Supramolecular X-shaped homopolymers and block polymers by midsegment complementary hydrogen bonds: Design of bifunctional initiators with interactive sites for metal-catalyzed living radical polymerization

Article abstract of DOI:10.1021/ma300479s

Two polymer chains, each carrying a complementary multiple hydrogen-bonding site at the center, led to supramolecular block copolymers where two segments are connected in an X-shaped configuration. For the precision synthesis of the pairing precursor polymers, a set of bifunctional initiators were designed, X-DAD-X and Y-ADA-Y, that consist of three hydrogen bond-forming sites in series (DAD or ADA; D = hydrogen donor; A = hydrogen acceptor) and two initiating sites (X or Y) for ruthenium-catalyzed living radical polymerization; thereby, DAD and ADA units are complementary to each other upon supramolecular association. Typically, DAD is of an -NH-C-N-C-NH- unit derived from 2,6-diaminopyridine, and ADA is of a -CO-NH-CO- unit from 2′-deoxyuridine. These functionalized initiators initiated living radical polymerizations of methyl methacrylate and styrene to give homopolymers (PMMA and PS, respectively) with a DAD or an ADA unit at the backbone center. Upon mixing in CDCl ₃, for example, a pair of complementary polymers, such as PMMA-DAD-PMMA and PS-ADA-PS, formed supramolecular X-shaped block polymers via the multiple hydrogen bond association between their midchain DAD and ADA sites, as demonstrated by in situ ¹H NMR (association constant ～10²). ¹H DOSY analysis also supported the interaction: the association products gave a single diffusion constant, thus behaving as one polymer chain, while each component chain had different diffusion constants.

Full text of DOI:10.1021/ma300479s

Article
pubs.acs.org/Macromolecules
Supramolecular X-Shaped Homopolymers and Block Polymers by
Midsegment Complementary Hydrogen Bonds: Design of
Bifunctional Initiators with Interactive Sites for Metal-Catalyzed
Living Radical Polymerization
Sang-Ho Lee, Makoto Ouchi,* and Mitsuo Sawamoto*
Department of Polymer Chemistry, Graduate School of Engineering, Kyoto University, Katsura, Nishikyo-ku, Kyoto 615-8510, Japan
S
* Supporting Information
ABSTRACT: Two polymer chains, each carrying a comple-
mentary multiple hydrogen-bonding site at the center, led to
supramolecular block copolymers where two segments are
connected in an X-shaped configuration. For the precision
synthesis of the pairing precursor polymers, a set of
bifunctional initiators were designed, X-DAD-X and Y-ADA-
Y, that consist of three hydrogen bond-forming sites in series
(DAD or ADA; D = hydrogen donor; A = hydrogen acceptor)
and two initiating sites (X or Y) for ruthenium-catalyzed living
radical polymerization; thereby, DAD and ADA units are
complementary to each other upon supramolecular association. Typically, DAD is of an −NH−C−N−C−NH− unit derived
from 2,6-diaminopyridine, and ADA is of a −CO−NH−CO− unit from 2′-deoxyuridine. These functionalized initiators initiated
living radical polymerizations of methyl methacrylate and styrene to give homopolymers (PMMA and PS, respectively) with a
DAD or an ADA unit at the backbone center. Upon mixing in CDCl3, for example, a pair of “complementary” polymers, such as
PMMA-DAD-PMMA and PS-ADA-PS, formed supramolecular X-shaped “block” polymers via the multiple hydrogen bond
1
1
association between their midchain DAD and ADA sites, as demonstrated by in situ H NMR (association constant ∼102). H
DOSY analysis also supported the interaction: the association products gave a single diffusion constant, thus behaving as one
polymer chain, while each component chain had different diffusion constants.
induce specific aggregation behaviors (Scheme 1). A simple
strategy is to attach a pair of complementary interactive sites at
terminals of two different homopolymers (Scheme 1A).2 Their
blends (e.g., homopolymers A and B) led to microphase
separation akin to those in conventional covalently linked
diblock copolymers (A−B) of the same homopolymer
segments. As first demonstrated by Meijer,3 a strong dynamic
interaction of an extremely large association constant, such as
multiple complementary hydrogen bonds, generates “covalent
bond-like” linkages between end-functionalized polymeric
segments that in turn form supramolecular polymers behaving
nearly identically to the corresponding diblock and multiple
block copolymers. Related variants of this approach includes
random incorporation with interactive monomers (via pendent
groups) for homopolymers (Scheme 1B)4 or block copolymers
(Scheme 1C)5 to induce specific aggregation, though specially
designed noncommon monomers are required.
INTRODUCTION
■
In nature, the recognition between macromolecular chains
through selective noncovalent bond interaction is essential to
the “central dogma”, i.e., the replication, transcription, and
expression of genes, peptides, and other biomacromolecules. A
typical example is of course a DNA double helix, in which two
polymeric strands interacts with each other through a series of
complementary hydrogen bonds. The elegant shapes, precise
higher-order architecture, and the superb functions of natural
polymers tell us the importance of selective and dynamic
interactions at well-defined positions between macromolecular
chains.
