Aggregation and chain dynamics in supramolecular polymers by dynamic rheology: Cluster formation and self-Aggregation

Article abstract of DOI:10.1021/ma101962y

Supramolecular polyisobutylenes (PIB) bearing mono- and bifunctional chain ends with hydrogen-bonding units were prepared, and their association behavior in the melt state was investigated by dynamic rheology and compared to aggregation in solution, aimin

Full text of DOI:10.1021/ma101962y

10006 Macromolecules 2010, 43, 10006–10016
DOI: 10.1021/ma101962y
Aggregation and Chain Dynamics in Supramolecular Polymers by
Dynamic Rheology: Cluster Formation and Self-Aggregation
†
‡
‡
‡
‡
€
€
Florian Herbst, Klaus Schroter, Ilja Gunkel, Stefan Groger, Thomas Thurn-Albrecht,
Jochen Balbach,‡ and Wolfgang H. Binder*,†
†Institute of Chemistry, Division of Technical and Macromolecular Chemistry, Faculty of Natural Sciences II
(Chemistry, Physics and Mathematics), Martin-Luther University Halle-Wittenberg, Halle 06120, Germany,
‡
and Institute of Physics, Faculty of Natural Sciences II (Chemistry, Physics and Mathematics),
Martin-Luther University Halle-Wittenberg, Halle 06120, Germany
Received August 24, 2010; Revised Manuscript Received October 20, 2010
ABSTRACT: Supramolecular polyisobutylenes (PIB) bearing mono- and bifunctional chain ends with
hydrogen-bonding units were prepared, and their association behavior in the melt state was investigated by
dynamic rheology and compared to aggregation in solution, aiming at determining association dynamics in
the solid state. A preparation combining living cationic polymerization with either azide/alkyne “click”
reactions or nucleophilic substitution reactions enabled a full end group transformation to the final PIB
polymers, modified with either thymine or 2,6-diaminotriazine end groups as proven by NMR and MALDI
methods with molecular weights of ∼3500 and ∼10 000 g/mol. Stoichiometric mixtures of these polymers
bearing specifically interacting thymine/triazine moieties were prepared by solution blending and the
temperature-dependent dynamics investigated by rheological measurements. At temperatures of 20-60 °C
all samples display strongly thermoreversible aggregation with sheet-type or partially cross-linked structures,
which deaggregate at temperatures of ∼80 °C. More complex aggregates with bridged micellar structure were
formed from the respective bifunctional PIB’s bearing thymine and 2,6-diaminotriazine moieties. Thus, in
addition to specific linear aggregates, the formation of clusters and aggregates of different architecture has to
be taken into account to understand and control structure and mechanical properties of supramolecular
chains in the melt.
Introduction
rate of dissociation, assuming that the associative step is diffusion
limited with a constant on the order of (1.5-4) ꢀ 109 M-1 s-1
.
The field of hydrogen-bonded supramolecular polymers is a
strongly expanding area, enabling to introduce dynamic features
into polymers and polymeric architectures by use of specifically
designed noncovalent (supramolecular) interactions. Thus, beside
conventional (“covalently-linked”) polymers, hydrogen-bonded
polymers have gained in importance over the past two decades,1-5
most of all with respect to their dynamic properties, generating
e.g. self-healing polymers,6,7 dynamic composites,8,9 and complex
polymeric scaffolds for biomedical applications.10 In the past years,
we have demonstrated the organization of polymeric chains into
larger aggregates, such as supramolecular networks11,12 or highly
defined structures such as pseudoblock copolymers.13,14
Basically, the dynamics of such materials are determined by two
factors: (a) chain dynamics of the individual polymeric building
blocks and (b) the inherent dynamics of the connective hydrogen
bond, which links the polymeric chains in a defined supramolec-
ular architecture and position.15 In recent years, significant efforts
have been undertaken to achieve a deeper understanding of the
inherent dynamics of supramolecular bonds.
This concept was then transferred to hydrogen-bonded supramo-
lecular polymers by Meijer et al.,17 first determining the time scale
of opening dynamics of quadruply bound hydrogen bonds
(ureidopyrimidine dimers) in solution by spectroscopic techniques.
However, it turned out that the purely diffusion-controlled model
in this case is not applicable, as tautomeric equilibria change
the fraction of molecular collisions between the two associating
partners, thus reducing the number of successful dimerization
events. Another direct method to probe the supramolecular binding
in hydrogen-bonded polymers in solution was achieved by Vancso
et al.18-20 by probing the molecular force between two hydrogen-
bonded partners, one affixed onto an AFM tip and the other on a
surface, both being connected by transiently associating bifunctional
fragments via Meijer’s quadrupolar bond. Time-dependent effects
of the binding and chain elongation could thus be observed.
Detailed rheological studies of small-molecule associating frag-
ments interconnected by quadrupolar hydrogen bonds (most of all
measurements of zero-shear viscosities) by Cohen-Stuart et al.21,22
subsequently revealed the influence of chain stoppers and tempera-
ture on effective chain length and relaxation. Recently, Bouteiller et
al. have reported significant clustering of hydrogen bond moieties
(bis-ureas) in poly(isobutylene)-modified bis-urea samples23,24 fol-
lowing early examples of network formation as reported by
Stadler et al. in polymeric urazole systems.25,26 A significant
progress has been achieved in gelous network structures, consist-
ing of associating polymers, bound via grafted metal-complexing
sites, as discussed by Craig et al.27-30 These studies allowed an
estimation of the dynamics of the underlying supramolecular
Usually, the dynamics of aggregation in solution has been
studied assuming that the basic step of aggregate formation
within a simple supramoelcular system built from two associating
fragments (bearing the respective groups A and B) is diffusion-
limited, as described by Hammes et al.16 The thermodynamic
equilibrium constant of association/dissociation in a simple triple
hydrogen bond (adenine/uracile) can thus be transformed into the
*Corresponding author. E-mail: wolfgang.binder@chemie.uni-halle.de.
pubs.acs.org/Macromolecules
Published on Web 11/10/2010
r 2010 American Chemical Society

Article
Macromolecules, Vol. 43, No. 23, 2010 10007
Scheme 1. Chemical Structure and Abbreviations of the Supramolecular Mono- and Bifunctional Polyisobutylenes (PIBs) Bearing Thymine or
2,6-Diaminotriazine Moieties
bond via frequency-dependent dynamic mechanical measure-
ments of either the viscosity or the storage and loss moduli.
Scaling of the respective curves with the known time scale of
association in solution led to a superposition of all measured
viscosities, thus indirectly allowing the determination of the rates
of association.
Studies of chain dynamics in supramolecular polymers in the melt
state are rare, as direct measurement techniques and simple models
for chain dynamics are not available in dense materials. A recent
example was given by Rowan et al.,31 who described the formation
of supramolecular clusters in nucleobase-modified poly(THF) poly-
mers with low molecular weights (Mn ∼ 1400 g mol-1). Multiple
stacks were generated by the aromatic moieties of the attached
nucleobases. Gel-like network formation was proven by rheological
experiments.
The present publication reports on the synthesis and dynamic
rheology of mono- or bifunctionalized poly(isobutylenes) PIBs
with defined hydrogen-bonding motifs at the chain ends, namely
thymine and 2,6-diaminotriazine end group (see Scheme 1).
Blends of the different mono- and bifunctional PIBs were
investigated. The aim was to study the rheological behavior of the
associating system in the bulk and to compare this to the chain
dynamics of the unmodified polymer. As PIB can be prepared by
living cationic polymerization exact chain length control and low
polydispersities are possible. The presented system is highly
defined with respect to (a) its molecular weight, (b) its molecular
weight distribution, and (c) the number of interacting moieties
present at each chain end of the polymer, based on equimolar
supramolecular mixtures of the functionalized polyisobutylenes.
