Synthesis of linear, cyclic, figure-eight-shaped, and tadpole-shaped amphiphilic block copolyethers via t-Bu-P4-catalyzed ring-opening polymerization of hydrophilic and hydrophobic glycidyl ethers

Article abstract of DOI:10.1021/ma500494e

This paper describes the synthesis of systematic sets of figure-eight- and tadpole-shaped amphiphilic block copolyethers (BCPs) consisting of poly(decyl glycidyl ether) and poly[2-(2-(2-methoxyethoxy)ethoxy)ethyl glycidyl ether], together with the corresponding cyclic counterparts, via combination of the t-Bu-P₄-catalyzed ring-opening polymerization (ROP) and click cyclization. The clickable linear BCP precursors, with precisely controlled azido and ethynyl group placements as well as a fixed molecular weight and monomer composition (degree of polymerization for each block was adjusted to be around 50), were prepared by the t-Bu-P₄-catalyzed ROP with the aid of functional initiators and terminators. The click cyclization of the precursors under highly diluted conditions produced a series of cyclic, figure-eight-, and tadpole-shaped BCPs with narrow molecular weight distributions of less than 1.06. Preliminary studies of the BCPs self-assembly in water revealed the significant variation in their cloud points depending on the BCP architecture, though there were small architectural effects on their critical micelle concentration and morphology of the aggregates.

Full text of DOI:10.1021/ma500494e

Article
pubs.acs.org/Macromolecules
Synthesis of Linear, Cyclic, Figure-Eight-Shaped, and Tadpole-
Shaped Amphiphilic Block Copolyethers via t‑Bu‑P4‑Catalyzed Ring-
Opening Polymerization of Hydrophilic and Hydrophobic Glycidyl
Ethers
Takuya Isono,† Yusuke Satoh,† Kana Miyachi,† Yougen Chen,‡ Shin-ichiro Sato,‡ Kenji Tajima,‡
Toshifumi Satoh,*,‡ and Toyoji Kakuchi*,‡
†Graduate School of Chemical Sciences and Engineering, Hokkaido University, Sapporo 060-8628, Japan
‡Division of Biotechnology and Macromolecular Chemistry, Faculty of Engineering, Hokkaido University, Sapporo 060-8628, Japan
S
* Supporting Information
ABSTRACT: This paper describes the synthesis of systematic
sets of figure-eight- and tadpole-shaped amphiphilic block
copolyethers (BCPs) consisting of poly(decyl glycidyl ether)
and poly[2-(2-(2-methoxyethoxy)ethoxy)ethyl glycidyl ether],
together with the corresponding cyclic counterparts, via
combination of the t-Bu-P4-catalyzed ring-opening polymer-
ization (ROP) and click cyclization. The clickable linear BCP
precursors, with precisely controlled azido and ethynyl group
placements as well as a fixed molecular weight and monomer
composition (degree of polymerization for each block was
adjusted to be around 50), were prepared by the t-Bu-P4-
catalyzed ROP with the aid of functional initiators and
terminators. The click cyclization of the precursors under highly diluted conditions produced a series of cyclic, figure-eight-, and
tadpole-shaped BCPs with narrow molecular weight distributions of less than 1.06. Preliminary studies of the BCPs self-assembly
in water revealed the significant variation in their cloud points depending on the BCP architecture, though there were small
architectural effects on their critical micelle concentration and morphology of the aggregates.
weights have been never achieved, though there were several
reports about synthesizing tadpole- and figure-eight-shaped
BCPs, such as [cyclic poly(chloroethyl vinyl ether)]-block-
polystyrene,16 (cyclic-polystyrene)-block-poly(ethylene
oxide),17 [cyclic-poly(ethylene oxide)]-block-polystyrene,18 [cy-
clic-poly(N-isopropylacrylamide)]-block-poly(ε-caprolac-
tone),19 (cyclic-polystyrene)-block-[cyclic-poly(ε-caprolac-
tone)],20 [cyclic-poly(ethylene oxide)]-block-[cyclic-poly-
(tetrahydrofuran)],21 cyclic-[poly(ethylene oxide)-block-
polystyrene]2,22 and (cyclic-polystyrene)-block-poly(acrylic
acid).23 To provide a fundamental insight into the relationship
between the cyclic-containing topology and self-assembling
property, a concise and robust strategy is required for preparing
figure-eight- and tadpole-shaped BCPs possessing varied block
arrangements.
INTRODUCTION
■
Macromolecules possessing cyclic architectures, such as cyclic,
figure-eight-shaped, and tadpole-shaped polymers, are of great
interest due to the endless nature of the cyclic chain, which
endowed them with an increased glass transition temperature,
lower viscosity, and smaller hydrodynamic volume.1−6 In
addition, the cyclic block copolymers (BCPs) provided special
properties in relation to self-assembly, which are typically not
attainable with the corresponding linear counterparts.7−15 For
example, Tezuka and Yamamoto et al. reported an increase in
the thermal stability of micelles formed from cyclic poly-
(ethylene oxide)-block-poly(butyl acrylate) as compared to the
corresponding linear counterpart.7−9 Recently, Hawker et al.
