Self-healing of polymers via synchronous covalent bond fission/radical recombination

Article abstract of DOI:10.1021/cm202635w

The present paper is devoted to preparation of intrinsic self-healing polymeric materials used for structural applications. The authors introduced a novel healing chemistry based on dynamically reversible C-ON bonds, which imitates natural healing in livi

Full text of DOI:10.1021/cm202635w

Article
pubs.acs.org/cm
Self-Healing of Polymers via Synchronous Covalent Bond Fission/
Radical Recombination
Chan′e Yuan, Min Zhi Rong,* Ming Qiu Zhang,* Ze Ping Zhang, and Yan Chao Yuan
Key Laboratory for Polymeric Composite and Functional Materials of Ministry of Education, DSAPM Lab, School of Chemistry and
Chemical Engineering, Sun Yat-Sen (Zhongshan) University, Guangzhou 510275, P. R. China
S
* Supporting Information
ABSTRACT: The present paper is devoted to preparation of intrinsic self-
healing polymeric materials used for structural applications. The authors
introduced a novel healing chemistry based on dynamically reversible C−ON
bonds, which imitates natural healing in living bodies without affecting their
operations. To verify its feasibility, alkoxyamine moieties served as
intermolecular links in polystyrene. Upon heating, covalent bond fission and
radical recombination synchronously took place among alkoxyamine moieties.
Cracked parts were thus reconnected repeatedly, without losing integrity and
load bearing ability of the material even above Tg.
KEYWORDS: self-healing, synchronous reversibility, alkoxyamines, cross-linked polymer
Here we show the feasibility of using a dynamically reversible
C-ON bond as the healing chemistry, which combines the
aforesaid covalent bond breakage and reconnection into one
step. The possible material deformation in the course of crack
remending would thus be avoided even when the healing is
carried out above its glass transition temperature. As illustrated
in Scheme 1a, C-ON bond in an alkoxyamine moiety
INTRODUCTION
■
Bioinspired self-healing polymeric materials have attracted
more and more research interests.1−13 Unlike extrinsic self-
healing that functions in virtue of embedded healing agent,3−5
intrinsic self-healing depends only on recombination ability of
macromolecules themselves.6−13 To remend damaged polymer
constructed by covalent chemical bonds, thermally reversible
Diels−Alder (DA) intermonomer linkages were introduced.6
Accordingly, cracks in cross-linked polymer made from
cycloaddition between multifuran and multimaleimide were
repeatedly rebonded. During crack healing, however, the
polymer has to be heated up to retro-DA temperature (120
°C) for disconnecting DA bonds (i.e., de-cross-linking), which
is generally higher than the glass transition temperature, and
then cooled for reconnection of the DA bonds (DA reaction ≈
80 °C). It means that the material would inevitably lose its
shape as a result of (i) molecular cleavage during the first stage
of healing (i.e., retro-DA reaction/de-cross-linking) and (ii)
drastically increased deformability due to glass transition.14 It is
worth noting that on one hand, breakage of reversible bonds
and improved mobility of molecules are necessary for crack
healing,15 whereas on the other, distortion of end-use products
is not allowed for structural applications where mechanical
strength properties are critical even if self-healing is proceeding.
The requirements seem to be contradictory and are hard to be
satisfied simultaneously in conventional polymers.
Scheme 1. (a) Dissociation/Association of Alkoxyamines
Derivatives; (B) Dynamic De-Cross-Linking, Crossover
Cross-Linking, or Exchange Reactions of PS
In fact, natural healing in living bodies would not affect their
operations. Circulatory system keeps on working, for instance,
despite that blood clotting is being formed in an injured vessel.
Evidently, mimicking healing process of organisms to a higher
degree would further improve performance of artificial self-
healing materials. Innovative ideas about molecular structure
design and reversible mechanism of molecular fracture and
combination should be proposed accordingly.
