Figure-Eight-Shaped and Cage-Shaped Cyclic Polystyrenes

Article abstract of DOI:10.1021/acs.macromol.6b00093

Nonlinear polystyrenes (PS) with similar molecular weights but with different molecular structures having star-, figure-8-, and cage-shaped architectures were synthesized by combining atom transfer radical polymerization (ATRP) and click chemistry. Figure-8- and cage-shaped PS were fractionated by using a gradient normal phase liquid chromatography as confirmed by SEC-LS, MALDI-TOF MS, ¹H NMR, and FT-IR spectrometry. Their purities were estimated by using a two-dimensional liquid chromatography (2D-LC). The glass transition temperatures of these topologically different polymers were in the order of cage-, figure-8-, and star-shaped polymers possibly due to the multiple links that constrain the overall molecular diffusivity in the case of multicyclic polymers (figure-8, and cage). Monte Carlo simulation on the glass transition behavior of model system also agreed well with the experimental results.

Full text of DOI:10.1021/acs.macromol.6b00093

Article
pubs.acs.org/Macromolecules
Figure-Eight-Shaped and Cage-Shaped Cyclic Polystyrenes
Taeheon Lee,†,⊥ Joongsuk Oh, Jonghwa Jeong, Haeji Jung, June Huh,* Taihyun Chang,*
§,⊥
†
†
,‡
,§
,†
and Hyun-jong Paik*
†
Department of Polymer Science and Engineering, Pusan National University, Busan, 46241, Korea
Department of Chemical and Biological Engineering, Korea University, Seoul, 02841, Korea
‡
§
Department of Chemistry and Division of Advanced Materials Science, Pohang University of Science and Technology (POSTECH),
Pohang, 37673, Korea
*
S Supporting Information
ABSTRACT: Nonlinear polystyrenes (PS) with similar molecular
weights but with different molecular structures having star-,
figure-8-, and cage-shaped architectures were synthesized by
combining atom transfer radical polymerization (ATRP) and click
chemistry. Figure-8- and cage-shaped PS were fractionated by using
a gradient normal phase liquid chromatography as confirmed by
1
SEC-LS, MALDI−TOF MS, H NMR, and FT-IR spectrometry.
Their purities were estimated by using a two-dimensional liquid
chromatography (2D-LC). The glass transition temperatures of
these topologically different polymers were in the order of cage-,
figure-8-, and star-shaped polymers possibly due to the multiple links that constrain the overall molecular diffusivity in the case of
multicyclic polymers (figure-8, and cage). Monte Carlo simulation on the glass transition behavior of model system also agreed
well with the experimental results.
1
. INTRODUCTION
We recently reported the preparation of bicyclic polystyrene
PS) by using combination of ATRP and “click” chemistry
23,24
(
Topological control of polymers has provided new opportunities
in polymer science since it gives unprecedented properties
originated from nonlinear molecular architectures such as star-
like, comb-like, dendrimer, and cyclic polymers. Among
various architectural topologies of polymers, cyclic polymers
have attracted significant attention due to their unique
efficient strategy for cyclization based on first report from
25
Grayson group. In our precious reports, the 3-arm star PS was
synthesized by ATRP and used it as a precursor polymer for
bicyclic polymer after simple azidation of 3-arm star polymer.
ATRP techniques provide halogen terminal convertible into
versatile chain-end functionality such as azido-functionality
1
2
3
4
5
−8
properties associated with chain topologies.
Recently, the
26
clickable with alkyne-functionality. However, there is inherent
synthesis of multicyclic polymers has been reported vigorously
with the demand for new polymer materials from polymer
application fields. For example, the Tezuka group has pioneered
synthesis of different topological multicyclic poly(THF)s,
including fused, spiro, and bridged forms, using electro-
static self-assembly and covalent fixation under dilute condition.
Numerous researchers have endeavored to prepare various
topologically appealing multicyclic polymers including sun-
difficulty to preserve the chain-end functionality since ATRP
27
is not truly a living process. Because of imperfect chain-end
functionality of precursor polymer and linking reaction of
several polymer for multifunctional coupling agent, this system
produces high-molecular-weight impurities along with bicyclic
polymer. Therefore, it is necessary to fractionate the target
bicyclic polymer from the byproducts.
6,7
9
10
1
1,12
13−17
18,19
In this paper, we report the synthesis of two topologically
different multicyclic polymers, namely, figure-8-shaped (8-cyc)
and cage-shaped (C-cyc) polymers (Scheme 1), and compare
their glass transition behavior in terms of chain topological
effect. The 8- and C-cyc polymers were successfully synthesized
by combining atom transfer radical polymerization (ATRP) and
click” chemistry with difunctional and tetrafunctional coupling
agents, and purified by fractionation with a gradient normal
phase liquid chromatography (NPLC). Their purities were
shaped,
tadpole-shaped,
8-shaped,
manacle-
2
0
21
shaped, and paddle-shaped polymers through controlled
radical polymerization or anionic polymerization. Their complex
molecular topologies involving cycles in various fashions offer
a new opportunity to tune thermomechanical and dynamical
properties without changing chemical species of the materials.
For instance, the molecular topological constraints such as
internal cross-links and physical knots can result in nontrivial
effects on the molecular diffusivity and the glass transition
temperature. However, very few works have been conducted
for the structure−property correlations of cyclic polymers
due to the difficulty in preparing those molecules with high
“
Received: January 14, 2016
Revised: April 22, 2016
8
,22
purity.
