the influence of different ratios of thiol to allyl was
investigated. Additionally, irradiation was conducted two to
three times in an attempt to increase the degree of function-
alization. 400 mg of polymer was dissolved in 4.00 mL of
14,000
21
All molecular weights are reported in g mol
.
a
Degree of polymerization (DP) (subscript in sample name) and molec-
1
ular weight calculated from H NMR.
b
Number and weight average molecular weights and dispersity
obtained from DMF-SEC using PMMA standards (see Supporting Infor-
mation Fig. S8 for SEC spectra).
2
374
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
1
FIGURE 2 Functionalized allyl groups after irradiation with UV determined from H-NMR spectra. a) Functionalized allyl groups
after each radiation step (sample names in the graph include DP of pMAC followed by the thiol:allyl ratio used. Lines have been
added between data points acquired for the same sample to guide the eye. b) Final functionalization degree as a function of DP
and in relation to the thiol:allyl ratio. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
order to minimize gradient effects due to high absorbance
values as a result of added initiator, each vial was regularly
shaken during irradiation, as continuous stirring was not
possible. It should however be noted that such gradient
effects might still have been present in the samples, as the
absorbance values in step 2 and 3 were higher than 0.5.
irradiation, the functionalization of the allyl groups was lim-
ited to between 32 and 48%, even if the degree of function-
alization increased slightly upon additional radiation. It
therefore seemed like the functionalization of pMAC was hin-
dered, either by the radical-scavenging properties of the cat-
echol or by steric hindrance. A dependence on molecular
weight for the degree of functionalization could not be seen,
although generally, a higher amount of thiol gave a higher
functionalization of allyl groups, where a 3:2 ratio of thiol to
allyl resulted in functionalization of 41 to 48% of the allyl
groups after the three steps of irradiation. As a comparison,
(
Supporting Information). Due to the high absorbance values,
it is therefore not possible to draw any reliable conclusions
regarding the amount of irradiation in steps 2 and 3.
The amount of functionalized allyl groups was determined
1
using H-NMR analysis of crude samples after each radiation
step (Supporting Information Fig. S10). In Figure 2, it can be
seen that even after repeated addition of initiator and UV-
FIGURE 4 UV-Vis spectra of dilute solutions of FeCl
3 2
6 H 0
31
mixed with pMAC(allyl29-co-dopa17) (Fe :dopa ratio 1:3) at dif-
ferent pH. Aqueous buffer solutions with pH 4, 7, and 10
(Scharlau) were used. For pH 5.6 and 8.3, pH was adjusted
using 2 M NaOH or acetic acid (37%). The pH of all solutions
was measured using a pH meter. Peaks evolving at 578 and
500 nm correspond to the formation of bis and tris complexes,
g
FIGURE 3 Comparison of T for pMAC and pMAC(allyl/dopa)
acquired from DSC measurements as a function of DP (Sup-
porting Information Figs. S11–S16). Labels corresponding to
the amount of allyl groups that have been functionalized with
dopa have been added to facilitate comparison of the samples.
3
1
respectively, between Fe
ions and catechol as pH is
[
Color figure can be viewed in the online issue, which is avail-
increased. [Color figure can be viewed in the online issue,
which is available at wileyonlinelibrary.com.]
able at wileyonlinelibrary.com.]
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378
2375

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 5 Results from rheology: a) oscillatory frequency sweep, 0.1% strain; b) oscillatory strain sweep, 1 Hz; c) oscillatory time
sweep, constant frequency of 1 Hz and three consecutive 60 s cycles of 0.1% strain followed by 100% strain; and d) photograph of
31
a free-standing gel after complex formation with Fe . [Color figure can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
Moulay and Mehdaoui functionalized poly(acrylic acid) with
catechol through oxidative decarboxylation and found that
the degree of functionalization was limited to 30% (35%
tions between dopa groups on adjacent polymer chains,
through e.g. stacking of the aromatic rings or hydrogen
bonding. Additionally, the bulkiness of the dopa side chains
could restrict the movements of the polymer chains even fur-
ther. No crystallization could be detected for either unfunc-
tionalized pMAC or for the functionalized pMAC(allyl/dopa).
