Modular Domain Structure: A Biomimetic Strategy for Advanced Polymeric Materials

Article abstract of DOI:10.1021/ja039127p

A long lasting challenge in polymer science is to design polymers that combine desired mechanical properties such as tensile strength, fracture toughness, and elasticity into one structure. A novel biomimetic modular polymer design is reported here to add

Full text of DOI:10.1021/ja039127p

Published on Web 01/29/2004
Modular Domain Structure: A Biomimetic Strategy for
Advanced Polymeric Materials
Zhibin Guan,* Jason T. Roland, Jane Z. Bai, Sharon X. Ma,
Theresa M. McIntire, and Maianh Nguyen
Contribution from the Department of Chemistry, 516 Rowland Hall, UniVersity of California,
IrVine, California 92697-2025
Received October 20, 2003; E-mail: zguan@uci.edu
Abstract: A long lasting challenge in polymer science is to design polymers that combine desired mechanical
properties such as tensile strength, fracture toughness, and elasticity into one structure. A novel biomimetic
modular polymer design is reported here to address this challenge. Following the molecular mechanism
used in nature, modular polymers containing multiple loops were constructed by using precise and strong
hydrogen bonding units. Single-molecule force-extension experiments revealed the sequential unfolding
of loops as a chain is stretched. The excellent correlation between the single-molecule and the bulk properties
successfully demonstrates our biomimetic concept of using modular domain structure to achieve advanced
polymer properties.
Introduction
strength, toughness, and sometimes elasticity as well, three
properties that are rarely found in one synthetic polymer.4 Recent
Mechanical properties are among the most fundamental
properties of polymeric materials.1 Despite the great progress
made in recent years in polymeric materials science and
technologies,2,3 it remains a great challenge to design a polymer
that combines important mechanical properties such as tensile
strength, fracture toughness, and elasticity into one structure,
because these properties usually require different molecular
mechanisms and, hence, are generally considered exclusive to
each other.4 Nature, on the other hand, has evolved complex
and elegant polymeric materials that combine these properties
fulfilling a myriad of functions. Nature achieves this by precisely
controlling the molecular and nanoscopic structures of biom-
acromolecules such as proteins. The secondary and tertiary
structures of proteins are not only important for their biological
functions but also critical for their superior material properties.
For example, silks,5 cell adhesion proteins,6 and connective
proteins existing in both soft and hard tissues such as muscle,7-10
seashell,11 and bone12 exhibit a remarkable combination of high
single-molecule nanomechanical studies revealed that the
combination of these mechanical properties in natural materials
arise from their unique molecular and nanoscopic structures.7-10
The biopolymers in these systems share a common modular
structure composed of a linear array of nanodomains, in which
each domain is held together by secondary interactions, e.g.,
hydrogen-bonding, hydrophobic, and van der Waals interactions.
One marvelous example of this modular design is titin, a giant
protein of muscle sarcomere that has 200-300 repeating
modules (Figure 1). Sequential unfolding of the titin domains
results in a saw-tooth pattern in the force-extension curve, with
each peak corresponding to the unfolding of an individual
domain.5-12 The saw-tooth shaped force-extension curves
reflect a constant force sustained over the entire extension, which
makes the polymer strong, along with a large area under the
force-extension curve, making it tough as well.11 In addition,
when the external force is removed, the unfolded domains of
modular proteins will refold automatically, making them elastic.
Here we report the first application of this elegant molecular
mechanism commonly employed in nature to address a long-
standing challenge in material science: to design polymers
having precise molecular nanostructures that combine high
strength, toughness, and elasticity. In contrast to natural
polymers, synthetic polymeric materials usually lack precise
higher ordered structures, despite the fact that significant
progress has been made in designing short oligomers with well-
defined secondary structures.13-20 Furthermore, there is a gap
in connecting the single-molecule properties with the bulk
(1) Hearle, J. W. S. Polymers and Their Properties, Vol. 1: Fundamentals of
Structure and Mechanics; 1982.
(2) Krejchi, M. T.; Atkins, E. D. T.; Waddon, A. J.; Fournier, M. J.; Mason,
T. L.; Tirrell, D. A. Science 1994, 265, 1427-1432.
(3) Rathore, O.; Sogah, D. Y. J. Am. Chem. Soc. 2001, 123, 5231-5239.
(4) Booth, C., Price, C., Eds. ComprehensiVe Polymer Science: The Synthesis,
Characterization, Reactions, and Applications of Polymers, Vol. 2: Polymer
Properties; 1989.
(5) Oroudjev, E.; Soares, J.; Arcdiacono, S.; Thompson, J. B.; Fossey, S. A.;
Hansma, H. G. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 6460-6465.
(6) Law, R.; Carl, P.; Harper, S.; Dalhaimer, P.; Speicher, D. W.; Discher, D.
E. Biophys. J. 2003, 84, 533-544.
(7) Rief, M.; Gautel, M.; Oesterhelt, F.; Fernandez, J. M.; Gaub, H. E. Science
1997, 276, 1109-1112.
(8) Kellermayer, M. S. Z.; Smith, S. B.; Granzier, H. L.; Bustamante, C. Science
1997, 276, 1112-1116.
(11) Smith, B. L.; Schaffer, T. E.; Viani, M.; Thompson, J. B.; Frederick, N.
A.; Kind, J.; Belcher, A.; Stucky, G. D.; Mors, D. E.; Hansma, P. K. Nature
1999, 399, 761-763.
(12) Thompson, J. B.; Kindt, J. H.; Drake, B.; Hansma, H. G.; Morse, D. E.;
Hansma, P. K. Nature 2001, 414, 773-776.
(9) Li, H.; Oberhauser, A. F.; Fowler, S. B.; Clarke, J.; Fernandez, J. M. Proc.
