Lewis acid adducts of narrow band gap conjugated polymers

Article abstract of DOI:10.1021/ja110968m

We report on the interaction of Lewis acids with narrow band gap conjugated copolymers containing donor and acceptor units. Examination of the widely used poly[(4,4-bis(2-ethylhexyl)cyclopenta-[2,1-b:3,4-b′]dithiophene)-2, 6-(diyl-alt-benzo[2,1,3]thiadiaz

Full text of DOI:10.1021/ja110968m

ARTICLE
pubs.acs.org/JACS
Lewis Acid Adducts of Narrow Band Gap Conjugated Polymers
Gregory C. Welch and Guillermo C. Bazan*
Center for Polymers and Organic Solids, Department of Chemistry and Biochemistry, and Department of Materials, The University of
California, Santa Barbara, California 93106, United States
S Supporting Information
b
ABSTRACT: We report on the interaction of Lewis acids with
narrow band gap conjugated copolymers containing donor and
acceptor units. Examination of the widely used poly[(4,4-bis(2-
ethylhexyl)cyclopenta-[2,1-b:3,4-b0]dithiophene)-2,6-(diyl-alt-
benzo[2,1,3]thiadiazole)-4,7-diyl] (1) shows weaker binding
with B(C6F5)3 when compared with a small molecule that
contains a cyclopenta-[2,1-b:3,4-b0]dithiophene (CDT) unit
flanked by two benzo[2,1,3]thiadiazole (BT) fragments. Studies on model compounds representative of 1, together with a
comparison between B(C6F5)3 and BBr3, indicate that the propensity for Lewis acid coordination is decreased because of steric
encumbrance surrounding the BT nitrogen sites. These observations led to the design of chromophores that incorporate an acceptor
unit with a more basic nitrogen site, namely pyridal[2,1,3]thiadiazole (PT). That this strategy leads to a stronger B-N interaction
was demonstrated through the examination of the reaction of B(C6F5)3 with two small molecules bis(4,4-bis(hexyl)-4H-cyclopenta-
[2,1-b;3,4-b0]dithiophene)-4,7-pyridal[2,1,3]thiadiazole (8) and bis{2-thienyl-(4,4-bis(hexyl)-4H-cyclopenta[2,1-b;3,4-b0]dithiophene)}-
4,7-pyridal[2,1,3]thiadiazole (9) and two polymer systems (poly[(4,4-bis(2-ethylhexyl)cyclopenta-[2,1-b:3,4-b0]dithiophene)-2,6-
diyl-alt-([1,2,5]thiadiazolo[3,4-c]pyridine)-4,7-diyl] (10) and poly[(4,4-bis(2-ethylhexyl)cyclopenta-[2,1-b:3,4-b0]dithiophene)-
2,6-diyl-alt-(40,70-bis(2-thienyl)-[1,2,5]thiadiazolo[3,4-c]pyridine)-5,5-diyl] (11). From a materials perspective, it is worth pointing
out that through the binding of B(C6F5)3, new NIR-absorbing polymers can be generated with band gaps from 1.31 to 0.89 eV.
A combination of studies involving ultraviolet photoemission spectroscopy and density functional theory shows that the narrowing
of the band gap upon borane coordination to the pyridal nitrogen on PT is a result of lowering the energies of both the highest
occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of the optically relevant fragments;
however, the LUMO is decreased to a greater extent, thereby giving rise to the narrowing of the gap.
’ INTRODUCTION
attract additional interest as they have potential to be incorpo-
rated into organic NIR photodetectors and NIR photovoltaic and
electrochromic devices with the potential to outperform current
inorganic counterparts.23,24
Push-pull organic chromophores and related polymeric
systems are an important class of semiconducting and light
harvesting materials that have been studied and utilized in various
technologies.1 Their molecular structures contain electron rich
and electron deficient moieties linked via a π-conjugated bridge.