Received: September 4, 2011
Revised: October 12, 2011
Published: October 21, 2011
© 2011 American Chemical Society
5076
dx.doi.org/10.1021/cm202635w|Chem. Mater. 2011, 23, 5076−5081

Chemistry of Materials
Article
injected into the solution of methacryloyl chloride (7.8 g, 75 mmol)
and diol (7.3 g, 25 mmol) in anhydrous tetrahydrofuran (50 mL). The
reaction was carried out with stirring under argon for 10 h at 25 °C,
and then evaporated to dryness. The residue was partitioned between
saturated sodium chloride solution and dichloromethane in which the
aqueous layer was extracted with dichloromethane for several times.
The collected organic layers were dried with anhydrous sodium sulfate
and evaporated to dryness. The crude product was purified by flash
chromatography eluting with 1:20 ethyl acetate/hexane (v/v) to give
dimethacrylic ester as pale yellow oil (9.4 g, 88% yield). FT-IR (KBr/
cm−1): 3107−2856, 1719, 1637, 1327, 1166, 763, 701; 1H NMR
(CDCl3): δ/ppm 0.72 (3H, CH3), 1.16 (3H, CH3), 1.31 (3H, CH3),
1.39 (3H, CH3), 1.49 (3H, CH3), 1.42−2.02 (4H, 2CH2), 1.95 (3H,
CH3), 2.01 (3H, CH3), 4.63 (1H, CH), 5.29 (1H, CH), 5.49 (1H,
CH), 5.52 (1H, CH), 5.82 (1H, vinyl), 5.98 (1H, vinyl) 6.05 (1H,
vinyl), 6.23 (1H, vinyl), 7.25−7.35 (5H, aromatic). 13C NMR
(CDCl3): δ/ppm 18.25, 18.59, 21.46, 22.23, 34.20, 34.47, 44.57,
44.81, 59.42, 60.55, 66.88, 84.25, 125.18, 125.89, 127.68, 127.82,
136.12, 136.59, 140.27, 140.49, 166.77, 166.93. Mass spectrum for
C25H35NO5 [M+1]+ found 430.4.
Preparation of Reversibly Cross-Linked PS. Benzoyl peroxide
(58.1 mg, 0.24 mmol) and certain amount of cross-linker
dimethacrylic ester were dissolved in styrene (7.8 g, 75 mmol) and
then charged into a tube. The tube was evacuated and backfilled with
argon for three times. The mixture was cured for 24 h at 80 °C and
postcured for 1 h at 100 °C. Three molar feeding ratios of styrene over
cross-linker dimethacrylic ester were used: 5/1, 7.5/1, and 9.4/1. FT-
IR spectra of the products present following characteristic peaks (KBr/
cm−1): 3026−2975, 1602, 1364, 1149, 1117, 1043, 699, 758, in which
there was no peak of double bond at 1650 cm−1, indicating that the
polymer has been completely cured. The control material, irreversibly
cross-linked PS excluding alkoxyamine moiety, was made following
similar procedures by using ethyleneglycol dimethacrylate (EGDMA)
as cross-linker.
Characterization. Fourier transform infrared (FT-IR) spectra
were recorded by a Bruker EQUINOX55 Fourier transformation
infrared spectrometer coupled with an infrared microscope spec-
trometer. 1H NMR and 13C NMR spectra were measured by a
VARIAN Mercury-Plus 300 (300 MHz) with CDCl3 as solvent. Mass
spectra were obtained by Shimadzu LCMS-2010A. Electron spin
resonance (ESR) spectroscopy study was performed on a JEOL JES-
FA200 spectrometer equipped with nitrogen heating setup operating
at 8.85 Hz. Modulation frequency and amplitude were 100 kHz and
0.1 mT, respectively. Dynamic mechanical analysis (DMA) was
conducted on a Mettler Toledo DMA/SDTA861 using single
cantilever mode under 1 Hz at a heating rate of 2 °C min−1 in
nitrogen. Thermomechanical analysis (TMA) was performed on TA/
DMA2980 instrument in penetration mode under a pressure of 0.35
MPa at a heating rate of 2 °C min−1 in nitrogen. Thermal degradation
habits of the materials were determined by a TA-Q50 thermogravim-
eter (TGA) in N2 at a heating rate of 10 °C min−1. To evaluate healing
efficiency of the materials, double cleavage drilled compression
(DCDC) tests14 were conducted using a SANS CMT 6000 universal
tester. The specimen was a column of a rectangular cross-section
(length L = 50 mm, width w = 16 mm, thickness t = 9.1 mm) with a
circular hole (radius R = 2 mm) drilled through its center. Prior to the
