Quantitative mass determination of conjugated polymers for single molecule conformation analysis: Enhancing rigidity with macrocycles

Article abstract of DOI:10.1039/c0cc00670j

A combination of advanced synthetic and single molecule spectroscopic techniques allowed us to demonstrate that macrocycles, covalently bound to a conjugated polymer backbone, raise the effective chain persistence length by at least a factor of 5.

Full text of DOI:10.1039/c0cc00670j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Quantitative mass determination of conjugated polymers for single
molecule conformation analysis: enhancing rigidity with macrocyclesw
a
b
b
b
c
¨
Klaus Becker, Gerald Gaefke, Jasmin Rolffs, Sigurd Hoger* and John M. Lupton*
Received 30th March 2010, Accepted 11th May 2010
First published as an Advance Article on the web 4th June 2010
DOI: 10.1039/c0cc00670j
A combination of advanced synthetic and single molecule
spectroscopic techniques allowed us to demonstrate that
macrocycles, covalently bound to a conjugated polymer back-
bone, raise the effective chain persistence length by at least a
factor of 5.
periphery in controlling p-system shape through intramolecular
interactions. This approach leads to chains extended over
B185 nm in the solid state. In order to conclusively prove
such chain extension, we present a molecular ruler approach to
calibrating the GPC molecular weight of polymer samples by
synthesizing a family of well-defined oligomeric structures and
deriving the underlying mass scaling relationship.
Conjugated polymers are versatile autofluorescent model
systems to study fundamental aspects of polymer physics.
The relevance of elementary material parameters such as
topology and conformation to applications in organic electronics
is becoming ever more apparent.1 For example, the degree of
chain extension impacts on photophysical stability in materials
as diverse as phenylene-vinylenes and polyfluorenes: the more
bent a chain, the greater the degree of random electronic
fluctuations on the chain and in the immediate environment.
Single molecule techniques have revealed how such bending
reduces the overall photochemical stability,2 an effect which is
masked in ensemble spectroscopy but is crucially important to
understanding bulk material stability. Several concepts exist to
potentially improve the rigidity of polymer chains for electronic
applications, such as templating in zeolites1b or rotaxanes,3
dendronization,4 or decoration with bulky brush-like side-
groups.5 Large persistence lengths, in excess of 60 nm, have
been observed in individual polymer chains with supra-
molecular self-assembly of dendritic side-groups by using
atomic force microscopy.4h In this case, rigidity is enforced
by intermolecular interactions. These prior avenues generally
involve insulating, optically inert templates, which can lessen
the overall material performance. We recently demonstrated a
further route to isolating conjugated wires by encapsulating in
covalently-bound conjugated macrocycles.6 As the peripheral
rings are chromophores themselves, the molecular system
exhibits light-harvesting functionality,7 with the ability to
funnel and accumulate excitation energy.6a By considering
the evolution of single chain conformation with length, we
now demonstrate the passive functionality of the macrocyclic
Fig. 1 illustrates the structures of the bare 1 and decorated 2
polymers, both poly(p-phenylene-ethynylene-butadiynylene)s,
of which 2 displays distinct light-harvesting functionality as
described in detail in ref. 6a. Both polymers were fractionated
by gel permeation chromatography (GPC) as described in the
ESI.w In order to determine the chain shape, we studied the
luminescence of single polymer molecules under irradiation
with polarized light, as sketched in the top right of Fig. 1.5,8
An individual p-conjugated chain segment will absorb light of
a particular polarization. The more non-collinear segments are
present, the more polarizations of light the macromolecule
couples to: the anisotropy decreases. Such single molecule
studies of polarization anisotropy have advantages over
ensemble fluorescence depolarization often used to infer chain
a Photonics and Optoelectronics Group, Department of Physics and
CeNS, Ludwig Maximilians University of Munich, 80799 Munich,
Germany
b
´
Kekule-Institut fur Organische Chemie und Biochemie der
Universitat Bonn, 53121 Bonn, Germany.