demonstrated the control of a microphase-separated structure
in cylinder-forming poly(ethylene oxide)-block-poly(styrene)
via macrocyclization, which enabled a ca. 30% decrease in the
domain spacing.12 Although many synthetic approaches for
cyclic BCPs have been established,3−5 the synthesis of cyclic-
containing BCPs, such as figure-eight- and tadpole-shaped
BCPs, is still a remaining and challenging task. A
comprehensive sets of such BCPs, i.e., linear, cyclic, figure-
eight-shaped, and tadpole-shaped BCPs, possessing comparable
chemical structures, monomer compositions, and molecular
Among the wide range of BCP systems, polyether-based
amphiphilic BCPs, which are obtained via the ring-opening
polymerization (ROP) of epoxy monomers, have been
recognized as the most traditional but important class of
Received: March 7, 2014
Revised: April 8, 2014
Published: April 16, 2014
© 2014 American Chemical Society
2853
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
Scheme 1. Synthesis of Cyclic, Figure-Eight-Shaped, and Tadpole-Shaped Amphiphilic Block Copolyethers (BCPs) by
Combining the t-Bu-P4-Catalyzed Ring-Opening Polymerization (ROP) and Intramolecular Click Cyclization
BCP materials due to their nonionic character, good chemical
stability, biocompatibility, and low toxicity, and such BCPs
were commercialized for applications as a nonionic surfactant,
functional fluid, emulsifier, and carrier for drug delivery.24−26
Indeed, there has been considerable efforts to investigate the
self-assembly of polyether-based BCPs, such as the poly-
(ethylene oxide) (PEO)/poly(propylene oxide) (PPO), PEO/
poly(butylene oxide) (PBO), and poly(glycidol)/PPO BCP
systems, in both the bulk and solution states during the past
several decades.25,27−30 These elaborated studies have revealed
that the block arrangement, e.g., AB type diblock or ABA and
BAB type triblocks,31,32 as well as macromolecular architec-
tures, e.g., linear and macrocyclic structures,33,34 affect the BCP
self-assembling properties, such as micelle size, morphology,
critical micelle concentration, critical micelle temperature, and
cloud point in aqueous solution. However, there is less
information available about the topological effects on the self-
assembly. The synthesis of polyether-based BCPs possessing a
macrocyclic architecture was achieved by the acetalization of
PEO-b-PPO-b-PEO or PEO-b-PBO-b-PEO triblock copoly-
mers through the reaction of a hydroxyl end group with
dichloromethane under Williamson conditions and high
dilution.33,34 However, this methodology is hard to use for
the synthesis of polyether-based BCPs with figure-eight- and
tadpole-shaped architectures. The intramolecular click reactions
of α-azido,ω-ethynyl- (or α-ethynyl,ω-azido-) functionalized
precursors have emerged as a powerful mean for constructing
cyclic polymers and cyclic BCPs.35,36 Such a highly efficient
method should be expandable for the synthesis of polyether-
based BCPs with figure-eight- and tadpole-shaped architectures
by combining the appropriately designed precursor BCPs with
precisely controlled azido and ethynyl group placements as well
as a fixed molecular weight and monomer composition. Toward
this contribution, we recently demonstrated that 1-tert-butyl-
4,4,4-tris(dimethylamino)-2,2-bis[tris(dimethylamino)-
phosphoranylidenamino]-2Λ5,4Λ5-catenadi(phosphazene) (t-
Bu-P4)-catalyzed ROP was an effective method to synthesize
well-defined polyethers with various end-functionalities from a
wide range of substituted epoxy monomers, which was further
applied for the precise synthesis of linear diblock copo-
lyethers.37−39 In addition, a combination of the t-Bu-P4-
catalyzed ROP and intramolecular click cyclization was found
to be an efficient route for the synthesis of cyclic and figure-
eight-shaped polyethers.40 We now describe a novel approach
to provide systematic sets of figure-eight- and tadpole-shaped
amphiphilic BCPs via the intramolecular click reaction of linear
BCP precursors bearing azido and ethynyl functionalities at
designed positions, together with the corresponding linear and
cyclic counterparts, as shown in Scheme 1. The clickable linear
BCP precursors were precisely synthesized based on the t-Bu-
P4-catalyzed ROP of decyl glycidyl ether (1a) as a hydrophobic
monomer and 2-(2-(2-methoxyethoxy)ethoxy)ethyl glycidyl
ether (1b) as a hydrophilic monomer with the combination
of functional initiators and terminators. We initially examined
the homopolymerization and block copolymerization of 1a and
1b to ensure the well-controlled nature of the t-Bu-P4-catalyzed
ROP system. The intramolecular click reaction of α-azido,ω-
ethynyl-functionalized poly-1a-block-poly-1b was then exam-
ined to produce the cyclic BCP and to provide fundamental
insight into the cyclization conditions. Finally, the combination
of the t-Bu-P4-catalyzed ROP and intramolecular click
cyclization was applied to clickable linear precursors with
various di-, tri-, and pentablock sequences for producing figure-
eight-shaped BCPs with four different block arrangements and
tadpole-shaped BCPs with two different block arrangements. A
preliminary study of the self-assembling properties in water was
performed using the obtained linear, cyclic, figure-eight-shaped,
and tadpole-shaped BCPs.
RESULTS AND DISCUSSION
■
To confirm the polymerization properties of decyl glycidyl
ether (1a) and 2-(2-(2-methoxyethoxy)ethoxy)ethyl glycidyl
ether (1b), the t-Bu-P4-catalyzed ROPs of 1a and 1b were
conducted before the synthesis of the block copolymers. The
polymerizations of 1a and 1b were carried out using 3-phenyl-
1-propanol (2a) as the initiator at the [monomer]0/[2a]0/[t-
Bu-P4]0 ratio of 25/1/1 in toluene at room temperature
(Scheme 2a). The monomer conversions of 1a and 1b reached
>99% within 20 h to produce the narrowly dispersed poly-1a
(Ia) and poly-1b (Ib) with the number-average molecular
weights determined from the NMR (Mn,NMRs) of 5710 and
5750 g mol−1, respectively. The Mns (Mn,clcds) calculated from
2854
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
Scheme 2. (a) Homopolymerization and (b) Block
Copolymerization for Decyl Glycidyl Ether (1a) and 2-(2-(2-
Methoxyethoxy)ethoxy)ethyl Glycidyl Ether (1b)
1
Figure 2. H NMR spectrum of IIa in CDCl3 (400 MHz).
of 1a and 1b with the opposite monomer addition sequence
also produced the corresponding block copolymer, IIb, with
the Mn,NMR of 21 800 g mol−1 and Mw/Mn of 1.05 (Figure 1b).