In synthetic polymers, on the other hand, intermolecular
interaction between different homopolymers is more or less
repulsive, rather than attractive, and thereby they are generally
incompatible and undergo macrophase separation. Block
copolymers do form macroscopically homogeneous microphase
separated structures (lamella, rod, cylinder, etc.),1 but the
patterns are limited and strongly dependent on segment
combination and composition. Equally serious, the intermo-
lecular interaction therein is mostly static and nonresponsive to
external stimuli.
Despite successful formation of supramolecular and dynamic
polymers, however, the structural designs in these examples are
yet inferior to those for natural polymers. In particular, the
Received: March 8, 2012
Revised: April 8, 2012
Published: April 19, 2012
Over the past few years, efforts have been directed to
introduce dynamic interactive sites into polymer chains to
© 2012 American Chemical Society
3702
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710

Macromolecules
Article
Scheme 1. Previous Researches on Interactions between Polymers through Introduction of Dynamic Interaction Sites on the
Chains
Scheme 2. Bifunctional Initiators for Living Radical Polymerization Carrying Complementary H-Bond and Polymers from the
Initiators
Scheme 3. Living Radical Polymerization with 2,6-Diamide Pyridine-Based Chlorine Initiator of a Donor (D)−Acceptor (A)
−Donor (D) Configuration (Cl-DAD-Cl) and Thymine-Based Bromine One of an Acceptor (A)−Donor (D)−Acceptor (A)
Counterpart (Br-ADA-Br)
3703
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710

Macromolecules
Article
purified by passing through purification columns (Solvent Dispensing
System, Glass Contour) and bubbled with dry nitrogen for more than
15 min immediately before use. n-Monobutylamine (TCI; >98%) was
degassed by reduced pressure before use. n-Octane and tetralin
(internal standard for gas chromatography) were dried over calcium
chloride and distilled from calcium hydride.
complementary interactive sites are mostly at chain termini or
at randomly placed pendent groups, while those in natural
polymers are built in far more precisely in both position and
sequence. Precise introduction of interactive sites at well-
defined positions on synthetic polymers should lead to more
advanced and versatile regular structures derived from effective
and unique self-assembling.
Synthesis of Cl-DAD-Cl Initiator. Dry dichloromethane (130
mL) and triethylamine (7.4 mL, 0.053 mol, 2.5 equiv) and were mixed,
sequentially in this order, to 2,6-diaminopyridine (2.3 g, 0.020 mol, 1.0
equiv) placed in an argon-filled 250 mL round-bottom flask. The
solution was cooled to 0 °C, and α-chlorophenylacetyl chloride (8.0
mL, 0.050 mol, 2.4 equiv) was added slowly under an argon
atmosphere. The solution was allowed to reach room temperature and
stirred for overnight. An excess amount of distilled water (200 mL)
was introduced to quench the reaction, and the organic layer was
washed with NaHCO3 saturated solution (100 mL, twice). The
organic layer was dried over MgSO4 for several hours, and the filtered
solution was evaporated, followed by silica gel chromatography with
ethylacetate/hexane mixture eluent. Finally, the product was purified
by recrystallization with dichloromethane and hexane mixture (Cl-
To approach this idea, our strategy is directed to design
bifunctional initiators for living radical polymerization that carry
a complementary hydrogen-bond units (X∼∼D∼∼X and
X∼∼A∼∼X; X: initiating site; D and A: complementary H-
bond units; ∼∼: polymer chain) (Scheme 2). Given
quantitative initiation from the two X sites, such initiators
would generate “well-defined” polymers bearing an in-chain
interactive site at the middle or center of a backbone.
Additionally, sequential block copolymerization from these
initiators would give similar central functions now embedded
into an ABA-block architecture, which would induce cooper-
ative self-aggregation by hydrogen bonds and by microphase
separation inherent of block copolymers. Alternatively, periodic
introduction of interactive units along the main chain in a
regular interval might also be achieved through the chain
extension of homotelechlic and midchain functionalized
oligomers with narrow molecular weight distributions
(MWDs): e.g., nX∼∼D∼∼X → −(∼∼D∼∼)n−.
Thus, in this paper, we report the design of a sets of
bifunctional initiators (Cl-DAD-Cl and Br-ADA-Br; Scheme 3)
for metal-catalyzed living radical polymerization6 in which three
complementary hydrogen-bonding sites are embedded in
series: Cl-DAD-Cl is a 2,6-diamide pyridine-based dichloride
of a donor (D)-acceptor (A)-donor (D) configuration and Br-
ADA-Br is a thymine-based dibromide of an acceptor (A)-donor
(D)-acceptor (A) configuration.7 Obviously, metal-catalyzed
living radical polymerization from these initiators will give
various well-defined homopolymers and block polymers with
midchain complementary hydrogen-bonding sites through
which interesting supramolecular assembly would form.