¨
¨
Fig. 1 Structures of the bare polymer 1 and the macrocycle-clad
polymer 2. Polarization anisotropy r for different average chain
lengths hni of 1 [panels (a)–(e)] and 2 [(f)–(h)]. GPC values for hni
are given in orange, corrected values in green. For 1, as the weight
increases, the anisotropy decreases due to chain folding. Such a strong
reduction is not found for 2 because of improved chain rigidity. The
cartoons derive from illustrative bead-on-a-string simulations of
chain folding. The sketch in the top right illustrates the polarization
anisotropy measurement.
E-mail: hoeger@uni-bonn.de; Fax: +49 228-735662;
Tel: +49 228-736127
c Department of Physics and Astronomy, University of Utah,
Salt Lake City, UT 84112, USA.
E-mail: lupton@physics.utah.edu; Fax: +1 801-581-4801;
Tel: +1 801-581-6408
w Electronic supplementary information (ESI) available: Synthesis,
characterization, GPC, fluorescence and photostability. See DOI:
10.1039/c0cc00670j
ꢀc
This journal is The Royal Society of Chemistry 2010
4686 | Chem. Commun., 2010, 46, 4686–4688

View Article Online
shape,4c as they provide a direct image of the statistical
distribution of chain shapes within the polymer sample. In
addition, the technique is insensitive to excitation energy
transfer effect, as only the anisotropy in excitation is probed.
Single chain samples were formed by serial dilution to a
concentration of the order 10ꢁ10 mol Lꢁ1 in toluene, mixing
with a polystyrene matrix, and subsequent deposition on a
quartz substrate.8f,g The samples were mounted under vacuum
to maximize photostability, and imaged in a fluorescence
microscope.8g The shape of the chain, which—in contrast to
poly(phenylene-vinylene)s2b,c—consists of straight conjugated
segments, can be inferred from the maximum and minimum
fluorescence intensities observed under rotation of the incident
laser polarization plane (excitation at 457.8 nm).8c The polar-
ization anisotropy is given by r = (Imax ꢁ Imin)/(Imax + Imin).
r = 1 denotes an extended chain, whereas r = 0 corresponds
to a random coil structure absorbing light of any polarization.
As the molecules are immobilized, chain shape is independent
of temperature.8g
Fig. 2 Relation between GPC (vs. PS) and calculated molecular
weight for oligomers of 1 (blue) and 2 (green). Data from ref. 10c
are also shown (red). The exponents are 1.23 (blue), 1.00 (green) and
1.38 (red).
Fig. 1 displays histograms of the polarization anisotropy
r recorded for at least 150 single polymer chains of each
fraction. For the bare polymer 1, panels (a)–(e), there is a
clear decrease in anisotropy as the chain length increases. This
effect is much less pronounced for the macrocycle polymer 2,
shown in panels (f)–(h). Although the concept of template-
enhanced rigidity is readily demonstrated in principle using
polarization anisotropy, such as described here and previously
shown for polystyrene-grafted polythiophenes,5 it is extremely
hard to actually prove in practice since accurate knowledge of
the chain length9 is required. Note that a range of powerful
techniques exist to conclusively demonstrate the presence of
truly isolated chains.8h In contrast to conventional shape
determination by, e.g., light scattering techniques, concentration
effects can be excluded in single molecule studies: the
concentration of single chains is, in effect, infinitively low.