test, a blade was wedged into the central hole of the DCDC specimen
to create precracks 0.5 mm deep at the upper and lower crowns. Then,
a restraining clamp was placed around the specimen and axial
compression was applied. The lateral confinement offered by the
clamp prevented the cracks from growing beyond a length, l (l/R =
0.8), from the crowns of the hole. Afterward, the clamp was removed
and the specimen was reloaded in compression at a crosshead rate of
0.2 mm/min to record the critical force, σcrit, and crack length, l, of the
subsequent fracture event. The test was paused at increments of 60 μm
in crosshead displacement and, after an equilibration period of 10 s,
the peak load over that period was recorded and a photo was taken for
determining crack length. After the test, the specimen was placed in a
preheated oven at 130 °C in argon for 2.5 h, followed by cooling to
room temperature. Finally, the healed specimen was tested again
homolytically cleaves upon heating and produces transient
carbon-centered and persistent nitroxide radicals in equal
amounts16−21 with very high frequency factor (e.g., 2.4 × 1014
s−1 in solution20). Meanwhile, the carbon-centered radicals
would be rapidly trapped by the nitroxide radicals.17 As a result,
under certain homolysis temperature the C-ON bonds in
alkoxyamines frequently cleave but immediately recombine.
Moreover, the recombination of radicals involves crossover or
exchange reaction between radicals that belong to different
alkoxyamine moieties before the chain breakage.22 Taking
advantage of this dynamic equilibrium, nitroxide-mediated
radical polymerization has become one of the most useful
methods for living radical polymerization20,21,23,24 and various
well-defined and complex molecular architectures, many of
which are difficult to prepare via traditional methods, have been
synthesized.21,25−28 For example, a reversible sol−gel was
obtained from linear poly(methacrylic ester)s containing
alkoxyamine units on the side chain.29 Upon heating, the
cross-linked gel in solvent (10 wt % anisole solution of
polymer) was formed through radical exchange reactions of
alkoxyamine moieties, releasing alkoxyamine molecules. The
retro-reaction, i.e., de-cross-linking, occurred in the presence of
an excess amount of alkoxyamine molecules. Although this kind
of sol−gel conversion took place in solvent rather than in solid,
it theoretically proved that the reversible reaction between
alkoxyamine moieties may be useful to construct a reversible
cross-linked structure. Nevertheless, in a hard polymer solid,
molecular mobility is greatly restricted. As a result, release and
incorporation of alkoxyamine molecules are impossible.
To create a novel thermally reversible polymer solid with
crack remendability, in this work, we directly incorporate
alkoxyamine linkages into cross-linked polymer solid as cross-
linkers that cleave and reconnect without byproduct. Because of
synchronous scission of alkoxyamine moieties and recombina-
tion of resultant radicals, the cross-linked networks would not
be completely disrupted during dissociation/association of
alkoxyamines derivatives, so that the polymer is allowed to
maintain its integrity and load bearing ability whenever the
cracked parts are repaired.
EXPERIMENTAL SECTION
■
Chemicals. Hydroxy-2,2,6,6-tetramethylpiperidinyloxy (98%) and
methacryloyl chloride (98%) were purchased from Aldrich and used as
received. Styrene (99%) was obtained from Alfa Aesar and purified by
distillation under reduced pressure in the presence of calcium hydride.
Linear PS (GPPS 685D) was provided by Dow Chemical. All the other
reagents and solvents were commercial products and used without
further purification.