Herein we present the precision control of radical polymer-
izations with these designed initiators as well as the
complementary interaction between resultant homopolymers
of methyl methacrylate (MMA) and styrene (St). Quite
recently, Lourtie and Bernard have reported a similar design
of bifunctional initiators for reversible addition−fragmentation
chain transfer polymerization,8 but ours is perhaps the first
example for metal-catalyzed living radical polymerization or
atom transfer radical polymerization. Note that our focus is to
introduce interaction sites at well-defined internal positions on
different polymer chains for the effective interaction without
changing the main polymer structures as well as to synthesize
supramolecular miktoarm star polymers (or X-shaped block
polymers).
1
DAD-Cl: 4.7 g, 54% yield). H NMR (Figure S1: 500 MHz, CDCl3,
r.t.) δ (ppm): 8.72 (s, 2H), 7.95 (d, 2H), 7.73 (t, 1H), 7.51 (m, 4H),
7.40 (m, 6H), 5.49 (s, 2H).
Synthesis of 2-Bromo-2-methylpropanoic Anhydride.9
A
mixture of 2-bromoisobutyric acid (20 g, 0.12 mol, 1.0 equiv),
dehydrated THF (200 mL), and triethylamine (20 mL, 0.15 mol, 1.2
equiv) were placed in an argon-filled 250-mL round-bottom flask. The
solution was cooled to 0 °C, and 2-boromoisobutyryl bromide (18.05
mL, 0.146 mol, 1.2 equiv) was slowly added under argon. The solution
was then stirred for 3 h at room temperature and filtered. The filtrated
solution was evaporated to dryness, and the residue was dissolved in
methylene chloride, followed by washing with NaHCO3 saturated
solution (100 mL, three times). The organic layer was dried over
Na2SO4 for several hours, and the filtered solution was evaporated.
Finally, the product was purified by recrystallization with hexane
1
(21.20 g, 55% yield). H NMR (Figure S2: 500 MHz, CDCl3, r.t.) δ
(ppm): 2.00 (s, 12H).
Synthesis of Br-ADA-Br Initiator. A mixture of 2-bromo-2-
methylpropanoic anhydride (21 g, 0.067 mol, 5.0 equiv), 4-
(dimethylamino)pyridine (2.4 g, 0.020 mol, 1.5 equiv), 2-deoxyuridine
(3.0 g, 0.013 mol, 1.0 equiv), and dry pyridine (40 mL) were
introduced into an argon-filled round-bottom flask. The solution was
stirred for 20 h and then evaporated, followed by silica gel
chromatography with an ethylacetate/hexane (4/6 v/v) mixed eluent.
Finally, the product was purified by recrystallization with a
dichloromethane and hexane mixture (Br-ADA-Br: 5.40 g, 77%
yield). 1H NMR (Figure S3: 500 MHz, CDCl3, r.t.) δ (ppm): 8.97 (s,
1H), 7.66 (d, 1H), 6.34 (m, 1H), 5.79 (m, 1H), 5.34 (m, 1H), 4.60−
4.32 (m, 3H), 2.68−2.58 (m, 2H), 1.96 (m, 12H).
Polymerization Procedures. Polymerization was carried out by
the syringe technique under dry argon in baked glass tubes equipped
with a three-way stopcock or in sealed glass vials. A typical procedure
for MMA with Cl-DAD-Cl/Ru(Cp*)Cl(PPh3)2/n-BuNH2 is given. In
a round-bottom flask (50 mL) filled with argon was placed
Ru(Cp*)Cl(PPh3)2 (9.3 mg, 0.012 mmol), toluene (0.43 mL), n-
octane (0.16 mL), MMA (1.28 mL, 12.0 mmol), n-BuNH2 solution
(0.30 mL, 400 mM in toluene), and a solution of Cl-DAD-Cl (0.83
mL, 145.3 mM in THF) were sequentially added; the total volume was
3.00 mL. Immediately after mixing, aliquots (0.50−1.0 mL each) of the
solution were injected into baked glass tubes, which were then sealed
(except when a stopcock was used) and placed in an oil bath kept at 80
°C. In predominant intervals, the polymerization mixture was
terminated by cooling to −78 °C in dry ice−methanol. Monomer
conversion was determined from residual monomer concentration
measured by gas chromatography with n-octane (for MMA) or tetralin
(for St) as an internal standard. The quenched solutions were
evaporated to dryness to give the products, which were subsequently
dried overnight under vacuum at room temperature.
