Triggered Structural Control of Dynamic Covalent Aromatic Polyamides: Effects of Thermal Reorganization Behavior in Solution and Solid States

Article abstract of DOI:10.1021/acs.macromol.5b01778

Thermally rearrangeable aromatic polyamides (TEMPO-PA) and random copolyamides (TEMPO-PA-COOH) incorporating alkoxyamine moieties in the main chain were synthesized, and the effects of thermal reorganization behavior on their solution and solid-state structures were investigated. The hydrodynamic radius in solution decreased as the solution temperature increased because of the dissociation of the alkoxyamine unit. Additionally, the dry density of the thin films decreased as the fabrication temperature increased because of the suppression of polymer aggregation caused by the thermally induced radical crossover reaction. In addition, at the film surface of the random copolyamide containing hydrophobic TEMPO and hydrophilic 3,5-diaminobenzoic acid (DABA) units, the hydrophilicity decreased as the fabrication temperature increased. This is because hydrophobic TEMPO and hydrophilic DABA units tend to be discretely aggregated near the film surface to minimize the surface energy and suppress the hydrogen bonding via a radical crossover reaction during the thin-film fabrication process. The present study clearly shows that both the solution structure and the solid-state molecular aggregation structure of the dynamic covalent polymers can be easily controlled by a thermal trigger, and it provides a new method for controlling the higher-order structure of polymer solutions and solids.

Full text of DOI:10.1021/acs.macromol.5b01778

Article
pubs.acs.org/Macromolecules
Triggered Structural Control of Dynamic Covalent Aromatic
Polyamides: Effects of Thermal Reorganization Behavior in Solution
and Solid States
Motohiro Aiba, Tomoya Higashihara, Minoru Ashizawa, Hideyuki Otsuka,
and Hidetoshi Matsumoto*
†
‡
†
†
,
†
†Department of Organic and Polymeric Materials, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro-ku, Tokyo 152-8552,
Japan
‡
Department of Polymer Science and Engineering, Faculty of Engineering, Yamagata University, 4-3-16 Jonan, Yonezawa City,
Yamagata 992-8510, Japan
*
S Supporting Information
ABSTRACT: Thermally rearrangeable aromatic polyamides (TEMPO-PA) and random copolyamides (TEMPO-PA-COOH)
incorporating alkoxyamine moieties in the main chain were synthesized, and the effects of thermal reorganization behavior on
their solution and solid-state structures were investigated. The hydrodynamic radius in solution decreased as the solution
temperature increased because of the dissociation of the alkoxyamine unit. Additionally, the dry density of the thin films
decreased as the fabrication temperature increased because of the suppression of polymer aggregation caused by the thermally
induced radical crossover reaction. In addition, at the film surface of the random copolyamide containing hydrophobic TEMPO
and hydrophilic 3,5-diaminobenzoic acid (DABA) units, the hydrophilicity decreased as the fabrication temperature increased.
This is because hydrophobic TEMPO and hydrophilic DABA units tend to be discretely aggregated near the film surface to
minimize the surface energy and suppress the hydrogen bonding via a radical crossover reaction during the thin-film fabrication
process. The present study clearly shows that both the solution structure and the solid-state molecular aggregation structure of
the dynamic covalent polymers can be easily controlled by a thermal trigger, and it provides a new method for controlling the
higher-order structure of polymer solutions and solids.
INTRODUCTION
reported that the molecular weight distribution of dynamic
covalent polymers reached the theoretical value of approx-
imately 2.0 through the reversible dissociation and association
of 2,2,6,6-tetramethylpiperidinyloxy (TEMPO) units in the
■
Organic polymers are very important materials for various
applications, and the development of new technologies requires
advanced polymeric materials exhibiting high performance and
critical properties. Although the polymer properties can be
tailored and enhanced by chemical modification and organic−
inorganic hybridization with existing polymers, the develop-
ment of novel polymers remains an attractive way to provide
valuable materials for advanced applications from the
7
main chain. Moreover, they revealed that the radical crossover
reaction between two different dynamic polymerspolyester
and polyurethaneinduced their hybridization via “polymer
8
scrambling” at the main chain level.
Among the various polymeric materials, wholly aromatic
polyamides (PAs), such as poly(m-phenyleneisophthalamide)
1
,2
perspective of both science and engineering.
Dynamic
covalent polymers have become some of most the attractive
materials because their primary structures and properties can be
controlled by external stimuli, even after the polymerization,
Received: August 11, 2015
Revised: February 21, 2016
3
−6
and they have been used in various fields.
Otsuka et al.
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 1. Chemical structures of TEMPO-PA and TEMPO-PA-COOH.
and poly(p-phenyleneterephthalamide), are widely used as
pump−thaw cycles. The microwave power was 0.5 mW, the field
modulation was 0.2 mT with a time constant of 0.03 s, and the number
of scans was 16. The relative intensities were calculated using the
integral areas in the spectra.
engineering plastics because of their high mechanical strength
9
,10
and flame resistance.
Their outstanding physical properties
and chemical stabilities arise from their rigid aromatic backbone
and secondary amide linkages, which result in extended rod-like
structures that interact with each other via hydrogen bonds and
strong π−π stacking interactions. However, disrupting the
hydrogen bonds between the secondary amide linkages by N-
alkylation or three-center hydrogen bonding to suppress the
hydrogen bonding affects both the thermal/mechanical proper-
The thermal analysis (TGA) was performed in a nitrogen
atmosphere using a RIGAKU Thermo plus EVO TG8120 thermal
analyzer (Akishima, Japan) for thermogravimetry (TG) and differential
thermal analysis (DTA). Before the evaluation, all of the samples were
preheated at 80 °C in the TG/DTA furnace to remove any moisture
and residual solvents. The samples were subsequently cooled to 30 °C
and then heated to 1000 °C at a heating rate of 10 °C/min. The
inherent viscosity was measured at 30 °C in DMSO at a polymer
11−14
ties and the packing mode of the polymer main chain.
−1
Moreover, control over the aggregation structures of PAs is also
critical for determining the physical properties in the solid state.