Synthesis of 4-Hydroxy-1-((2′-hydroxy-1′-phenylethyl)oxy)-
2,2,6,6-tetramethylpiperidine (Diol). The diol is a derivative of
2,2,6,6-tetramethylpiperidinyloxy (TEMPO), which was prepared and
purified as previously reported.30 The product was characterized by
infrared and nuclear magnetic resonance (1H NMR and 13C NMR)
spectroscopy to confirm its molecular structure. FT-IR (KBr/cm−1):
1
3295, 2982, 1453, 1373, 1046. H NMR (CDCl3): δ/ppm 1.21 (3H,
CH3), 1.28 (3H, CH3), 1.31 (3H, CH3), 1.54 (3H, CH3), 1.31−1.98
(4H, 2CH2), 3.72 (1H, CH), 4.01 (1H, CH), 4.20 (1H, CH), 5.26
(1H,CH), 7.26−7.34 (5H, aromatic). 13C NMR (CDCl3): δ/ppm
21.65, 21.94, 33.38, 34.99, 49.08, 49.25, 61.22, 62.19, 63.31, 70.38,
84.46, 127.08, 128.21, 128.56, 138.91. Mass spectrum for C17H27NO3
[M+1]+ found 294.3.
Synthesis of 4-Methacryloyloxy-1-((2′-methacryloyloxy-1′-
phenylethyl)oxy)-2,2,6,6-tetramet-hylpiperidine (Dimetha-
crylic Ester). The synthesis of dimethacrylic ester followed the
published procedures.26 Triethylamine (5.1 g, 50 mmol) was slowly
5077
dx.doi.org/10.1021/cm202635w|Chem. Mater. 2011, 23, 5076−5081

Chemistry of Materials
Article
Scheme 2. Synthetic Route of Alkoxyamine-Based Cross-Linked Polystyrene
following the above procedure. Healing efficiency was calculated from
the ratio of the fracture toughness of the healed specimen over that of
the virgin specimen.31
each temperature the signal intensity is kept on certain level
within a fixed testing time. Therefore, the ESR spectra can be
used to describe radical concentration in the material, as
suggested by Sylvain et al.32 To highlight reversibility of the C-
ON bonds, Figure 2a gives the normalized absorptions
RESULTS AND DISCUSSION
■
To verify our consideration mentioned in the Introduction, we
synthesize cross-linked polystyrene (PS) in which alkoxyamine
moiety acts as the reversible cross-linker (Scheme 1b). First, the
reaction between diol and methacryloyl chloride produces
dimethacrylic ester (Scheme 2). Then cross-linked PS is yielded
via bulk polymerization of styrene with dimethacrylic ester as
cross-linking agent and benzoyl peroxide (BPO) as initiator,
respectively. Molecular structures of the related substances are
confirmed by infrared and nuclear magnetic resonance (1H
NMR and 13C NMR) spectroscopy (see Experimental Section),
respectively. The resultant cross-linked PS possesses similar
mechanical properties as conventional cross-linked PS with
ethyleneglycol dimethacrylate (EGDMA) as cross-linking
agent. When the molar feeding ratio of styrene over cross-
linker is fixed at 10/1, for example, flexural strength and
modulus of the former are 99.4 MPa and 5.0 GPa, whereas
those of the latter are 93.0 MPa and 5.2 GPa.
Electron spin resonance (ESR) spectroscopy measurement, a
technique for studying free radicals, is carried out at different
temperatures to examine fission behavior of the cross-linker
dimethacrylic ester in cross-linked PS. As shown in Figure 1,
Figure 2. (a) Normalized absorption lines obtained from ESR data of
reversibly cross-linked molar PS (with molar feeding ratio of styrene
over cross-linker dimethacrylic ester of 7.5/1), measured during
heating−cooling cycles between 20 and 130 °C. (b) Relative ESR
signal intensities calculated from integral area of the normalized
absorption lines.
recorded by cyclic ESR measurements between 20 and 130
°C. Accordingly, relative intensity of the radical signal estimated
from the area under each curve in Figure 2a is plotted in Figure
2b. It is seen that when temperature alternates between 20 and
130 °C, the amount of free radicals fluctuates as a result of
scission of alkoxyamine moiety and recombination of the
radicals. These processes are fully reversible, despite the fact
that both peak and valley values of the relative intensity slightly
increases with the thermal cycles. The latter phenomenon
might be attributed to permanent dissociation of a few
alkoxyamines caused by irreversible combination of methylene
radicals, which results in remainder of nitroxide radicals. This
can be detected by the temperature dependence of the
absorption at around 3247 G (Figure 2a). It peaks up at 130
Figure 1. Typical ESR spectra of reversibly cross-linked PS (with
molar feeding ratio of styrene over cross-linker dimethacrylic ester of
7.5/1) measured at various temperatures.