Natl. Acad. Sci. U.S.A. 2000, 97, 6527-6531.
(10) Marszalek, P. E.; Lu, H.; Li, H.; Carrion-Vazquez, M.; Oberhauser, A. F.;
Schulten, K.; Fernandez, J. M. Nature 1999, 402, 100-103.
9
2058
J. AM. CHEM. SOC. 2004, 126, 2058-2065
10.1021/ja039127p CCC: $27.50 © 2004 American Chemical Society

A Biomimetic Strategy for Advanced Polymeric Materials
A R T I C L E S
Polymerization reactions were carried out via prepolymer
formation followed by chain extension. Poly(tetramethylene
glycol) (PTMG, 1 equiv, with a number average molecular
weight, Mn, of 1400 g/mol) was injected slowly by a syringe
pump into a solution of 2 equiv of 1,6-diisocyanatohexane to
form a prepolymer. The PNB-protected UPy monomer 4 was
added as a chain extender to complete the polymerization
process, providing a fully protected control polymer B that is
unable to form UPy dimers because the quadruple hydrogen
bonding sites are blocked by the PNB protection (Scheme 1b).
Polymerization reactions with the free UPy monomer 5 were
carried out in the same manner, providing a deprotected polymer
A in which the free UPy units dimerize to form loops along
polymer chains (Scheme 1b). Proton nuclear magnetic resonance
spectroscopy (1H NMR) confirmed the formation of UPy dimers
in polymer A, showing three characteristic peaks for the
hydrogen-bonding protons of the UPy dimer at 10.28, 11.95,
and 13.20 ppm. For the control polymer B, only one intramo-
lecularly hydrogen bonded proton was observed at 9.2 ppm.
Comparison of the 1H NMR spectrum of polymer A with those
of other UPy dimer systems reported in the literature revealed
that the UPy units exist as keto tautomers in our system.23 The
number-averaged molecular weights of the polymers are typi-
cally about 70 000 g/mol as measured by gel permeation
chromatography (GPC) using polystyrene standards.
Figure 1. Concept of biomimetic modular polymer design. (a) A small
section of titin, which has 200-300 repeating immunoglobulin (Ig) domains.
(b) The design of modular polymer containing multiple loops held by
secondary forces.
physical properties for synthetic materials. In this report, we
use titin as a model to design the first synthetic polymers having
well-defined modular structures by using strong hydrogen-
bonding units (Figure 1). We also report our investigation of
their nanomechanical properties at the single-molecule level
through atomic force microscopy (AFM) and the correlation of
the single-molecule properties with macroscopic material prop-
erties.
Results and Discussion
Based on molecular modeling and single-molecule studies,
the six hydrogen bonds between â-strands A′ and G in the
immunoglobulin (Ig) module of titin play a critical role in its
mechanical stability.21 In our biomimetic design, a strong
quadruple hydrogen bonding motif, 2-ureido-4-pyrimidone
(UPy), was employed to direct the formation of loops along a
polymer chain. Meijer and co-workers have shown beautifully
that UPy dimerizes strongly with a dimerization constant Kdim
> 108 M-1 in toluene.22 Based on the magnitude of the
dimerization constant, the free energy required to break the UPy
dimer is more than 10.9 kcal/mol. This is comparable to protein
unfolding energy and lower than typical covalent bond energies,
and, therefore, suits our biomimetic study.
The synthesis of the UPy-containing monomers began with
the alkylation of ethyl acetoacetate with allyl bromide followed
by condensation with guanidine carbonate to afford an iso-
cytosine, which was further treated with allyl isocyanate to
provide compound 2 (Scheme 1a). The isocytosine ring was
protected as para-nitrobenzyl (PNB) ether for improved solubil-
ity and also for further synthesis of a control polymer to be
used in comparative studies. Hydroboration of the diolefin 3
afforded the protected UPy-containing monomer 4. The PNB
protection was removed by catalytic hydrogenation to give the
free UPy monomer 5. Both the protected UPy monomer 4 and
the free UPy monomer 5 will be incorporated into polymer
chains for comparative studies.
Both the control polymer (B) and the modular polymer having
multiple loops (A) were subjected to single-molecule force-
extension studies using AFM. The single-molecule nanome-
chanical properties for many biopolymers including proteins,7-10
polysaccharides,24 DNAs,25 and synthetic polymers26 have been
studied in aqueous environment using AFM. For our polymers,
water is a poor solvent and disrupts the hydrogen bonding
between the UPy dimers. Previous studies have shown that
stretching a single polymer chain in its collapsed conformation
(in poor solvents or in air) gave curves that are not related to
the polymer secondary structure due to the Rayleigh-Plateau
instability.27 To obtain meaningful force spectra reflecting the
inherent structure of our polymers, our single-chain stretching
experiments were done in toluene, which is a good solvent for
the polymer and should preserve the hydrogen bonding in the
UPy dimers because of its low polarity. A special liquid-cell
reservoir was custom-built for our AFM studies to accommodate
the low surface tension of toluene in sealing the liquid cell.
We followed similar protocols reported previously7,11,26,28 in
single-chain studies of our polymers using AFM. Either the
modular polymer A or the control polymer B was allowed to
be adsorbed from a dilute solution (10-5-10-7 M) onto a freshly
coated gold surface. The sample was probed in our custom-
built liquid cell using toluene as the solvent. Following other
studies, physisorption was used to pick up a polymer chain by
the AFM tip (Si3N4 tips; Park Scientific Instrument). When the
(13) Hill, D. J.; Mio, M. J.; Prince, R. B.; Hughes, T. S.; Moore, J. S. Chem.
ReV. 2001, 101, 3893-4011.