Poly(isobutylene) was chosen as the central polymer, as it is liquid
at room temperature (Tg ∼ -70 °C) and thus easily allows
rheological investigations in bulk without the necessity of added
solvents. We find strong effects of supramolecular aggregates on
the rheological properties incomparison to the nonfunctionalized
PIB. But in general, the formation of supramolecular aggregates
in the bulk is more complex than expected from the very selective
interactions of the hydrogen-bonding units in solution and is
characterized by a variety of structures.
Synthesis of 6-(4-Ethynylbenzyl)-1,3,5-triazine-2,4-diamine (9).
Dicyandiamide (17.70 mmol, 1.49 g), 4-ethynylphenylacetonitrile
(4.41 mmol, 600 μL), and sodium hydroxide (2.21 mmol, 0.14 g)
were added to dry isopropanol (60 mL). The mixture was heated
under reflux conditions for 24 h, during which the formation of a
precipitatewas observed. Isopropanolwas thendecanted, and the
residue was heated two times under reflux conditions with
isopropanol (70 mL) for 10 min. After drying in high vacuum
919.0 mg (4.08 mmol, 92.5%) of 9 was obtained as a beige solid.
1H NMR (400 MHz, DMSO-d6): δ (ppm) 7.39 (d, 2H), 7.27 (d,
2H), 6.62 (s, 4H), 4.10 (s, 1H), 3.65 (s, 2H). 13C NMR (100 MHz,
DMSO-d6): δ (ppm) 175.6, 166.9, 139.9, 131.3, 129.1, 119.4, 83.3,
80.1, 44.2.
Synthesis of Thymine-Functionalized PIBs via “Click” Reac-
tions (4, 5). A general procedure for the synthesis of thymine-
functionalized PIBs via “click” reactions is shown via the
synthesis of monofunctional thymine-functionalized PIB 4b
with Mn = 2750 g/mol (azido-functionalized starting material).
In an one-neck round-bottom flask azido-functionalized PIB
(0.31 mmol, 852 mg) was dissolved in toluene (17 mL). Water (8
mL), isopropanol (8 mL), compound 8 (0.62 mmol, 102 mg),
TBTA (0.03 mmol, 16 mg), and DIPEA (0.62 mmol, 112 μL)
were added. The mixture was sparged with nitrogen for 30 min,
bromotris(triphenylphosphine)copper(I) (0.03 mmol, 29 mg)
was added, and the flask was sealed with a rubber septum.
Subsequently, the flask was placed in the microwave oven and
irradiation was started. Microwave conditions were SPS method
(“pulse-no pulse-pulse”), reaction temperature 90 °C, ΔT = 6
°C, reaction time 24 h, and an irradiation power of 100 W. After the
reaction was finished, the solvent was removed and n-hexane was
added. The polymer was precipitated into a 10-fold excess of
methanol. Pure thymine-functionalized PIB was obtained by col-
umn chromatography (SiO2) in chloroform (Rf(4b) = 0.2). Yield:
708 mg (83%), Mn(GPC) = 3300 g/mol, Mn(NMR) =3200g/mol. 1H
NMR (500 MHz, CDCl3): δ (ppm) 7.80 (s, 1H), 7.37 (s, 1H), 4.97
(s, 2H), 1.91 (s, 5H), 1.44 (bs, 105H), 1.11 (bs, 315H), 0.99 (s, 15H).
13C NMR (125 MHz, CDCl3): δ (ppm) 163.3, 150.7, 143.8, 140.2,
119.7, 111.2, 59.5, 58.8, 58.2, 55.7, 51.5, 42.8, 42.2, 38.2, 37.8, 37.8,
34.8, 32.6, 32.4, 31.2, 30.8, 30.8, 29.2, 25.5, 12.3.
Synthesis of 2,6-Diaminotriazine-Functionalized PIBs via
“Click” Reactions (6a,b, 7). A general procedure for the synthe-
sis of 2,6-diaminotriazine-functionalized PIBs via “click”
reactions is shown via the synthesis of monofunctional 2,6-
diaminotriazine-functionalized PIB 6a with Mn = 2600 g/mol
(azido-functionalized starting material). In an one-neck round-
bottom flask azido-functionalized PIB (0.93 mmol, 2.42 g) was
dissolved in toluene (30 mL). Water (10 mL), isopropanol
(10 mL), TBTA (0.19 mmol, 99 mg), and DIPEA (0.14 mmol,
25 μL) were added. The mixture was sparged with nitrogen for
30 min, copper iodide triethylphosphite (0.09 mmol, 33 mg) and
compound 9 (1.12 mmol, 251 mg) were added, and the flask was
sealed with a rubber septum.
Experimental Section
Abbreviations. Tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]-
amine (TBTA), N,N-diisopropylethylamine (DIPEA).
Materials. All materials were obtained from Sigma-Aldrich and
used without further purification if not mentioned otherwise.
Copper iodide triethylphosphite [CuI(P(OEt)3)] was synthesized
according to Langille et al.32 5-Methyl-1-(prop-2-ynyl)-1H-pyrim-
idine-2,4-dione (8) was synthesized according to Lindsell et al.33
Thymine-functionalized PIB (3) was synthesized according to
Binder et al.34

10008 Macromolecules, Vol. 43, No. 23, 2010
Herbst et al.
Subsequently, the flask was placed in the microwave oven and
irradiation was started. Microwave conditions were SPS method
(“pulse-no pulse-pulse”), reaction temperature 90 °C, ΔT 6 °C,
reaction time 24 h, and irradiation power 50 W. After reaction was
finished, the solvent was removed and chloroform was added. The
suspension was filtered, and the solvent was removed. Nonfunctio-
nalized polymer chains were separated by column chromatography
(SiO2) in chloroform (Rf(6a)=0,Rf(azido-functionalized PIB)=1).
After unreacted azido-functionalized PIB was eluted, the solvent
mixture was changed to chloroform/methanol = 10/1 and 2,6-
diaminotriazine-functionalized PIB was eluted (Rf(6a) = 1).
The solvent was removed; the residue was dissolved in a small
amount of n-hexane and precipitated in a 10-fold excess of
and referred to the -CH3 signal of the PIB chain at 1.10 ppm.
The dimerization constant for the 2,6-diaminotriazine group
was determined using 2,6-diaminotriazine-functionalized PIB
6a. The dimerization constant for the thymine group was
determined using thymine-functionalized PIB 3. The associa-
tion constant for the thymine/2,6-diaminotriazine system was
determined by using thymine-functionalized PIB 3 (as host [H])
and 2,6-diaminotriazine-functionalized PIB 6a (as guest [G]).
GPC spectra were recorded on a Viscotek GPCmax VE 2002
solvent/sample module using a HHRH Guard-17369 and
GMHHR-N-18055 column in THF at 40 °C and a refractive
index detector. Polyisobutylene standards were obtained from
PSS (Polymer Standards Service) and used for conventional
external calibration. Standards with a molecular weight of 340,
1650, 7970, 26 300, 61 800, and 87 600 g/mol were used for
calibration.