For example, the properties related to the transport of small
molecules through polymer thin films/membranes mostly
depend on the free volume. Therefore, it is important to
control the molecular aggregation structure to afford the
desired properties (e.g., mechanical properties, transparency,
concentration of 0.5 g L . Gel permeation chromatography (GPC)
was performed using a Viscotek GPC-1000 system (Malvern
Instruments, Malvern, UK) equipped with a TDA 302 triple detector
and a TSK-GEL α-M column with a conventional calibration curve
based on polystyrene standards. DMF with 0.05 M LiBr was used as
the carrier solvent at a flow rate of 0.6 mL/min. The number- and
weight-averaged molecular weights (M and M , respectively) were
n
w
calculated based on the light-scattering data.
15−18
and permselectivity).
To the best of our knowledge,
To evaluate the radical stability and thermal rearrangement behavior
in solution, 1H NMR and GPC measurements were performed,
respectively. TEMPO-PA (M = 46 400, M /M = 3.30) was dissolved
however, the effects of the thermal reorganization of dynamic
covalent polymers on their solution and solid-state structures
remain unclear.
n
w
n
−1
in DMSO-d
°
at a concentration of 6 mg mL and then heated at 130
C under an air or argon atmosphere. H NMR measurements were
6
1
Here, we report the synthesis of a new aromatic PA
conducted with the samples collected at 16 and 24 h. After dissolving
(
TEMPO-PA) with alkoxyamine moieties (TEMPO units) in
the TEMPO-PA (M = 46 400, M /M = 3.30) in DMF with 0.05 M
n
w
n
the main chain and random coPA (TEMPO-PA-COOH)
composed of a hydrophobic TEMPO unit and a hydrophilic
−
1
LiBr at a concentration of 2 mg mL , the polymer solution was
degassed by four freeze−pump−thaw cycles and then heated at 130 °C
under an argon atmosphere. GPC measurements were conducted for
the samples collected at 0.5, 1, 2, 4, 16, and 24 h.
3,5-diaminobenzoic acid (DABA) unit (see Figure 1). Addi-
tionally, the fundamental properties and effects of radical
crossover reactions on the polymer aggregation behavior were
investigated in detail. Furthermore, the effect of the thermal
reorganization process on the phase separation behavior of
TEMPO-PA-COOH was carefully examined by time-resolved
water contact angle measurements and atomic force micros-
copy (AFM).
Dynamic light-scattering (DLS) measurements were performed
using a Wyatt DynaPro NanoStar (Wyatt Technology, Santa Barbara,
CA). The hydrodynamic diameter (R ) and the diffusion coefficient
H
(D) of the prepared polymers in DMSO at a concentration of 2.0 wt %
were calculated based on a CONTIN analysis using DYNAMICS
software (Wyatt Technology, Santa Barbara, CA). To avoid reactions
between the alkoxyamine moieties and oxygen, the polymer solutions
were filtered through a 0.45 μm membrane filter after four freeze−
pump−thaw cycles and transferred to quartz cells in a glovebox. The
viscosities of DMSO at 70, 100, and 130 °C were determined to be
1.277, 0.855, and 0.408 mPa s, respectively, by the viscometer.
Additionally, the refractive indices were found to be 1.452, 1.431, and
1.392, respectively, by interferometry.
The polymer densities in the thin films were determined using an
Alfa Mirage SD-200L electronic densimeter (Osaka, Japan) at ambient
temperature (24−25 °C). Ethanol was selected as the nonsolvent.
The water contact angles of the thin films were measured by the
sessile drop method using a DropMaster 500 (Kyowa Interface
Science Co., Niiza, Japan) at ambient temperature (24−25 °C). A
purified water droplet with a volume of a 2.0 μL was used as the probe
liquid, and the contact angles were calculated via a contour curve-
fitting method using FAMAS software (Kyowa Interface Science Co.,
Niiza, Japan).
EXPERIMENTAL SECTION
■
Materials. TEMPO, styrene, Pd−C, 4-nitrobenzoyl chloride,
DABA, and isophthaloyl chloride were purchased from TCI (Tokyo,
Japan). Dehydrated pyridine, dehydrated tetrahydrofuran (THF), N-
methylpyrrolidone (NMP), N,N-dimethylformamide (DMF), and
dimethyl sulfoxide (DMSO) were obtained from Wako Pure Chemical
Industries (Osaka, Japan). NMP was dried over calcium hydride and
distilled under nitrogen. As a reference, a hydrophilic random
copolyamide (PA-20, Figure S1) without dynamic covalent bonding
18
was prepared according to a previous report.
Characterization. Proton nuclear magnetic resonance ( H NMR)
1
spectra were recorded on a Bruker DPX300S spectrometer (Karlsruhe,
Germany) in DMSO-d calibrated to tetramethylsilane as an internal
6
standard (δH 0.00). Fourier transform infrared (FT-IR) spectra were
measured on a Horiba FT-720 spectrometer (Kyoto, Japan). Electron
spin resonance (ESR) measurements were performed using a JEOL
JES-X320 T-band ESR spectrometer (Akishima, Japan) equipped with
a JEOL DVT temperature controller. The samples were measured in
DMSO at a polymer concentration of 10 wt % in 0.8 mm glass
capillaries, which were sealed after being degassed by four freeze−
The surface morphology was characterized using AFM (SPA400, SII
Nanotechnology, Tokyo, Japan) in phase-contrast mode.
Synthesis of the TEMPO-Based Diol Compound (TEMPO-
OH). The diol was successfully prepared according to a previous
6
report. The obtained TEMPO-OH was fully characterized by IR and
B
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 1. Synthesis of (a) TEMPO-PA and (b) TEMPO-PA-COOH
1
NMR spectroscopy. H NMR (300 MHz, CDCl , δ, 25 °C): 1.21 (s,
(C−H stretching), 1693 (−CO− stretching), 1623 (N−H
bending), 1602 (−CC− stretching), 1517 (−CC− stretching).
Synthesis of PA Containing the TEMPO Unit (TEMPO-PA).