3
8
under N atmosphere). We additionally believe that further
2
functionalization could have been achieved if the samples
would have been irradiated further but, due to the low
increase in functionalized allyl groups in the third irradiation
step, further irradiation was not conducted in this study.
31
Gel Formation with Fe
As evidenced in literature, DOPA has a high affinity towards
Thermal Properties
3
1
After the successful functionalization with dopa, the thermal
properties of pMAC and pMAC(allyl/dopa) were investigated
using DSC, Figure 3. For the allyl-functionalized pMAC it
could be seen that the glass-transition temperature (Tg)
increased slightly with increasing polymer length, as more
energy is required to move longer chain segments and
because the probability of chain entanglements increases.
ions such as Fe . Depending on the pH of the surrounding
3
1
environment, Fe
can form complexes with up to three
dopa moieties. Below pH 5, there are mainly monocom-
plexes, but as pH is increased, both bis and tris complexes
5
become more probable.
3
1
The complexation between Fe
and synthesized dopa-
functional polymers was demonstrated for pMAC(allyl /
2
9
Due to the similar functionalization degrees, no conclusion
dopa17) using UV-Vis-absorption spectroscopy. Analysis of
the spectra in Figure 4 show that as the pH is raised from
3.9 to 7 there is an evolution of a peak around 578 nm. Con-
tinuing the increase in pH, furthermore gives an increase in
intensity for a peak around 500 nm. In accordance with
regarding how different amounts of dopa affect the T can be
g
drawn from the DSC data. It is, however, clear that the func-
tionalization with dopa-thiol increases the T of the materials
g
drastically. This increase in T is probably a result of interac-
g
2
376
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
literature, these peaks at 578 and 500 nm correspond to the
formation of bis and tris complexes, respectively.
thiol amount only gave a slight increase in the functionaliza-
tion degree of the allyl groups of pMAC, even after repeated
UV-exposure, implying that there might be a maximum func-
tionalization degree for the system. It was additionally dem-
onstrated that the synthesized dopa-functional material
could be used to produce a self-healing gel via complex for-
4
31
The shown complexation between Fe -ions and dopa was
envisioned as a way of producing self-healing gels. In order
to test this theory, a solution of pMAC(allyl /dopa ) and
2
9
17
31
FeCl ꢀ6 H O, with a dopa:Fe ratio of 3:1, was prepared in
31
3
2
mation with Fe ions.
methanol. 2 M NaOH was used to increase the pH of the
solution, resulting in the formation of a free-standing gel.
The gel properties were then analyzed using oscillatory rhe-
ology with 8 mm parallel plates. A frequency scan from 0.01
to 10 Hz at 0.1% [Fig. 5(a)] showed that the storage modu-
Conclusively, TEC chemistry can be used as a simple and
effective method to synthesize dopa-functional polymers,
without having to protect the catechol. In this work, allyl-
functional PC, pMAC, was used to demonstrate the concept,
but we foresee that this route can be effective also for other
allyl-functional polymers, thereby expanding the possibilities
for dopa-functional materials.
0
00
lus, G , was higher than the loss modulus, G , confirming that
the material exhibited viscoelastic properties. From the com-
plex viscosity, it could additionally be seen that the material
was shear thinning, which means that it flowed easier at
higher shear rates. Additionally, when running a frequency
scan from 10 to 0.01 Hz, a cross-over frequency between G
and G was detected (Supporting Information Fig. S17). This
behavior is not uncommon for physically cross-linked
0
0
0
ACKNOWLEDGMENT
Wilhelm Beckers Jubileumsfond is acknowledged for financial
support.
3
9–41
gels.
At low frequencies, slow shear rates, the dynamic
bonds have sufficient time to reform during deformation and
the material exhibits flow. At higher frequencies, above the
cross-over frequency, the bonds are however not given suffi-
cient time to break and reform during deformation, resulting
in an elastic gel-like response.