1
methanol. Yield: 1.03 g (43%), Mn(NMR) = 3960 g/mol. H
NMR (500 MHz, CDCl3): δ (ppm) 7.78 (d, 2H), 7.71 (s, 1H),
7.39 (d, 2H), 5.31 (bs, 4H), 4.34 (t, 2H), 3.79 (s, 2H), 1.44 (bs,
130H), 1.11 (bs, 390 H), 0.99 (s, 15H). 13C NMR (125 MHz,
CDCl3): δ (ppm) 177.5, 167.0, 147.4, 136.9, 129.6, 129.2, 125.7,
119.2, 59.6, 58.9, 58.3, 55.8, 51.3, 44.8, 42.4, 38.2, 37.9, 37.9,
34.9, 32.7, 32.5, 31.3, 30.9, 30.9, 29.3, 25.7.
Sample Preparation for Rheological and SAXS Measure-
ments. Pure PIB samples (M1-M3, M5-M7, M9-M11; see
Table 2) were prepared by dissolving the polymer in chloroform.
The solution was filtered through a 0.2 μm PTFE filter, and the
solvent was removed. The sample was dried in high vacuum to
constant weight and annealed at 40 °C for 72 h. For mixtures
containing thymine- and 2,6-diaminotriazine-functionalized
PIBs (M4, M8, and M12) the amounts needed for an equimolar
mixture were calculated from the molecular weight determined
via NMR. The amounts were weighed, and the polymers were
dissolved in chloroform. After filtration through a 0.2 μm PTFE
filter the solutions were combined, and the solvent was removed.
Samples were dried in high vacuum to constant weight and
annealed at 40 °C for 72 h.
Methods. NMR spectra were obtained from a Varian Gemini
2000 (400 MHz) and Varian Unity Inova 500 (500 MHz)
spectrometer in CDCl3 (liquid NMR). Molecular weights of
the PIBs via NMR were determined as follows: For monofunc-
tional PIBs the integral of the initiator at 1.00 ppm was set to
15 H (two overlapping signals with 9 and 6 H). For bifunctional
PIBs the integral of the initiator at 7.17 ppm was set to 3 protons.
The resonance of the -CH2- groups of the polymer chain
was integrated, and the resulting value was divided by 2 to yield
the degree of polymerization (DP). The DP was multiplied by
56.1 g/mol, the molecular weight of the repeat unit, and the
molecular weights of the initiator fragment and the introduced
end group(s) were added (the end groups conform to the
residues “R” in Scheme 1). The added molecular weights of
the fragments are (in g/mol) 148.5 for TMPCl (without Cl),
287.3 for DCCl (without Cl), 41 for allyl group, 197.2 for the
thymine group introduced via the chloromethyl ether synthesis
route, 248.3 for thymine group introduced via the “click” reaction,
and 309.4 for the 2,6-diaminotriazine group. Complete functiona-
lization of the monofunctional polymer chains was proven by
integrating the signal of the initiator at 1.00 ppm (7.10 ppm for
bifunctional PIBs) and setting this value to 15 H (bifunctional:
3 H). The integration of the characteristic resonance of the
thymine end group at 7.80 ppm leads to a value of 1 H (bifunc-
tional: 2 H), thus indicating a complete end-group functionali-
zation. The integration of the characteristic resonance of the
2,6-diaminotriazine end group at 3.79 ppm leads to a value of 2 H
(bifunctional: 4 H) and indicates a complete functionalization.
NMR titration experiments (“Job plots”) were performed on a
Bruker Avance III spectrometer. The proton frequency was
600.13 MHz, and a BBI probe head was used. Temperature was
controlled via a BVT3200 controller. The flip angle was 40°,
relaxation delay (RD) was 5 s, and 32 scans were performed for
each sample. Temperature calibration was performed using neat
methanol and a calibration formula published by Merbach
et al.35 Chemical shifts were recorded in ppm (δ; parts per million)
MALDI-TOF-MS experiments were performed on a Bruker
Autoflex III system operating in reflectron and linear modes.
Ions were formed by laser desorption (smart beam laser at 355,
532, 808, and 1064 ( 5 nm; 3 ns pulse width; up to 2500 Hz
repetition rate), accelerated by voltage of 20 kV and detected as
positive ions. Baseline subtraction and smoothing of the re-
corded spectra were performed using a Savitzky-Golay algo-
rithm. The instrument was calibrated with poly(ethylene glycol)
(PEG) standards (Mn = 2000 g/mol and Mn = 4000 g/mol)
applying a quadratic calibration method with an error of 1-
2 ppm. PEG standards (calibration standards) were dissolved
in THF at a concentration of 20 mg/mL. trans-2-[3-(4-tert-
Butylphenyl)-2-methyl-2-propenylidene]malononitrile (DCTB)
(matrix) and lithium trifluoroacetate (LiTFA) (salt) were dis-
solved in THF at a concentration of 20 mg/mL. For PEG
standards the ratio of matrix:analyte:salt was 100:10:1, and
1 μL of the solution was spotted on a MALDI-target plate.
Samples of 2,6-diaminotriazine-functionalized PIBs were pre-
pared by dissolving the polymer in THF at a concentration of
20 mg/mL. Dithranol was used as matrix and dissolved in THF
at a concentration of 20 mg/mL. LiTFA was used as salt and
dissolved at a concentration of 10 mg/mL. The ratio of matrix:
analyte:salt was 100:10:1; the measurement was performed in
deflection mode using HPC (high-precision calibration) method.
For thymine-functionalized PIBs DCTP was used as matrix.
Small-angle X-ray scattering experiments (SAXS) were per-
formed using a Rigaku rotating anode. The X-ray beam was
˚
monochromated with an Osmic X-ray optics (λ = 1.54 A). The
size of the beam on the sample was ∼300 μm. Samples were
placed in a chamber that was evacuated to a pressure of ∼10-1
mbar. A Siemens 2D detector with 1024 ꢀ 1024 channels was
used to count the scattered intensity. Typical measurements
took some minutes. Before every measurement the temperature
was equilibrated for 15 min.
Rheological measurements were performed on an Anton Paar
(Physica) MCR 501 using parallel plates (diameter 8 mm). The
sample temperature was regulated by thermoelectric cooling/
heating in a Peltier chamber under a nitrogen atmosphere. At
each temperature the sample was equilibrated for 20 min before
the measurement started. All measurements were performed in
dynamic mode. After the rheological measurement at 80 °C the
measurement at 20 °C was repeated (after temperature equili-
bration for 30 min). Usually, the results did not differ more than
10% from the first measurement at 20 °C.
Results and Discussion
Synthesis of the Polymeric Compounds. Allyl-functionalized
PIBs were synthesized via living carbocationic polymerization
as described elsewhere.36 2-Chloro-2,4,4-trimethylpentane
(TMPCl) was used for the synthesis of monofuntional PIBs, and
1-tert-butyl-3,5-bis(1-chloro-1-methylethyl)benzol (DCCl) was
used for the synthesis of bifunctional PIBs. The transforma-
tion of the allyl-functionalized PIB to the azido-functionalized

Article
Macromolecules, Vol. 43, No. 23, 2010 10009
Table 1. Molecular Weight Data of the PIB Building Blocks 1-7
PDId
yieldb [%]
a
entry
polymer
Mn(theor) [g/mol]
PDI(theor)
Mn(GPC)e [g/mol]
Mn(NMR) [g/mol]
1
2
3
4
5
6
7
8
9
10
1a
1b
2
3500
10000
3000
3479
11612
2914
3310
2825
7925
3625
3680
10600
2730
3720
10900
2870
3600
11400
3200
3500
3960
7900
3700
1.23
1.25
1.27
1.13
1.23
1.16
1.34
86
93
97
75
73
83
70
43
55
91
3
1.20
1.24
1.23
1.29
1.12
1.11
1.23
3500
4a
4b
5
6a
6b
7
10900
3300
3600
c
c
-
-
-
c
-
c
c
-
c
-
a PDI of the azido- or hydroxyl-functionalized precursor determined by GPC. b Yield of the isolated product. c Not possible due to interaction with the
GPC column. d Determined via GPC. e Polyisobutylene standards were used for calibration.