there is no evident absorption until 100 °C, and then the
relative ESR signal intensity drastically increases with a rise in
temperature, indicating that homolytic cleavage of alkoxyamine
moiety (C-ON bond) starts at about 125 °C. It is noted that at
5078
dx.doi.org/10.1021/cm202635w|Chem. Mater. 2011, 23, 5076−5081

Chemistry of Materials
Article
°C, but becomes a wide shoulder centered at 3258 G under 20
°C, manifesting that the radicals at elevated temperature consist
of two types of radicals (i.e., methylene and nitroxide radicals)
but only remainder of nitroxide radicals at room temperature
(Scheme 1).
Although radicals are detected as of 125 °C and the amount
of radicals increases with increasing temperature, the temper-
ature of homolytic cleavage cannot be unlimitedly raised. At
∼250 °C thermal degradation of linear PS derived from scission
of alkoxyamine moieties is triggered (see the Supporting
Information and Figure 3). Nevertheless, the reversibly cross-
linked PS keeps stable below 200 °C.
abnormality reflects dual role of the reversible cross-linkage in
the polymer. On one hand, the cross-links are in a dynamic
equilibrium state of random dissociation/association resulting
from the synchronous cleavage and recombination of the C-
ON bonds. The relative molecular weight between cross-links
has to rapidly change so that its effect on Tg is not statistically
definite enough as compared with the irreversibly cross-linked
PS. On the other hand, the cross-links still take the
responsibility of networking PS chains. Otherwise, complete
cleavage of cross-linked sites should result in transformation
from cross-linked PS to the linear version that has glass
transition at ∼100 °C. This argument is supported by the
dependence of area under tan δ peak on cross-linker content of
the materials. For either irreversibly or reversibly cross-linked
PS, the area under tan δ peak decreases with a rise in cross-
linker content, because higher cross-linking density leads to less
entanglement and friction between PS chains. Additionally, the
storage modulus curve of reversibly cross-linked specimen
shows an elastic plateau above Tg similar to irreversibly cross-
linked PS (Figure 4b), which strongly suggests the existence of
cross-linked network even at a temperature of as high as 200
°C. However, the lower storage modulus at temperatures above
Tg of reversibly cross-linked PS in comparison with irreversibly
cross-linked PS reveals that cross-link density of the former
should be lower than the later. This behavior is consistent with
the dynamic equilibrium feature of reversible cross-linkage.
A careful survey of Figure 4a indicates that the PS reversibly
cross-linked by dimethacrylic ester with molar feeding ratio of
styrene over cross-linker of 7.5/1 exhibits two maxima during
glass transition. In addition to the maximum corresponding to
Tg at ∼125 °C as mentioned above, another one appears at
about 105 °C, which is quite close to the Tg of linear PS. It
might be related with the heterogeneous network structure
derived from dynamic cross-linking/de-cross-linking processes.
That is, the maximum at 105 °C should originate from segment
motion of free linear PS chains generated by random
disconnection of the reversible cross-links. When the polymer
involves more reversible cross-linker (e.g., molar feeding ratio
of styrene over cross-linker = 5/1); however, lengths of the
disconnected chains produced by fission of alkoxyamine
moieties have to be rather short, which cannot afford to form
to a new damping peak. Accordingly, only single peak is
observed on the loss factor spectrum. As for the polymer with
less reversible cross-linker (e.g., molar feeding ratio of styrene
over cross-linker = 10/1), the cleft chains are long (or flexible)
enough to be immediately reconnected, so that the tan δ peak is
still unimodal.
Figure 3. Pyrolytic behaviors of linear PS, reversibly cross-linked PS,
and irreversibly cross-linked PS. The numerals in parentheses
represent molar feeding ratio of styrene over cross-linker.