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
a
Scheme 1. Overall Scheme for Synthesis of 8-cyc and C-cyc Polymers
a
Key: (A)Synthesis of 4-arm star polymer by ATRP from tetra-functional initiator, (B) synthesis of 8-cyc PS (6) between 4-arm star polymer and
di-functional coupling agent by “click” chemistry, and (C) synthesis of C-cyc PS (9) between 4-arm star polymer and di-functional coupling agent by
“click” chemistry.
examined by using two-dimensional liquid chromatography
2D-LC). The glass transition behavior of these high-purity
multicyclic polymers was investigated by differential scanning
calorimetry (DSC) and compared with Monte Carlo simulation
of model cyclic chain systems.
The eluent was kept at A for the first 100 min then the eluent was
changed from A to B linearly for the next 200 min. In the second LC,
a polypore mixed bed column (Agilent, 5 μm, 250 × 4.6 mm) was used
for SEC separation and THF was used as the eluent. Flow rate of the
second-D SEC was set at 1.6 mL/min for a fast analysis. Temperature
of the second-D SEC column was controlled at 70 °C by a column
oven (Futecs, AT-4000). NPLC- setup for the purification of the
as-prepared sample was the same as the first dimension NPLC of the
2D-LC analysis except for the flow rate and the solvent gradient
program. Flow rate was set at 0.5 mL/min and the eluent was kept at
A for the first 30 min then changed linearly to B during the next
30 min. Monomer conversion was determined by HP 5890 gas
chromatography equipped with HP101 column (methyl silicone fluid,
(
2. METHODS
Materials. Styrene (Junsei, 99.5%) was purified before use by
passage through and alumina column to remove the inhibitor and
stabilizer. Copper(I) bromide (CuBr, Aldrich, 98.0%) was purified by
stirring with glacial acetic acid, followed by filtering and washing the
solid four times with ethanol and twice with diethyl ether. The solid
was dried under vacuum for 2 days. Tetrahydrofuran (THF, J. T. Baker,
1
25 m × 0.32 mm × 0.30 μm). H NMR spectra were obtained on a
Varian Unity Inova 500 MHz superconducting FT-NMR spectrometer
1
00%) was distilled over CaH . N,N,N′,N″,N″-Pentamethyldiethylene-
2
triamine (PMDETA, 98%), pentaeryethritol (99.0%), 1,5-dipentanediol
96.0%), triethylamine (TEA, 99.5%), copper(II) bromide (CuBr2,
8.0%), 2-bromoisobutyryl bromide (98.0%), dithranol (90%), and
using CDCl as solvent. Infrared spectra were recorded on a Nicole
3
(
9
6700 FT-IR spectrophotometer. Differential scanning calorimeter (DSC)
analyses were performed with TA Instruments Q 100, which operated at
10 °C/min under nitrogen gas. A Bruker Autoflex speed mass spectrometer
equipped with a 2k Hz smartbeam-II laser was used for MALDI−TOF MS
experiments. The instrument operated at an accelerating potential of 20 kV
in positive mode. Mass calibration was performed using homemade PS
standards. Dithranol and DCTB were used as MALDI matrix. NaTFA
were used as cationization agent. They were dissolved in THF and a small
aliquot was deposited on a MALDI plate.
NaTFA (98%) were purchased from Aldrich and used as received.
Magnesium sulfate (MgSO , Junsei), propargyl bromide (TCI), sodium
4
azide (NaN , Junsei) and all other chemicals were used as received.
3
Instruments. For size exclusion chromatography (SEC) analysis,
three PS/DVB columns (Agilent Polypore 300 × 7.5 mm, Waters
Styragel HR4 300 × 7.8 mm, and Jordi mixed bed 300 × 8.0 mm)
were used. A Viscotek TDA302 detector was used for differential
refractometry (RI) and light scattering (LS) detection. Eluent was
THF delivered by a Bischoff HPLC compact pump at a flow rate of
Synthesis of Pentaerythritol Tetrakis(2-bromoisobutyrate) (1).
Pentaerythritol (4.0 g, 29.4 mmol) was added to the N purged round-
2
0
.7 mL/min. For 2D-LC analysis, the two LC systems were connected
by an electronically controlled 2-position, 10 port switching valve
Alltech, SelecPro). The first dimension of 2D-LC was solvent gradient
bottom flask with 100 mL of dried THF and then TEA (19.7 mL,
141 mmol) added to the flask. 2-Bromoisobutyryl bromide (17.4 mL,
141 mmol) was injected dropwise to the above flask at 0 °C over
30 min. The reaction mixture was reacted at room temperature for
(
normal phase liquid chromatography (NPLC) and the second
dimension was SEC. In the first-D LC, a bare silica column (Nucleosil,
the next 12 h. After reaction, reaction mixture was washed with H O/
2
5
μm, 100 Å pore, 150 × 4.6 mm) was used and THF/n-hexane
mixture was used as the eluent. Flow rate of the first-D LC was set at
.05 mL/min to synchronize with the second-D SEC separation. The
solvent gradient program for the first LC separation was set as follows
solvent A, THF/n-hexane = 40/60; solvent B, THF/n-hexane = 60/40),
diethyl ether. The organic phase was dried over MgSO . Diethyl ether
was removed using a rotary evaporator. The product was recrystallized
4
0
in hexane. Thus 12.25 g of the white crystalline product was obtained
1
(yield: 57%). H NMR (400 MHz, CDCl ): δ = 4.33 (s, 8H), 1.94
3
(
(s, 24H).