Scheme 2. Synthesis of Mono- and Bifunctional Thymine- and 2,6-Diaminotriazine-Functionalized PIB via the Azide/Alkyne “Click” Reaction
PIB was accomplished according to a method developed by
Binder et al.11,34
functionalization. Because of the strong polarity change after
reaction with the thymine as well as the 2,6-diaminotriazine
group(s), it was possible to obtain the pure compounds 4-7
after column chromatography, thus removing eventually pre-
sent impurities or incompletely functionalized polymers. The
purity and structure (and complete functionalization in the case
of bifunctional PIBs) were confirmed by NMR spectroscopy
(Figure 1, see also Supporting Information) and MALDI-
TOF-MS analysis (see Figure 1). All compounds showed the
expected resonances as well as the expected absolute molecular
weights in the respective mass spectra. The MALDI-TOF-MS
spectrum for monofuntional thymine-functionalized PIB 4b is
shown in Figure 1b indicating one main series, where each series
of ions is separated by 56 Da, the mass of the repeating unit. In
agreement with observations made in our group,34 this signal
Critical for the following investigations was the synthesis
of pure polymeric material, truly containing the hydrogen-
bonding end groups (thymine and 2,6-diaminotriazine)
affixed to each polymer chain quantitatively. We therefore
employed two strategies for the synthesis of the thymine
(3-5) and triazine (6, 7) modified polyisobutylenes, either
relying on the previously developed combination of living
cationic polymerization with activated chloromethyl ethers
(3)34 or via a combination of the living cationic polymeriza-
tion with azide/alkyne “click” chemistry (4-7).11,37-41 As
the first synthetic strategy via chloromethyl ethers is only
applicable for monofunctionalized polyisobutylenes up to a
molecular weight of ∼10 000 g mol-1, the “click” approach
was probed for the synthesis of bifunctional modified PIBs of
medium and higher molecular weights.
Thus, the thymine (4a,b, 5) and 2,6-diaminotriazine-
functionalized PIBs (6a,b, 7) were prepared by azide/alkyne
“click” reaction starting from the azido-functionalized PIB
species and the corresponding alkyne, bearing the respective
supramolecular group (see Scheme 2).
Azido-functionalized PIBs were reacted in a “click”-type
reaction with the alkyne-bearing hydrogen-bonding motives
8 and 9 to yield the final 2,6-diaminotriazine (6a,b, 7) or
thymine functionalized (4a,b, 5) PIB polymer, respectively.
In all cases, microwave irradiation proved advantageous to
achieve a complete conversions of the azide and increases the
yield of the final, fully substituted polymer. Both the respec-
tive mono- and bifunctionalized polymers (bearing one or
two hydrogen-bonding motives at their chain ends) were
prepared.
can be assigned to the expected ion [M Ag2]þ assuming
3
exchange of the acidic CO-NH-CO proton of the thymine
group with the Agþ ion of the silver salt. The theoretical m/z
value for the species [M Ag2]þ (n = 36), 2540.3 Da, is in good
3
agreement with the experimental value of 2541.0 Da. The
MALDI-TOF-MS spectra for monofunctional 2,6-diamino-
triazine-functionalized PIB 6a is shown in Figure 1a. The
spectrum shows one main series, where again each series of
ions was separated by 56 Da, the mass of the isobutylene
repeating unit. Signals of the main series can be assigned to
the respective [M H]þ ions of 6a, while the second minor series
3
can be assigned to [M Li]þ ions. The theoretical m/z value for a
3
species [M H]þ (n = 30) is 2050.117 Da, which again is in
excellent agreement with the measured value of 2050.592 Da.
The same holds true for the theoretical m/z value for a species
3
[M Li]þ (n = 30) is 2056.126 Da, a value that agrees well with
the signal at 2056.598 Da.
3
Characterization of Functionalized PIBs. Critical for the
following rheological investigations was the synthesis of pure
polymeric material, truly containing the hydrogen-bonding
end groups quantitatively. We therefore confirmed the struc-
ture via 1H and 13C NMR spectroscopy as well as MALDI-
TOF-MS measurements.
Association in Solution: NMR Titration Experiments. The
thermodynamic association constants in solution were de-
termined via NMR titration experiments using the respective
monofunctional polymers. Thus, the association constant
(Kassn) for the triple hydrogen bond thymine/2,6-diamino-
triazine and the corresponding dimerization constants (Kdim
)
Chromatographic and NMR spectroscopic data are shown
in Table 1, revealing an excellent match between the Mn(GPC)
and Mn(NMR) data, thus indicating a complete end group
were determined using the functionalized PIBs 3 and 6a. One
representative graph is shown in Figure 2 (for more see
Supporting Information). The obtained values are in good

10010 Macromolecules, Vol. 43, No. 23, 2010
Herbst et al.
Figure 1. (a) MALDI spectrum of 2,4-diamonitriazine-functionalized PIB 6a, (b) MALDI spectrum of thymine-functionalized PIB 4b, (c) 13C NMR
spectrum of 2,4-diamonitriazine-functionalized PIB 6a, and (d) 13C NMR spectra of the thymine-functionalized PIB 4b.
same kind. Accordingly, one would expect also in the melt
dominant specific interactions and only to a much lesser
extent self-aggregation phenomena.
Association of Monofunctional Polymers Studied by Melt
Rheology. Any change of molecular weight or architecture of
the chain molecules leads to strong changes in the rheological
properties. Hence, rheology is well suited to reveal associa-
tion behavior of PIB equipped with supramolecular entities
at its chain ends. Additionally, temperature-dependent mea-
surements give important information on the chain dynamics
itself as well as on the temperature-dependent association of the
individual polymeric chain ends.
In a first step monofunctional polymers and supramolec-
ular mixtures (consisting of equimolar mixtures of the func-
tionalized polymers with respect to the end group) are
investigated by frequency sweep measurements at different
Figure 2. NMR titration experiment for the determination of the
temperatures. The supramolecular interaction is based on a
association constant in solution for the thymine/2,6-diaminotriazine
triple hydrogen bond, namely the thymine/2,6-diaminotria-
interaction (functionalized PIBs 3 and 6a). The concentration of PIB 3
zine interaction (Figure 3). The possible modes of associa-
was held constant at 0.022 mol/L.
tion in the liquid state for an A/B-type supramolecular
agreement with literature values obtained from low molec-
ular weight compounds. Thus, the association constant
(Kassn) for the thymine-2,6-diaminotriazine interaction in
solution is determined as Kassn = 1087 M-1 (870 M-1 in the
literature42), for the thymine/thymine interaction Kdim as
3.8 M-1 (4.3M-1 in the literature42), and for the triazine/triazine
interaction Kdim as 1.7 M-1 (2.2 M-1 in the literature42). The
measurements confirmed the much stronger specific interac-
tion of the complementary thymine-triazine groups in
solution, compared to the self-aggregation of groups of the
polymer are schematically depicted in Figure 3. The desired
aggregation between the complementary chain ends A and B
as well as self-aggregation between A ends or B ends are
possible. On the basis of the association constants from the
studies in solution, one would expect an overwhelming
contribution of aggregation between the complementary
chain ends.