(14) Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. Chem. ReV. 2001, 101,
3219-3232.
(15) Nowick, J. S. Acc. Chem. Res. 1999, 32, 287-296.
(16) Cornelissen, J. J. L. M.; Donners, J. J. J. M.; de Gelder, R.; Graswinckel,
W. S.; Metselaar, G. A.; Rowan, A. E.; Sommerdijk, N. A. J. M.; Nolte,
R. J. M. Science 2001, 293, 676-680.
(17) Hirschberg, J. H.; Brunsveld, L.; Ramzi, A.; Vekemans, J. A.; Sijbesma,
R. P.; Meijer, E. W. Nature 2000, 407, 167-170.
(23) Beijer, F. H.; Sijbesma, R. P.; Kooijman, H.; Spek, A. L.; Meijer, E. W. J.
Am. Chem. Soc. 1998, 120, 6761-6769.
(18) Zimmerman, S. C.; Wendland, M. S.; Rakow, N. A.; Zharov, I.; Suslick,
K. S. Nature 2002, 418, 399-403.
(24) Rief, M.; Oesterhelt, F.; Heymann, B.; Gaub, H. E. Science 1997, 275,
1295-1297.
(19) Lehn, J. M. Angew. Chem. 1988, 100, 91-116.
(20) Berl, V.; Huc, I.; Khoury, R. G.; Krische, M. J.; Lehn, J.-M. Nature 2000,
407, 720-723.
(25) Smith, S. B.; Cui, Y.; Bustamante, C. Science 1996, 271, 795-799.
(26) Ortiz, C.; Hadziioannou, G. Macromolecules 1999, 32, 780-787.
(27) Haupt, B. J.; Senden, T. J.; Sevick, E. M. Langmuir 2002, 18, 2174-
2182.
(21) Lu, H.; Schulten, K. Biophys. J. 2000, 79, 51-65.
(22) Sijbesma, R. P.; Beijer, F. H.; Brunsveld, L.; Folmer, B. J. B.; Hirschberg,
J. H. K. K.; Lange, R. F. M.; Lowe, J. K. L.; Meijer, E. W. Science 1997,
278, 1601-1604.
(28) Li, H.; Linke, W. A.; Oberhauser, A. F.; Carrion-Vazquez, M.; Kerkvliet,
J. G.; Lu, H.; Marszalek, P. E.; Fernandez, J. M. Nature 2002, 418, 998-
1002.
9
J. AM. CHEM. SOC. VOL. 126, NO. 7, 2004 2059

A R T I C L E S
Guan et al.
Scheme 1. Synthesis of (a) the UPy-Containing Monomers and (b) the UPy-Containing Polymersa
a In modular polymer A, the green loop stands for the PTMG linker.
tip was retracted, extension curves such as the ones shown in
Figure 2 were recorded.
to the polymer chain. Therefore, the peak force corresponds to
the force required to break a UPy dimer at a specific pulling
rate, and the spacing between the peaks represents the gain in
length during the unfolding.31
Further evidence for the single-chain stretching comes from
fitting the force-extension data with the wormlike chain (WLC)
model (eq 1), which predicts the relationship between the
extension of a single polymer chain (x) and the entropic restoring
force generated (F(x)).7,32 The adjustable parameters of the WLC
model are the persistence length, b, and the contour length of
the polymer, L. The k and T in eq 1 are the Boltzmann constant
and temperature, respectively.
The single-chain force-extension curves for both the control
polymer (B) and the modular polymer (A) are shown in Figure
2. For the control polymer B that does not have the modular
structure, we typically only observed a single peak29 that is
characteristic for entropic extension of a random coil chain
(Figure 2b). On the contrary, the modular polymer A consis-
tently showed distinct sawtooth patterns in the force-extension
curve (Figure 2a and c),30 similar to those observed in titin and
other modular biopolymers.7-10 The big contrast for the force-
extension curves observed for the polymers A and B strongly
suggests the sequential unfolding of the UPy dimers along the
modular polymer A chain. As the polymer chain was stretched,
the force rose gradually until one UPy dimer could not sustain
the force and broke instantaneously, adding additional length
F(x) ) (kT/b)[0.25(1 - x/L)-2 - 0.25 + x/L]
(1)
Both the control polymer (B) and the modular polymer (A) could
be fit nicely with the WLC model with a persistence length of
0.2 nm (Figure 2b and c, solid lines). Persistence lengths ranging
(29) In pulling the protected UPy control polymer B, we most time observed
only a single peak. At about 0.1% rate, we observed two or more peaks on
one curve. For those showing multiple peaks, they could not be fitted with
the wormlike chain (WLC) model for a single chain, indicating that the
multiple peaks originated from multiple chain adsorptions.
(30) In pulling the deprotected UPy modular polymer A, the success rate of
picking up a chain is about 2%. In one set of experiments with a total of
48 force-extension curves, the statistical information is as follows: 6 curves
show two peaks, 16 curves show three peaks, 16 curves show four peaks,
7 curves show five peaks, 2 curves show six peaks, and 1 curve shows
seven peaks. Among all the unfolding peaks observed from this pool of
curves, 52 peaks show peak force around 100 pN and 61 peaks show peak
force around 200 pN. Of this pool of curves, about 70% of the final
extension is in the range of 200-300 nm.
(31) Whereas the single-molecule stretching data could be explained by other
alternative mechanisms, we believe our current explanation is the simplest
and most straightforward one to account for the experimental data. The
difficulty here is that there is no definitive way to rule out any other
possibility. The adsorption was done in very dilute polymer solution to
avoid aggregation. Before collecting the force-extension data, the AFM
tip was gradually pulled away from the surface step by step to get rid of
weak nonspecific interactions and finally leave one single chain strongly
adsorbed on the substrate and AFM tip. We did pick up multiple chains
from time to time, but in those cases, the multiple peaks could not be fitted
with the WLC single-chain model.