B
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 1. (a) SEC-LS traces of 2, 3, as-synthesized 8-cyc PS (mixture of 5 and 6), as-synthesized C-cyc PS (mixture of 8 and 9) and (b) solvent
gradient NPLC chromatograms of 4-arm star precursor (3), as-synthesized 8-cyc PS (mixture of 5 and 6), and as-synthesized C-cyc PS (mixture of
8
and 9).
Synthesis of 1,5-Bis(2-propynyloxy)pentane (4). Sodium hydride
The molecular weight of the obtained polymer was measured by
(
NaH, 1.04 g, 43.2 mmol) was added to the round-bottom flask
containing a solution of 1,5-pentanediol (1.5 g, 14.4 mmol) in DMF
50 mL). Propargyl bromide (3.85 mL, 43.2 mmol) was added slowly
to the flask with stirring at 0 °C. After addition of propargyl bromide,
the reaction was carried out at 60 °C for 24 h. The reaction mixture
was extracted with ethyl acetate followed by washing with deionized
SEC-LS (M = 6700 and M /M = 1.05). The chain-end functionality
n
w
n
1
of 3 was determined by H NMR (chain-end functionality =92%).
Synthesis of Figure-8-Shaped (8-cyc) PS (6). CuBr (3.10 g,
21.6 mmol) was added to a Schlenk flask, which was subsequently
(
evacuated and backfilled with N three times. Deoxygenated PMDETA
2
(4.51 mL, 21.6 mmol) and THF (200 mL) were added to the flask via
syringes. 3 (0.5 g, 0.06 mmol) and 4 (33.4 mg, 0.19 mol) put in 20 mL
of THF and were degassed, respectively. Then, those solutions were
added dropwise to the solution of CuBr/PMDETA at 35 °C using
a syringe pump at the rate of 0.2 mL/h. After complete addition, the
reaction was allowed to proceed at 35 °C for 5 h. The reaction mixture
was passed through a neutral alumina column to remove the copper
catalyst. THF was removed using a rotary evaporator. Pure 6 was
separated from its mixture with branched PS (5) using a gradient
NPLC (M = 6700 and M /M = 1.02).
water several times. The organic phase was dried over MgSO and the
4
volatiles were removed in vacuo. The crude product was further
purified by column chromatography using hexane/ethyl acetate (3/1).
After drying, 1.68 g of the brown oil was obtained in 65% yield.
1
H NMR (CDCl ): δ = 4.13 (s, 4H), 3.52 (s, 4H), 2.40 (s, 2H), 1.63
3
(
s, 4H), 1.44 (s, 2H).
Synthesis of Tetrakis(2-propynyloxymethyl)methane (7). NaH
1.06 g, 44.1 mmol) was added to the round-bottom flask containing
a solution of pentaerythritol (1.0 g, 7.34 mmol) in DMF (100 mL).
Propargyl bromide (3.93 mL, 44.1 mmol) was added slowly to the
reaction vessel with stirring at 0 °C. After addition of propargyl
bromide, the reaction was carried out at 60 °C for 24 h. The reaction
mixture was extracted with ethyl acetate followed by washing with
deionized water several times. The organic phase was dried over
(
w
w
n
Synthesis of Cage-Shaped (C-cyc) PS (9). CuBr (3.10 g, 21.6 mmol)
was added to a Schlenk flask, which was subsequently evacuated and
backfilled with N three times. Deoxygenated PMDETA (4.51 mL,
2
2
1.6 mmol) and THF (200 mL) were added to the flask via syringes. 3
(0.5 g, 0.06 mmol) and 7 (26.7 mg, 0.09 mmol) put in 20 mL of THF and
MgSO and the volatiles were removed in vacuo. The crude product
4
were degassed, respectively. Then, those solutions were added dropwise
into the solution of CuBr/PMDETA at 35 °C using a syringe pump at
the rate of 0.2 mL/h. After complete addition, the reaction was allowed
to proceed at 35 °C for 5h. The reaction mixture was passed through a
neutral alumina column to remove the copper catalyst. THF was removed
using a rotary evaporator. Pure 9 was separated from its mixture with
branched PS (8) using a gradient NPLC (M = 6700 and M /M = 1.01).
was further purified by column chromatography using hexane/ethyl
acetate (3/1). After drying, 1.24 g of the brown oil was obtained
1
in 60% yield. H NMR (400 MHz, CDCl ): δ = 4.12 (s, 8H), 3.53
3
(
s, 8H), 2.40 (s, 4H).
Synthesis of 4-Arm Star Polystyrene (2). CuBr (47 mg, 0.33 mmol),
CuBr (7.31 mg, 0.03 mmol), PMDETA (0.752 mL, 0.36 mmol) and
2
w
w
n
pentaerythritol tetra(2-bromoisobutyrate) (206 mg, 0.36 mmol) were
added to a dried 100 mL Schlenk flask. The flask was sealed with
glass stopper then evacuated and backfilled with N2 three times.