The intramolecular click cyclization of α-ethynyl,ω-azido
end-functionalized polymers is one of the most successful
methods to effectively produce cyclic polymers.35,36 We also
demonstrated that the combination of the t-Bu-P4-catalyzed
ROP and click cyclization is an effective method to produce
well-defined cyclic polyethers.38,40 Utilizing the established
procedure, the cyclic amphiphilic BCP, III, was synthesized
through three reaction steps as shown in Scheme 3: (1) the
block copolymerization of 1a and 1b initiated from 6-azido-1-
hexanol (2b) to form the α-azido-functionalized poly-1a-block-
poly-1b (3a), (2) the modification of the terminal hydroxyl
group into the propargyl group to form the α-azido,ω-ethynyl-
functionalized poly-1a-block-poly-1b (3b), and (3) the intra-
molecular click cyclization. The t-Bu-P4-catalyzed ROP of 1a
using 6-azido-1-hexanol (2b) as the initiator was performed
with the [1a]0/[2b]0/[t-Bu-P4]0 ratio of 50/1/1, and the
monomer conversion of 1a reached >99% after 20 h of
polymerization. The NMR and SEC measurements for the
aliquot sample verified the formation of the α-azido-function-
alized poly-1a with the number-average molecular weight
(Mn,NMR) of 11 000 g mol−1 and Mw/Mn value of 1.05. The
subsequent addition of 50 equiv of 1b (with respect to 2b) to
the polymerization mixture allowed the block copolymerization
to produce 3a with Mn,NMR of 22 000 g mol−1. The SEC trace of
3a was completely shifted to the higher molecular weight
region as compared to that of the poly-1a obtained by the first
polymerization while retaining the narrow Mw/Mn of 1.05
(Figure 3a). The hydroxyl ω-chain end of 3a was then treated
with propargyl bromide in the presence of sodium hydride to
produce 3b (Mn,NMR = 21 900 g mol−1, Mw/Mn = 1.04). The
quantitative introduction of the propargyl group was verified by
comparing the integration ratio of the signals due to the
methylene adjacent to the azido group (proton a in Figure 3b)
at 3.26 ppm and the methylene adjacent to the ethynyl group
(proton A) at 4.40 ppm. The α-azido,ω-ethynyl-functionalized
linear precursor, 3b, was subjected to the intramolecular click
cyclization with the CuBr/N,N,N′,N″,N″-pentamethyldiethyle-
netriamine (PMDETA) catalyst system in DMF at 120 °C. To
ensure highly diluted conditions, the DMF solution of 3b (40 g
L−1) was slowly (0.3 mL h−1) added to the catalyst solution in
DMF (27.5 mL). After completing the addition, propargyl-
functionalized Wang resin (PSt−CCH)42 was added to the
reaction mixture to completely remove the unreacted 3b, and
the absence of the starting material 3b after the reaction was
confirmed by an FT-IR analysis, in which the absorption at
2100 cm−1 due to the azido group completely disappeared
the initial monomer-to-initiator ratio and monomer conversion
matched the Mn,NMRs of Ia and Ib, indicating the controlled
nature of the t-Bu-P4-catalyzed ROP systems. The NMR and
matrix-assisted laser desorption/ionization mass spectral
analyses of the obtained polymers revealed that the polymer-
izations of 1a and 1b proceeded without side reactions, such as
the transfer reaction to the monomer41 (Table S1, Figures S1−
S3). Based on the results, the linear amphiphilic diblock
copolymers, IIa and IIb, were then synthesized via the
sequential block copolymerization of 1a and 1b, as shown in
Scheme 2b. Unless otherwise noted, the number-average
degree of polymerization (DP) of the targeted BCP (poly-
1a/poly-1b) is adjusted to be around 100 (50/50). A poly-1a
with the Mn,NMR of 10 800 g mol−1 and Mw/Mn of 1.03 was first
prepared by the t-Bu-P4-catalyzed ROP of 1a using 2a as the
initiator with the perfect monomer conversion at the [1a]0/
[2a]0/[t-Bu-P4]0 ratio of 50/1/1. The polymerization was
further continued by the subsequent addition of 50 equiv of 1b
with respect to 2a to afford IIa with the Mn,NMR of 21 900 g
mol−1 and Mw/Mn of 1.04 in 90% yield. The SEC trace of poly-
1a was completely shifted to the high molecular weight region
after the subsequent polymerization of 1b, indicating the
successful block copolymerization of 1a and 1b (Figure 1a).
1
The H NMR spectrum of IIa showed signals due to both the
poly-1a and poly-1b along with the minor signals due to the 3-
phenyl-1-propoxy group at 2.68 and 1.85 ppm, confirming that
the obtained IIa possesses an initiator moiety at the α-chain
end (Figure 2). Noteworthy is that the block copolymerization
Figure 1. SEC traces of (a) IIa and (b) IIb. The dashed lines show the
poly-1a and poly-1b obtained by the first polymerization.
2855
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
Scheme 3. Synthesis of Cyclic Amphiphilic Block Copolyether
Figure 3. (a) SEC traces of 3a (dashed line, poly-1a obtained by the first polymerization; solid line, 3a obtained by the second polymerization), 3b,
1
and III (dashed line, before purification; solid line, after the purification using preparative SEC). (b) H NMR spectra of 3a, 3b, and III in CDCl3
(400 MHz).