(32) Bustamante, C. Science 1994, 265, 1599-1600.
9
2060 J. AM. CHEM. SOC. VOL. 126, NO. 7, 2004

A Biomimetic Strategy for Advanced Polymeric Materials
A R T I C L E S
Figure 2. AFM single chain force-extension data. Both the modular polymer A and the control polymer B were subjected to single-chain force-extension
studies in toluene. Whereas a single stretching peak was usually observed for the control polymer B, the characteristic sawtooth pattern was observed for
the modular polymer A. In Figure 2b and c, all scattered dots represent experimental data and the solid lines are results from WLC fitting. (a) An overlay
of representative force-extension curves for modular polymer A. Sawtooth patterned curves were consistently obtained.30 (b) One representative single-
chain force-extension curve for the control polymer B, in which only one peak was observed. (c) A single-chain force-extension curve for the modular
polymer A shows the characteristic sawtooth pattern with three peaks.
from 0.22 to 0.4 have been reported for various synthetic
polymers and natural proteins based on WLC fitting of single-
chain stretching data.26,33,34 These values are typically lower
than those obtained from classical bulk scattering measure-
ments.35 This discrepancy in persistence length is generally
accounted for by the limitation of the WLC model;36 neverthe-
less, this model has been successfully applied to fit force-
extension data for many natural and synthetic polymer systems.
For our polymers A and B, the major chemical composition of
the main chains is the flexible polyether (PTMG). A persistence
length of 0.2 nm is reasonable for our system when compared
with the 0.4 nm values for titin and other proteins, which should
have larger persistence lengths because of their more rigid
polypeptide backbones.
Biological modular polymers such as titin show a constant
increase in contour length (∆L) between consecutive peaks in
their force-extension curves due to the sequential unfolding
of identical modules as the polymer is stretched. In contrast,
the gain in contour length upon opening of each loop is not
constant for our modular polymer A (Figure 2a and c). The
nonconstant ∆L values are attributed to three factors for the
modular polymer A: (1) polydispersity of the PTMG linker (PDI
≈ 2.2 by GPC), (2) partial chain extension during prepolymer
formation, and (3) possible random dimerization of UPy units
on polymer chains. In addition to the polydisperse nature of
the PTMG, during the prepolymer formation in the first step of
polymerization, partial chain extension may occur to elongate
the loop length. Furthermore, one UPy unit may dimerize with
another nonadjacent UPy unit on the chain, which will random-
ize the loop size. Similar kinds of “misfolding” have been
observed in biological modular polymers.37 Even though the
∆L values are not constant, the consistent stepwise increase in
contour length suggests the sequential unfolding of UPy modules
along the polymer A chain as it was stretched.
A more subtle observation was that two peak forces were
consistently observed in the force spectra of the polymer A
(Figure 2a), one at a lower force of ∼100 pN and another at a
higher force of ∼200 pN.30 We propose this is due to different
UPy dimer topologies formed along polymer chains. During
polymerization to form the modular polymer A, the UPy units
can be connected by the diisocynate prepolymer in head-to-
head, head-to-tail, or tail-to-tail fashions. These different con-
nectivities will lead to the formation of the following three
different UPy dimer topologies, I-III (Scheme 2): The
unfolding of dimers I and II should require a stronger force
than the unfolding of dimer III because, for both I and II, the
quadruple hydrogen bonds need to be broken cooperatively,
whereas, for III, the module can be “peeled off” gradually. We
propose that the higher force peaks are attributed to the
unfolding of types I and II UPy dimers and the lower force
peaks are attributed to the unfolding of type III UPy dimer.
The mechanical properties of the polymers were tested in
bulk as well to correlate with the single-chain nanomechanical
studies. Polymer samples were cast into thin films and subjected
to stress-strain analyses using Mini-Instron. Figure 3 shows
the comparison of stress-strain data for the modular polymer
A, the control polymer B, and a regular polyurethane (PU, Mn
∼1000,000 g/mol) that does not contain any UPy units. On a
stress-strain curve, the maximum stress at break gives the
tensile strength of the material, while the whole area covered
under the curve reflects the total energy required to break the
material, that is, the toughness. The introduction of protected
UPy units into the polyurethane enhanced its tensile strength
(33) Bemis, J. E.; Akhremitchev, B. B.; Walker, G. C. Langmuir 1999, 15,
2799-2805.
(34) Senden, T. J.; di Meglio, J.-M.; Auroy, P. Eur. Phys. J. B 1998, 3.
(35) Flory, P. J. Principles of Polymer Chemistry; 1953.
(36) Rief, M.; Pascual, J.; Saraste, M.; Gaub, H. E. J. Mol. Biol. 1999, 286,
553-561.
(37) Oberhauser, A. F.; Marszalek, P. E.; Carrion-Vazquez, M.; Fernandez, J.
M. Nat. Struct. Biol. 1999, 6, 1025-1028.
9
J. AM. CHEM. SOC. VOL. 126, NO. 7, 2004 2061

A R T I C L E S
Guan et al.
Scheme 2. Three Possible Topologies for UPy Dimers I-IIIa
a The flexible loops in green color in the drawings are the PTMG linkers. The arrows represent the directions of forces applied to the UPy dimer modules.