An overview on samples and sample compositions for
the rheological measurements is provided in Table 2. Various
monofunctional PIBs with molecular weights of ∼3500 g/mol

Article
Macromolecules, Vol. 43, No. 23, 2010 10011
reveals a huge increase of the viscosity for sample M3 compared
to the nonfunctionalized polymer M1. Additionally, a constant
zero shear rate viscosity was not reached (see Figure 5a)
although M3 consists of a low molecular weight PIB below
the entanglement molecular weight Mc.43,44
Nevertheless, the viscosity at an angular frequency of
0.1 s-1 is displayed in Table 2 to allow a rough estimate for
the viscosity increase, which is higher by a factor of ∼33 for
M3 in comparison to M1.
Although the dimerization constant of the 2,6-diaminotria-
zine group in the solution (as determined by NMR titration
experiments, vide supra) (Kdim = 1.7 M-1) is comparable to
that of the thymine group (Kdim = 3.8 M-1), the viscosity
increase in the melt is significantly higher in M3 by a factor of
∼6 compared to M2, implying a stronger association or larger
aggregation number of the polymers bearing the 2,6-diamino-
triazine groups in the melt state. As it is well known that
melamine and melamine-based system (like the 2,6-diamino-
triazine end group of 6b in M3) undergo a strong aggregation in
the solid state,45 we assume a similar aggregation into larger
clusters in the case of the polymers containing thymine moi-
eties, as in sample M2 (cf. also SAXS measurements below).
Possible structures are shown in Scheme 3a (for M2) as well as
the formation of clusters for M3 (Scheme 3b). The observed
increase of the viscosity and the fact that no terminal flow is
observed (Figure 5a) can thus clearly be assigned to the
formation of strong, larger aggregates formed due to the
presence of the 2,6-diaminotriazine groups. Similar aggregated
structures have been observed before in multifunctional net-
work-forming supramolecular polymers containing uradiazole
cross-linkers46 or urea-based hydrogen bonds.47
To what extent self-aggregation is still present in case of
the specifically assembling polymer was measured using the
pseudo-homopolymer mixture M4. For the equimolar mix-
ture of thymine- and 2,6-diaminotriazine-functionalized PIB
M4 the viscosity in comparison to M1 is increased by a factor
of ∼39.5 (at 20 °C). Although the viscosity values for M4 at
lower temperatures are very similar to those of M3, there are
significant differences. M4 shows a Newtonian behavior at
low frequencies (see Figure 5b). Since M4 is an equimolar
mixture of PIB 6a and 3 (and thus an equimolar mixture of
thymine and 2,6-diaminotriazine groups), each functional
group is able to undergo the supramolecular interaction with
its counterpart. This specific interaction is strong and ex-
pected to suppress the formation of any other aggregates
(e.g., melamine-based cluster). This fact obviously limits the
aggregation number or size of the self-aggregates and hence
also their effect on viscosity. Accordingly, the zero shear rate
viscosity of M4 stays below the (extrapolated) value for M3
(compare parts a and b of Figure 5). But also for M4 more
than two chains are found to aggregate. Otherwise, the
viscosity increase by a factor of 39.5 compared to M1 could
not be understood. For a linear unentangled polymer one
would only expect an increase by a factor of 2 for a 2-fold
increase in molecular weight by combining two chains at one
end. However, as demonstrated by Flory et al.,44 a nonlinear
increase of viscosity was observed at molecular weights (PIB)
below 17 000 g mol-1. Therefore, the present increase is also
partially explained via this unnormal behavior.
Figure 3. Possible modes of association in A/B-type supramolecular
polymers in solution. A = thymine groups; B = 2,6-diaminotriazine
groups. The inset shows the molecular structure of the hydrogen-
bonding groups used.
Table 2. Sample Composition and Rheology Data for Homopolymers
and Polymer Mixtures M1-M11
η0 [Pa s] for T [°C]a
3
entry name compound 1 compound 2
20
40
60
80
1
M1
M2
M3
M4
M5
M6
M7
M8
M9
M10
M11
M12
1a
3
299
1610
61
201
17
37
6
9
2
3
6a
6a
1b
4a
6b
4a
2
9980b 2110b
574b 254b
4
3
11800 1420
241
149
56
65
5
2430
504
6
154000b 3300
42 400 7670
55000 8410
527
119
407
188
6
7
1890
1050
15
2370b 292b
8
6b
9
306
c
59
c
10
11
12
5
-
-
-
-
-
-
c
c
c
7
-
8450
c
c
5
7
25000b 1210b
a Values for the zero-shear rate viscosity. b Sample shows “flowlike”
behavior, but terminal flow is not yet reached; i.e., the viscosity is not the
true zero shear rate viscosity. Instead, the viscosity at 0.1 s-1 is displayed.
c No flow at this temperature in our frequency range.
(M1-M3) and ∼10 000 g/mol (M5-M7) as well as bifunc-
tional PIBs with molecular weights of ∼3500 g/mol (M9-M11)
were investigated. These are either pure PIBs (allyl-functiona-
lized PIBs) with nonassociative end groups (M1, M5, M9), pure
thymine- (M2, M6, M10), or pure 2,6-diaminotriazine-functio-
nalized PIBs (M3, M7, M11). Additionally, exact 1/1 stoichio-
metric mixtures (with respect to the associative hydrogen-
bonding end group) of the thymine-functionalized PIBs and
2,6-diaminotriazine-functionalized PIBs were prepared (M4,
M8, M12), enabling the desired A-B-type connection of the
PIB chains.
Table 2 also contains the measured zero-shear rate viscosities
at 20, 40, 60, and 80 °C. Plots of the viscosity vs temperature
are shown in Figure 4a for samples with a molecular
weight of ∼3500 g mol-1 and in Figure 4b for samples with
∼10 000 g mol-1. As anticipated, the viscosity decreases for all
polymers and mixtures with increasing temperature as a result
of increasing mobility of the polymer chains.
For the series of monofunctional PIBs with a molecular
weight of ∼3500 g/mol, the nonfunctionalized PIB (M1)
shows the lowest viscosity. With the introduction of the
thymine moiety (M2) at the PIB chain ends the measure-
ments show an increase of the melt viscosity by a factor of
∼5.3 at 20 °C. This can be attributed to the formation of
supramolecular aggregates due to the weak interaction
between the hydrogen bonds of the thymine moieties.
A similar (self-associative) interaction was also expected for
pure 2,6-diaminotriazine-functionalized PIB M3. Figure 4a
The comparable temperature dependence in the viscosity of
M3 and the nonfunctionalized polymer M1 shows that the
aggregates in M3 are stable over the studied temperature range.
The temperature dependence is governed by the chain dynamics.
Contrarily, for M4 the viscosity decreases with increasing tem-
perature stronger than for the nonfunctionalized polymer M1.
Hence, the dynamics of the supramolecular interaction gain in
importance. The associating hydrogen bonds are able to undergo

10012 Macromolecules, Vol. 43, No. 23, 2010
Herbst et al.
Figure 4. Viscosity vs temperature for (a) monofunctional PIBs (M1-M4) with a molecular weight of ∼3500 g/mol and (b) monofunctional PIBs
(M5-M8) with a molecular weight of ∼10 000 g/mol, (c) master curve of the storage and loss modulus for M8, and (d) temperature dependency of the
horizontal shift factor aT for M1-M8. The two lines mark the extremes in the temperature dependence.
Figure 5. Viscosity (real part of the viscosity), storage (G0), and loss modulus (G00) vs angular frequency at 20 °C (a) for M3 and M2 and (b) for M4.