(Figure S4). Although ca. 7% of a high molecular weight elution
peak due to byproducts formed by intermolecular click reaction
was observed in the SEC trace, the cyclic diblock copolymer,
III (Mn,NMR = 22 300 g mol−1), was isolated after the
purification using preparative SEC as a light brown waxy
solid in 37.0% yield. The cyclized product III displayed a
unimodal elution peak with the Mw/Mn value of 1.04 (Figure
3a). A decrease in the elution volume was observed from the
starting materials 3b to III (Figure 3a), indicating the formation
of a cyclic product having a hydrodynamic volume smaller than
the linear precursor. The ratio between the Mn,SECs at the peak
top of III and 3b, i.e., Mn,p(cyclic)/Mn,p(linear) = ⟨G⟩, was
determined to be 0.80 (Table S2), which was in good
agreement with the reported value for the monocyclic
poly(butylene oxide) (⟨G⟩ = 0.69−0.8140). This is strong
evidence for the intramolecular cyclization to form the cyclic
polymer. Furthermore, the shrinking factor, g′, based on the
intrinsic viscosities of III ([η]cyclic = 8.0 mL g−1) and 3b
([η]linear = 13.1 mL g−1), i.e., g′ = [η]cyclic/[η]linear = 0.61 (Table
S2), also supported the decrease in the hydrodynamic volume
azido group (proton a; 3.26 ppm), ethynyl proton (proton B;
2.50 ppm), and methylene adjacent to the ethynyl group
(proton A; 4.40 ppm) of 3b were replaced after the click
reaction by a methine proton of the newly formed triazole ring
(proton B′; 7.60 ppm) and two methylene protons adjacent to
the triazole ring (protons a′ and A′; 4.33 and 4.80 ppm). Thus,
the desired cyclic block copolyether, III, was obtained by the
intramolecular click cyclization of the α-azido,ω-ethynyl-
functionalized linear precursor.
The intramolecular click cyclization was further applied to
produce four types of figure-eight-shaped BCPs using the
appropriate linear precursors with azido and ethynyl function-
alities, as depicted in Schemes 4 and 5. First, a triblock
copolymer bearing two azido groups at the chain center, 4a,
was prepared by the t-Bu-P4-catalyzed sequential block
copolymerization of 1a (50 equiv) and following 1b (50
equiv) using 2,2-bis((6-azidohexyloxy)methyl)propane-1,3-diol
(2c) as the initiator. The progress of the block copolymeriza-
tion was followed by SEC measurements, and the elution peak
of the polymer obtained by the first polymerization of 1a
shifted to a higher molecular weight region by the second
polymerization of 1b while retaining a narrow molecular weight
1
due to cyclization. The H NMR comparison of 3b and III
confirmed that the signals due to the methylene adjacent to the
2856
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
Scheme 4. Synthesis of Figure-Eight-Shaped Amphiphilic Block Copolyethers Consisting of the Same Two Cyclic Units
distribution (Figure 4a). The presence of the initiator moiety
was confirmed by the signal due to the methylene proton
adjacent to the azido group (proton a in Figure 4b). In a similar
fashion, the pentablock copolymers, 5a and 6a, were prepared
by the sequential block copolymerizations using 2c as the
initiator with the following monomer addition sequence: 25
equiv of 1a → 50 equiv of 1b → 25 equiv of 1a for 5a and 25
equiv of 1b → 50 equiv of 1a → 25 equiv of 1b for 6a. The
SEC results clearly demonstrated the progress of the sequential
block copolymerizations (Figure S5). For all the block
copolymerizations, the full conversion of the monomer was
ensured by the NMR measurement of the aliquot of the
reacting mixture before adding the next monomer. The Mn,NMR
values of 4a, 5a, and 6a (21 800, 22 100, and 22 000 g mol−1 for
4a, 5a, and 6a, respectively) well agreed with the predicted
value (22100 g mol−1), and the Mw/Mn values were in the
range of 1.04−1.08. The treatment of 4a, 5a, and 6a with
propargyl bromide in the presence of sodium hydride produced
4b, 5b, and 6b, which was confirmed by the signals due to the
ethynyl proton (proton B in Figure 4b) and methylene proton
adjacent to the ethynyl group (proton A). The α,α′-
diazido,ω,ω′-diethynyl linear BCP precursors 4b, 5b, and 6b
were then subjected to the click cyclization using the same
conditions for the synthesis of III. The figure-eight-shaped
BCPs, IV (Mn,NMR = 21 800 g mol−1, Mw/Mn = 1.06), V
(Mn,NMR = 22 300 g mol−1, Mw/Mn = 1.03), and VI (Mn,NMR
=
22 200 g mol−1, Mw/Mn = 1.03), were obtained as a light brown
waxy solid by purification using an alumina column and
preparative SEC in 39.6−46.0% yields. In the SEC traces of the
crude products, ca. 14%, 16%, and 21% of high molecular
weight byproducts were detected for the click cyclization of 4b,
5b, and 6b, respectively. The 1H NMR analyses of the isolated
figure-eight-shaped BCPs confirmed the absence of the azido
and ethynyl groups as well as the formation of triazole rings
1
(Figure 4b, Figures S6 and S7). For example, the H NMR
spectrum of IV exhibited a signal due to the triazol methine
(proton B′) at 7.56 ppm and methylenes adjacent to the
triazole ring (proton a′ and A′) at 4.33 and 4.79 ppm, while the
signals due to the propargyl group and methylene adjacent to
the azido group disappeared (Figure 4b). The IR analysis also
confirmed the absence of the azido group in the isolated
products (Figure S8). Moreover, the cyclized structures for the
2857
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
Scheme 5. Synthesis of Figure-Eight-Shaped Amphiphilic Block Copolyethers Consisting of Two Different Cyclic Units
Figure 4. (a) SEC traces of 4a (dashed line, poly-1a obtained by the first polymerization; solid line, 4a obtained by the second polymerization), 4b,
and 4c (dashed line, before purification; solid line, after the purification by preparative SEC). (b) 1H NMR spectra of 4a, 4b, and 4c in CDCl3 (400
MHz).
obtained BCPs were established by the longer elution time in
the SEC trace as compared to the linear precursors as well as
the decrease in the intrinsic viscosity. For example, the SEC
trace shifted to the lower molecular weight region after the click
2858
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
Figure 5. (a) SEC traces of 7c, 7d (dashed line, before purification; solid line, after the purification using preparative SEC), 7e, and VII (dashed line,
1
before purification; solid line, after the purification using preparative SEC). (b) H NMR spectra of 7c, 7d, 7e, and VII in CDCl3 (400 MHz).
reaction of 4b to form IV (Figure 4a), and the ⟨G⟩ value was
calculated to be 0.80 (Table S3), which was in good agreement
with that of the figure-eight-shaped poly(butylene oxide) (⟨G⟩
= 0.65−0.8340). The g′ values of IV, V, and VI were in the
range of 0.46−0.56 (Table S3) and were lower than that of III,
indicating the more compact conformations of IV, V, and VI
than the monocyclic III. Thus, the three types of figure-eight-
shaped BCPs with different distributions of hydrophilic and
hydrophobic blocks were obtained by the double click
cyclization of the multiblock type precursors having two
ethynyl groups at the ω-chain ends and two azido groups at the
chain center.