Figure 3. Stress-strain curves for the control and modular polymers. The curve in the bottom (magenta) is for the polyurethane (PU) made from PTMG
and HDI without any UPy units. The curve in the middle (blue) is the control polymer B containing the protected UPy units. The curve at the top (red) is
for the modular polymer A containing UPy dimerized loops. The modular polymer A could not be broken by the Mini-Instron at its maximum load. A
stretching-retraction cycle for the polymer A is shown in the inset, which shows a huge hysteresis, indicating great energy dissipation during the cycle.
and toughness. Even though the PNB protection blocked the
UPy from dimerization to form loops, π-π stacking, dipole-
dipole interactions, and weak residual hydrogen bonding
between the protected UPy units should increase the tensile
strength and toughness of the protected polymer B. For polymer
A, in which UPy units can dimerize to form loops along the
chains, the stress-strain curve shows dramatic differences from
the curve of the control polymer (B). It shows a sigmoid stress-
strain curve characteristic for elastic polymers. After a pseudo
yield region, the sample becomes much stiffer with a significant
increase in modulus and strength. Due to its high strength and
toughness, the ultimate tensile strength at break and the fracture
toughness could not be obtained for polymer A because the
sample could not be broken even at the maximum load of the
Mini-Instron for the smallest sample specimen we could prepare.
Nevertheless, quantitative comparisons can be made at 950%
strain for the three samples. At 950% strain, the tensile stresses
are 58.2, 17.4, and 6.0 MPa, and the energy absorptions per
unit volume are 191.5, 78.5, and 17.7 MPa, respectively, for
the modular polymer A, the control polymer B, and the PU
samples. This comparison clearly shows that the modular
polymer A is significantly stronger and tougher than either the
control polymer B or the simple PU. The polymer A is also
very elastomeric, as evidenced by the high strain up to 900%
and the complete recovery to its original length in three
consecutive extension-retraction experiments. The huge hys-
teresis accompanied in one extension-retraction cycle (inset
in Figure 3) further reveals the great energy dissipation
capability of the system, an important feature for high toughness.
The bulk mechanical data correlate well with our single-chain
force-extension observation, which successfully demonstrates
our biomimetic concept: the introduction of modular structures
held by sacrificial weak bonds into a polymer chain can
successfully combine the three most fundamental mechanical
properties, i.e., high tensile strength, toughness, and elasticity,
into one polymer.
Classical polymer theory has predicted that many physical
properties of bulk polymers can be explained in terms of the
behavior of single molecules.1,35 For very large macromolecules,
contributions from inter- and intramolecular interactions to bulk
physical properties become difficult to distinguish, and molec-
ular and macroscopic mechanical properties merge.1 In bulk
films, the UPy units may dimerize within the same chain to
form intramolecular loops or dimerize between chains to form
intermolecular loops. From a macroscopic property point of
view, both breaking intramolecular loops and breaking inter-
molecular UPy dimers require force and absorb energy. As has
been discussed previously for the adhesive biopolymers in
9
2062 J. AM. CHEM. SOC. VOL. 126, NO. 7, 2004

A Biomimetic Strategy for Advanced Polymeric Materials
A R T I C L E S
biominerals such as seashell and bone,11,12 both the successive
opening of intrachain loops or folded domains within a single
molecule and the successive release of sacrificial interchain
bonds should contribute to the extremely high toughness of these
materials. In our modular polymer A, the unfolding of both
intramolecular and intermolecular loops should contribute to
the significantly enhanced mechanical properties, i.e., the
combination of high tensile strength, fracture toughness, and
elasticity.
cooled to -78 °C, and 2.5 M n-butyllithium (45.0 mL, 113 mmol)
was added dropwise via syringe pump over 1 h. The reaction was
allowed to warm to room temperature and stir overnight. The yellow
solution was quenched with DI H2O (300 mL) and extracted with ether
(3 × 200 mL). The combined extracts were dried over MgSO4,
concentrated in Vacuo, and purified by flash chromatography (silica,
4% acetone/hexanes) to give 1 as a colorless oil (10.22 g, 57%).1H
NMR (500 MHz, CDCl3) δ 1.29 (t, J ) 7.1 Hz, 3H), 2.36 (m, 2H),
2.65 (t, J ) 7.2 Hz, 2H), 3.45 (s, 2H), 4.21 (q, J ) 7.1 Hz, 2H), 5.02
(m, 2H), 5.81 (m, 1H); 13C NMR (125 MHz, CDCl3) δ 14.68, 27.82,
42.45, 49.78, 61.79, 115.97, 137.19, 167.57, 180.93, 202.41. HRMS
m/z: (NH3 - CI) calcd for C9H14O3, 170.0943; found, 170.0944.
2-Amino-6-but-3-enyl-1H-pyrimidin-4-one. Guanidine carbonate
(5.41 g, 30.0 mmol) was added to a solution of 1 (10.22 g, 60.06 mmol)
and ethanol (500 mL). The reaction was heated to reflux for 14 h and
then cooled to room temperature.38 The ethanol was removed in Vacuo,
and the resulting solid was collected by vacuum filtration to give
2-amino-6-but-3-enyl-1H-pyrimidin-4-one as white crystals (4.2 g,
43%).1H NMR (500 MHz, DMSO) δ 2.28 (m, 2H), 4.96 (d, J ) 12.2
Hz, 1H), 5.03 (d, J ) 17.3 Hz, 1H), 5.38 (s, 1H), 5.81 (m, 1H), 6.47
(s, 2H), 10.65 (s, 1H); 13C NMR (125 MHz, CDCl3) δ 31.54, 35.95,
99.88, 100.31, 115.15, 137.84, 155.71, 163.97. HRMS m/z: (NH3 -
CI) calcd for C8H11N3O, 165.0902; found, 165.0907.
1-Allyl-3-(6-but-3-enyl-4-oxo-1,4-dihydro-pyrimidin-2-yl)-urea (2).