Deoxygenated styrene (20 mL, 175 mmol) and anisole (5 mL) were
added to the flask via syringes. Then, the flask was placed in an oil bath
at 100 °C. At 20% monomer conversion (measured by gas chromato-
graphy), the polymerization was quenched by exposing to air. The
reaction mixture was diluted with THF and passed through a neutral
alumina column to remove the copper catalyst. The polymer was
obtained by precipitation in methanol. The polymer was collected
by filtration and dried under vacuum. The molecular weight of
Monte Carlo Simulation. Dynamic Metropolis Monte Carlo
28,29
method with the 8-site bond fluctuation model
is used to
investigate the glass transition behavior of star, 8-cyc and C-cyc
polymers. Topologically monodisperse chains, each of which has a
given architecture with a fixed number of 36 bonds, were generated on
a L × L × L simulation box with L = 100a under the periodic boundary
condition where a is the unit lattice spacing. In order to model the
glassy behavior of polymer chain, we employ the potential proposed
30,31
by Wittkop et al.
without nonbonded potential for isolating chain
architecture effect. The bonded potential U (r) was given as
b
the obtained polymer was determined by SEC-LS (M = 6700 and
w
⎧
⎪
⎪
⎪
4
.461ϵ for r = (±2, 0, 0)
.201ϵ for r = (±2, ±1, 0)
M /M = 1.05).
w
n
Synthesis of Azido-Terminated 4-Arm Star Polystyrene (3).
Sodium azide (97.3 mg, 1.50 mmol) was added to a solution of 2
2
⎪
−0.504ϵ for r = (±3, 0, 0)
−3.275ϵ for r = (±3, 0, 0)
(
2 g, 0.54 mmol) in DMF (10 mL). The reaction was conducted at
Ub(r) = ⎨
room temperature for 24 h. The reaction mixture was diluted with
chloroform and washed with deionized water several times. The organic
⎪
⎪
phase was dried with MgSO . After filtration of solid, solvent was
⎪−3.478ϵ for r = (±2, ±2, ±1)
4
evaporated. The polymer was obtained by precipitation in methanol.
The polymer was collected by filtration and dried under vacuum.
⎪
⎩
−1.707ϵ for r = (±3, ±1, 0)
(1)
C
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 3. MALDI−TOF MS spectra of 2, 3, 6, and 9: (a) overall
spectra and (b) expanded spectra of near the maxima of peak envelopes.
Figure S1. The bromo chain-end functionality of 2 was
determined as ∼92% by comparing the methyl protons
(
0.97−0.70 ppm) at the initiator part to methine proton
(
4.59−4.36 ppm) adjacent to the terminal bromide group in
1
the H NMR spectrum. The bromo chain-end groups of 2 were
quantitatively transformed to azido groups by simple nucleo-
1
philic substitution reaction as shown in the H NMR spectrum
(
Figure S2). The azido-transformation in the precursor polymer
(
3) was also characterized by the appearance of peak due to
−1
azide group at 2100 cm in the FT IR spectrum (Figure S3).
Di- (4) and tetra- (7) alkynyl coupling agents were synthesized
using propagyl bromide (Figures S4 and S5).
1
Figure 2. H NMR spectra of 3, 6, and 9.
8
-cyc (6) and C-cyc (9) PS were prepared by a “click”
coupling reaction between 3 and coupling agents, 4 or 7
respectively, under a dilute condition (2.5 g/L) to minimize the
intermolecular coupling reaction resulting in branched high-
molecular-weight polymers. The multicyclic polymer (6 or 9)
was formed along with branched polymers (5 or 8, respectively),
as evidenced by SEC (Figure 1a). As shown in Figure 1a, 2
(25.19 min) and 3 (25.21 min) are not much different since
azidation does not change the topology. However, cyclization
of 3 changed the elution time of the multicyclic polymers, 6
(25.4 min) and 9 (25.6 min). The relative delay in the elution
of multicyclic polymers, 6 and 9 is due to their reduced
hydrodynamic volume. The larger shift for 9 than 6 indicates that
where r = (x,y,z) represent the set of all permutations of eigen values of
the bond vector in the bond fluctuation model and ε is the unit of energy.
A lattice occupation density is set to be ϕ = 0.5 for polymer chains in the
simulation box, which corresponds to a polymer melt in the bond
32
fluctuation model. Each system of randomly generated configuration
6
was initially equilibrated at k T/ε = ∞ for 1 × 10 Monte Carlo steps.
B
7
(
MCS) and then annealed at a given k T/ε for 2 × 10 MCS.
B
3. RESULTS AND DISCUSSION
The 4-arm star PS (2) was synthesized by ATRP using the tetra-
functional initiator, pentaerythritol tetrakis(2-bromoisobutyrate)
(1), of which the chemical structure was characterized in
D
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 1. Symbol, Structure, and Relative Mass Difference (m/z) Values of the Peaks Observed in MALDI−TOF Mass Spectra
symbol of peak series
description
structure
relative m/z difference from N symbol in Scheme 1
U
N
M
H
E
4-arm star PS with 4-terminal double bonds
4-arm star PS with 4-terminal azide groups
metastable peak of N
I−(PS)n−4−(C H )
−172 (+36)
0
2
3
3
3
6
6
6
9
8
4
7
4
I−(PS) −(N )
n
3
I−(PS) −(N ) −(N)
−25−27
−41
+360 (+48)
+540 (+20)
+720 (−8)
+288 (−24)
n
3 3
4-arm star PS with 3-terminal azide groups
8-cyc PS
I−(PS) −(N ) −H
n 3 3
I−(PS) −(2N −4)
2
n
3
T
S
tadpole-shaped PS
I−(PS) −(2N −4)−(N −4)
n 3 3 2
4-arm star PS with 4-terminal triazole ring
C-cyc PS
I−(PS) −(N −4)
n 3 4
C
I−(PS) −(4N −7)
n 3
Figure 4. 2D contour plots of (a) mixture of 5 and 6, (b) 6, (c) mixture of 8 and 9, and (d) 9: the first D, NPLC; the second D, SEC.