3
−
CH , 3H), 1.28 (s, CH , 3H), 1.31 (s, −CH , 3H), 1.54 (s, −CH ,
3
3
3
3
3
4
H), 1.31−1.98 (m, −CH − 4H), 3.72 (dd, −CH−, J = 11.1 Hz, 1H),
2
.01 (dd, −CH−, J = 11.1 Hz, 1H), 4.20 (t, −CH−, 6.0 Hz, 1H), 5.26
TEMPO-NH (532 mg, 1.0 mmol) was dissolved in dried NMP (3.0
2
−1
(
(
m, −CH−, 1H), 7.26−7.34 (m, ArH, 5H). IR (film), ν (cm ): 3295
mL) at rt, and the solution was cooled to −78 °C. Isophthaloyl
chloride (203 mg, 1.0 mmol) was subsequently added, and the
temperature was slowly raised to rt. The solution was then stirred for 3
h and precipitated with methanol to obtain a white fibrous polymer.
The obtained polymer was filtered off, washed several times with
methanol, and dried in a vacuum oven for 48 h at 40 °C. Yield: 615.4
O−H stretching), 3111−2855 (C−H stretching), 1456 (CC
stretching).
Synthesis of the TEMPO-Based Dinitro Compound (TEMPO-
NO ). 4-Nitrobenzoyl chloride (6.96 g, 37.5 mmol), TEMPO-OH
3.52 g, 12.5 mmol), and dehydrated THF (50 mL) were added to a
two-necked 100 mL flask under a nitrogen atmosphere. After complete
dissolution, dehydrated pyridine (3.96 g, 50.0 mmol) was injected into
the solution. After the solution was stirred at room temperature (rt)
for 12 h, a white precipitate was removed by filtration, and the filtrate
was evaporated. The crude product was purified by silica gel
chromatography using CHCl3 to give TEMPO-NO2 as a white
powder. Yield: 6.20 g (84%). H NMR (300 MHz, CDCl , δ, 25 °C):
2
(
1
mg (93%). M = 46 400, M /M = 3.30. H NMR (300 MHz, DMSO-
n
w
n
d , δ, 40 °C): 0.73 (s, 3H), 1.29 (s, −CH , 3H), 1.40 (s, −CH , 3H),
6
3
3
1.52 (s, −CH , 3H), 1.73−2.01 (m, −CH −, 4H), 4.51 (dd, −CH −, J
3
2
2
= 12.2 Hz, 1H), 4.76 (dd, −CH −, J = 12.2 Hz, 1H), 5.11−5.16 (m,
2
−CH−, 2H), 7.27−7.46 (m, ArH, 5H), 7.71 (t, ArH, J = 7.5 Hz, 1H),
7.86−7.95 (m, ArH, 8H), 8.16 (d, ArH, J = 7.5 Hz, 5H), 8.57 (s, ArH,
1
13
3
1H), 10.71 (s, −NH−, 1H). C NMR (75 MHz, DMSO-d , δ, 40
6
0
.79 (s, −CH , 3H), 1.23 (s, CH , 3H), 1.36 (s, −CH , 3H), 1.45 (s,
°C): 17.08, 20.70, 28.87, 29.96, 33.2, 44.18, 48.41, 60.22, 65.80, 66.33,
83.63, 119.70, 124.41, 124.85, 127.28, 127.67, 127.89, 128.13, 128.80,
130.08, 131.18, 134.79, 139.93, 143.62, 143.70, 164.93 (CO),
165.10 (CO), 165.46 (CO), 165.50 (CO). IR (KBr), ν
3
3
3
−
CH , 3H), 1.67−2.07 (m, −CH − 4H), 4.59 (dd, −CH −, J = 11.1
3
2
2
Hz, 1H), 4.90 (dd, −CH −, J = 11.1 Hz, 1H), 5.10 (t, −CH−, J = 6.0
2
Hz, 1H), 5.24−5.33 (m, −CH−, 1H), 7.27−7.38 (m, ArH, 5H), 8.06−
.30 (m, ArH, 8H). 13C NMR (75 MHz, CDCl , δ, 25 °C): 21.16,
−1
8
2
1
1
(cm ): 3295 (N−H stretching), 2973−2872 (C−H stretching), 1716
3
1.27, 34.01, 34.13, 44.61, 44.67, 60.68, 60.79, 67.30, 68.55, 84.22,
23.74, 123.77, 127.79, 128. 37, 128.54, 130.84, 130.85, 135.61,
36.06, 139.92, 139.96, 150.78, 164.47 (CO), 164.65 (CO). IR
(−CO− stretching), 1671 (−CO− stretching), 1601 (−CC−
stretching), 1527 (−N−H bending).
Synthesis of a Random Copolyamide (CoPA) Containing
−1
(
KBr), ν (cm ): 3111−2855 (C−H stretching), 1711 (−CO−
TEMPO and DABA Units (TEMPO-PA-COOH). TEMPO-NH (2.35
2
stretching), 1607 (−CC− stretching), 1526 (−NO asymmetric
g, 4.4 mmol) and DABA (669 mg, 4.4 mmol) were dissolved in dried
NMP (18.7 mL) at rt. After the solution was cooled to −78 °C,
isophthaloyl chloride (1.79 g, 8.8 mmol) was added. The temperature
was slowly raised to rt, and then the solution was stirred for 3 h. A
viscous reaction solution was subsequently precipitated with methanol
to give a white fibrous polymer. The obtained polymer was filtered off,
2
stretching), 1349 (−NO symmetric stretching).
2
Synthesis of the TEMPO-Based Diamine Compound
(
TEMPO-NH ). Under a hydrogen atmosphere, TEMPO-NO (4.43
2
2
g, 7.5 mmol), Pd−C (887 mg, 20 wt %), and ethyl acetate (300 mL)
were injected into a two-necked 500 mL flask. After stirring at rt for 24
h, the solution was filtered and evaporated to give the crude product.