EXPERIMENTAL SECTION
■
Materials. Monomers (MMA and styrene: Aldrich; purity >99%)
were dried overnight over calcium hydride and purified by distillation
from calcium hydride before use. 2,6-Diaminopyridine (Wako; >97%),
(+)-2′-deoxyuridine (Wako; >98%), 2-boromoisobutyryl bromide
(Wako; >95%), triethylamine (Wako; 99%), 4-(dimethylamino)-
pyridine (Aldrich; 99%), Ru(Cp*)Cl(PPh3)2 (Cp*: cyclopentadienyl,
Aldrich), α-chlorophenylacetyl chloride (Aldrich; 90%), and 2-
bomoisobutyric acid (TCI; >98%) were used as received. Dichloro-
methane (Wako; dehydrated) and pyridine (Wako; dehydrated) were
bubbled with dry argon for more than 15 min immediately before use.
Toluene and THF (Kishida Kagaku; purity 99.5%) was dried and
Cleavage of PMMA Obtained from Br-ADA-Br. A baked glass
tube was charged with 0.064 g (1.02 × 10−5 mol) of PMMA (Mn,SEC
=
3704
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710

Macromolecules
Article
Scheme 4. Syntheses of Bifunctional Initiators, Cl-DAD-Cl (a) and Br-ADA-Br (b)
Figure 1. 1H NMR spectra (CDCl3, room temperature) from 7.5 to 11.5 ppm of Br-ADA-Br (bottom) and the mixtures with Cl-DAD-Cl of various
ratios ([Cl-DAD-Cl]/[Br-ADA-Br] = 0.5, 1, 2, 4, 8, and 16) for calculation of Kass: [Br-ADA-Br] = 2.0 mM (constant); [Cl-DAD-Cl] = 0, 1, 4, 8, 16,
24, and 32 mM.
6260, Mw/Mn = 1.13) and 0.055 g of sodium methoxide (1.02 × 10−3
mol) with 9.0 mL of THF and 1.0 mL of dry methanol. The tube was
sealed and heated at 80 °C for 17 h. The solvent was evaporated,
followed by reprecitipation in methanol from THF solution. The
product was subsequently dried overnight under vacuum and analyzed
with SEC.
Cleavage of PS from Cl-DAD-Cl. A baked glass tube was charged
with 0.04 g (2.05 × 10−5 mol) of PS (Mn,SEC = 2180, Mw/Mn = 1.05),
4.0 mL of 1,4-dioxane, and 2.0 mL of 12 mol/L HCl solution in 1,4-
dioxane. The tube was sealed and heated at 100 °C for 48 h. The
product was reprecipitated in methanol and dried overnight under
vacuum for SEC analysis.
Measurements. Mn and Mw/Mn of polymers were measured by
size exclusion chromatography (SEC) at 40 °C in THF or CHCl3 as
an eluent. For the THF−SEC, three polystyrene-gel columns [LF-404
(from Shodex); pore size, 3000 Å; 4.6 mm i.d. × 250 cm] were
connected to a DU-H2000 pump, a 74S-RI refractive-index detector,
and a 41-UV ultraviolet detector (all from Shodex), and a flow rate was
set to 0.3 mL/min. For the CHCl3−SEC, three polystyrene gel
columns [K-805 L (from Shodex); pore size 20−1000 Å; 8.0 mm i.d.
× 30 cm] were connected to a PU-980 pump and a 930-RI refractive
index detector, and a 970-UV ultraviolet detector (all from Jasco) and
flow rate was set to 1.0 mL/min. The columns were calibrated against
13 standard poly(MMA) (PMMA) samples (Polymer Laboratories;
Mn = 620−1 200 000; Mw/Mn = 1.06−1.22) for obtained PMMA
3705
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710

Macromolecules
Article
a
Table 1. Living Radical Polymerizations of MMA or St with Cl-DAD-Cl or Br-ADA-Br
b
b
entry
initiator
monomer
MMA
catalyst
cocatalyst
temp (°C)
time (h)
conv (%)
Mn
Mw/Mn
1
Cl-DAD-Cl
RuCp*
n-BuNH2
80
3
7
41
70
90
24
41
99
46
76
98
61
78
90
73
95
100
28
59
98
5100
7900
9300
2400
4200
9200
4600
7300
9200
6900
9400
10300
7900
11800
12900
3900
7800
9200
1.16
1.13
1.11
1.31
1.57
1.27
1.13
1.11
1.14
1.17
1.15
1.15
1.21
1.24
1.24
1.12
1.13
2
3
19
3
4
RuInd
RuCp*
RuCp*
RuInd
RuCp*
n-Bu3N
80
100
80
5
7
6
48
9
7
St
n-BuNH2
n-BuNH2
n-Bu3N
8
22
75
3
9
10
11
12
13
14
15
16
17
18
Br-ADA-Br
MMA
9
93
3
80
6
9
St
n-Bu2NH
100
8
22.5
75
1.14
a
b
[Monomer]0 = 4.0 M; [initiator]0 = 40 mM; [Ru(Cp*)Cl(PPh3)2]0 or [Ru(Ind)Cl(PPh3)2]0 = 4.0 mM; [cocatalyst]0 = 40 mM in toluene. By size
exclusion chromatography calibrated with PMMA or PS standards.