To further ascertain dynamic feature of the reversibly cross-
linked PS, we compared its temperature dependences of loss
factor, measured by dynamic mechanical analysis (DMA), with
those of irreversibly cross-linked PS (with EGDMA as cross-
linker) in Figure 4a. For the latter polymers, their glass
The dynamic cross-linking behavior, or synchronous
reversibility, has become an intrinsic performance of the
dimethacrylic ester cross-linked PS, as demonstrated by the
multiple cyclic DMA scans within temperature range between
20 and 200 °C (Figure 5). The marginal increase of Tg with
testing cycles is probably due to appearance of small amount of
irreversible connection between methylene radicals. Never-
theless, most of the reversible links must still remain; otherwise,
their Tg should be significantly increased to the values of
irreversibly cross-linked PS. The subsequent multiple remend-
ability tests also support the argument. According to the results
of Figure 1, it is known that crack healing of the PS can be
conducted at a temperature higher than 125 °C, allowing
synchronous fission of the cross-linkages and radical recombi-
nation. Therefore, we set the healing temperature at 130 °C for
quantitative characterization of the ability of mechanical
Figure 4. Temperature dependences of (a) loss factor of reversibly
cross-linked PS and irreversibly cross-linked PS, and (b) storage
modulus of linear PS, reversibly cross-linked PS and irreversibly cross-
linked PS. The numerals in parentheses represent molar feeding ratio
of styrene over cross-linker.
transition temperatures, Tg, determined from tan δ peak
positions, significantly increase with the cross-linker content as
usual, because of the increase in cross-linking density. On the
contrary, the reversibly cross-linked PS shows similar Tg at
around 125 °C regardless of dimethacrylic ester content. This
5079
dx.doi.org/10.1021/cm202635w|Chem. Mater. 2011, 23, 5076−5081

Chemistry of Materials
Article
Figure 7. Temperature dependences of dimensional change for linear
PS, reversibly cross-linked PS and irreversibly cross-linked PS. The
numerals in parentheses represent molar feeding ratio of styrene over
cross-linker. The curves were measured by TMA.
Figure 5. DMA spectra of reversibly cross-linked PS measured by
repeated scans between 20 and 200 °C. Molar feeding ratio of styrene
over cross-linker dimethacrylic ester: (a) 5/1, (b) 7.5/1, and (c) 10/1.
material can stand against pressure at elevated temperature (see
the Supporting Information, Figure S1). These suggest that the
concept of synchronous covalent bond fission/radical recombi-
nation indeed works in developing self-healing polymer, which
is robust enough even when macromolecular chains are
temporarily cleft.
For the reversibly cross-linked PS with molar feeding ratio of
styrene over cross-linker of 5/1, low healing efficiency of 26.4%
is obtained, which seems to be contrary to expectation because
the polymer containing more alkoxyamine moieties. This can
be explained by the fact that the higher cross-linking density of
the polymer reduces range of motion of radicals. Consequently,
exchange reactions (scrambling) of alkoxyamine moieties
(Scheme 1b) across the crack interfaces are limited. In fact,
healing efficiency of the polymer can be increased to 70.1%
when healing temperature is raised to 145 °C. The increased
radical concentration and reduced restraint on range of motion
of radicals account for the improvement.
property restoration in terms of double cleavage drilled
compression (DCDC) tests (see the Experimental Section).
Figure 6 depicts typical crack length dependences of stress of
virgin and healed specimens measured by multiple tests under
With respect to the reversibly cross-linked PS with molar
feeding ratio of styrene over cross-linker of 10/1, the measured
healing efficiency (32.6%) is also low. It should be attributed to
the insufficient amount of radicals despite that range of motion
of the radicals is large enough as a result of lower cross-linked
density.
Considering that thermal treatment itself might contribute to
the healing effect, control DCDC tests were performed on
irreversibly cross-linked PS. No healing was detected when the
specimens with cracks were heated at a temperature 20 °C
higher than Tg of the polymer. It is thus confirmed that
chemical reconnection of fractured bonds on crack surfaces
dominates the crack healing.
Figure 6. Typical critical stress, σcrit, versus crack length, l/R, of
reversibly cross-linked PS specimen with molar feeding ratio of styrene
over cross-linker dimethacrylic ester of 7.5/1. The numerals in
parentheses represent healing efficiency. Crack healing temperature
and time: 130 °C, 2.5 h.
fixed healing time of 2.5 h. At molar feeding ratio of styrene
over cross-linker of 7.5/1, the specimens can be repeatedly
healed, and the efficiency of the first healing event is 75.9%.