Article
Macromolecules, Vol. 43, No. 23, 2010 10013
Scheme 3. Possible Structures Formed by Aggregation via Hydrogen
Bonds of Monofunctional (a-c) and Bifunctional Polymers (d-g)
Only M3 shows a temperature dependence similar to those of
the nonfunctionalized PIBs, which can be assigned to the
formation of nondynamic clusters, which were stable up to
temperatures of 90 °C (as revealed by SAXS measurement)
and thus covering the whole temperature range of the
rheology experiment.
SAXS Measurements of Monofunctional Polymers. In
order to check for structural signs of the self-aggregation
of samples with solely pendant thymine and 2,6-triazine
moieties, samples M2-M4 were investigated by SAXS mea-
surements. While for pure thymine-functionalized PIB M2
no excess scattering could be observed (possibly due to a lack
of contrast, see Supporting Information), SAXS measure-
ments for M3 and M4 in Figure 6 clearly show temperature-
dependent excess peaks. For pure 2,6-diaminotriazine-func-
tionalized PIB M3 the ratio between the positions of the
√ √
observed scattering peaks is 1: 2: 3, which can be assigned
to a BCC lattice (body-centered cubic). A BCC grid is typical
for block copolymers made of a long and a short block.49
Since the 2,6-diaminotriazine-functionalized PIB 6a (M3)
can be seen as a block copolymer where one block is the PIB
chain and the other one is the 2,6-diaminotriazine moiety,
this observation is in agreement with the structure of the
polymer. It is noteworthy that the BCC structure is destroyed
only at temperatures above 90 °C, thus explaining the high
values for the melt viscosity up to temperatures of 80 °C and the
lack of terminal flow. The BCC structure is formed again after
cooling to 20 °C, indicating a reversible cluster formation. The
observed periodicity of the BCC structure d=2π/qmax ∼6.5 nm
compares reasonably well with the estimated end-to-end dis-
tance of the PIB chains (2 nm as calculated with the ideal chain
model), taking into account that the chains might in reality be
considerably stretched.
For the equimolar mixture M4 of thymine- and 2,6-diami-
notriazine-functionalized PIB the SAXS measurement reveals
excess scattering, but no well-ordered structure. The presence
of thymine groups in M4 suppresses the formation of a BCC
structure of the 2,6-diaminotriazine-functionalized PIB chains
and can thus be seen as an additional indication for a selective
thymine-2,6-diaminotriazine interaction in the solid state.
For the monofunctional analogues with a higher molecular
weight (M6-M8) SAXS measurements were performed in
order to prove whether there is an influence of the molec-
ular weight on the structure formation. For pure thymine-
functionalized PIB M6 again only a very weak excess scattering
signal could be observed (see Supporting Information). In
contrast to M3 (Figure 6b), the higher molecular weight
analogue M7 shows no well-ordered periodic structure
(Figure 6d). The equimolar mixture M8 of monofunc-
tional thymine- and 2,6-diaminotriazine-functionalized
PIB chains reveals (Figure 6c) again a peak, but in com-
parison to the low molecular weight analogue sample M4
(Figure 6a) at lower q. This difference reflects the higher
molecular weight of M8. The different results in SAXS
measurements for samples M6-M8 again are compatible
with a selective thymine-2,6-diaminotriazine interaction
in the solid state.
an exchange of their counterparts on the same time scale as the
external mechanical perturbation and hence the viscosity addi-
tionally decreases. This proves the specific aggregation of the
polymers via the hydrogen-bonding moieties used as shown in
Scheme 3c.
For the monofunctional samples M5-M8 with a molecular
weight of ∼10 000 g mol-1 the melt viscosity in dependence on
temperature is displayed in Figure 4b. In comparison to the
lower molecular weight analogues (M1-M4) the values for the
melt viscosity are about 1 order of magnitude larger, suggesting
an entangled melt. This is somewhat surprising as the critical
molecular weight for entanglement (Mc) was reported to be
16 020 or 17 000 g mol-1 for poly(isobutylene).43,44 Despite
their higher viscosities, the curves for M5, M7, and M8 reveal a
similar shape in comparison to their lower molecular weight
analogues. The increased melt viscosity values for M7 and M8
indicate cluster formation (M7) as well as a supramolecular
A-B-type aggregation (M8).
A significant difference is observed for pure thymine-
functionalized PIB (M6) at 20 °C, where the highest viscosity
within this series is observed. In this sample, the huge drop of
the viscosity by increasing the temperature to 40 °C indicates
the presence of only weak interactions between thymine-
thymine groups. Since the molecular weight of the thymine-
functionalized PIB in M6 is ∼11 000 g/mol, the interaction of
the polymer chains between spatial separate supramolecular
aggregates can be seen as a possible cause for the unexpected
high values for the melt viscosity at 20 °C. The fact that for
M6 a zero shear viscosity was not observed within our fre-
quency range confirms this assumption. By increasing the
temperature and thus increasing the chain mobility, such an
interaction is no longer possible and a huge drop of the melt
viscosity is observed.
Although the supramolecular interaction introduced by
the attachment of hydrogen-bonding moieties onto the PIB
chain is strongly temperature dependent, master curves for
the monofunctional samples M1-M8 could be generated by
using simple horizontal frequency-temperature superposi-
tion. This was possible since the shape of the storage and loss
modulus vs angular frequence curves remained similar with
increasing temperature.48 As an example, the master curve of
M8 is displayed in Figure 4c as well as the shift factors for
M1-M8 in Figure 4d. Figure 4d reveals a stronger tempera-
ture dependence of the shift factors for the samples contain-
ing a hydrogen-bonding moiety (dynamic interactions) in
comparison to the nonfunctionalized samples M1 and M5.
Association of Bifunctional Polymers Studied by Melt
Rheology. In addition to the monofunctional samples, the
respective bifunctional PIB samples with a molecular weight
of ∼3500 g/mol were investigated by dynamic rheology. In
these samples the formation of more complex supramolecular
assemblies is possible. For bifunctional thymine-functionalized
PIB M10 or bifunctional 2,6-diaminotriazine-functionalized
PIB M11 the formation of a large variety of more complex
aggregates, such as supramolecular chains or cross-linked
clusters can be expected (Scheme 3d-f). The presence of both

10014 Macromolecules, Vol. 43, No. 23, 2010
Herbst et al.
Figure 6. SAXS measurements for (a) M4 at 20 °C (solid line) and 120 °C (dashed line), (b) M3 at 20 °C (solid line) and 120 °C (dashed line), (c) M8 at
20 °C (solid line) and 120 °C (dashed line), and (d) M7 at 20 °C (solid line) and 120 °C (dashed line).
Figure 7. Frequency dependence of storage and loss modulus for M10-M12 at (a) 20 and (b) 80 °C.
bis-thymine- and bis-2,6-diaminotriazine-functionalized
PIB chains in M12 should lead to a preferred formation of
supramolecular (-A-B-)n chains, possibly accompanied by
aggregation of those chain-end connections (Scheme 3g).
Pure allyl-functionalized PIB M9 reveals similar values for
the melt viscosity in comparison to the monofunctional ana-
logue M1 with a similar molecular weight, since the presence of
a second allyl end group does not lead to a significant change in
the melt viscosity (see Table 2).