To determine the molecular weight we approach the
problem from the bottom up, making well-defined oligomers
of size comparable to the polymer.10 We recently reported
oligomeric phenylene-butadiynylenes up to the 17mer and
showed how the correction factor for the overestimation of
the mass determined by GPC against a polystyrene (PS)
standard depends on length.10c The GPC and calculated
weights from ref. 10c are plotted against each other on a
double logarithmic scale in Fig. 2 (red). Assuming the overall
chain shape does not change significantly over the length range
studied, the GPC weight should depend on (calculated) chain
length according to a power law,11 as observed in Fig. 2.12 The
question in calibrating the GPC of 1 and 2 is over what range
of chain length the power law can be extrapolated.
succeeded in polymerizing the dimer building block up to
the 6mer (to form a 12mer). However, these data (green)
follow a clear power law, which we extrapolate to higher
molecular weights. As a consequence of the failed polymerization
of large oligomers of 2, long monodisperse chains of 2 are not
available. To study the effect of chain extension in the single
chain anisotropy experiments, we therefore have to work with
the polydisperse polymer 2, necessitating the molecular ruler
approach for calibrating the polymer GPC described here.
The average chain length hni of the fractionated polymer
samples as determined by GPC and corrected using the power
law extrapolation is depicted in Fig. 1 in orange and green,
respectively. Polymer 1 exhibits significant chain folding
(i.e. histogram points at low r) as the average chain length is
increased 3.5-fold from 4 to 14 monomers. For the longest
chain samples of 1 with, on average, 70 units, all chains display
some distortion; most chains allow absorption of light of any
polarization (r - 0), indicating random folding. Note that this
chain folding is not visible in GPC: if the overall chain
structure (which controls the radius of gyration) changed
within the molecular weight range studied we would observe
a change in the power law dependence of Fig. 2.11 We note
therefore that the chains may be more extended in solution
than in the solid state matrix used in the single molecule
experiments, as previously suggested by Huser et al.8b
In contrast, as the size of 2 is increased, the chains remain
much more rigid, with only a small shift of anisotropy to lower
values upon a B25-fold increase in molecular weight [panels
(f–h)]. How can we relate these microscopic data to the notion
of a ‘‘persistence length’’? First, it is crucial to read the results
of Fig. 1 both in terms of the individual chain and with regards
to the distribution of chain conformations. Even for the more
flexible polymer 1, fully extended chains of B154 nm length
exist (hni E 70, r - 1). Note that the polydispersity (D) of
these samples is quite low (D = 1.30, ref. 6a). Even though the
most extended chains may arise from the shortest molecules in
the polymer sample, this approximate figure for the single
chain length should be sufficiently representative of the entire
sample. For this particular single chain, the ‘‘persistence
length’’ is thus B10ꢂ greater than what is derived from
We have now succeeded in demonstrating the relationship
between GPC-determined molecular weight and the degree of
polymerization for 1 up to the 64mer.13 We synthesized the
octamer of 1 and, following polymerization, obtained the
corresponding peak molecular weights of the higher oligomers,
i.e. multiples of the octamer, in GPC. The corresponding data
are shown in Fig. 2 in blue. The dependence follows a power
law over the entire weight range. A similar relation will hold
for 2 since both polymers have identical backbone structure.
Polymerization of large oligomers of 2 failed; we only
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 4686–4688 | 4687

View Article Online
interpretative models in ensemble measurements on comparable
structures.9 Regarding the effect of encapsulation, we can
make a further quantitative statement: the r-distribution of 2
for hni E 84 can be found in 1 between hni E 14–20. The
average persistence length is thus raised by at least a factor
of 5 upon encapsulation. In addition, we find that 30% of the
longest chains of 2 are not bent (r > 0.8), i.e. have a
persistence length in excess of B185 nm. We note that the
microscopic origin of the rigidity enhancement of 2 remains
under investigation. It is conceivable that, during processing,
solvent molecules are incorporated between the backbone and
the ring, further enhancing stiffening.
3 (a) M. J. Frampton and H. L. Anderson, Angew. Chem., Int. Ed.,
2007, 46, 1028–1064; (b) S. M. E. Simpkins, B. M. Kariuki and
L. R. Cox, J. Organomet. Chem., 2006, 691, 5517–5523.
4 (a) T. Sato, D. L. Jiang and T. Aida, J. Am. Chem. Soc., 1999, 121,
10658–10659; (b) A. F. Zhang, L. J. Shu, Z. S. Bo and
A. D. Schluter, Macromol. Chem. Phys., 2003, 204, 328–339;
¨
(c) W. S. Li, D. L. Jiang and T. Aida, Angew. Chem., Int.