The mild decline in healing efficiency with healing/refracture
cycles might originate from heterogeneity of dynamic
recombination of alkoxyamine moieties at the fractured surfaces
or reduction of reversible alkoxyamine moieties due to
permanent connection of methylene radicals. When the healing
time is prolonged to 6 and 9 h, the healing efficiencies are 75.3
and 75.7%, respectively. It means that 2.5 h is sufficient for
crack healing under 130 °C, as thermal equilibrium between
dissociation and association of alkoxyamine moieties is
reached.32
It is worth noting that the crack healing temperature (130
°C) is higher than Tg of the polymer, but the specimens do not
show creep distortion during healing. Thermal mechanical
analysis (TMA) and DMA data further manifest that at the
rubbery state the reversibly cross-linked PS has similar
deformation resistance as irreversibly cross-linked PS (Figure
7 and Figure 4b). Simple load bearing test also confirm the
CONCLUSIONS
■
The synchronous bond fission/radical recombination feature of
alkoxyamines has been successfully introduced into cross-linked
PS. It enables the polymer to repeatedly self-heal cracks without
deformation induced by chain scission, and provides the
polymer with mechanical stability at elevated temperature as
well. Moreover, the healing can be conducted within a wide
temperature range (i.e., between the onset temperatures of
homolysis and pyrolysis). These characteristics allow damaged
products made from the polymer to be unconsciously repaired
using the heat out of operating environment, where temper-
ature always has to be fluctuating in practice.
Because bond fission/radical recombination temperature of
alkoxyamines can be tuned by changing their molecular
5080
dx.doi.org/10.1021/cm202635w|Chem. Mater. 2011, 23, 5076−5081

Chemistry of Materials
Article
structures (e.g., by altering side groups),24 new self-healing
polymers with even wider healing temperature range or room
temperature healability might be produced accordingly. On the
other hand, because alkoxyamine moieties can be easily
incorporated into various monomers, a series of existing
polymers would be provided with self-healing property based
on C-ON bonds. When alkoxyamines are attached to the
monomers carrying double bonds, for example, all radical or
ionic chain polymerizations are allowed to proceed. If
alkoxyamines are linked with different functional groups (like
hydroxyl/carboxyl, or isocyanate/hydroxyl), their polymer-
izations could proceed via step polymerization. Additionally,
from the point of view of molecular structure, alkoxyamines are
able to take effect no matter whether they appear in linear or
cross-linked polymers. In this context, the family of self-healing
polymers would be remarkably expanded to meet actual
application requirements.
(17) Veregin, R. P. N.; Georges, M. K.; Hamer, G. K.; Kazmaier, P.
M. Macromolecules 1995, 28, 4391.
(18) Rowan, S. J.; Cantrill, S. J.; Cousin., G. R. L.; Sanders., J. K.M.;
Stoddart, J. F. Angew. Chem., Int. Ed. 2002, 114, 938.
(19) Rowan, S. J.; Cantrill, S. J.; Cousin, G. R. L.; Sanders, J. K.M.;
Stoddart, J. F. Angew. Chem., Int. Ed. 2002, 41, 898.
(20) Abdul Mannan, Md.; Ichikawa, A.; Miura, Y. Polymer 2007, 48,
743.
(21) Hawker, C. J.; Bosman, A. W.; Harth, E. Chem. Rev. 2001, 101,
3661.
(22) Otsuka, H.; Aotani, K.; Higaki, Y.; Takahara, A. J. Am. Chem.
Soc. 2003, 125, 4064.
(23) Georges, M. K.; Veregin, R. P. N.; Kazmaier, P. M.; Hamer, G.
K. Macromolecules 1993, 26, 2987.
(24) Miura, Y.; Nakamura, N.; Taniguchi, L. Macromolecules 2001,
34, 447.
(25) Wu, D. Y.; Meure, S.; Solomon, D. Prog. Polym. Sci. 2008, 33,
479.