The next target is the figure-eight-shaped BCP consisting of
two different cyclic units, VII, which is in contrast to the above-
prepared figure-eight-shaped BCPs having the same two cyclic
units. Our synthetic strategy for VII is based on the double
click cyclization of a linear diblock copolymer of 1a and 1b
bearing two ethynyl groups at the ω-chain end and two azido
groups at the α-chain end and the chain center (7e), as shown
in Scheme 5a. To synthesize the diblock copolymer having two
azido groups at the α-chain end and the chain center (7d), we
first prepared the α,ω-diazido,ω-hydroxyl poly-1a (7b), in
which the hydroxyl group would be used as the initiation site
for the subsequent chain extension. We newly designed and
synthesized 1-(((1-azido-3-(1-ethoxyethoxy)propan-2-yl)oxy)-
methyl)-4-(bromomethyl)benzene (2d) as the terminator for
introducing both the azido and hydroxyl groups at the ω-chain
end. 2d was prepared by the ring-opening of ethoxyethyl
glycidyl ether using sodium azide in the presence of ammonium
chloride followed by the treatment with an excess amount of
α,α-dibromo-p-xylene, as shown in Scheme 5b. An α-azido end-
functionalized poly-1a (7a, Mn,NMR = 10 900, Mw/Mn = 1.02)
was treated with 2d in the presence of sodium hydride, and the
ethoxyethyl group was then deprotected under acidic
1
conditions to give 7c. The H NMR spectrum of 7c showed
the signals due to the benzyl protons (protons A and D in
Figure 5b) at 4.70 ppm as well as methylene protons adjacent
to the azido group (protons a) at 3.26 ppm, from which the
quantitative introduction of an azido and a hydroxyl group at
the ω-chain end was verified. After a rigorous dehydration, 7c
was utilized as a macroinitiator for the synthesis of a diblock
copolymer of 1a and 1b with two azido groups at the α-chain
end and the junction point of the two blocks, 7d. The
polymerization was carried out at the [1b]0/[7c]0/[t-Bu-P4]0
ratio of 50/1/1 to give 7d in 77.7% yield. The SEC trace of 7c
shifted to the higher molecular weight region after the
polymerization (Figure 5a), which confirmed that the polymer-
ization was initiated from the hydroxyl group of 7c. The Mn,NMR
and Mw/Mn of the isolated 7d were determined to be 22 000 g
mol−1 and 1.06, respectively. 7d was then further converted to
2859
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
Scheme 6. Synthesis of Tadpole-Shaped Amphiphilic Block Copolyethers
the requisite linear precursor, 7e (Mn,NMR = 22 100 g mol−1,
Mw/Mn = 1.05), by the treatment with 1-bromomethyl-3,5-
bis(2-propynyloxy)benzene (2e) in the presence of sodium
hydride. The double click cyclization of 7e and SEC
fractionation led to the desired VII with the Mn,NMR and Mw/
Mn of 22 200 g mol−1 and 1.04, respectively, in 38.7% yield.
Here, ca. 38% of high molecular weight byproducts formed
through the intermolecular click reaction were detected in the
SEC trace of the crude product. The 1H NMR and IR
spectroscopic analyses proved that there were no unreacted
azido and ethynyl groups (Figure 5b and Figure S8), implying
that the obtained product should have a doubly cyclized
BCPs consisting of a hydrophilic cyclic unit and a hydrophobic
tail, VIII (Mn,NMR = 21 900 g mol−1, Mw/Mn = 1.04), in 56.2%
yield. In a similar fashion, the tadopole-shaped BCPs consisted
of a hydrophobic cyclic unit and a hydrophilic tail, IX (Mn,NMR
= 22 400 g mol−1, Mw/Mn = 1.06), was obtained in 32.7% yield.
Although 23% and 20% of byproducts were formed during the
cyclization reaction of 8c and 9c, respectively, as indicated by
the SEC traces of the crude products, these were removed
through the SEC fractionation process. The chemical structures
of VIII and IX were confirmed by IR and NMR analyses, in
which the disappearances of the azido and ethynyl groups as
well as the formation of the triazol ring were observed (Figures
S9−S11). The cyclized architectures for VIII and IX were
further verified by the ⟨G⟩ values (0.88 for VIII and 0.91 for
IX; Figure S12 and Table S4) as well as the g′ values (0.88 for
VIII and 0.68 for IX, Table S4).
1
structure. In addition, the H NMR spectrum of VII showed
two signals due to the triazole methine protons (proton J′ in
Figure 5b) at 7.76 and 7.64 ppm. The increase in the SEC
elution volume (Figure 5a; ⟨G⟩ = 0.75) as well as decrease in
the intrinsic viscosity (g′ = 0.60) also provided evidence of the
doubly cyclized architecture of the product. Therefore, the
synthesis of a series of figure-eight-shaped amphiphilic BCPs
with four different types of hydrophobic/hydrophilic orienta-
tions has been achieved via the double click cyclization of
diazido,diethynyl-functionalized linear precursors.