samples, and 13 standard polystyrene (PS) samples (Polymer
Laboratories; Mn = 500−3 840 000; Mw/Mn = 1.01−1.14) for obtained
PS samples, respectively. Polymer samples were purified by preparative
SEC in CHCl3 at room temperature on polystyrene gel preparative
column [K-5002 (from Shodex); exclusion limit = 5 × 103; particle
size = 15 μm; 5.0 cm i.d. × 30 cm; flow rate = 10 mL/min] that were
connected to a PU-2086 precision pump, a RI-2031 refractive-index
detector, and UV-2075 ultraviolet detector (all from Jasco) . 1H NMR
and 1H DOSY were recorded in CDCl3 at room temperature or 30 °C
on JEOL JNM-ECA500 spectrometer, operating at 500.16 (1H) MHz.
n-Bu3N). Both complexes catalyzed living polymerization to
give fairly controlled polymers (Figures S5 and S6). However,
RuCp* obviously gave narrower MWDs (Mw/Mn < 1.2: entries
1−3, 7−12, and 16−18 in Table 1), and this Cp* catalyst was
used throughout this study for polymer preparation because
molecular weight control, or quantitative initiation from both
halogens, is vital in placing the interactive site at the chain
center.
3. Structural Analysis of Obtained Polymers. Five
samples of PMMA or PS were thus prepared to study the
interaction between pairs of polymers (Table 2). All the
RESULTS AND DISCUSSION
■
1. Design of Complementary Hydrogen-Bonding
Bifucntional Initiators. Cl-DAD-Cl and Br-ADA-Br were
prepared via the acylation of 2,6-diaminopyridine and 2′-
deoxyuridine, respectively, as illustrated in Scheme 2. For the
latter, an anhydride (2-bromo-2-methylpropanoic anhydride)
was employed for the acylation because it turned out more
effective and less prone to side reactions than the
corresponding acid bromide (2-boromoisobutyryl bromide).
The complementary interaction between Cl-DAD-Cl and Br-
ADA-Br was evaluated by 1H NMR to determine the
association constant (Ka) (Figure 1).10 From the same stock
solutions of the components in CDCl3, a series of mixtures
were prepared at varying ratios ([Cl-DAD-Cl]/[Br-ADA-Br] =
0.5, 1, 2, 4, 8, 12, and 16 molar ratio). The chemical shift of the
NH proton in Br-ADA-Br progressively shifted downfield, from
7.98 to 11.03 ppm, as the Cl-DAD-Cl amount was increased.
From these changes, Ka was estimated to be 83.3 M−1 (see
Supporting Information, Figure S4 and Table S1), comparable
to those for similar three-point complementary hydrogen bond
pairs.11 The two initiators were therefore found to undergo
relatively strong complementary interaction via their DAD and
ADA units.
Table 2. Samples of ADA- or DAD-Functionalized Polymers
a
in This Work
d
d
f
g
sample code
Mn,SEC
Mw/Mn
Mn,NMR
Fn(DAD or ADA)
b
e
e
DAD-PS-1
2830
1.08
3300
2770
2060
3850
16800
1.17
0.96
0.73
0.94
1.10
b
DAD-PS-2
2870
2830
4080
15400
1.05
1.18
1.06
b
DAD-PMMA
b
ADA-PS
c
e
e
ADA-PMMA
1.17
a
All polymer samples were purified by preparative SEC to remove low
b
molecular weight residues. [Monomer]0 = 4.0 M, [initiator]0 = 40
mM, [Ru(Cp*)Cl(PPh3)2]0 = 4.0 mM, [n-BuNH2]0 = 40 mM in
c
toluene at 80 °C for MMA and 100 °C for St. [MMA]0 = 4.0 M, [Br-
ADA-Br]0 = 20 mM, [Ru(Cp*)Cl(PPh3)2]0 = 2.0 mM, [n-BuNH2]0 =
d
20 mM in toluene at 80 °C. Measured by size-exclusion
chromatography calibrated with PMMA or PS standards in THF.
e
f
CHCl3 was used as an eluent. Calculated with 1H NMR.
g
Determined by Mn,NMR/Mn,SEC
.
samples were purified with preparative SEC, and the
functionality (Fn) in terms of DAD or ADA was determined
1
by H NMR. For example, a PS sample from Cl-DAD-Cl
2. Ruthenium-Catalyzed Living Radical Polymeriza-
tion. Cl-DAD-Cl and Br-ADA-Br were then employed for
living radical polymerizations of MMA and styrene (Table 1).