Allyl isocyanate (3.15 mL, 35.6 mmol) was added to a solution of
2-amino-6-but-3-enyl-1H-pyrimidin-4-one (4.20 g, 23.4 mmol) and
pyridine (75 mL). The reaction was heated to reflux for 4 h.38 After
cooling to room temperature, the pyridine was removed in vacuo by
codistillation with toluene. The crude solid was recrystallized from a
minimum amount of ethanol to give 2 as white crystals (4.84 g, 77%).1H
NMR (500 MHz, CDCl3) δ 2.41 (m, 2H), 2.57 (m, 2H), 3.90 (m, 2H),
5.17 (m, 1H), 5.29 (d, J ) 18.8 Hz, 1H), 10.43 (s, 1H), 12.00 (s, 1H),
13.14 (s, 1H); 13C NMR (125 MHz, CDCl3) δ 31.25, 32.41, 42.65,
106.69, 116.31, 117.47, 134.68, 135.57, 151.98, 155.09, 157.10, 173.52.
HRMS m/z: (NH3 - CI) calcd for C12H16N4O2, 248.1273; found,
248.1270.
Conclusion
A novel biomimetic modular polymer design is reported here
to address the challenge of designing synthetic polymers with
a combination of mechanical strength, toughness, and elasticity.
Following the molecular mechanism used in the skeletal muscle
protein titin, modular polymers containing multiple loops were
constructed by using precise and strong hydrogen-bonding units.
Single-molecule force-extension studies were conducted for
both the control polymer and the polymer containing many
folded loops, which revealed the sequential unfolding of loops
as a modular polymer chain is stretched. Bulk stress-strain
experiments demonstrated that the modular polymer containing
many loops successfully combine high mechanical strength,
toughness, and elasticity. The excellent correlation between the
single-molecule and the bulk properties successfully demon-
strates our biomimetic concept of using a modular domain
structure for advanced polymer properties. We are currently
developing modular polymers containing new better-defined
nanodomain structures. The molecular parameters of the nan-
odomains will be varied to systematically investigate the
relationship between molecular structures and the single-
molecule mechanical properties, which will be correlated with
the nanoscopic and macroscopic mechanical properties of the
materials. Insight into the relationship between the molecular
properties of polymeric materials and their macroscopic per-
formance should allow us to begin to develop rational materials
designs: working up from molecular design to select desired
materials characteristics to perform in particular applications.
1-Allyl-3-[4-but-3-enyl-6-(2-nitrobenzyloxy)pyrimidin-2-yl]-
urea (3). 2-Nitrobenzyl bromide (6.94 g, 32.1 mmol) was added to a
mixture of 2 (3.99 g, 16.1 mmol), potassium carbonate (4.44 g, 32.1
mmol), and DMF (70 mL). The reaction was heated to 70 °C for 24
h.38-40 After cooling to room temperature, the inorganic salts were
removed by vacuum filtration. DMF was removed in Vacuo by
codistillation with toluene. The resulting brown oil was diluted with
DI H2O (250 mL) and extracted with CHCl3 (5 × 150 mL). The organic
extracts were combined, dried over MgSO4 and concentrated in Vacuo
to an orange solid. Tritration in hot ether gave 3 as a white solid (5.56
Experimental Section
I. Synthesis and Characterization of Monomers and Polymers.
1
General. H NMR spectra were recorded at 400 and 500 MHz, and
13C NMR spectra were recorded at 125 and 100 MHz on Bruker
instruments. 1H and 13C NMR chemical shifts are reported as δ values
in ppm relative to TMS or residual solvent. Mass spectral data was
obtained on a Micromass autospec spectrometer. Combustion analyses
were performed by Atlantic Microlab (Norcross, GA). Gel permeation
chromatography (GPC) was carried out using an Agilent 1100 Series
GPC-SEC Analysis System along with a mixed bed Plgel Mixed-C
column from Polymer Labs. The eluent was CHCl3/DMF (95:5), and
a flow rate of 0.500 mL/min was used. The calibration was performed
using polystyrenes from Aldrich as standards. All commercial reagents
were used as received with the following exceptions: the solvents CH2-
Cl2, THF, and toluene were obtained from an alumina filtration system.
Liquid chromatography was performed using forced flow (flash
chromatography) of the indicated solvent system on Fisher silica gel
60 (230-400 mesh). Moisture sensitive reactions were carried out under
a nitrogen atmosphere using flame-dried glassware and standard syringe/
septa techniques.
1
g, 90%). H NMR (500 MHz, CDCl3) δ 2.44 (m, 2H), 2.72 (m, 2H),
4.00 (t, J ) 4.5 Hz, 2H), 5.13 (m, 2H), 5.14 (d, J ) 10.3 Hz, 1H),
5.26 (d, J ) 17.1 Hz, 1H), 5.75 (s, 2H), 5.85 (m, 1H), 5.93 (m, 1H),
6.29 (s, 1H), 7.51 (m, 1H), 7.65 (m, 2H), 8.13 (d, J ) 8.2 Hz, 1H),
9.26 (s, 1H); 13C NMR (125 MHz, CDCl3) δ 32.39, 36.92, 42.74, 65.26,
100.42, 115.94, 116.23, 125.49, 129.21, 129.49, 132.78, 134.15, 135.11,
137.22, 154.43, 157.76, 170.09. HRMS m/z: (NH3 - CI) calcd for
C19H21N5O4, 383.1594; found, 383.1598.
1-[4-(4-Hydroxybutyl)-6-(2-nitrobenzyloxy)pyrimidin-2-yl]-3-(3-
hydroxypropyl)-urea (4). A solution of 3 (5.30 g, 13.8 mmol in 20
mL of THF) was added to a flask charged with a 0.5 M solution of
9-borabicyclo[3.3.1]nonane (9-BBN) (74.7 mL, 37.3 mmol). The
reaction stirred at room temperature for 2 h. Ethanol (22.4 mL), 6 N
NaOH (7.5 mL), and 30% H2O2 (15.0 mL) were added sequentially.