For the preliminary study on regarding effect of the cyclic
architecture, we investigated the self-assembly of the linear
(IIa), cyclic (III), figure-eight-shaped (IV, V, VI, and VII), and
The synthesis of the tadpole-shaped BCPs were attempted as
illustrated in Scheme 6. Poly-1a having a 3-phenyl-1-propoxy α-
chain end (Ia′, Mn,NMR = 11 100 g mol−1, Mw/Mn = 1.05) was
modified by the treatment with 2d and following deprotection
of the ethoxyethyl group, providing ω-azido,ω-hydroxyl-
functionalized poly-1a, 8a. The t-Bu-P4-catalyzed ROP of 1b
using 8a as a macroinitiator followed by treatment with
propargyl bromide led to the poly-1a-block-poly-1b with an
ethynyl group at the ω-chain end and an azido group at the
junction point between the two blocks, 8c (Mn,NMR = 22 400 g
mol−1, Mw/Mn = 1.05). An analogue of 8c with the opposite
block sequence, 9c (Mn,NMR = 22 100 g mol−1, Mw/Mn = 1.05),
was prepared in a similar fashion, beginning with poly-1b
having the 3-phenyl-1-propoxy α-chain end (Ib′, Mn,NMR = 10
900 g mol−1, Mw/Mn = 1.06). The click cyclization of 8c and
following SEC fractionation gave the desired tadopole-shaped
Figure 6. Temperature dependence of optical transmittance at 300 nm
obtained for 0.50 g L−1 aqueous solution of linear (IIa), cyclic (III),
figure-eight-shaped (IV, V, VI, and VII), and tadpole-shaped BCPs
(VIII and IX).
2860
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
Table 1. Molecular Characteristic, Critical Micelle Concentration (CMC), Hydrodynamic Diameter (2Rh), and Cloud Point
(Tc) of Linear (IIa), Cyclic (III), Figure-Eight-Shaped (IV, V, VI, and VII), and Tadpole-Shaped BCPs (VIII and IX)
a
b
c
d
e
f
sample
architecture
linear
Mn,NMR (g mol−1
)
Mw/Mn
DPa/DPb
CMC (mg L−1
)
2Rh (nm)
Tc (°C)
IIa
III
IV
21 900
22 300
21 800
22 300
22 200
22 200
21 900
22 400
1.04
1.04
1.06
1.03
1.03
1.04
1.04
1.06
50/51
51/50
50/48
52/48
50/50
50/50
51/49
51/50
1.4
1.8
1.5
1.0
1.4
1.2
1.4
3.4
123
166
138
179
171
221
83
83
87
75
80
67
85
cyclic
figure-eight shape
V
VI
VII
VIII
IX
g
tadpole shape
n.d.
170
59
a
b
c
Determined by 1H NMR. Determined by SEC in THF. Number-average degree of polymerizations of decyl glycidyl ether (DPa) and triethylene
d
glycol methyl glycidyl ether (DPb) in the copolymer were determined by 1H NMR. Determined by steady-state fluorescence method using pyrene
e
f
as a prove at 25 °C. Determined based on multiangle DLS measurements (concentration, 0.50 g L−1; temperature, 25 °C). Determined by
g
turbidimetric analysis (concentration, 0.50 g L−1). Tc was defined by the temperature at which the transmittance of sample solution reached 50%. Tc
was not observed upon the heating up to 90 °C.
tadpole-shaped (VIII and IX) amphiphilic BCPs in water. The
micelle solutions (concentration 0.50 g L−1) were prepared by
the direct dissolution of the BCPs in pure water at room
temperature by sonication. The dynamic light scattering (DLS)
experiment was employed to investigate the size of the
aggregates (Figure S13). The intensity-average size distribution
of every BCP aqueous solution displayed a monomodal peak
corresponding to the aggregates formed through the self-
assembling process. Multiangle DLS measurements for the
solutions revealed the linear dependence of the relaxation
frequency (Γ = 1/τ) on the square of the wavevector (q2),
which clearly indicated the Brownian diffusive motion (see
Supporting Information). The slope is equal to the diffusion
coefficient (D) of the aggregates in water, from which the
hydrodynamic diameter (2Rh) was calculated by the Stokes−
Einstein relation and was in the range of 83−221 nm (Table 1).
The Rh values of the BCPs were apparently greater than those
of the fully extended chain lengths (approximately 36 nm for
linear BCP, 18 nm for cyclic and figure-eight-shaped BCPs, 27
nm for tadpole-shaped BCPs, in which the DP and molecular
length of a monomer unit were assumed to be 100 and 0.358
nm,43 respectively), suggesting that the aggregates of every
BCP were large compound micelles or vesicles. The actual
morphologies of the aggregates were then confirmed by
transmission electron microscopy (TEM), in which all of the
samples were observed without staining. The TEM images
revealed the presence of hard-sphere-like nanoparticles or its
agglomerates in all case (Figure S14), leading to the conclusion
that all of the aggregates consisted of large compound micelles.
Therefore, the cyclic architecture of the BCP as well as the
block arrangement would have a small impact on the
morphology of the aggregate in water. The critical micelle
concentrations (CMC) for the BCPs in water were determined
by a fluorescence technique using pyrene as the fluorescence
probe at 25 °C (Figure S15). There is no distinctive difference
in the CMC values among the BCPs (1.0−3.4 mg L−1, Table
1), indicating that the CMC value is not significantly affected
by the cyclic topology as well as the block arrangement. Booth
et al. reported that the CMCs of cyclic poly(ethylene oxide)-
block-poly(propylene oxide) and its linear precursor were very
similar.34 The cloud point (Tc) for the aggregates, a measure of
their thermal stability,42 were determined by the variable-
temperature UV−vis absorption measurements of the aqueous
BCP solutions. The transmittance at 300 nm was monitored
during the continuous heating at the 1.0 °C min−1 heating ratio,
and Tc was defined as the temperature at which the
transmittance of the sample solution reached 50%. Every
BCP, except for VIII, exhibited the characteristic phase
transition phenomenon at the Tc ranging from 59 to 87 °C,
whereas VIII exhibited no such phase transition up to 90 °C.