Then, the crude product was purified by silica gel chromatography
using ethyl acetate/hexane (1/5, v/v) and dried under vacuum at 70
C for 24 h. Yield: 3.50 g (88%). H NMR (300 MHz, CDCl , δ, 25
washed several times with methanol, and dried in a vacuum oven for
1
48 h at 40 °C. Yield: 4.03 g (97%). H NMR (300 MHz, DMSO-d , δ,
6
40 °C): 0.70 (s, 3H), 1.13 (s, −CH , 3H), 1.26 (s, −CH , 3H), 1.37 (s,
3
3
1
°
°
−CH , 3H), 1.53−1.98 (m, −CH −, 4H), 4.48 (dd, −CH −, J = 12.2
3
3
2
2
C): 0.76 (s, 3H), 1.21 (s, −CH , 3H), 1.35 (s, −CH , 3H), 1.44 (s,
Hz, 1H), 4.74 (dd, −CH −, J = 12.2 Hz, 1H), 5.07−5.13 (m, −CH−,
3
3
2
−
(
CH , 3H), 1.61−2.01 (m, −CH −, 4H), 4.04 (s, −NH , 4H), 4.47
2H), 7.26−7.43 (m, ArH, 5H), 7.67 (t, ArH, J = 7.5 Hz, 2H), 7.83−
7.92 (m, ArH, 8H), 8.16 (s, ArH, 6H), 8.52−8.68 (m, ArH, 3H),
3
2
2
dd, −CH −, J = 12.0 Hz, 1H), 4.90 (dd, −CH −, J = 11.4 Hz, 1H),
2
2
13
5
.06 (t, −CH−, J = 6.0 Hz, 1H), 5.16−5.25 (m, −CH−, 1H), 6.63 (t,
10.60. 10.66 (s, −NH−, 4H). C NMR (75 MHz, DMSO-d , δ, 40
6
ArH, J = 7.5 Hz, 4H), 7.34−7.39 (m, ArH, 5H), 7.75−7.84 (m, 4H).
°C): 20.86, 33.51, 44.27, 44.36, 60.02, 60.26, 65.93, 66.04, 66.48,
66.53, 83.64, 116.47, 116.99, 119.81, 124.50, 124.93, 127.40, 127.82,
128.03, 128.28, 128.91, 130.28, 131.12, 131.15, 131.29, 131.61, 134.98,
135.06, 139.69, 139.72, 140.16, 143.70, 143.72, 143.78, 143.80, 165.07
(CO), 165.23 (CO), 165.33 (CO), 165.46 (CO), 165.63
(CO), 165.69 (CO), 165.75 (CO), 167.25 (CO). IR (KBr),
1
3
C NMR (75 MHz, CDCl , δ, 25 °C): 21.30, 21.34, 34.14, 34.22,
3
4
1
5.01, 45.12, 60.66, 60.91, 66.32, 66.63, 84.55, 114.04, 119.95, 120.42,
28. 05, 128.37, 131.84, 131.96, 140.86, 151.11, 151.14, 166.58 (C
−1
O), 166.72 (CO). IR (KBr), ν (cm ): 3475 (N−H asymmetric
stretching), 3369 (N−H symmetric stretching), 3227, 2973−2939
C
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
Figure 2. H NMR spectra of (a) TEMPO-PA and (b) TEMPO-PA-COOH in DMSO-d at 40 °C.
6
−1
ν (cm ): 3295 (N−H stretching), 2973−2872 (C−H stretching),
719 (−CO− stretching), 1684 (−CO− stretching), 1674
−CO− stretching), 1597 (−CC− stretching), 1526 (−N−H
bending).
Fabrication of TEMPO-PA and TEMPO-PA-COOH Films. The
The ESR spectroscopy measurements of TEMPO-PA were
first performed at different temperatures to determine the
homolytic cleavage behavior of the C−ON bonds in the
alkoxyamine moieties. Figure 3 shows the ESR spectra of
1
(
prepared TEMPO-PA and TEMPO-PA-COOH were dissolved in
DMSO to obtain 10 wt % solutions. The polymer solutions were
subsequently degassed by freeze−pump−thaw cycles and filtered
through a 0.45 μm poly(vinylidene fluoride) membrane filter in a
glovebox. The polymer solutions were coated onto clean glass
substrates by a spin-coating method to generate the as-cast films with
thicknesses of approximately 2.0 μm. The as-cast films were then
heated to various temperatures (70, 100, and 130 °C) for 3 h and
dried under vacuum for at least 8 h before the film properties were
characterized.
RESULTS AND DISCUSSION
■
Figure 3. ESR spectra of TEMPO-PA measured at different
temperatures (rt, 70, 100, and 130 °C) in DMSO.
Preparation, Cleavage Behavior, and Thermal Proper-
ties of TEMPO-PA and TEMPO-PA-COOH. The diamine
monomer, TEMPO-NH , was prepared in four steps according
2
TEMPO-PA at different temperatures, and the relative
intensities of the radical signal estimated from the area are
summarized in Table 1. As the temperature increased from rt to
1
to Scheme 1, and its molecular structure was confirmed by H
NMR, 13C NMR, and FTIR spectra (see the Experimental
Section). TEMPO-PA was subsequently synthesized by the
low-temperature polycondensation of isophthaloyl chloride
Table 1. Relative ESR Intensities Calculated from the
Integral Areas in the Spectra Shown in Figure 3
with TEMPO-NH . In addition, TEMPO-PA-COOH, which
2
was composed of TEMPO and DABA units, was synthesized by
the polycondensation reaction between TEMPO-NH , DABA,
2
temperature [°C]
25
70
100
130
and isophthaloyl chloride. The chemical structures of the
prepared TEMPO-PA and TEMPO-PA-COOH were fully
characterized by 1H NMR, C NMR, and FTIR spectra
intensity [au]
1.00
2.76
4.77
26.3
13
(
Figure 2 and Figure S2). For TEMPO-PA, the characteristic
130 °C, the ESR signal intensity also increased, indicating that
the equilibrium of the alkoxyamine moiety shifted toward the
dissociated state. Although the relative intensities gradually
increased from 25 to 100 °C, a drastic increase from 100 to 130
°C can be observed because the rate constant of the
dissociation reaction of the alkoxyamine is expressed in terms
of the Arrhenius equation. Additionally, the equilibrium
constant of the alkoxyamine moiety at 130 °C was
approximately 5 times higher than that at 100 °C, as revealed
by comparing the intensities at each temperature.