the shrinkage of the chain size is larger for 9 than 6. The
appearance of broad multimodal peaks (21.5−24.7 min) on SEC
trace of a mixture of 5 and 6 indicates the formation of complex
branched byproducts along with 6. 3 has 90% of azido-
9 respectively, by fractionation with gradient NPLC as shown in
Figure 1b. While 3 eluted as a sharp single peak at V ∼ 5 min,
E
for 5 and 6 and for 8 and 9 they elute over a wide elution time
(t ) indicating that they contain a number of byproducts. The
E
1
functionality predicted by H NMR spectroscopy (Figure S2).
main peaks labeled as 6 and 9 in Figure 1b are the elution peaks
of the target products, 8-cyc PS and C-cyc PS, respectively.
9 eluted earlier than 6 despite the identical molecular weight
due to the different architectures that affects the interaction of
polymer chains with the stationary phase differently. To obtain
the high purity cyclic polymers, they are fractionated by collect-
Thus, lacked precursor polymer must produce byproducts;
on coupling with alkynyl groups of 4 produced 5 which could
have reactive acetylene groups. These acetylene groups could
react further with precursor polymers resulting in the formation
of dimeric or multimeric branched high-molecular-weight
polymers. Similarly, the mixture of 8 and 9 also showed broad
multimodal peaks after click reaction between 3 and 7. To
remove high-molecular-weight branched polymer, cyclic poly-
mers were separated from both mixtures of 5 and 6 and of 8 and
ing the effluents over the t between the small vertical bars
E
shown in the Figure 1b.1
Figure 2 shows the H NMR spectra of 3 and separated
cyclic polymers, 6 and 9, from each mixture. As a result of
E
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
click” coupling reaction, the methine proton at 3.98 ppm
Article
“
in the N -terminated star polymer (Figure 2a) appeared at
3
5.12 ppm due to the neighboring triazole unit in the cyclic
polymer (Figure 2, parts b and c). Also a new peak at 4.35 ppm
appears for the methylene protons in the linker adjacent to the
triazole unit.
Figure 3 shows the results of MALDI−TOF MS analyses
of the synthesized polymers. In Figure 3a, the overall MALDI−
MS peak envelops of precursors and cyclic polymers show
nearly the same molecular weight distribution centered around
6
.0k in agreement with the molecular weight determined by
1
SEC-LS and H NMR. Additionally, expanded spectra of near
the maxima of peak envelope were analyzed in Figure 3b. For 2,
the major peak series is U (see Table 1) and they were formed
upon elimination of HBr during the MALDI process leaving a
double bond at the terminal styrene unit. It is also well-known
that a few subsidiary peaks can show up depending on the
sample preparation and MALDI process due to the reaction of
3
3
terminal Br. The spectrum of 3 shows major peak designated
as N (target structure with 4 azide groups) and some minor
peaks. H seems to be the species having one less azide group. It
is likely to have been formed when an arm is terminated during
the ATRP. M is the metastable species (−26 m/z) that is
34
always observed for the azido-functionalized polymers. For 6,
some byproduct species are still found in addition to the main
target 8-cyc (E series). T and S series can be assigned to the
tadpole shaped (a ring with two tails) and 4-arm star PS with
4-terminal triazole rings. The m/z value of E series for 50mer
was found to be 6171.5 in excellent agreement with the
+
calculated value (m/z .: 6171.7, C443H464O N Na ) indicat-
cal
12 12
Figure 5. Glass transition temperatures for 3 (a), 6 (b), and 9 (c).
ing that the target 8-cyc PS was synthesized successfully.
However, the 8-cyc PS was not fractionated perfectly by
HPLC since the byproducts elute very close to the major peak
as shown in Figure 1b. On the other hand, the MALDI MS
spectrum of 9 appears to be very pure as expected from the
well-isolated HPLC peak in Figure 1b. In addition, the m/z
value of 9 was measured as 6099.5 that is in good agree-
characteristics of 8-cyc polymer where the chain termini are
absent.
In order to further investigate the chain architecture effect
on the glass transition temperature, we simulated model bulk
systems of star, 8-cyc, and C-cyc polymers by employing
a dynamic Monte Carlo method with bond fluctuation
model. The glass transition of the model chain is reflected
by introducing the bonded potential eq 1 that energetically
favors a certain bond vector. In the model systems (star, 8-cyc,
and C-cyc), the number of bonds per chain is kept constant so
as to isolate the architecture effect on the glass transition
behavior. The detail of simulation method is described in
the Simulation section. For the determination of the glass
transition temperature, we used the Vogel−Fulcher relation
ment with the calculated m/z value of 6099.6 for 50mer
+
(
C438H452O N Na ).
12
12
2D-LC was carried out to assess the purity of 6 and 9.
Figure 4a and 4c show the 2D-LC chromatograms of 5 and 6
and of 8 and 9, respectively, before the NPLC fractionation.