Although the difference in the polymer structures among the
BCPs is only the macromolecular architecture, a significant
variation in Tc was observed, which should provide an
interesting insight into the thermal stabilities of the aggregates
depending on the cyclic architecture. The cyclic amphiphilic
BCP III showed a Tc at 87 °C, which was 4 °C higher than that
of the linear counterpart IIa (Tc = 83 °C). The increased Tc
value of III with respect to IIa should be ascribed to the
restricted intermicelle linking due to the looped conformation
of the hydrophilic block at the shell region. In contrast, the
figure-eight-shaped BCP IV, which can be regarded as a dimeric
analogue of III, showed a lower Tc value of 75 °C. A significant
difference in the Tc was observed between VIII (Tc > 90 °C)
and IX (Tc = 59 °C), though both of them were categorized as
having a tadpole-shaped architecture. Given that the aggregate
of VIII and IX is surrounded by the shell of hydrophilic cyclic
and linear poly-1b, respectively, these results suggest that the
cyclic architecture of the shell-forming block contributes to the
increase in the Tc by reducing the intermicelle linking as a
consequence of their restricted chain mobility and lower
entanglement nature. These results also implied that the cyclic
architecture of the core-forming block led to a decrease in the
Tc with respect to the corresponding linear diblock counterpart.
The similar Tc values among the figure-eight-shaped BCPs
consisting of discrete hydrophobic and hydrophilic cyclic units,
VII and IIa, are understandable due to the balancing out of two
such effects. On the other hand, the lower Tc values for the
figure-eight-shaped BCPs, IV, V, and VI, with respect to IIa
could not be explained by such effects due to their complicated
chain packing at the interface of the micelle core and shell.
Given that V and VI are regarded as the analogue of the poly-
1b-block-poly-1a-block-poly-1b and poly-1a-block-poly-1b-block-
poly-1a triblock copolymers, respectively, an interesting
comparison can be made between their Tc values. The
relatively low Tc values of the triblock copolymers with
hydrophobic end blocks with respect to those of the opposite
block arrangement were reported for the poly(ethylene oxide)/
poly(propyrene oxide) and poly(ethylene oxide)/poly(butylene
oxide) triblock copolymer systems. This effect is related to the
intermicelle linking through the hydrophobic end blocks.32
2861
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
(4) Laurent, B. A.; Grayson, S. M. Chem. Soc. Rev. 2009, 38, 2202−
2213.
(5) Jia, Z.; Monteiro, M. J. J. Polym. Chem., Part A: Polym. Chem.
Therefore, the lower Tc value of VI than that of V should be
ascribed to the enhanced intermicelle linking due to the
hydrophobic end block.
2012, 50, 2085−2097.
(6) Kricheldorf, H. R. J. Polym. Chem., Part A: Polym. Chem. 2010, 48,
251−284.
CONCLUSION
■
We have demonstrated the synthesis of systematic sets of
figure-eight- and tadpole-shaped amphiphilic BCPs together
with the corresponding linear and cyclic BCP counterparts via
the t-Bu-P4-catalyzed ROP of hydrophobic and hydrophilic
glycidyl ethers. The well-controlled nature of the ROP system
allowed us to attain a series of azido- and ethynyl-functionalized
di-, tri-, and multiblock copolymers with a predicted molecular
weight, monomer composition, and narrow molecular weight
distribution, which led to produce the corresponding cyclic-
containing amphiphilic BCPs via the intramolecular click
cyclization. Our preliminary studies on the self-assembly of
the linear, cyclic, figure-eight-shaped, and tadpole-shaped
amphiphilic BCPs in water revealed the significant variation
in the Tc values depending on the cyclic topology, though there
were small topological effects on their CMC value and
morphology of the aggregates. However, further detailed
investigations should be necessary to fully understand the
aqueous self-assembling properties associated with the cyclic
architectures. Considering the functional group loading capacity
of the glycidyl ether-based BCPs, the present synthetic strategy
provides a wide array of BCP systems with cyclic-containing
architectures, which would be used as a model system for
investigating the topological effect of the BCP self-assembly in
both the solution and bulk states. On the basis of this versatile
synthetic strategy, we are currently trying to investigate the
topological effects on the microphase separation leading to
various nanostructures.
(7) Honda, S.; Yamamoto, T.; Tezuka, Y. J. Am. Chem. Soc. 2010,
132, 10251−10253.
(8) Honda, S.; Yamamoto, T.; Tezuka, Y. Nat. Commun. 2013, 4,
1574.
(9) Heo, K.; Kim, Y. Y.; Kitazawa, Y.; Kim, M.; Jin, K. S.; Yamamoto,
T.; Ree, M. ACS Macro Lett. 2014, 3, 233−239.
(10) Zhu, Y.; Gido, S. P.; Iatrou, H.; Hadjichristidis, N.; Mays, J. W.
Macromolecules 2003, 36, 148−152.
(11) Takano, A.; Kadoi, O.; Hirahara, K.; Kawahara, S.; Isono, Y.;
Suzuki, J.; Matsushita, Y. Macromolecules 2003, 36, 3045−3050.
(12) Poelma, J. E.; Ono, K.; Miyama, D.; Aida, T.; Satoh, K.; Hawker,
C. J. ACS Nano 2012, 6, 10845−10854.
(13) Iatrou, H.; Hadjichristidis, N.; Meier, G.; Frielinghaus, H.;
Monkenbusch, M. Macromolecules 2002, 35, 5426−5437.
(14) Borsali, R.; Minatti, E.; Putaux, J.-L.; Schappacher, M.; Deffieux,
A.; Viville, P.; Lazzaroni, R.; Narayanan, T. Langmuir 2003, 19, 6−9.
(15) Minatti, E.; Viville, P.; Borsali, R.; Schappacher, M.; Deffieux, A.;
Lazzaroni, R. Macromolecules 2003, 36, 4125−4133.
(16) Beinat, S.; Schappacher, M.; Deffieux, A. Macromolecules 1996,
29, 6737−6743.
(17) Dong, Y.-Q.; Tong, Y.-Y.; Dong, B.-T.; Di, F.-S.; Li, Z.-C.
Macromolecules 2009, 42, 2940−2948.
(18) Wang, G.; Hu, B.; Fan, X.; Zhang, Y.; Huang, J. J. Polym. Chem.,
Part A: Polym. Chem. 2012, 50, 2227−2235.