The reaction was heated to 50 °C for 2 h.41 After cooling to room
temperature, a saturated solution of potassium carbonate was added
3-Oxo-hept-6-enoic Acid Ethyl Ester (1). THF (100 mL) was added
to a flask charged with NaH (3.23 g, 134 mmol). The resulting
suspension was cooled to 0 °C and ethyl acetoacetate (12.0 mL, 94.0
mmol) was added via syringe pump over 1 h. The clear solution was
(38) Ky Hirschberg, J. H. K.; Beijer, F. H.; van Aert, H. A.; Magusin, P. C. M.
M.; Sijbesma, R. P.; Meijer, E. W. Macromolecules 1999, 32, 2696-2705.
(39) Bochet, C. G. J. Chem. Soc., Perkin Trans. 1 2002, 125.
(40) Zehavi, U.; Amit, B.; Patchornik, A. J. Org. Chem. 1972, 37, 2281-2285.
9
J. AM. CHEM. SOC. VOL. 126, NO. 7, 2004 2063

A R T I C L E S
Guan et al.
(50 mL). The organic layer was separated, dried over potassium
carbonate, and concentrated in vacuo to brown oil. Purification by flash
chromatography (silica, CHCl3 to 5% MeOH/CHCl3) gave a slightly
yellow solid which was recrystallized from a minimum amount of
acetone to give 4 as white crystals (5.79 g, 54%). 1H NMR (500 MHz,
DMSO) δ 1.43 (m, 2H), 1.63 (m, 4H), 2.60 (t, J ) 7.5 Hz, 2H), 3.26
(m, 2H), 3.41 (m, 2H), 3.47 (m, 2H), 4.42 (t, J ) 5.1 Hz, 1H), 4.51 (t,
J ) 5.1 Hz, 1H), 5.72 (s, 2H), 6.44 (s, 1H), 7.64 (m, 1H), 7.77 (m,
2H), 8.13 (d, J ) 7.9 Hz, 1H), 9.19 (t, J ) 5.4 Hz, 1H), 9.48 (s, 1H);
13C NMR (125 MHz, DMSO) δ 24.30, 31.896, 32.54, 36.15, 36.41,
58.48, 60.40, 64.21, 98.91, 124.89, 129.33, 129.78, 131.80, 134.14,
147.48, 153.97, 157.64, 169.12, 171.15. HRMS m/z: (NH3 - CI) calcd
for C19H26N5O6, 420.1883; found, 420.1876.
GPC using CHCl3/DMF (95:5) as the eluent measured the average
molecular weights as Mn ) 68 000 g/mol, Mw ) 123 000 g/mol; Mw/
1
Mn ) 1.80, Tg ) 6.92 °C. H NMR (500 MHz, CDCl3) δ 1.25, 1.32,
1.48, 1.61, 1.92, 2.52, 3.15, 3.41, 4.06-4.11, 4.40-5.36, 5.82, 10.2,
11.8, 13.1.
II. Single-Chain Force-Extension Measurements by Atomic
Force Microscopy. Preparation of Physiadsorped Polymers onto
Au-Coated Silicon Wafers. Polymers A and B were adsorbed onto
freshly gold coated silicon wafers (1.2 cm × 1.2 cm) by immersion in
a solution of polymer (0.2 mg/mL) in CHCl3/DMAc (95:5) for 8 h.
The sample was removed from the solution and sonicated (3 × 5 min)
in CHCl3. Samples were dried under a stream of nitrogen and used
immediately.
1-[6-(4-Hydroxybutyl)-4-oxo-1,4-dihydropyrimidin-2-yl]-3-(3-hy-
droxypropyl)-urea (5). Palladium on carbon (10 wt. percent, 5.3 mg)
was added to a solution of 4 (341 mg, 1.20 mmol) in methanol (10
mL). The solution was stirred under an atmosphere of hydrogen for 12
h. Nitrogen was bubbled through the solution of 12 h, and the palladium
was removed by filtration through Celite. The solution was concentrated
in vacuo to a pink solid. Tritration with acetone followed by
recrystallization from n-butanol gave pure 5 (284 g, 83%) as a white
Atomic Force Microscopy Force-Extension Experiments. A CP
Research atomic force microscope equipped with a 100 µm scanner
from Park Scientific Instruments was used to carry out the AFM single-
chain stretching experiments. ProScan software (version 1.6) was used
for all of the data acquisition. Force-extension experiments were carried
out in toluene (filtered by 0.02 µm membrane). The prepared sample
was loaded into a custom-built liquid reservoir filled with filtered
toluene at room temperature. A micro-liquid-cell (purchased from Park
Scientific Instruments) equipped with silicon nitride cantilevers (mi-
crolevers, purchased from www.store.veeco.com) was allowed to
approach the surface. We used the microlever C, which has a force
constant in the range of 14.4 to 15.7 pN/nm as calibrated by the Asylum
Research, to collect the force-extension data. Force versus distance
curves were acquired using a spectroscopy mode (ML06C cantilever)
with a frequency of 0.3-0.5 Hz and a distance of 0.2-0.7 µm,
correlating to pulling speeds of 0.06-0.35 µm/s.