Photoexcitation leads to intramolecular charge-transfer from the
electron rich fragment in the molecule, referred to as the donor
(D), to the electron deficient segment, referred to as the acceptor
(A).2 Such a redistribution of electron density leads to absorption
bands that can extend from the ultraviolet to well into the near-
infrared regions of the spectrum, depending on the electronic
offset between the donor and acceptor components and the
overall delocalization length.3-10 The resulting materials have
found utility in many organic electronic and optical applications
including use as nonlinear optics,11-13 as light emitters for
OLEDs,14,15 as sensitizers in DSSCs16-18 and more recently as
light harvesting materials for bulk-heterojunction (BHJ) solar
cells.19 With respect to the latter, the incorporation of D/A
conjugated copolymers with optical band gaps near 1.4 eV results
in BHJ devices with excellent spectral overlap with the terrestrial
solar spectrum and ultimately enables efficient photon harvesting
and increased photocurrent generation.20-22 Organic materials
that absorb into the near-infrared (NIR) have also begun to
Due to the broad range of applications that take advantage of
push-pull organic chromophores, there exists a demand for new
strategies for designing materials with tunable optical and
electronic properties. By far the most common method is to
modify the electron donating or accepting character of the D or A
fragments, respectively, by alteration of the molecular
framework.5,25-27 While this strategy has yielded many novel
narrow band gap materials, the synthesis can often require
multiple steps, and thus, new, general, and simpler approaches
to modulate optical properties are desired. Methods such as
protonation28-35 or metal ion complexation36-39 at basic sites,
or chemical doping,40,41 can greatly alter the optical and electro-
nic properties of a material, although this can lead to charged
species and loss of semiconducting behavior, respectively, traits
that may impact device performance.
A wide range of conjugated polymers and related oligomers
contain heteroatoms with available lone pairs of electrons.20 Our
Received: December 16, 2010
Published: March 04, 2011
r
2011 American Chemical Society
4632
dx.doi.org/10.1021/ja110968m J. Am. Chem. Soc. 2011, 133, 4632–4644
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Adduct Formation between 5,50-Bis(benzo[2,1,3]thiadiazole)-3,30-di-n-dodecylsilylene-2,20-bithiophene (BT/DTS/
BT) and B(C6F5)3 as Described in Reference 42
group recently demonstrated that through the selective binding
of Lewis acids to an accessible nitrogen atom on the acceptor
unit, electron density can be removed from the π-conjugated
system, resulting in a narrowing of the band gap.42 In the case of
5,50-bis(benzo[2,1,3]thiadiazole)-3,30-di-n-dodecylsilylene-2,20-
bithiophene (BT/DTS/BT), shown in Scheme 1, one observes a
change in the absorption maximum (λmax) from ∼500 to
∼650 nm upon binding of B(C6F5)3 to the nitrogen atoms in
the outer two benzo[2,1,3]thiadiazole (BT) units. The degree of
change in the absorption properties can be tuned by varying the
strength of the Lewis acid, with the strongest Lewis acid leading
Figure 1. Structures of cyclopentadithiopene-co-benzothiadiazole co-
polymer 1 and related small molecules 2 and 3. R2 = 2-ethylhexyl,
R1 = hexyl.
to the smallest band gap, i.e., most red-shifted absorption
spectrum. Recent reports have demonstrated a similar phenom-
enon for benzo[1,2-c:4,5-c0]bis([1,2,5]thiadiazole) (BBTD)25
and thiazolyl43 based π-conjugated organic systems. We there-
fore envisioned extending this methodology to related polymeric
derivatives, with a focus on B(C6F5)3. This Lewis acid has found
applications in various fields of chemistry owing to its strong
acidity (comparable to BF3 and BCl3), resistance to B-C bond
cleavage, and tolerance to air and moisture.44-46 For example,
B(C6F5)3 has been widely used as cocatalyst in the polymeriza-
tion of olefins,47 a primary catalyst for organic transformations,48
and, more recently, as a key component for metal-free catalysis
via frustrated Lewis pairs.49-51 Herein we examine the interac-
tion of B(C6F5)3 with D/A conjugated polymers to untangle
electronic and steric contributions to the binding equilibrium
and modification of optical properties. Furthermore, we report
on the development of a new class of materials incorporating a
pyridal-like nitrogen into the π-conjugated backbone to increase
the propensity toward Lewis adduct formation.
pyridine (Py) regenerates the original absorption spectrum of
1, which we attribute to formation of the Py-B(C6F5)3 adduct56
and the release of the parent polymer. These data should be
compared against the situation with BT/DTS/BT which can
bind 2 equiv of B(C6F5)3 under similar experimental conditions.
Our initial thoughts centered on the possibility that the differ-
ence in reactivity could be attributed to the more sterically
hindered BT nitrogen atoms, which are doubly flanked in 1 vis
a vis BT/DTS/BT.