(19) Wang, X.; Liu, T.; Liu, S. Biomacromolecules 2011, 12, 1146−
1154.
(20) Shi, G.-Y.; Sun, J.-T.; Pan, C.-Y. Macromol. Chem. Phys. 2011,
212, 1305−1315.
(21) Hatakeyama, F.; Yamamoto, T.; Tezuka, Y. ACS Macro Lett.
2013, 2, 427−431.
(22) Fan, X.; Huang, B.; Wang, G.; Huang, J. Macromolecules 2012,
45, 3779−3786.
ASSOCIATED CONTENT
■
S
* Supporting Information
(23) Lonsdale, D. E.; Monteiro, M. J. J. Polym. Chem., Part A: Polym.
Chem. 2011, 49, 4603−4612.
Experimental section and additional data (1H NMR, MALDI-
TOF MS, IR, SEC, DLS, TEM, and CAC data). This material is
available free of charge via the Internet at http://pubs.acs.org.
(24) Mark, H. F., Bikales, N. M., Overberger, C. G., Menges, G.,
Kroschwitz, J. I., Eds.; Encyclopedia of Polymer Science and Engineering;
John Wiley & Sons, Inc.: New York, 1985; Vol. 6.
(25) Schmolka, I. R. J. Am. Oil Chem. Soc. 1977, 54, 110−116.
(26) Gupta, S.; Tyagi, R.; Parmar, V. S.; Sharma, S. K.; Haag, R.
Polymer 2012, 53, 3053−3078.
AUTHOR INFORMATION
Corresponding Authors
*E-mail satoh@poly-bm.eng.hokudai.ac.jp (T.S.).
*E-mail kakuchi@poly-bm.eng.hokudai.ac.jp (T.K.).
■
(27) Booth, C.; Attwood, D. Macromol. Rapid Commun. 2000, 21,
501−527.
(28) Brocas, A.-L.; Gervais, M.; Carlotti, S.; Pispas, S. Polym. Chem.
2012, 3, 2148−2155.
Notes
The authors declare no competing financial interest.
(29) Mai, S.-M.; Fairclough, J. P. A.; Terrill, N. J.; Turner, S. C.;
Hamley, I. W.; Matsen, M. W.; Ryan, A. J.; Booth, C. Macromolecules
1998, 31, 8110−8116.
ACKNOWLEDGMENTS
■
This work was financially supported by the MEXT program
“Strategic Molecular and Materials Chemistry through
Innovative Coupling Reactions” of Hokkaido University, the
MEXT Grant-in-Aid for Challenging Exploratory Research
(Grant 25620089), and the MEXT Grant-in-Aid for Scientific
Research on Innovative Areas “Advanced Molecular Trans-
formation by Organocatalysts”. T.I. was funded by the JSPS
Fellowship for Young Scientists.
(30) Halacheva, S.; Rangelov, S.; Tsvetanov, C. Macromolecules 2006,
39, 6845−6852.
(31) Yu, G.-E.; Yang, Y.-W.; Yang, Z.; Attwood, D.; Booth, C.
Langmuir 1996, 12, 3404−3412.
(32) Altinok, H.; Yu, G.-E.; Nixon, S. K.; Gorry, P. A.; Attwood, D.;
Booth, C. Langmuir 1997, 13, 5837−5848.
(33) Yu, G.-E.; Yang, Z.; Attwood, D.; Price, C.; Booth, C.
Macromolecules 1996, 29, 8479−8486.
(34) Yu, G.-E.; Garrett, C. A.; Mai, S.-M.; Altinok, H.; Attwood, D.;
Price, C.; Booth, C. Langmuir 1998, 14, 2278−2285.
(35) Laurent, B. A.; Grayson, S. M. J. Am. Chem. Soc. 2006, 128,
4238−4239.
REFERENCES
■
(1) Hadjichristidis, N.; Pitsikalis, M.; Pispas, S.; Iatrou, H. Chem. Rev.
2001, 101, 3747−3792.
(36) Lansdale, D. E.; Bell, C. A.; Monteiro, M. J. Macromolecules
2010, 43, 3331−3339.
(37) Misaka, H.; Sakai, R.; Satoh, T.; Kakuchi, T. Macromolecules
2011, 44, 9099−9107.
(2) Hadjichristidis, N.; Iatrou, H.; Pitsikalis, M.; Mays, J. Prog. Polym.
Sci. 2006, 31, 1068−1132.
(3) Yamamoto, T.; Tezuka, Y. Polym. Chem. 2011, 2, 1930−1941.
2862
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863

Macromolecules
Article
(38) Misaka, H.; Tamura, E.; Makiguchi, K.; Kamoshida, K.; Sakai,
R.; Satoh, T.; Kakuchi, T. J. Polym. Sci., Part A: Polym. Chem. 2012, 50,
1941−1952.
(39) Kwon, W.; Roh, Y.; Kamoshida, K.; Kwon, K. H.; Jeong, Y. C.;
Kim, J.; Misaka, H.; Shin, T. J.; Kim, J.; Kim, K.-W.; Jin, K. S.; Chang,
T.; Kim, H.; Satoh, T.; Kakuchi, T.; Ree, M. Adv. Funct. Mater. 2012,
22, 5194−5208.
(40) Isono, T.; Kamoshida, K.; Satoh, Y.; Takaoka, T.; Sato, S.-i.;
Satoh, T.; Kakuchi, T. Macromolecules 2013, 46, 3841−3849.
(41) Hans, M.; Keul, H.; Moeller, M. Polymer 2009, 50, 1103−1108.
(42) Rao, J.; Zhang, Y.; Zhang, J.; Liu, S. Biomacromolecules 2008, 9,
2586−2593.
(43) Rixman, M. A.; Dean, D.; Ortiz, C. Langmuir 2003, 19, 9357−
9372.
2863
dx.doi.org/10.1021/ma500494e | Macromolecules 2014, 47, 2853−2863