In the case of M10, M11, and M12 the presence of bifunc-
tional polymer chains leads to a rubberlike material which is not
flowingat20and40°C, while in the case of M11 even at 60 °Cno
flowing was observed (see Supporting Information). Surpris-
ingly, pure bis-thymine-functionalized PIB M10 exhibits higher
values for the storage modulus (Figure 7a) than M12, although a
stronger interaction is expected for the bis(2,6-diaminopyridine)
sample. For bis-2,6-diaminotriazine-functionalized M11
the smallest values for the storage modulus were observed,
although the monofunctional analogue M3 revealed the stron-
gest effect on storage modulus within the series M1-M4 due to
the formation of strong clusters. At 60 °CM10 and M12 flow but
with large differences in the values for the melt viscosity. While
for pure thymine-functionalized PIB M10 a value of 2370 Pa s
3
was observed at 60 °C, the equimolar mixture of thymine- and
2,6-diaminotriazine-functionalized PIB chains M12 reveals a
value of 25 000 Pa s. This observation leads to the assumption
3
that the supramolecular aggregates in M10 and M12 differ from
each other. As already mentioned, the undirected supramolecu-
lar interaction between two or more thymine groups (or 2,6-
diaminotriazine groups) can lead to a network as displayed
in Scheme 3e, while the more directed interaction between
thymine and 2,6-diaminotriazine groups in M12 leads preferred
to the formation of long supramolecular (-A-B-)n chains.
The higher values for M10 for the storage modulus at 20 and
40 °C can thus be explained by the higher rigidity of the supra-
molecular network in comparison to the extended, sheetlike

Article
Macromolecules, Vol. 43, No. 23, 2010 10015
supramolecular aggregates shown in Scheme 3g. By increasing
the temperature (>60 °C), the weak thymine-thymine interac-
tion is released to a higher extent than the stronger thymine-2,6-
diaminotriazine interaction, thus explaining the significantly
lower viscosities of M10 in comparison to M12 at 60 °C.
For linear high molecular weight PIB the plateau modulus
was reported to be in the range of 2.5 ꢀ 105-3.2 ꢀ
105 Pa.43,50 In comparison, samples M10 (∼1 ꢀ 107 Pa)
and M12 (∼1 ꢀ 106 Pa) reveal much higher values for the
plateau modulus. This observation can be explained by the
formation of supramolecular cross-links as a result of undi-
rected interactions like those displayed in Scheme 3e. These
undirected interactions are especially existent in M10, since
here only thymine groups are present.
M12 reveals a lower plateau value than M10, since the
undirected interactions are suppressed by the strong and
directed thymine-2,6-diaminotriazine interaction, and thus
the amount of supramolecular cross-links is reduced. The
fact that the plateau modulus for M12 is still higher than
the reported plateau value for the linear PIB may indicate
that also for M12 additional supramolecular cross-links
between different end group aggregates are present. How-
ever, the network formation of M10 was not observed for the
bis(2,6-diaminotriazine)-telechlic PIB M11 to the same ex-
tent. Until now, it cannot be specified why M11 reveals the
smallest values for the storage modulus within this series as
the formation of clusters (like in M3) should lead to the
formation of a strongly cross-linked network as displayed
in Scheme 3e. A possible explanation would be the partial
formation of aggregates like in Scheme 3f, which would lead
to a nonbridged cross-linked structure. The fact that M11
polymers by a single hydrogen bond. Obviously, larger aggre-
gates of chains are formed due to less specific interactions. The
formation of more complex aggregates such as microphase-
separated cluster-type assemblies is in some cases even reflected
by the appearance of regular peaks in the small-angle scattering
pattern. Nevertheless, the specific aggregation between thymine
or 2,6-diaminotriazine end groups clearly showed up in the
rheological properties. Therefore, not only the strength of the
individual hydrogen bond but also secondary effects such as
unspecific clustering of the H-bonds have to be taken into
account in order to achieve further understanding and control
of supramolecular polymers in the melt state.
Acknowledgment. We are thankful for grants DFG BI 1337/
7-1 (F.H., W.H.B.) and DFG TH 1281/5-1 as well as grants DFG
INST 271/249-1, INST 271/247-1, and INST 271/248-1 (W.H.B.)
for financial support. Significant investments into our NMR
infrastructure from the European Regional Development Fund
(ERDF) by the European Union are gratefully acknowledged.
J.B. and S.G. acknowledge support by ProNet-T3 (BMBF).
Supporting Information Available: Detailed information
(synthesis, column chromatography, NMR data) for compound 9
as well as thymine-functionalized PIBs 4, 5 and 2,6-diaminotriazine-
functionalized PIBs6a-c,7; MALDI-TOF-MS spectra for thymine-
functionalized PIB 5 and 2,6-diaminotriazine-functionalized PIB 7;
SAXS measurements of M2, M6, M10, M11, M12; temperature-
dependent SAXS measurement for M3; detailed description of the
NMR titration experiments, rheology data for M10, M11, and M12
at 40 and 60 °C; glass transition temperatures for M1-M12
measured via DSC. This material is available free of charge via the
Internet at http://pubs.acs.org.
does not flow at 60 °C and reveals at 80 °C with 8450 Pa s
3
a much higher viscosity than M10 (292 Pa s) and M12
(1210 Pa s) indicates that the formation of strong clusters
are also present up to high temperatures in M11.
3
References and Notes
3
(1) Binder, W. H.; Zirbs, R. In Advances Polymer Science; Binder, W.,
Ed.; Springer: Berlin, 2007; Vol. 207, pp 1-78.
(2) Bouteiller, L. In Advances Polymer Science; Binder, W. H., Ed.;
Springer: Berlin, 2007; Vol. 207, pp 79-112.
(3) Brunsveld, L.; Folmer, B. J. B.; Meijer, E. W.; Sijbesma, R. P.
SAXS Measurements of Bifunctional Polymers. For poly-
meric samples containing bifunctional PIBs M10-M12
SAXS measurements do not reveal any ordered structure
(see Supporting Information). According to the monofunc-
tional samples M2 and M6 for bis-thymine-functionalized
M10 no excess scattering was observed. Although low molec-
ular weight 2,6-diaminotriazine-functionalized PIB M3
reveals a BCC structure, for the bifunctionalized analogous
M11 no ordered structure was observed, which can be
assigned to the reduced mobility of the polymer chain if
one end is already attached to some kind of aggregate.
Chem. Rev. 2001, 101, 4071–4098.
(4) ten Brinke, G.; Ruokolainen, J.; Ikkala, O. In Advances Polymer
Science; Binder, W. H., Ed.; Springer: Berlin, 2007; Vol. 207, pp
113-177.
(5) Xu, H.; Srivastava, S.; Rotello, V. In Advances Polymer Science;
Binder, W. H., Ed.; Springer: Berlin, 2007; Vol. 207, pp 179-198.
(6) Cordier, P.; Tournilhac, F.; Soulie-Ziakovic, C.; Leibler, L. Nature
2008, 451, 977–980.
ꢀ
(7) Montarnal, D.; Cordier, P.; Soulie-Ziakovic, C.; Tournilhac, F.;
Leibler, L. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 7925–
7936.
(8) Kautz, H.; van Beek, D. J. M.; Sijbesma, R. P.; Meijer, E. W.
Conclusion
Supramolecular polymers bearing mono- and bifunctional
chain ends with hydrogen-bonding units were prepared, and
their association behavior was investigated by dynamic rheology
in the melt state and SAXS. The synthetic approach combined
living cationic polymerization with either azide/alkyne “click”
reactions or nucleophilic substitution reactions and enabled a full
end-group transformation to the final polymers, modified with
either thymine- or 2,6-diaminotriazine end groups.