After the click reaction, the presence of the large amount of
byproduct is apparent. The purity of 6 is only 39% poorer than
9
(81%). It seems to be a natural result since the probability
of making 6 requires two linkers to react with a 4-arm star
together while 9 requires only one linker to react with a 4-arm
star. After the purification by NPLC fractionation (Figure 2b),
most of the byproducts disappeared. The small amount of
seemingly tadpole-shaped byproduct in 8-cyc PS observed in
MALDI−MS (T series) was not detected in 2D-LC indicating
that it elutes very close to 6.
35−37
given by
D(T) = D exp[−A/(T − T)]
(2)
∞
o
where D(T) is the self-diffusion constant at a temperature T
and the D , A, and T are the fit parameters for Vogel−Fulcher
∞
0
relation. Here the T , referred to as Vogel−Fulcher temper-
0
ature, can be interpreted as a thermodynamic limit of glass
transition temperature where the relaxation time of polymer
becomes infinity. In the present work, the glass transition
behavior of star, 8-cyc, and C-cyc is investigated in terms of the
After fractionation of 3, 6, and 9 using the NPLC, an isolated
effect of chain topology on thermal behavior was investigated.
Figure 5 shows the DSC curves of 3, 6 and 9. The glass transi-
tion temperatures of these topologically different PSs are in the
order of 9 (C-cyc, T = 114.5 °C), 6 (8-cyc, T = 83.9 °C), and
Vogel−Fulcher temperature T . The self-diffusion constants
0
g
g
were estimated from the mean-square displacement g (t) of the
3
(4-arm star, T = 80.8 °C). This glass transition behavior is
1
g
chain segments as a function of time t:
reflective of more severe topological constraints for molecular
motion in 9 (C-cyc polymer) that contains two focal points,
as compared to 6 (8-cyc) and 3 (star polymer) having a
2
g (t) = ⟨[R (t) − R (0)] ⟩
s
s
(3)
1
single focal point. It can be also inferred that higher T of 6
where R (t) is the position vector of segment at time t and
g
s
in comparison with that of 3 is attributed to the architectural
the bracket ⟨ ⟩ stands for the average over all segments in
F
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 6. (a−c) Double logarithmic plots of g /6t vs t for the simulated systems for (a) star, (b) 8-cyc, and (c) C-cyc. (d) Self-diffusion constants for
1
star, 8-cyc, and C-cyc as a function of temperature.
the simulated system. Since the mean square displacement
is also in agreement with the experimental results. The diffusion
of the overall polymers with these three architectures, which
have commonly 4 bond linked focal point, is governed by the
motion of focal point that is the slowest segmental motion. The
C-cyc polymers that has two focal points needs more thermal
energy preventing the vitrification than star and 8-Cyc
polymers, both of which have only one focal point. Given
that the slowest motion of such focal point predominantly
determine the diffusivity of the overall chain, the glass transition
behavior of 8-Cyc is nearly the same as that of star whose
dynamics is only slightly enhanced by the local motion of four
chain ends.
should show a diffusive behavior, g (t) = 6Dt, at t > τ where
1
τ is the relaxation time of the chain, we can estimate the
self-diffusion constant D from a plateau in the plot of
g1/t vs t.
Figure 6 presents double logarithmic plots of g /6t vs t for
1
the simulated systems for 3 (Figure 6a), 6 (Figure 6b) and 9
(
Figure 6c), respectively. The g /6t curves for 3, 6, and 9 show
1
̃
well-defined plateau regimes at T(=k T/ε) = 3.0−15.0 after
B
6
sufficiently long time simulations (∼O (10 ) MCS), whereas
no clear plateaus (or crossover to the diffusive regime) were
7
̃
yet reached at T < 2.5 even after 2 × 10 MCS. Figure 6d
shows the self-diffusion constants for 3, 6, and 9 as a function of
̃
T estimated from Figure 6a-6c. The solid lines represent
4
. CONCLUSION
Vogel−Fulcher fits eq 2 with three adjustable parameters
including Vogel−Fulcher temperature, which give D
=
=
=
In this paper, we describe the synthesis of figure-8-shaped
and cage-shaped PSs by combining ATRP and “click” coupling
reaction. The structures of figure-8-shaped and cage-shaped
PS separated from branched high-molecular-weight polymer
∞
̃
.000116, A = 4.165, and T = 0.900 for star, D
0
0
0
o
∞
̃
.000119, A = 4.290, and T = 0.933 for 8-cyc, and D
.000099, A = 4.452, and T = 1.034 for C-cyc, respectively.
o
∞
̃
o
1
byproducts were characterized by SEC, FT-IR, H NMR, and
̃
The Vogel−Fulcher analysis reveals that T for 9 is higher
o
MALDI−TOF. Their purities were estimated from 2D-LC
analysis. Glass transition temperatures of 4-arm star polymer,
figure-8-shaped, and cage-shaped polymer were measured by
̃
than T for 3 and 6, which is reflective of more restrictive bond
o
constraints for 9, as also manifested by smaller self-diffusion
̃
constant for 9. It is also found from the simulation that T of 6
DSC. T s of multicyclic polymers (figure-8-shaped, and cage-
o
g
̃
̃
shaped polymer) were higher than precursor 4-arm star polymer
owing to the topological difference. The observed glass transition
behavior also agreed well with the Monte Carlo simulation
results based on Vogel−Fulcher analysis.
is nearly the same with T of 3 although T of 6 is slightly higher.
The simulation results for three different chain architectures
o
o
showed that the difference in T s between star and 8-shape is
g
much smaller than that between 8-shape and cage-shape, which
G
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(
11) Fan, X.; Wang, G.; Huang, J. Synthesis of Macrocyclic Molecular
ASSOCIATED CONTENT
■
Brushes with Amphiphilic Block Copolymers as Side Cains. J. Polym.