Ed., 2004, 43, 2943–2947; (d) J. Y. Wang, J. M. Han, J.
Yan, Y. Ma and J. Pei, Chem.–Eur. J., 2009, 15,
3585–3594; (e) H. Frauenrath, Prog. Polym. Sci., 2005, 30,
325–384; (f) P. Samorı, C. Ecker, I. Gossl, P. A. J. De Witte,
´
J. J. L. M. Cornelissen, G. A. Matselaar, M. B. J. Otten,
A. E. Rowan, J. M. Nolte and J. P. Rabe, Macromolecules, 2002,
35, 5290–5294; (g) J. M. Lupton, P. Schouwink, P. E. Keivanidis,
A. C. Grimsdale and K. Mullen, Adv. Funct. Mater., 2003, 13,
¨
154–158; (h) V. Percec, C.-H. Ahn, G. Ungar, D. J. P. Yeardley,
¨
M. Moller and S. S. Sheiko, Nature, 1998, 391, 161–164.
The sketched chains in Fig. 1 are derived from bead-on-
a-string simulations following Hu et al.,8c where the polymer
stiffness has been adjusted to account for intramolecular
templating. Each bead corresponds to a linear transition
dipole; the sketched conformation results in the average
anisotropy found in the histogram, providing a visualization
of the average conformation.
5 (a) T. Sugimoto, Y. Ebihara, K. Ogino and M. Vacha, Chem-
PhysChem, 2007, 8, 1623–1628; (b) T. Sugimoto, S. Habuchi,
K. Ogino and M. Vacha, J. Phys. Chem. B, 2009, 113,
12220–12226.
6 (a) K. Becker, P. G. Lagoudakis, G. Gaefke, S. Hoger and
¨
J. M. Lupton, Angew. Chem., Int. Ed., 2007, 46, 3450–3455;
(b) S. Hoger, A.-D. Meckenstock and H. Pellen, J. Org. Chem.,
¨
Chem., 1999, 37, 2685–2698; (d) S. Hoger and A.-D. Meckenstock,
¨
1997, 62, 4556–4557; (c) S. Hoger, J. Polym. Sci., Part A: Polym.
Encapsulation of the backbone increases rigidity by reducing
conformational freedom. Single molecule studies have
previously revealed that elongated chains exhibit greater
photostability.2b,c Consequently, the encapsulated polymer
sample 2, when exposed to UV laser excitation in solution,
displays enhanced resilience to photodegradation with respect
to 1.13,14
¨
Tetrahedron Lett., 1999, 39, 1735–1736; (e) M. Fischer and
S. Hoger, Tetrahedron, 2003, 59, 9441–9446.
¨
7 (a) G. Calzaferri, S. Huber, H. Maas and C. Minkowski, Angew.
Chem., Int. Ed., 2003, 42, 3732–3758; (b) V. S. Y. Lin,
S. G. Dimagno and M. J. Therien, Science, 1994, 264,
1105–1111; (c) D. Gust, T. A. Moore and A. L. Moore, Acc.
Chem. Res., 2001, 34, 40–48.