Subsequent rheological measurements were conducted on
either pure mono- or bifunctionalized polymers or of equimolar
mixtures generating A/B-type aggregates. It could be demon-
strated that supramolecular polymers and their mixtures show a
strong and reversible temperature dependence of their mechan-
ical behavior. The temperature dependence of the rheological
properties is stronger than for the nonfunctionalized PIB homo-
polymers. This clearly demonstrates the additional effect of the
dynamics of the hydrogen bonds compared to solely chain
dynamics. Generally, the effects as e.g. the increase in viscosity
is not compatible with a simple specific connection of two
Macromolecules 2006, 39, 4265–4267.
(9) Wisse, E.; Spiering, A. J. H.; Pfeifer, F.; Portale, G.; Siesler, H. W.;
Meijer, E. W. Macromolecules 2009, 42, 524–530.
(10) Dankers, P. Y. W.; Harmsen, M. C.; Brouwer, L. A.; Van Luyn,
M. J. A.; Meijer, E. W. Nature Mater. 2005, 4, 568–574.
ꢀ
(11) Binder, W. H.; Petraru, L.; Roth, T.; Groh, P. W.; Palfi, V.; Keki,
S.; Ivan, B. Adv. Funct. Mater. 2007, 17, 1317–1326.
€
(12) Hackethal, K.; Dohler, D.; Tanner, S.; Binder, W. H. Macromo-
lecules 2010, 43, 1761–1770.
(13) Binder, W. H.; Bernstorff, S.; Kluger, C.; Petraru, L.; Kunz, M.
Adv. Mater. 2005, 17, 2824–2828.
(14) Kunz, M. J.; Hayn, G.; Saf, R.; Binder, W. H. J. Polym. Sci., Part
A: Polym. Chem. 2004, 42, 661–674.
(15) Cates, M. E. Macromolecules 1987, 20, 2289–2296.
(16) Hammes, G. G.; Park, A. C. J. Am. Chem. Soc. 1968, 90, 4151–
4157.
(17) Sontjens, S. H. M.; Sijbesma, R. P.; van Genderen, M. H. P.;
Meijer, E. W. J. Am. Chem. Soc. 2000, 122, 7487–7493.
€
(18) Embrechts, A.; Schonherr, H.; Vancso, G. J. J. Phys. Chem. B 2008,
112, 7359–7362.
(19) Vancso, G. Angew. Chem., Int. Ed. 2007, 46, 3794–3796.

10016 Macromolecules, Vol. 43, No. 23, 2010
Herbst et al.
€
ꢀ
(20) Zou, S.; Schonherr, H.; Vancso, G. J. Angew. Chem., Int. Ed. 2005,
(36) Kennedy, J. P.; Ivan, B. Designed Polymers by CARBOCATIONIC
MACROMOLECULAR ENGINEERING - Theory and Practice;
Carl Hanser Verlag: Munich, 1992.
44, 956–959.
(21) Knoben, W.; Besseling, N. A. M.; Bouteiller, L.; Cohen-Stuart,
M. A. Phys. Chem. Chem. Phys. 2005, 7, 2390–2398.
(22) Knoben, W.; Besseling, N. A. M.; Cohen-Stuart, M. A. J. Chem.
Phys. 2007, 126, 024907–9.
(37) Binder, W. H.; Enders, C.; Herbst, F.; Hackethal, K. Synthesis and Self
Assembly of Hydrogen-Bonded Supramolecular Polymers. In Complex
Macromolecular Architectures: Synthesis, Characterization, and Self
Assembly; Hadjichristidis, N., Tezuka, Y., Hirao, A., Eds.; Wiley-VCH:
Berlin, 2010, in press.
(23) Courtois, J.; Baroudi, I.; Nouvel, N.; Degrandi, E.; Pensec, S.;
ꢀ
Ducouret, G.; Chaneac, C.; Bouteiller, L.; Creton, C. Adv. Funct.
Mater. 2010, 20, 1803–1811.
(24) Pensec, S.; Nouvel, N.; Guilleman, A.; Creton, C.; Boue, F. o.;
(38) Binder, W. H.; Kluger, C. Curr. Org. Chem. 2006, 10, 1791–1815.
(39) Binder, W. H.; Sachsenhofer, R. Macromol. Rapid Commun. 2007,
28, 15–54.
ꢀ
Bouteiller, L. Macromolecules 2010, 43, 2529–2534.
€
(25) Muller, M.; Dardin, A.; Seidel, U.; Balsamo, V.; Ivan, B.; Spiess,
(40) Binder, W. H.; Sachsenhofer, R. Macromol. Rapid Commun. 2008,
29, 952–981.
H. W.; Stadler, R. Macromolecules 1996, 29, 2577–2583.
(26) Schirle, M.; Hoffmann, I.; Pieper, T.; Kilian, H.-G.; Stadler, R.
Polym. Bull. 1996, 36, 95–102.
(27) Loveless, D. M.; Abu-Lail, N. I.; Kaholek, M.; Zauscher, S.; Craig,
S. L. Angew. Chem., Int. Ed. 2006, 45, 7812–7814.
(28) Serpe, M. J.; Craig, S. L. Langmuir 2007, 23, 1626–1634.
(29) Yount, W. C.; Loveless, D. M.; Craig, S. L. J. Am. Chem. Soc. 2005,
127, 14488–14496.
(41) Binder, W. H.; Zirbs, R. “Click” Chemistry in Macromolecular
Synthesis; John Wiley & Sons, Inc.: New York, 2002.
(42) Beijer, F. H.; Sijbesma, R. P.; Vekemans, J. A. J. M.; Meijer, E. W.;
Kooijman, H.; Spek, A. L. J. Org. Chem. 1996, 61, 6371–6380.
(43) Aharoni, S. M. Macromolecules 1986, 19, 426–434.
(44) Fox, T. G.; Flory, P. J. J. Phys. Chem. 1951, 55, 221–234.
(45) Gong, H.; Krische, M. J. J. Am. Chem. Soc. 2005, 127, 1719–1725.
(46) Dardin, A.; Boeffel, C.; Spiess, H. W.; Stadler, R.; Samulski, E. T.
Acta Polym. 1995, 46, 291–299.
(30) Yount, W. C.; Loveless, D. M.; Craig, S. L. Angew. Chem., Int. Ed.
2005, 44, 2746–2748.
(31) Sivakova, S.; Bohnsack, D. A.; Mackay, M. E.; Suwanmala, P.;
Rowan, S. J. J. Am. Chem. Soc. 2005, 127, 18202–18211.
(32) Langille, N. F.; Jamison, T. F. Org. Lett. 2006, 8, 3761–3764.
(33) Lindsell, E. W.; Murray, C.; Preston, P. N.; Woodman, T. A. J.
Tetrahedron 2000, 56, 1233–1245.
(47) Wisse, E.; Govaert, L. E.; Meijer, H. E. H.; Meijer, E. W. Macro-
molecules 2006, 39, 7425–7432.
(48) van Beek, D. J. M.; Spiering, A. J. H.; Peters, G. W. M.; te
Nijenhuis, K.; Sijbesma, R. P. Macromolecules 2007, 40, 8464–
8475.
(34) Binder, W. H.; Kunz, M. J.; Kluger, C.; Hayn, G.; Saf, R.
Macromolecules 2004, 37, 1749–1759.
(35) Ammann, C.; Meier, P.; Merbach, A. J. Magn. Reson. 1982, 46,
319–321.
(49) Hamley, I. W. The Physics of Block Copolymers; Oxford University
Press: Oxford, 1998.
(50) Fetters, L. J.; Lohse, D. J.; Richter, D.; Witten, T. A.; Zirkel, A.
Macromolecules 1994, 27, 4639–4647.