The thermal stabilities of the TEMPO-PA and TEMPO-PA-
COOH were subsequently evaluated under a nitrogen
atmosphere by TGA. As seen in Figure 4, TEMPO-PA and
TEMPO-PA-COOH exhibit the following two-step weight-loss
behavior: The region from 250 to 400 °C corresponds to the
decomposition of the ester group, carboxylic acid group, and
alkoxyamine moiety (Figure S3, detailed discussions are
included in the Supporting Information), whereas the region
from 500 to 700 °C corresponds to the decomposition of the
peak from the amide linkage is observed at 10.71 ppm (denoted
as A) in the H NMR spectrum, and the integral values for the
1
peaks are consistent with the expected structure. Although
different ordered structures (head to head/head to tail) could
be formed by the polycondensation because of the asymmetric
structure of TEMPO-NH , the splitting pattern of the
2
isophthaloyl chloride moiety (labeled as R, S, and T) was not
1
observed in the H NMR spectra. For TEMPO-PA-COOH, the
characteristic peaks of the amide proton (labeled as U) and the
aromatic protons corresponding to the DABA unit (labeled as
V, W, X, Y, and Z) were confirmed. TEMPO-PA and TEMPO-
PA-COOH exhibited good solubilities in polar solvents, such as
DMF, DMSO, DMAc, and NMP. The molecular weights of
TEMPO-PA were estimated to be M = 46 400 and M =
n
w
1
52 900 (M /M = 3.30) by GPC. For TEMPO-PA-COOH,
w n
the inherent viscosity (η ) measured at a concentration of 0.5
g dL in DMSO at 30 °C was calculated to be 0.91, which is
consistent with a high-molecular-weight polymer.
inh
−1
D
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
reaction, the M and M values became higher and lower than
n
w
the original values, probably because the change in the
molecular distribution from broad to narrow led to the
formation of the middle-weight polymer. Indeed, Figure 5a
shows that the molecular distribution became narrower as the
peak tops shifted from higher to lower molecular weights. As
seen in Figure 5b, the M , M , and M /M values seemed to
n
w
w
n
achieve equilibrium after 1−24 h.
For TEMPO-PA-COOH, the TEMPO unit sequence in the
polymer main chain can be redistributed, and it is likely
possible to change the alignment of the TEMPO and DABA
units along the polymer chain after the thermal reorganization
reaction of TEMPO-PA-COOH. In general, the sequence of
the random polymer can be identified by the peaks of the
Figure 4. TG curves of TEMPO-PA and TEMPO-PA-COOH.
polymer main chain. Although the reversible reaction of the
alkoxyamine moiety was confirmed beginning at 70 °C, the 5%
weight-loss temperatures (Td5%) for TEMPO-PA and TEMPO-
PA-COOH were determined to be 259 and 275 °C,
respectively.
Radical Stabilities and Thermal Reorganization Be-
haviors of TEMPO-PA and TEMPO-PA-COOH. To
investigate the stability of the alkoxyamine moiety, TEMPO-
1
3
carbon nuclei in the amide linkages in the C NMR
2
0
spectrum. However, the sequential rearrangement of
1
3
TEMPO-PA-COOH could not be determined by the
C
NMR spectrum because no changes in any of the amide peaks
were observed after heating TEMPO-PA-COOH in DMSO-d
6
at a concentration of 100 mg mL− for 2 h (Figure S5). This
observation can probably be attributed to the reversible
reaction of the nitroxide with the styryl radical being limited
to the alkoxyamine moieties and a TEMPO unit always existing
next to a DABA unit.
1
−1
PA solutions in DMSO-d at a concentration of 6 mg mL
6
were heated at 130 °C under an air or argon atmosphere and
1
monitored by H NMR (Figure S4). Under the argon
Solution Structures of TEMPO-PA and TEMPO-PA-
COOH. The chemical shift provides information about
hydrogen bonding behavior. Generally, a shift of the amide
proton to a lower magnetic field occurs when hydrogen
atmosphere, no other peaks were seen, even after 24 h,
indicating that no side reactions in addition to the
recombination of nitroxide with the styryl radicals occurred.
In contrast, obvious side reactions occurred after 16 h under
the air atmosphere, as revealed by the emergence of
characteristic peaks other than those of TEMPO-PA. As
reported by Matyjaszewski et al., this side reaction involved the
irreversible reaction of the styryl radical with oxygen and the
21
bonding becomes strong. Thus, the chemical shifts of the
peaks assigned to the protons in the amide linkages of
TEMPO-PA and TEMPO-PA-COOH at different temperatures
were measured to evaluate the hydrogen bonding behavior.
1
9
1
following oxidation of nitroxide by peroxyl radicals. There-
fore, TEMPO-PA and TEMPO-PA-COOH should be treated
under an argon atmosphere.
Figure 6 shows the H NMR spectra of TEMPO-PA and
TEMPO-PA-COOH at different temperatures. The chemical
shifts of the amide protons move to lower magnetic field as the
temperature increases, indicating that the hydrogen bonding
between the amide linkages weakens as the temperature rises.
The DLS measurements of TEMPO-PA and TEMPO-PA-
COOH in DMSO at a concentration of 2.0 wt % were
subsequently performed to evaluate the polymer solution
structures at different temperatures. The correlation functions
Furthermore, to evaluate the effects of the thermal
reorganization behavior of the alkoxyamine moieties on Mn,
M , and M /M , TEMPO-PA (M = 46 400, M /M = 3.30)
w
w
n
n
w
n
was heated at 130 °C in DMF solution under an argon
atmosphere. No noticeable color change was observed during
the reaction, although the nitroxide radical appears red. As
reported by Otsuka et al., the M /Mn value reached
are shown in Figure S6. The hydrodynamic diameter, R , and
w
H
approximately 2.0, which is the theoretically expected value of
the step-growth polymerization within 1 h, and no additional
changes were observed after 1 h because the reaction reached
diffusion coefficient, D, were assessed by a CONTIN analysis
and are summarized in Table 2. These values were calculated
assuming that the polymer possesses the shape of a hard sphere.