Sci., Part A: Polym. Chem. 2011, 49, 1361−1367.
(
*
S
Supporting Information
The Supporting Information is available free of charge on the ACS
Publications website at DOI: 10.1021/acs.macromol.6b00093.
1H NMR spectra of initiator (1), coupling agents (4, 7),
12) Liu, L.; Parameswaran, S.; Sharma, A.; Grayson, S. M.;
Ashbaugh, H. S.; Rick, S. W. Molecular Dynamics Simulations of
Linear and Cyclic Amphiphilic Polymers in Aqueous and Organic
Environments. J. Phys. Chem. B 2014, 118, 6491−6497.
star polymers (2, 3) and FT-IR spectra of star polymers
(
13) Lord, S. J.; Sheiko, S. S.; LaRue, I.; Lee, H.-I.; Matyjaszewski, K.
(
2, 3) and coupling agents (4, 7) (PDF)
Tadpole Conformation of Gradient Polymer Brushes. Macromolecules
004, 37, 4235−4240.
2
AUTHOR INFORMATION
(14) Wan, X.; Liu, T.; Liu, S. Synthesis of Amphiphilic Tadpole-
Shaped Linear-Cyclic Diblock Copolymers via Ring-Opening Polymer-
ization Directly Initiating from Cyclic Precursors and Their
Application as Drug Nanocarriers. Biomacromolecules 2011, 12,
■
*
*
*
Corresponding Authors
(H.P.) Telephone: +82-51-510-2402. E-mail: hpaik@pusan.ac.kr.
(T.C.) Telephone: +82-54-279-2109. E-mail: tc@postech.ac.kr.
(J.H.) Telephone: +82-2-3290-5977. E-mail: junehuh@korea.
1
146−1154.
(15) Doi, Y.; Ohta, Y.; Nakamura, M.; Takano, A.; Takahashi, Y.;
ac.kr.
Matsushita, Y. Precise Synthesis and Characterization of Tadpole-
Shaped Polystyrenes with High Purity. Macromolecules 2013, 46,
1075−1081.
Author Contributions
These authors contributed equally to this work
⊥
(16) Oike, H.; Uchibori, A.; Tsuchitani, A.; Kim, H.-K.; Tezuka, Y.
Notes
Designing Loop and Branch Polymer Topology with Cationic Star
Telechelics through Effective Selection of Mono- and Difunctional
Counteranions. Macromolecules 2004, 37, 7595−7601.
The authors declare no competing financial interest.
(17) Shi, G.-Y.; Tang, X.-Z.; Pan, C.-Y. Tadpole-shaped Amphiphilic
ACKNOWLEDGMENTS
■
Copolymers Prepared via RAFT Polymerization and Click Reaction. J.
Polym. Sci., Part A: Polym. Chem. 2008, 46, 2390−2401.
This work was supported by the Active Polymer Center for
Pattern Integration (No. R11-2007-0056091) and Basic Science
Research Program through the National Research Foundation
of Korea (NRF) grant funded by the Korean government
MSIP) (2013R1A2A2A01068818). T.C. acknowledges the
support from NRF-Korea (2015R1A2A2A01004974). J.H.
acknowledges support from NRF-Korea (2013R1A1A2064112).
(18) Schappacher, M.; Deffieux, A. Controlled Synthesis of Bicyclic
″
Eight-Shaped″ Poly(chloroethyl vinyl ether)s. Macromolecules 1995,
28, 2629−2636.
(
(19) Fan, X.; Huang, B.; Wang, G.; Huang, J. Synthesis of
Amphiphilic Heteroeight-Shaped Polymer Cyclic-[Poly(ethylene
oxide)-b-polystyrene]
5, 3779−3786.
20) Tezuka, Y.; Ohashi, F. Synthesis of Polymeric Topological
via “Click” Chemistry. Macromolecules 2012,
2
4
(
REFERENCES
■
Isomers through Double Metathesis Condensation with H-Shaped
(
1) Schaefgen, J. R.; Flory, P. J. Synthesis of Multichain Polymers and
Investigation of their Viscosities. J. Am. Chem. Soc. 1948, 70, 2709−
718.
2) Bhattacharya, A.; Misra, B. N. Grafting: A Versatile Means to
Modify Polymers: Techniques, Factors and Applications. Prog. Polym.
Sci. 2004, 29, 767−814.
3) Hawker, C. J.; Frechet, J. M. J. Preparation of Polymers with
Controlled Molecular Architecture. A New Convergent Approach to
Dendritic Macromolecules. J. Am. Chem. Soc. 1990, 112, 7638−7647.
4) Semlyen, J. A. Introduction: Cyclic Polymers - The First 40 Years. In
Cyclic Polymers; Semlyen, J. A., Ed.; Springer: Dordrecht, The
Netherlands, 2002; pp 1−46.
5) Hadziioannou, G.; Cotts, P. M.; ten Brinke, G.; Han, C. C.; Lutz,
Telechelic Precursors. Macromol. Rapid Commun. 2005, 26, 608−612.
(21) Lonsdale, D. E.; Monteiro, M. J. Various Polystyrene Topologies
2
(
Built from Tailored Cyclic Polystyrene via CuAAC Reactions. Chem.
Commun. 2010, 46, 7945−7947.