(26) Higaki, Y.; Otsuka, H.; Takahara, A. Macromolecules 2004, 37,
1696.
ASSOCIATED CONTENT
(27) Amamoto, Y.; Higaki, Y.; Matsuda, Y.; Otsuka, H.; Takahara, A.
J. Am. Chem. Soc. 2007, 129, 13298.
(28) Yamaguchi, G.; Higaki, Y.; Otsuka, H.; Takahara, A.
Macromolecules 2005, 38, 6316.
(29) Higaki, Y.; Otsuka, H.; Takahara, A. Macromolecules 2006, 39,
2121.
(30) Hawker, C. J.; Barclay, G. G.; Dao, J. J. Am. Chem. Soc. 1996,
118, 11467.
(31) Brown, E. N.; Sottos, N. R.; White, S. R. Exp. Mech. 2002, 42,
372.
(32) Sylvain, M.; Christophe, L. M.; Paul, T.; Hanns, F.
Macromolecules 2000, 33, 4403.
■
S
* Supporting Information
Figure S1, described in the text. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: cesrmz@mail.sysu.edu.cn (M.Z.R.); ceszmq@mail.
sysu.edu.cn (M.Q.Z.).
ACKNOWLEDGMENTS
■
The authors thank the support of the Natural Science
Foundation of China (Grants 20874117, 50903095,
51073176 and U0634001), Doctoral Fund of Ministry of
Education of China (Grant 20090171110026), and the Science
and Technology Program of Guangdong Province (Grant
2010B010800021).
REFERENCES
■
(1) Mauldin, T. C.; Kessler, M. R. Int. Mater. Rev. 2010, 55, 317.
(2) Yuan, Y. C.; Yin, T.; Rong, M. Z.; Zhang, M. Q. Express Polym.
Lett. 2008, 2, 238.
(3) White, S. R.; Sottos, N. R.; Geubelle, P. H.; Moore, J. S.; Kessler,
M. R.; Sriram, S. R.; Brown, E. N.; Viswanathan, S. Nature 2001, 409,
794.
(4) Trask, R. S.; Bond, I. P. Smart Mater. Struct. 2006, 15, 704.
(5) Toohey, K. S.; Sottos, N. R.; Lewis, J. A.; Moore, J. S.; White, S.
R. Nat. Mater. 2007, 6, 581.
(6) Chen, X. X.; Dam, M. A.; Ono, K.; Mal, A.; Shen, H.; Nutt, S. R.;
Sheran, K.; Wudl, F. Science 2002, 295, 1698.
(7) Boul, P. J.; Reutenauer, P.; Lehn, J.-M. Org. Lett. 2005, 7, 15.
(8) Cordier, P.; Tournilhac, F.; Soulie-
2008, 451, 977.
́
Ziakovic, C.; Leibler, L. Nature
(9) Burattini, S.; Colquhoun, H. M.; Greenland, B. W.; Hayes, W.
Faraday Discuss. 2009, 143, 251.
(10) Kalista, S.; Ward, T. C.; Oyetunji, Z. Mech. Adv. Mater. Struct.
2007, 14, 391.
(11) Varley, R. J.; van der Zwaag, S. Polym. Int. 2010, 59, 1031.
(12) Skrzeszewska, P. J.; Sprakel, J.; A. de Wolf, F.; Fokkink, R.;
Stuart, M. A. C.; van der Gucht, J. Macromolecules 2010, 43, 3542.
(13) (a) Ghosh, B.; Urban, M. W. Science 2009, 323, 1458. (b) Ling,
J.; Rong, M. Z.; Zhang, M. Q. J. Mater. Chem. DOI:10.1039/
C1JM13467A.
(14) Plaisted, T. A.; Nemat-Nasser, S. Acta Mater. 2007, 55, 5684.
(15) Wool, R. P.; O’Connor, K. M. J. Appl. Phys. 1981, 52, 5953.
(16) Veregin, R. P. N.; Georges, M. K.; Kazmaier, P. M.; Hamer, G.
K. Macromolecules 1993, 26, 5316.
5081
dx.doi.org/10.1021/cm202635w|Chem. Mater. 2011, 23, 5076−5081