REFERENCES AND NOTES
1
2
3
4
J. H. Waite, M. L. Tanzer, Science 1981, 212, 1038–1040.
J. H. Waite, Int. J. Adhes. Adhes. 1987, 7, 9–14.
L. A. Burzio, J. H. Waite, Biochemistry 2000, 39, 11147–11153.
N. Holten-Andersen, M. J. Harrington, H. Birkedal, B. P. Lee,
From the strain sweep measurement [Fig. 5(b)], it can be
seen that at low shear rates, up to around 1% strain, the gel
exhibited constant moduli and viscosity. Above 1% strain,
there was, however, a drop in moduli and viscosity and at
P. B. Messersmith, K. Y. C. Lee, J. H. Waite, Proc. Natl. Acad.
Sci. USA 2011, 108, 2651–2655.
5
S. W. Taylor, D. B. Chase, M. H. Emptage, M. J. Nelson, J. H.
Waite, Inorg. Chem. 1996, 35, 7572–7577.
M. J. Harrington, A. Masic, N. Holten-Andersen, J. H. Waite,
P. Fratzl, Science 2010, 328, 216–220.
M. J. Sever, J. J. Wilker, Abstr. Pap. Am. Chem. Soc. Div.
Fuel Chem. 2001, 222, U567–U567.
J. Yu, W. Wei, M. S. Menyo, A. Masic, J. H. Waite, J. N.
Israelachvili, Biomacromolecules 2013, 14, 1072–1077.
4.32% strain the material displayed viscous properties,
meaning that the material could no longer be considered a
gel. After the strain was removed, the material, however,
returned to the gel state.
6
7
The self-healing characteristics of the gel were demonstrated
using a time sweep test, where the strain was alternated
between 0.1% and 100% strain at a constant frequency of 1
Hz, Figure 5(c). When the material was subjected to 100%
8
9
L. Li, W. Smitthipong, H. B. Zeng, Polym. Chem. 2015, 6,
3
53–358.
00
0
strain, the gel became viscous with G > G . Once the strain
1
0 D. S. Hwang, Y. S. Choi, H. J. Cha, In Marine Biomaterials:
was once again reduced to 0.1%, the gel-like character was
Characterization, Isolation and Applications; Kim, S.-K., Ed.;
CRC Press Boca Raton, FL, 2013, pp 289-310.
0
00
fully regained, with G > G . This procedure could be
repeated multiple times, without losses in moduli, confirm-
ing the self-healing character of the gel. The reformation of
the gel was furthermore found to be very fast, as the moduli
recovered within the 6 s passing between measuring points
of the two extremes.
11 R. L. Strausberg, R. P. Link, Trends Biotechnol. 1990, 8, 53–
57.
12 D. S. Hwang, Y. Gim, H. J. Yoo, H. J. Cha, Biomaterials
2
1
1
007, 28, 3560–3568.
3 S. W. Taylor, Anal. Biochem. 2002, 302, 70–74.
4 L. Nakonieczna, W. Przychodzen, A. Chimiak, Liebigs Ann.
Chem. 1994, 1055–1058.
CONCLUSIONS
1
5 S. N. Banerjee, C. Ressler, J. Org. Chem. 1976, 41, 3056–
058.
In the search for a fast and facile synthetic route towards
dopa-functional polymers, TEC chemistry was identified as a
promising candidate. A linear PC, pMAC, with accessible allyl
side groups, was considered a suitable platform for further
functionalization. It was found that up to 48% of these allyl
groups could be functionalized with a nonprotected dopa-
functional thiol within short reaction times using UV-
initiated TEC. The results showed that an increase in the
3
1
6 B. H. Hu, P. B. Messersmith, Tetrahedron Lett. 2000, 41,
5
795–5798.
1
1
7 M. J. Sever, J. J. Wilker, Tetrahedron 2001, 57, 6139–6146.
8 A. K. Sinha, A. Sharma, B. P. Joshi, Tetrahedron 2007, 63,
960–965.