The RH values of TEMPO-PA and TEMPO-PA-COOH
7
equilibrium. In contrast, after the thermal rearrangement
Figure 5. (a) GPC RI curves for TEMPO-PAs with heat treatment in DMF at 130 °C for 0, 0.5, and 24 h. (b) Plots of M and M /M as a function
n
w
n
of time for TEMPO-PAs with heat treatment in DMF at 130 °C for 0, 0.5, and 24 h.
E
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
Figure 6. H NMR spectra of (A) TEMPO-PA and (B) TEMPO-PA-COOH at (a) 25 °C, (b) 70 °C, (c) 100 °C, and (d) 130 °C in DMSO-d . The
6
peaks with asterisks are attributed to the amide protons.
Table 2. Hydrodynamic Diameters and Diffusion Coefficients of TEMPO-PA and TEMPO-PA-COOH at Different
Temperatures
TEMPO-PA
TEMPO-PA-COOH
temperature [°C]
25
70
100
4.13
9.68
130
1.83
53.6
25
70
100
3.10
10.3
130
1.35
53.6
a
RH [nm]
10.15
1.04
7.62
2.91
5.93
1.77
4.46
4.33
a
[10−8 m s−1]
2
D
a
The hydrodynamic diameters and diffusion coefficients were calculated using the Einstein−Stokes equation.
Figure 7. FTIR spectra of (a) TEMPO-PA and (b) TEMPO-PA-COOH films fabricated at different temperatures. The dotted lines (a, b, c, and d)
−1
indicate 1682, 1536, 1672, and 1534 cm , respectively.
decreased as the solution temperature increased, as did not the
D values. This can be explained by the fact that the molecular
weights of both TEMPO-PA and TEMPO-PA-COOH in
solution at high temperature are smaller than those at low
temperature, mainly because a large amount of the alkoxyamine
moieties in the main chain dissociate at high temperature (as
shown in Figure 3 and Table 1). Although suppressing the
hydrogen bonding would increase the RH value at high
temperatures, the contribution of the molecular weight is
more substantial than that of hydrogen bonding in solution.
These results clearly show that the solution structure of
dynamic covalent polymers can be controlled by the temper-
ature-induced reorganization behavior.
Solid-State Structures of TEMPO-PA and TEMPO-PA-
COOH. To investigate the effect of the radical crossover
reaction on the molecular aggregation structures, TEMPO-PA
and TEMPO-PA-COOH thin films were prepared at the three
different temperatures: 70, 100, and 130 °C. Because the
unexpected reaction of the alkoxyamine moieties occurred
readily in the presence of oxygen, the thin-film fabrication
process was conducted in a glovebox after the polymer
solutions were degassed by four freeze−pump−thaw cycles.
Figure 7 shows the FTIR spectra of TEMPO-PA and
TEMPO-PA-COOH used to evaluate the hydrogen bonding
behavior in the thin films. Generally, two characteristic peaks of
the amide linkage, i.e., the CO stretching and the N−H in-
plane bending, are shifted to higher and lower wavenumbers,
respectively, as the hydrogen bonding in the amide linkage
22
weakens. For TEMPO-PA, the peak corresponding to CO
−1
stretching shifts from 1682 to 1685 cm as the thin-film
fabrication temperature increases, whereas that of N−H in-
−1
plane bending shifts from 1536 to 1530 cm . This result
indicates that the hydrogen bonds between the amide linkages
are suppressed when the thin film is fabricated at high
temperatures. For TEMPO-PA-COOH, similar shifts of the
peaks attributed to the CO stretching and N−H in-plane
bending are also confirmed. The dry densities were
subsequently measured to estimate the molecular aggregation
15
structure and are summarized in Figure 8. As the thin-film
fabrication temperature increased, the dry densities of both
TEMPO-PA and TEMPO-PA-COOH clearly decreased from
1
.248 to 1.229 and from 1.301 to 1.282, respectively, indicating
that molecular aggregation strcture of the polymer thin film can
be controlled by the fabrication temperature. This result is in
good agreement with the FTIR spectra and implies that the
F
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
chemical structure. The water contact angles of the TEMPO-
PA and TEMPO-PA-COOH films fabricated at 70 °C
decreased linearly with increasing time. In contrast, two regions
(
an initial exponential decay and a subsequent linear decrease)
can be seen in the TEMPO-PA-COOH films fabricated at 100
and 130 °C. According to previous reports,
24,25
the initial
contact angle decay and subsequent linear decrease reflect the
surface separation induced by the water droplet and the change
in the droplet shape attributable to water evaporation,
respectively. As the initial decay cannot be seen in any
TEMPO-PA film, the separation likely arises not only from the
amide and ester linkages in the TEMPO unit but also from the
carboxylic acid group in the DABA unit. The experimental plot
for the time course of the water contact angles was fitted by the
26
equation
⎛
t ⎞
Figure 8. Dry densities of TEMPO-PA and TEMPO-PA-COOH films
fabricated at different temperatures (70, 100, and 130 °C).
⎜
⎟
θ(t) = (θini
−
− kt + θter
ter
⎠
(1)
τ
where θini and θter are the initial and terminal values of the
contact angles at t = 0 and in a quasi-equilibrium state,
respectively; τ is the time constant of the contact angle decay in
the initial stage; and k is the constant related to water
evaporation. It should be noted that τ is correlated with the
relaxation time of the interfacial mobility, and the difference in
the θini and θter values indicates the degree of surface
disruption of the polymer packing mainly occurs because of the
suppression of hydrogen bonding. To identify the effects of
thermal reorganization behavior on the hydrogen bonding and
dry density, the FTIR spectra and dry densities of PA-20
without the alkoxyamine unit were measured. Despite a change
in the membrane fabrication temperature, no differences in the
FTIR spectra or dry densities were noted (Figures S7 and S8),
indicating that the suppression of hydrogen bonding and
polymer aggregation behaviors are caused by the radical
crossover reaction. These results clearly show that the
aggregation structures of solid-state dynamic covalent polymers
can be controlled by temperature-induced reorganization
behavior during the fabrication process. At high temperatures,
the contribution of the suppression of hydrogen bonding might
become substantial in the coated solution by decreasing the
molecular weight and increasing the mobility of polymer chain.