Two ruthenium complexes, Ru(Cp*)Cl(PPh3)2 (RuCp*)12
and Ru(Ind)Cl(PPh3) (RuInd),13 were examined as catalysts in
conjunction with an amine cocatalyst (n-BuNH2, n-Bu2NH, or
(DAD-PS-1: Mn,SEC = 2830, Mw/Mn = 1.08) exhibited the
characteristic peak of a pyridine (3H) (Figure S7). Peak
integration led to Mn,NMR = 3300, and the number-average
functionality of DAD [Fn(DAD) = Mn,NMR/Mn,SEC] was
estimated to be 1.17, indicating near-quantitative introduction
of the central unit. Similarly, for a PMMA sample from Br-
3706
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710

Macromolecules
Article
Figure 2. 1H NMR spectra (CDCl3, room temperature) from 7.9 to 8.6 ppm of (a) DAD-PS-1 ([DAD-PS-1] = 2.0 mM), (b) ADA-PMMA ([ADA-
PMMA] = 2.0 mM), (c) 1:1 mixture of DAD-PS-1 and ADA-PMMA ([DAD-PS-1] = [ADA-PMMA] = 2.0 mM), and (d) 1:1 mixture of normal PS
and ADA-PMMA ([normal PS] = [ADA-PMMA] = 2.0 mM).
1
ADA-Br (ADA-PMMA: Mn,SEC = 15 400, Mw/Mn = 1.17), a
peak derived from the methine proton adjacent to the
deoxyribose ester [−CH(OCO−)−] was detected (Figure
S8), to give Mn,NMR = 16 800 and in turn Fn(DAD) = 1.10.
Overall, all the samples obtained in this study had nearly one
midchain ADA or DAD unit per chain.14
magnified H NMR spectra in the 7.9−8.6 ppm range for
DAD-PS-1 (a), ADA-PMMA (b), and their mixture (c). Upon
mixing, the thymine NH proton of ADA-PMMA clearly shifted
downfield from 8.14 (Figure 2b) to 8.33 ppm (Figure 2c),
indicating an interaction between the DAD and the ADA units.
On the other hand, no such a shift occurred when ADA-PMMA
was mixed with a DAD-free, commercial PS standard sample of
a similar molecular weight (Mn = 2630, Mw/Mn = 1.05) (Figure
2d).
The samples were then treated to cleave the bifunctional
initiator moiety from the polymer backbones; if the former is in
fact placed in the center of backbone as designed, molecular
weight should decrease by half after the cleavage. To perform
the cleavage experiment, two polymer samples were employed
apart from those listed in Table 2: PMMA obtained from Br-
ADA-Br (Mn,SEC = 6260, Mw/Mn = 1.13) and PS from Cl-DAD-
Cl (Mn,SEC = 2180, Mw/Mn = 1.05). The former was treated
with sodium methoxide (NaOCH3) in THF/MeOH (9/1 v/v)
(80 °C, 17 h) for transesterification of alkyl esters into methyl
esters. Note that this procedure cleaves the two ester linkages in
the ADA residue, whereas the pendent methyls in the PMMA
chains remain unchanged. For the latter, the amide linkages in
the DAD moiety were cleaved with trifluoroacetic acid in
dioxane at 100 °C for 48 h. After these cleavage experiments,
the overall molecular weights of the two samples (the peak tops
of SEC curves) were indeed reduced by half (Figures S9 and
S10).
As with the ADA- and the DAD-initiators (see section 1), a
series of mixtures of the two complementary polymers at
varying molar ratios (0.50 to 12) were further analyzed by H
1
NMR in CDCl3, to confirm their supramolecular interaction
and further to determine their association constant (Ka)
(Figure 3). For two pairs of polymers, ADA-PS/DAD-PS-2
(Figure 3a) and ADA-PS/DAD-PMMA (Figure 3b), the ADA’s
NH signals gradually and consistently shifted to downfield from
7.88 to 10.53 ppm as the relative amount of the DAD unit
increased, giving Ka = 54.3 and 116.2 M−1, respectively (see
also Figures S11 and S12). Both values are similar to those for
the initiator pairs (Ka = 83.3 M−1; section 1), but it is
interesting that the heteropolymer pair ADA-PS/DAD-PMMA
is more associative than the homopolymer pair and even the
initiator pair, though all the three systems involve the identical
ADA-DAD complementary functions.
4. Interaction between “Complementary” Polymer
Chains. The interaction between the ADA- and DAD-
functionalized polymers was examined by H NMR (Figure
Further analysis is now being conducted for this intriguing
phenomenon with a wider variety of polymer pairs, but it seems
no doubt that complementary hydrogen-bonding interactions
work well in solution even at the centers of polymer chains,
which are also strong enough to overcome repulsion between
different polymers such as PS and PMMA.