19 R. Bernini, E. Mincione, M. Barontini, G. Provenzano, L.
Setti, Tetrahedron 2007, 63, 9663–9667.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378
2377

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
2
0 S. Moulay, C.R. Chim. 2009, 12, 577–601.
31 C. E. Hoyle, T. Y. Lee, T. Roper, J. Polym. Sci. Part A: Polym.
Chem. 2004, 42, 5301–5338.
2
1 M. S. Menyo, C. J. Hawker, J. H. Waite, Soft Matter 2013, 9,
1
0314–10323.
32 K. L. Killops, L. M. Campos, C. J. Hawker, J. Am. Chem.
Soc. 2008, 130, 5062–5062.
22 L. Q. Xu, H. Jiang, K. G. Neoh, E. T. Kang, G. D. Fu, Polym.
Chem. 2012, 3, 920–927.
3 K. M. Mattson, A. A. Latimer, A. J. McGrath, N. A. Lynd, P.
Lundberg, Z. M. Hudson, C. J. Hawker, J. Polym. Sci. Part A:
Polym. Chem. 2015, 53, 2685–2692.
3
3 J. V. Olsson, D. Hult, Y. L. Cai, S. Garcia-Gallego, M.
Malkoch, Polym. Chem. 2014, 5, 6651–6655.
4 W. Chen, H. C. Yang, R. Wang, R. Cheng, F. H. Meng, W. X.
Wei, Z. Y. Zhong, Macromolecules 2010, 43, 201–207.
5 S. Onbulak, S. Tempelaar, R. J. Pounder, O. Gok, R. Sanyal,
A. P. Dove, A. Sanyal, Macromolecules 2012, 45, 1715–1722.
6 S. Tempelaar, L. Mespouille, P. Dubois, A. P. Dove, Macro-
molecules 2011, 44, 2084–2091.
7 X. L. Hu, X. S. Chen, Z. G. Xie, S. Liu, X. B. Jing, J. Polym.
Sci. Part A: Polym. Chem. 2007, 45, 5518–5528.
2
3
2
4 B. B. Uysal, U. S. Gunay, G. Hizal, U. Tunca, J. Polym. Sci.
Part A: Polym. Chem. 2014, 52, 1581–1587.
5 P. Antoni, M. J. Robb, L. Campos, M. Montanez, A. Hult, E.
Malmstrom, M. Malkoch, C. J. Hawker, Macromolecules 2010,
3
2
3
43, 6625–6631.
3
26 Y. Hed, K. Oberg, S. Berg, A. Nordberg, H. von Holst, M.
Malkoch, J. Mater. Chem. B 2013, 1, 6015–6019.
7 M. I. Montanez, L. M. Campos, P. Antoni, Y. Hed, M. V.
3
8 S. Moulay, R. Mehdaoui, React. Funct. Polym. 2004, 61, 265–
75.
2
2
Walter, B. T. Krull, A. Khan, A. Hult, C. J. Hawker, M. Malkoch,
Macromolecules 2010, 43, 6004–6013.
39 X. Z. Yan, D. H. Xu, J. Z. Chen, M. M. Zhang, B. J. Hu, Y. H.
Yu, F. H. Huang, Polym. Chem. 2013, 4, 3312–3322.
2
8 M. V. Walter, P. Lundberg, A. Hult, M. Malkoch, J. Polym.
Sci. Part A: Polym. Chem. 2011, 49, 2990–2995.
9 S. Garcia-Gallego, D. Hult, J. V. Olsson, M. Malkoch, Angew.
Chem. Int. Ed. 2015, 54, 2416–2419.
0 A. B. Lowe, Polym. Chem. 2010, 1, 17–36.
40 M. P. Algi, O. Okay, Eur. Polym. J. 2014, 59, 113–121.
2
41 O. Okay, In Supramolecular Polymer Networks and Gels; S.
Seiffert, Ed.; Springer International Publishing: Switzerland,
2015; pp 101–142.
3
2
378
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 2370–2378