(22) Hossain, M. D.; Lu, D.; Jia, Z.; Monteiro, M. J. Glass Transition
Temperature of Cyclic Stars. ACS Macro Lett. 2014, 3, 1254−1257.
(
(23) Jeong, J.; Kim, K.; Lee, R.; Lee, S.; Kim, H.; Jung, H.; Kadir, M.
A.; Jang, Y.; Jeon, H. B.; Matyjaszewski, K.; Chang, T.; Paik, H.-j.
Preparation and Analysis of Bicyclic Polystyrene. Macromolecules 2014,
(
4
(
7, 3791−3796.
24) Jeong, J.; Kim, H.; Lee, S.; Choi, H.; Jeon, H. B.; Paik, H.-j.
Improved Synthesis of Bicyclic Polystyrenes by ATRP and “Click”
Reaction. Polymer 2015, 72, 447−452.
(
P.; Strazielle, C.; Rempp, P.; Kovacs, A. J. Thermodynamic and
Hydrodynamic Properties of Dilute Solutions of Cyclic and Linear
Polystyrenes. Macromolecules 1987, 20, 493−497.
(25) Laurent, B. A.; Grayson, S. M. An Efficient Route to Well-
Defined Macrocyclic Polymers via “Click” Cyclization. J. Am. Chem.
Soc. 2006, 128, 4238−4239.
(
6) Tezuka, Y.; Fujiyama, K. Construction of Polymeric δ-Graph: A
(26) Matyjaszewski, K.; Xia, J. Atom Transfer Radical Polymerization.
Doubly Fused Tricyclic Topology. J. Am. Chem. Soc. 2005, 127, 6266−
270.
7) Oike, H.; Imaizumi, H.; Mouri, T.; Yoshioka, Y.; Uchibori, A.;
Tezuka, Y. Designing Unusual Polymer Topologies by Electrostatic
Self-Assembly and Covalent Fixation. J. Am. Chem. Soc. 2000, 122,
Chem. Rev. 2001, 101, 2921−2990.
27) Matyjaszewski, K. Atom Transfer Radical Polymerization
ATRP): Current Status and Future Perspectives. Macromolecules
012, 45, 4015−4039.
28) Carmesin, I.; Kremer, K. Static and Dynamic Properties of Two-
Dimensional Polymer Melts. J. Phys. (Paris) 1990, 51, 915−932.
29) Deutsch, H. P.; Binder, K. Interdiffusion and Self-Diffusion in
Polymer Mixtures: A Monte Carlo Study. J. Chem. Phys. 1991, 94,
294−2304.
30) Wittkop, M.; Ho
6
(
(
(
2
(
9
(
592−9599.
8) Kapnistos, M.; Lang, M.; Vlassopoulos, D.; Pyckhout-Hintzen,
(
W.; Richter, D.; Cho, D.; Chang, T.; Rubinstein, M. Unexpected
Power-Law Stress Relaxation of Entangled Ring Polymers. Nat. Mater.
008, 7, 997−1002.
9) Yamamoto, T.; Tezuka, Y. Topological Polymer Chemistry: A
Cyclic Approach Toward Novel Polymer Properties and Functions.
Polym. Chem. 2011, 2, 1930−1941.
10) Sugai, N.; Heguri, H.; Ohta, K.; Meng, Q.; Yamamoto, T.;
Tezuka, Y. Effective Click Construction of Bridged- and Spiro-
Multicyclic Polymer Topologies with Tailored Cyclic Prepolymers
kyklo-Telechelics). J. Am. Chem. Soc. 2010, 132, 14790−14802.
2
2
(
(
̈ ̈
lzl, T.; Kreitmeier, S.; Goritz, D. A Monte Carlo
Sudy of The Glass Transition in Three-Dimensional Polymer Melts. J.
Non-Cryst. Solids 1996, 201, 199−210.
(
(31) Grassberger, P. Pruned-Enriched Rosenbluth Method: Simu-
lations of θ Polymers of Chain Length up to 1 000 000. Phys. Rev. E:
Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1997, 56, 3682−
3693.
(
H
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(
32) Thompson, R. L.; McDonald, M. T.; Lenthall, J. T.; Hutchings,
L. R. Solvent Accelerated Polymer Diffusion in Thin Films.
Macromolecules 2005, 38, 4339−4344.
(
33) Kim, K.; Hasneen, A.; Paik, H.-j; Chang, T. MALDI-TOF MS
Characterization of Polystyrene Synthesized by ATRP. Polymer 2013,
4, 6133−6139.
5
(
34) Li, Y.; Hoskins, J. N.; Sreerama, S. G.; Grayson, S. M. MALDI−
TOF Mass Spectral Characterization of Polymers Containing an Azide
Group: Evidence of Metastable Ions. Macromolecules 2010, 43, 6225−
6228.
(
35) Vogel, H. The law of the relationship between viscosity of
liquids and the temperature. Phys. Z. 1921, 22, 645−646.
36) Fulcher, G. S. Analysis of Recent Measurements of The
Viscosity of Glasses. J. Am. Ceram. Soc. 1925, 8, 339−355.
37) Okun, K.; Wolfgardt, M.; Baschnagel, J.; Binder, K. Dynamics of
(
(
Polymer Melts above the Glass Transition: Monte Carlo Studies of the
Bond Fluctuation Model. Macromolecules 1997, 30, 3075−3085.
I
DOI: 10.1021/acs.macromol.6b00093
Macromolecules XXXX, XXX, XXX−XXX