1
2). For example, equimolar amounts of ADA-PMMA (Mn,SEC
=
15 400, Mw/Mn = 1.17) and DAD-PS-1 (Mn,SEC = 2830, Mw/Mn
= 1.08) were mixed in CDCl3 at 10 mM each, and the solution
was stirred overnight at room temperature. Figure 2 shows the
3707
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710

Macromolecules
Article
1
Figure 3. H NMR spectra (CDCl3, room temperature) from 7.5 to 11.5 ppm for calculation of Ka between ADA- and DAD-functionalized
polymers, (a) ADA-PS and DAD-PS-2 and (b) ADA-PS and DAD-PMMA: ADA-functionalized polymer (bottom) and the mixtures with DAD-
counterpart of various ratios ([DAD-polymer]/[ADA-polymer] = 0.5, 1, 2, 4, 8, and 12); [ADA-polymer] = 2.0 mM (constant), [DAD-polymer] =
0, 1, 4, 8, 16, and 24 mM.
1
5. H DOSY NMR Analyses for Interaction between
PS and methoxy group from PMMA had cleary different
mobility in that solution. This result indicates that there was no
interaction between ADA-PMMA and unfunctionalized PS in
this system.
The virtually single, very large population with DNMR clearly
different from those for the two components demonstrates the
efficient complementary pairing of PS and PMMA chains by
which the supramolecular structure thus formed could move as
a single large (supra)molecule. Further examination is needed
to clarify why the paired assembly is apparently more movable
than nonassociating PMMA.
1
Polymer Chains. As discussed in the preceding section, H
NMR spectroscopy is indeed a convenient way to examine
polymer−polymer interactions, but obviously it is less effective
and sensitive for higher molecular weight polymers where
signals of midchain functional groups (like ADA and DAD, one
per long backbone) are usually too weak for quantitative
1
measurement. Thus, diffusion-ordered H NMR spectroscopy
(DOSY NMR) was employed, which conveniently gives
diffusion coefficients (DNMR) of multiple solutes in solution.
In a typical run, a sample of ADA-PMMA of a relatively high
molecular weight (Mn,SEC = 15 400) was mixed with an
equimolar complementary sample DAD-PS-1 of a shorter
length (Mn,SEC = 2830) in CDCl3, and their interaction was
directly followed by DOSY NMR. Figure 4 illustrates a 2D
CONCLUSIONS
■
A pair of bifunctional initiators (Cl-DAD-Cl and Br-ADA-Br),
bearing complementary hydrogen-bonding interactive sites, was
designed for the precision synthesis of complementary
associative polymers. The ruthenium-catalyzed living radical
polymerization with these initiators led to well-defined
polymers (PS and PMMA) that almost quantitatively carried
DAD or ADA at the center of the backbone. 1H NMR analyses
revealed that the pair of polymers with DAD and ADA
undoubtedly interacted with each other in CDCl3, and the
association constants were comparable to that between
initiators, even with a pair of different polymers (PMMA and
PS). 1H DOSY NMR analyses also supported such a polymer−
polymer interaction, in which one molecular behavior was
observed. We are now studying advanced interaction behaviors
between polymers in terms of cooperative self-aggregation by
“hydrogen bond” and “block copolymer” or periodic
introduction of interaction sites.
1
correlation diagram of H NMR signals (abscissa) and DNMR
distribution (ordinate).
The diffusion data consist of two populations, and their
correlations with NMR signals indicate that the larger and
predominant distribution is for the expected supramolecular
PS-PMMA pair (DNMR = 2.00 × 10−9 m2/s), whereas the very
minor peak for nonassociating (free) ADA-PMMA chains
(Figure 4a, DNMR = 8.70 × 10−10 m2/s); the latter is probably
due to a minor deviation from a perfectly stoichiometric
mixture upon sampling. Importantly, no population was
detected for nonassociating DAD-PS, which should have a
larger DNMR = 4.00 × 10−9 m2/s according to separate
measurements (Figure S14). On the other hand, the mixture
between commercial PS standard (Mn = 2630, Mw/Mn = 1.05)
and ADA-PMMA (Mn,SEC = 15 400) did not behave as a single
molecule in DNMR (Figure 4b). The peak of phenyl group from
3708
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710

Macromolecules
Article
Figure 4. 1H DOSY NMR Spectra (CDCl3, 30 °C) of (a) 1:1 mixture of ADA-PMMA and DAD-PS-1 ([ADA-PMMA] = [DAD-PS-1] = 2.0 mM)
and (b) 1:1 mixture of ADA-PMMA and normal PS ([ADA-PMMA] = [normal PS] = 2.0 mM).
Science (JSPS) (M.O.) and the Global COE Program
“International Center for Integrated Research and Advanced
Education in Materials Science” of the Ministry of Education,
Culture, Sports, Science, and Technology, Japan.
ASSOCIATED CONTENT
■
S
* Supporting Information
Polymerization results including SEC curves, H NMR spectra
of polymers, details for Kass calculation, and H DOSY NMR
1
1
spectra of ADA-PMMA and DAD-PS-2. This material is
available free of charge via the Internet at http://pubs.acs.org.
REFERENCES
■
(1) Matsen, M. W.; Bates, F. S. Macromolecules 1996, 29 (4), 1091−
1098.