Surface Structures of TEMPO-PA and TEMPO-PA-
COOH. The surface reorganization of the TEMPO-PA and
TEMPO-PA-COOH films induced by liquid molecules was
investigated by water contact angle measurements. Figure 9
24−26
reorganization induced by water.
As the film fabrication
temperature increased, the τ value decreased from 6.77 to 5.87.
Based on the FTIR spectra and dry densities, the decrease in
the τ value is probably caused by the suppression of the
polymer chain aggregation. Meanwhile, the difference in the θini
and θter values increased from 2.2 to 3.1. Additionally, the
contact angles of the TEMPO-PA-COOH films increased with
increasing film fabrication temperature, whereas all TEMPO-
PA films showed the same contact angles. This observation
implies that more hydrophobic domains are formed in
TEMPO-PA-COOH, and simultaneously, the number of
DABA units involved in hydrophilic domains increases as the
film fabrication temperature increases. This interfacial behavior
will be discussed in detail after presenting the tapping mode
phase image obtained by AFM.
To clarify the phase separation between the hydrophilic and
hydrophobic domains, phase images of the surfaces of the
TEMPO-PA-COOH films were recorded with AFM in tapping
mode under ambient conditions (Figure 10). Although all of
the samples are random copolymers, nanoscaled phase
separations are clearly observed. The bright and dark regions
in the images are derived from the hard segments
corresponding to the TEMPO units and the soft segments
corresponding to the DABA units (Figure 10 (top)). The
threshold method was implemented to obtain binarized images,
as shown in Figure 10 (bottom), where the blue and red
regions are derived from hydrophilic and hydrophobic
segments corresponding to the DABA and TEMPO units,
respectively. The area fractions of the hydrophilic domains of
TEMPO-PA-COOH are summarized in Table 3. As seen in
Figure 10 (bottom) and Table 3, as the film fabrication
temperature increases, both the size of the hydrophilic domain
and the total area of the hydrophobic domains also increase.
Thus, both the DABA and TEMPO units tend to be discretely
aggregated near the film surface. The hydrophobic TEMPO
units are likely to be oriented toward the air to minimize the
surface free energy, which is induced by the thermal
reorganization behavior during the film fabrication process. In
Figure 9. Time courses of the water contact angles for (a) TEMPO-
PA and (b) TEMPO-PA-COOH films fabricated at different
temperatures (70, 100, and 130 °C) and (c) PA-COOH. The solid
curves are fitted by eq 1.
presents the time course of the water contact angles for the
TEMPO-PA and TEMPO-PA-COOH films fabricated at
different temperatures. Because the value of the contact angle
value is partly influenced by the surface roughness, the root-
mean-square roughness, RRMS, values of TEMPO-PA and
23
TEMPO-PA-COOH were determined by AFM. The RRMS
values for all of the films ranged from 0.2 to 0.4 nm, indicating
that the contact angle values mainly depend on the surface
G
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 10. AFM tapping mode phase images (above) and their binarized images obtained using the threshold method (bottom) for TEMPO-PA-
2
COOH films fabricated at (a) 70, (b) 100, and (c) 130 °C. Scan sizes are 250 × 250 nm .
Table 3. Area Fractions of the Hydrophobic Domains of
TEMPO-PA-COOH Films Fabricated at Different
Temperatures (70, 100, and 130 °C) Estimated by the
Threshold Method
the solution temperature increased because of the dissociation
of the alkoxyamine unit. Additionally, the dry density of the
thin films decreased as the fabrication temperature increased
because of the suppression of the polymer aggregation caused
by the radical crossover reaction. These results clearly show
that both the solution structure and the solid-state molecular
aggregation structure of the dynamic covalent polymers can be
easily controlled by a thermal trigger. In addition, the surface
molecular aggregation and phase separation behaviors of
dynamic covalent TEMPO-PA-COOH can be readily con-
trolled by applying thermal treatments during the film
fabrication. This is because hydrophobic TEMPO and
hydrophilic DABA units tend to be discretely aggregated near
the film surface to minimize the surface energy and suppress
hydrogen bonding via radical crossover reactions during the
thin-film fabrication process. These findings should provide a
new method for controlling the higher-order structure of
polymer solutions and solids.
temperature [°C]
70
100
130
area fraction of hydrophobic domains [%]
48.9
53.5
57.9
addition, the suppression of the hydrogen bonding in the casted
solution should also facilitate the separation of the hydrophobic
and hydrophilic domains. These results are in good agreement
with the difference between θ and θ of TEMPO-PA-COOH
ini
ter
and are probably related to the hydrophilic domain size. For the
TEMPO-PA-COOH film fabricated at 70 °C, no surface
reorganization was observed because the small hydrophilic
domains are surrounded and suppressed by the large number of
hydrophobic domains. Comparing the differences in the θ and
ini
θter values of the TEMPO-PA-COOH films fabricated at 100
and 130 °C revealed that increasing the size of the hydrophilic
domains helps to increase the degree of surface reorganization
induced by water. These results clearly show that the surface
molecular aggregation and phase separation behaviors of
dynamic covalent TEMPO-PA-COOH can also be easily
controlled by thermal treatments during film fabrication.
ASSOCIATED CONTENT
■
*
S
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.macro-
mol.5b01778.