8 (a) T. Ha, T. Enderle, D. S. Chemla, P. R. Selvin and S. Weiss,
Phys. Rev. Lett., 1996, 77, 3979–3982; (b) T. Huser, M. Yan and
L. Rothberg, Proc. Natl. Acad. Sci. U. S. A., 2000, 97,
11187–11191; (c) D. H. Hu, J. Yu, K. Wong, B. Bagchi,
P. J. Rossky and P. F. Barbara, Nature, 2000, 405, 1030–1033;
(d) S. S. Sartori, S. De Feyter, J. Hofkens, M. Van der Auweraer,
F. De Schryver, K. Brunner and J. W. Hofstraat, Macromolecules,
2003, 36, 500–507; (e) M. Forster, D. Thomsson, P. R. Hania and
I. G. Scheblykin, Phys. Chem. Chem. Phys., 2007, 9, 761–766;
We have demonstrated how accurate insight into both
conformation and molecular weight of large polymer samples
can be arrived at by careful calibration and application
of GPC scaling relationships using molecular ruler model
systems. Single chain polarization anisotropy provides accurate
insight into the microscopic distribution of conformations. In
contrast, ensemble techniques such as light scattering and
electron paramagnetic resonance spectroscopy do not reveal
the heterogeneity of the sample, require far greater sample
quantities, and cannot assess chain shape in the solid state as is
relevant for organic electronics.
¨
(f) M. J. Walter, N. J. Borys, G. Gaefke, S. Hoger and
J. M. Lupton, J. Am. Chem. Soc., 2008, 130, 16830–16831;
(g) K. Becker and J. M. Lupton, J. Am. Chem. Soc., 2006, 128,
6468–6479; (h) J. M. Lupton, Adv. Mater., 2010, 22, 1689–1721,
DOI: 10.1002/adma.200902306.
9 (a) U. H. F. Bunz, Chem. Rev., 2000, 100, 1605–1644; (b) A. Godt,
M. Schulte, H. Zimmermann and G. Jeschke, Angew. Chem., Int.
Ed., 2006, 45, 7560–7564; (c) P. Cotts, T. Swager and Q. Zhou,
Macromolecules, 1996, 29, 7323–7328.
10 (a) L. Jones, J. Schum and J. M. Tour, J. Org. Chem., 1997, 62,
1388–1410; (b) N. Aratani, A. Osuka, Y. Kim, D. Jeong and
D. Kim, Angew. Chem., Int. Ed., 2000, 39, 1458–1462;
Collaborative funding by the Volkswagen Foundation is
appreciated. JML gratefully acknowledges a Fellowship for
Science and Engineering from the David and Lucile Packard
Foundation.
(c) D. Mossinger, S.-S. Jester, E. Sigmund, U. Muller and
S. Ho
¨
¨
ger, Macromolecules, 2009, 49, 7974–7978.
¨
Notes and references
1 (a) B. J. Schwartz, Annu. Rev. Phys. Chem., 2003, 54, 141–172;
(b) T. Q. Nguyen, J. J. Wu, V. Doan, B. J. Schwartz and
S. H. Tolbert, Science, 2000, 288, 652–656.
11 V. N. Tsvetkov, Rigid-Chain Polymers, Plenum Publishers,
New York, 1989; I. Teraoka, Polymer Solutions, Wiley & Sons,
New York, 2002.
2 (a) M. A. Summers, P. R. Kemper, J. E. Bushnell, G. C. Bazan,
M. T. Bowers and S. K. Buratto, J. Am. Chem. Soc., 2003, 125,
5199–5203; (b) K. Becker, E. Da Como, J. Feldmann, F. Scheliga,
E. T. Csanyi, S. Tretiak and J. M. Lupton, J. Phys. Chem. B, 2008,
112, 4859–4864; (c) M. A. Summers, P. R. Kemper, J. E. Bushnell,
M. R. Robinson, G. C. Bazan, M. T. Bowers and S. K. Buratto,
J. Am. Chem. Soc., 2003, 125, 5199–5203.
12 The power law can only be applied over a limited range of molecular
weights as the exponent will change with increasing chain length.
13 See ESIw for further discussion.
14 At present, it cannot be excluded for certain that this enhanced
photostability is due to improved physical shielding of the
backbone from reactive oxygen species by the covalently-bound
macrocycle periphery.
ꢀc
This journal is The Royal Society of Chemistry 2010
4688 | Chem. Commun., 2010, 46, 4686–4688