AUTHOR INFORMATION
■
(2) (a) Feldman, K. E.; Kade, M. J.; de Greef, T. F. A.; Meijer, E. W.;
Kramer, E. J.; Hawker, C. J. Macromolecules 2008, 41 (13), 4694−
4700. (b) Feldman, K. E.; Kade, M. J.; Meijer, E. W.; Hawker, C. J.;
Kramer, E. J. Macromolecules 2010, 43 (11), 5121−5127. (c) De Greef,
T. F. A.; Kade, M. J.; Feldman, K. E.; Kramer, E. J.; Hawker, C. J.;
Meijer, E. W. J. Polym. Sci., Polym. Chem. 2011, 49 (19), 4253−4260.
(3) Folmer, B. J. B.; Sijbesma, R. P.; Versteegen, R. M.; van der Rijt, J.
A. J.; Meijer, E. W. Adv. Mater. 2000, 12 (12), 874−878.
(4) (a) Uzun, O.; Sanyal, A.; Jeong, Y.; Rotello, V. M. Macromol.
Biosci. 2010, 10 (5), 481−487. (b) Nair, K. P.; Breedveld, V.; Weck, M.
Macromolecules 2011, 44 (9), 3346−3357.
(5) (a) Mather, B. D.; Baker, M. B.; Beyer, F. L.; Berg, M. A. G.;
Green, M. D.; Long, T. E. Macromolecules 2007, 40 (19), 6834−6845.
(b) Bang, J.; Jeong, U.; Ryu, D. Y.; Russell, T. P.; Hawker, C. J. Adv.
Mater. 2009, 21 (47), 4769−4792. (c) Dobrosielska, K.; Takano, A.;
Matsushita, Y. Macromolecules 2010, 43 (2), 1101−1107.
Corresponding Author
*Tel +81-75-383-2603, Fax +81-75-383-2603, e-mail ouchi@
living.polym.kyoto-u.ac.jp (M.O.). Tel +81-75-383-2600, Fax
+81-75-383-2601, e-mail sawamoto@star.polym.kyoto-u.ac.jp
(M.S.)
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Takaya Terashima and Mr. Yuta Koda in our
laboratory for their helpful discussions and comments about
NMR analyses for the hydrogen bond interactions. This work
was supported by Grant-in-Aid for Young Scientists (A)
(23685024) from the Japan Society for the Promotion of
3709
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710

Macromolecules
Article
(6) (a) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001,
101 (12), 3689−3745. (b) Kamigaito, M.; Ando, T.; Sawamoto, M.
Chem. Rec. 2004, 4 (3), 159−175. (c) Ouchi, M.; Terashima, T.;
Sawamoto, M. Acc. Chem. Res. 2008, 41 (9), 1120−1132. (d) Ouchi,
M.; Terashima, T.; Sawamoto, M. Chem. Rev. 2009, 109 (11), 4963−
5050.
(7) (a) Fouquey, C.; Lehn, J. M.; Levelut, A. M. Adv. Mater. 1990, 2
(5), 254−257. (b) Gulikkrzywicki, T.; Fouquey, C.; Lehn, J. M. Proc.
Natl. Acad. Sci. U. S. A. 1993, 90 (1), 163−167.
(8) (a) Bernard, J.; Lortie, F.; Fenet, B. Macromol. Rapid Commun.
2009, 30 (2), 83−88. (b) Chen, S. B.; Bertrand, A.; Chang, X. J.;
Alcouffe, P.; Ladaviere, C.; Gerard, J. F.; Lortie, F.; Bernard, J.
Macromolecules 2010, 43 (14), 5981−5988.
(9) Ohno, K.; Wong, B.; Haddleton, D. M. J. Polym. Sci., Polym.
Chem. 2001, 39 (13), 2206−2214.
(10) Fielding, L. Tetrahedron 2000, 56 (34), 6151−6170.
(11) Meijer, E. W.; Brunsveld, L.; Folmer, B. J. B.; Sijbesma, R. P.
Chem. Rev. 2001, 101 (12), 4071−4097.
(12) Watanabe, Y.; Ando, T.; Kamigaito, M.; Sawamoto, M.
Macromolecules 2001, 34 (13), 4370−4374.
(13) Takahashi, H.; Ando, T.; Kamigaito, M.; Sawamoto, M.
Macromolecules 1999, 32 (11), 3820−3823.
(14) Only functionality of DAD-PMMA was lower than those of
other samples. The theoretical Mn of DAD-PMMA was calculated to
be about 1700 from the convesrion (12.8%) and the ratio of monomer
to initiator ([MMA]0/[[Cl-DAD-Cl]0 = 100). This is much lower than
the value by SEC (Mn,SEC = 2830). Thus, the lower Fn(DAD) might be
caused by an inexactness of the SEC analysis. For such a low molecular
weight polymer carrying a DAD unit at the middle point, a calibration
curve with PMMA standard might not be applicable.
3710
dx.doi.org/10.1021/ma300479s | Macromolecules 2012, 45, 3702−3710