13C NMR spectra of TEMPO-PA and TEMPO-PA-
COOH, FT-IR spectra of TEMPO-PA and TEMPO-PA-
COOH before and after heating, radical stability of
CONCLUSION
■
In the present study, TEMPO-PA and TEMPO-PA-COOH
were successfully synthesized by low-temperature polyconden-
sation. The fundamental properties and effects of the radical
crossover reaction of the prepared dynamic covalent polymers
on the solution, solid-state, and solid surface structures were
investigated. The hydrodynamic radius in solution decreased as
1
TEMPO-PA analyzed by H NMR spectra, sequential
1
3
rearrangement of TEMPO-PA-COOH monitored by C
NMR spectra, and correlation functions of TEMPO-PA
and TEMPO-PA-COOH (PDF)
H
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(
22) Skrovanek, D. J.; Howe, S. E.; Painter, P. C.; Coleman, M. M.
AUTHOR INFORMATION
■
*
Macromolecules 1985, 18, 1676−1683.
23) Cassie, A. B. D.; Baxter, S. Trans. Faraday Soc. 1944, 40, 546−
551.
Corresponding Author
E-mail matsumoto.h.ac@m.titech.ac.jp (H.M.).
(
(
24) Horinouchi, A.; Tanaka, K. RSC Adv. 2013, 3, 9446−9452.
Notes
(25) Oda, Y.; Horinouchi, A.; Kawaguchi, D.; Matsuno, H.; Kanaoka,
The authors declare no competing financial interest.
S.; Aoshima, S.; Tanaka, K. Langmuir 2014, 30, 1215−1219.
26) Horinouchi, A.; Atarashi, H.; Fujii, Y.; Tanaka, K. Macro-
molecules 2012, 45, 4638−4642.
(
ACKNOWLEDGMENTS
■
This study was partly supported by funding from the
Development of Human Resources in Science and Technology
of the Japan Science and Technology Agency (JST) and a
Grant-in-Aid for Scientific Research (C) (No. 15K05621) from
the Ministry of Education, Culture, Sports, Science and
Technology. The authors appreciate the kind support of Mr.
Kosuge, Department of Organic and Polymeric Materials,
Tokyo Institute of Technology, for performing the ESR
experiments and Dr. Ishikawa, Center for Advanced Materials
Analysis, Tokyo Institute of Technology, for collecting the DLS
measurements. The authors acknowledge the support of Ms.
Sudo and Prof. Kakimoto, Department of Organic and
Polymeric Materials, Tokyo Institute of Technology, for
conducting the GPC measurements.
REFERENCES
■
(
(
1) Hawker, C. J.; Wooley, K. L. Science 2005, 309, 1200−1205.
2) Ober, C. K.; Cheng, S. Z. D.; Hammond, P. T.; Muthukumar, M.;
Reichmanis, E.; Wooley, K. L.; Lodge, T. P. Macromolecules 2009, 42,
65−471.
3) Yuan, C.; Rong, M. Z.; Zhang, M. Q.; Zhang, Z. P.; Yuan, Y. C.
4
(
Chem. Mater. 2011, 23, 5076−5081.
(
(
(
4) Rao, J.; De, S.; Khan, A. Chem. Commun. 2012, 48, 3427−3429.
5) Rao, J.; Khan, A. Polym. Chem. 2013, 4, 2691−2695.
6) Wang, X.; Huang, J.; Chen, L.; Liu, Y.; Wang, G. Macromolecules
2
014, 47, 7812−7822.
7) Otsuka, H.; Aotani, K.; Higaki, Y.; Amamoto, Y.; Takahara, A.
Macromolecules 2007, 40, 1429−1434.
8) Otsuka, H.; Aotani, K.; Higaki, Y.; Takahara, A. J. Am. Chem. Soc.
003, 125, 4064−4065.
9) Espeso, J. F.; Lozano, A. E.; de la Campa, J. G.; Garcia-Yoldi, I.;
de Abajo, J. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 1743−1751.
10) García, J. M.; García, F. C.; Serna, F.; de la Pena, J. L. Prog.
Polym. Sci. 2010, 35, 623−686.
11) Shibasaki, Y.; Abe, Y.; Sato, N.; Fujimori, A.; Oishi, Y. Polym. J.
010, 42, 72−80.
12) Tanatani, A.; Yokoyama, A.; Azumaya, I.; Takakura, Y.; Mitsui,
(
(
2
(
(
̃
(
2
(
C.; Shiro, M.; Uchiyama, M.; Muranaka, A.; Kobayashi, N.; Yokozawa,
T. J. Am. Chem. Soc. 2005, 127, 8553−8561.
(
13) Shi, H.; Zhao, Y.; Zhang, X.; Zhou, Y.; Xu, Y.; Zhou, S.; Wang,
D.; Han, C. C.; Xu, D. Polymer 2004, 45, 6299−6307.
14) Schulze, M.; Michen, B.; Fink, A.; Kilbinger, A. F. M.
Macromolecules 2013, 46, 5520−5530.
15) Xie, W.; Ju, H.; Geise, G. M.; Freeman, B. D.; Mardel, J. I.; Hill,
A. J.; McGrath, J. E. Macromolecules 2011, 44, 4428−4438.
16) Shimazu, A.; Ikeda, K.; Miyazaki, T.; Ito, Y. Radiat. Phys. Chem.
000, 58, 555−561.
17) Lin, H.; Wagner, E.; Swinnea, J.; Freeman, B.; Pas, S.; Hill, A.;
Kalakkunnath, S.; Kalika, D. J. Membr. Sci. 2006, 276, 145−161.
18) Aiba, M.; Tokuyama, T.; Matsumoto, H.; Tomioka, H.;
Higashihara, T.; Ueda, M. J. Polym. Sci., Part A: Polym. Chem. 2014,
2, 3453−3462.
19) Telitel, S.; Amamoto, Y.; Poly, J.; Morlet-Savary, F.; Soppera,
O.; Lalevee, J.; Matyjaszewski, K. Polym. Chem. 2014, 5, 921−930.
20) Xie, G.; Pino, P.; Lorenzil, G. P. Macromolecules 1990, 23,
583−2588.
21) Yamauchi, K.; Kuroki, S.; Fujii, K.; Ando, I. Chem. Phys. Lett.
000, 324, 435−439.
(
(
(
2
(
(
5
(
́
(
2
(
2
I
DOI: 10.1021/acs.macromol.5b01778
Macromolecules XXXX, XXX, XXX−XXX