III. Stress-Strain Analysis using Instron. Silanization of Glass
Petri Dishes for Film Casting. Glass Petri dishes were cleaned with
a freshly prepared mixture of concentrated sulfuric acid and 30%
hydrogen peroxide (70:30, v:v) at 120 °C for 30 min. The dishes were
removed and washed with a large amount of DI water and rinsed with
acetone.42 The dishes were blown dry under a stream of nitrogen and
immediately used in the silanization procedures. The formation of the
trimethylsilyl layer on the glass Petri dishes was carried out in the vapor
phase under reduced pressure. The cleaned substrates and a small vial
(Fisher brand 12 × 35 mm2, 0.5 DR) containing 150 µL of trimethylsilyl
chloride were placed inside a vacuum desiccator (Wheaton dry-seal
desiccator, 100 mm). The desiccator was slowly evacuated using the
house vacuum, sealed tightly, and kept in this condition for 16 h at
room temperature.39 After completion of silanization, the Petri dishes
were removed, ultrasonicated in fresh acetone (5 × 5 min), dried under
a stream of nitrogen, and stored under nitrogen for future use.
Film Casting for Protected Polymer (B). Protected polymer B (1.41
g) was dissolved in chloroform (20 mL). The viscous solution was
ultrasonicated to remove any bubbles from the solution and then poured
into a TMS-treated glass Petri dish. The solvent was allowed to slowly
evaporate over 12 h, and the resulting film was dried at 60 °C in a
vacuum oven for 2 days. After drying, the film was stored under
nitrogen for future use.
1
solid. H NMR (500 MHz, DMSO) δ 1.42 (m, 2H), 1.60 (m, 4H),
2.36 (t, J ) 7.4, 2H), 3.20 (q, J ) 6.4, 2H), 3.40 (q, J ) 6.4, 2H), 3.45
(q, J ) 5.2, 2H), 4.38 (t, J ) 5.1, 1H), 4.52 (t, J ) 5.0, 1H), 5.75 (s,
1H), 7.45 (s, 1H), 9.71 (s, 1H), 11.49 (s, 1H); 13C NMR (125 MHz,
DMSO) δ 24.39, 32.34, 32.76, 36.58, 36.89, 58.83, 60.84, 104.41, 152,-
76, 155.93, 163.38, 168.15. HRMS m/z: (NH3 - CI) calcd for
C12H20N4O4: 284.1485; found, 284.1485.
Polymerization with Protected UPy-Containing Monomers to
Make the Control Polymer (B). PTMO (Mn ) 1400 g/mol) (0.70
mL, 0.50 mmol) was added dropwise to a solution of 1,6-diisocyana-
tohexane (168 µL, 1.00 mmol) and DMAc (0.50 mL). The solution
stirred at room temperature for 60 min, then a solution of 5 (209 mg,
0.500 mmol) in DMAc (0.50 mL) was added dropwise. The solution
stirred at room temperature for 18 h, resulting in a very viscous solution.
Precipitation from methanol (200 mL) provided polymer B as a white
elastomer.
GPC using CHCl3/DMF (95:5) as the eluent measured the average
molecular weights as Mn ) 69 000 g/mol, Mw ) 115 000 g/mol; Mw/
1
Mn ) 1.66, Tg ) 0.61 °C. H NMR (500 MHz, CDCl3) δ 1.25, 1.32,
1.48, 1.61, 1.92, 2.65, 2.14, 3.41, 4.06-4.20, 4.62-5.38, 5.72, 6.28,
7.49, 7.67, 8.12, 9.25.
Polymerization with Deprotected UPy-Containing Monomers to
Make the Modular Polymer (A). PTMO (Mn ) 1400 g/mol) (0.70
mL, 0.50 mmol) was added dropwise to a solution of 1,6-diiso-
cyanatohexane (168 µL, 1.00 mmol) and DMAc (0.50 mL). The
solution stirred at room temperature for 60 min, and then a solution of
6 (142 mg, 0.500 mmol) in DMAc (0.50 mL) was added dropwise.
The solution stirred at room temperature for 18 h, resulting in a very
viscous solution. Precipitation from methanol (200 mL) provided
polymer A as a colorless elastomer.
Film Casting for Deprotected Polymer (A). Deprotected polymer
A (1.41 g) was dissolved in a mixture of chloroform and methanol (20
mL, 40:1 CHCl3/MeOH). The viscous solution was ultrasonicated to
remove any bubbles from the solution and then poured into a TMS-
treated glass Petri dish. The solvent was allowed to slowly evaporate
over 12 h, and the resulting film was dried at 60 °C in a vacuum oven
for 2 days. After drying, the film was stored under nitrogen for future
use.
Stress-Strain Acquisition Using Instron. An Instron Universal
Testing Machine (model 5564) at Professor David Tirrell’s group at
(41) Brown, H. C.; Knights, E. F.; Scouten, C. G. J. Am. Chem. Soc. 1974, 96,
7765.
(42) Moon, J. H.; Shin, J. W.; Kim, S. Y.; Park, J. W. Langmuir 1996, 12,
4621.
9
2064 J. AM. CHEM. SOC. VOL. 126, NO. 7, 2004

A Biomimetic Strategy for Advanced Polymeric Materials
A R T I C L E S
Caltech was used to perform stress-strain testing. A uniform strain
rate of 100% gauge length per minute was chosen otherwise specified.
Samples were wrapped around by same material in the region of the
grips to prevent slip off or break at end. All experiments were performed
at room temperature.
tion (BYI Award to Z.G.), and the University of California at
Irvine. We thank Professor D. A. Tirrell and P. Nowatzki at
Caltech for their generous help on Mini-Instron tests, Professor
M. L. Mecartney and L. Ramirez at UCI for their help on Instron
tests, Professors H. G. Hansma and P. K. Hansma at UCSB for
consultation on AFM studies, and Professor D. A. Brant at UCI
for helpful discussions and insightful comments on the manuscript.
Acknowledgment. We are grateful for the financial support
from the National Science Foundation (CAREER Award to
Z.G., DMR-0135233), the Arnold and Mabel Beckman Founda-
JA039127P
9
J. AM. CHEM. SOC. VOL. 126, NO. 7, 2004 2065