To probe the role of steric hindrance, we employed the smaller
Lewis acid BBr3. Addition of 1 equiv of BBr3 to an o-DCB
solution of 1 resulted in an immediate color change from green to
faint yellow-green, while addition of 5 equiv gave a yellow-red
solution. The absorption spectra of these solutions (Figure 2b)
display complete disappearance of the absorption band due to
the parent polymer 1 and the emergence of broad bands with
λmax ∼ 1160 nm and absorption onsets (λonset) extending
beyond 2000 nm. Such spectra are reminiscent of doped
polymers.8 However, addition of excess Py leads to removal of
BBr3 and a return to the absorbance spectrum of 1. These
transformations are consistent with the borane forming a Lewis
adduct via a dative interaction with a heteroatom on the main
chain.25,42 Unfortunately, all attempts to isolate and fully char-
acterize 1/BBr3 adducts proved unsuccessful due to decomposi-
tion to unidentified products upon standing in solution for
several hours or when efforts were made to form films via spin
coating techniques. We attribute this instability to the reactivity
of the B-Br bond when the boron atom is bound to a Lewis base.
In order to understand the results obtained with 1, we
synthesized, examined, and compared the reactivity of two
related well-defined molecules, namely 5,50-bis(benzo[2,1,3]-
thiadiazole)-4,4-bis(hexyl)-4H-cyclopenta[2,1-b;3,4-b0]-
dithiophene (2) and bis(4,4-bis(hexyl)-4H-cyclopenta[2,1-b;3,4-b0]-
dithiophene)}-4,7-benzo[2,1,3]thiadiazole (3). Compounds 2
and 3, shown in Figure 1, contain the same structural units as in 1
and were synthesized by adapting literature procedures;3,42
complete details can be found in the Supporting Information.
’ RESULTS AND DISCUSSION
Addition of Lewis Acids to Polymer 1 and Related Small
Molecules. Initial efforts centered on studying the reaction
between poly[(4,4-bis(2-ethylhexyl)cyclopenta-[2,1-b:3,4-b0]-
dithiophene)-2,6-(diyl-alt-benzo[2,1,3]thiadiazole)-4,7-diyl] (1 in
Figure 1)52-54 and B(C6F5)3. This polymer has been previously
used in the fabrication of BHJ solar cells with power conversion
efficiencies near 5.5% and is a structural analogue to BT/DTS/BT,
except for the C-based cyclopentadithienyl framework and
the use of 2-ethylhexyl substituents.52,55 Addition of 1 equiv of
B(C6F5)3 to a green o-dichlorobenzene (o-DCB) solution of 1
results in no visible reaction or observable color change. On the
basis of our previous findings,42 a solution color change would
imply the formation of a new absorbing species in solution.
Analysis by UV-vis-NIR spectroscopy reveals a decrease of the
band centered at 780 nm and the appearance of a weak and broad
low energy transition in the 900-1700 nm range (Figure 2a).
Addition of up to 5 equiv B(C6F5)3 resulted in no further
significant change in the absorption spectra, with features
attributed to 1 still present. Exposure of these solutions to
4633
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Figure 2. UV-vis-NIR absorption spectra of 1 with varying equivalents of B(C6F5)3 (a) and BBr3 (b) in o-DCB solution at room temperature.
Figure 3. UV-vis-NIR absorption spectra of the small molecules 2 (a) and 3 (b) with no added Lewis acid (black), 1 equiv of B(C6F5)3 per BT unit
(red), and 1 equiv of BBr3 per BT unit (purple) in o-DCB solution at room temperature.
For 3, the BT nitrogen atoms exist in the same chemical
environment as in 1, as they are flanked by cyclopenta-[2,1-
b:3,4-b0]dithiophene (CDT) donor units, whereas the two BT
nitrogen atoms of 2 exist in different chemical environments: one
is flanked by a CDT unit while the other is located at the open
terminus of the molecule. The difference in accessibility of each
nitrogen atom was anticipated to provide insight into the
reactivity toward 3-coordinate boranes.
Addition of 2 mol equiv of B(C6F5)3 to 2 in o-DCB (i.e., one
boron center per BT unit) results in a color change from red
to blue. As shown by the UV-vis-NIR spectra in Figure 3a,
there is a complete disappearance of the absorption band at
λmax = 518 nm with the concomitant emergence of a new band
with λmax = 686 nm, attributed to the bis-B(C6F5)3 adduct.42
This complexation is illustrated in Scheme 2. Similarly, addition
of 2 mol equiv of the stronger and smaller Lewis acid BBr3 to 2
leads to an absorption band with λmax = 717 nm, which is further
red-shifted relative to what is observed with B(C6F5)3. These
trends and observations are similar to those previously obtained
with BT/DTS/BT; adduct formation is nearly complete with
2 equiv borane, and the stronger Lewis acid BBr3 gives rise to the
furthest narrowing of the optical transition.42
1 mol equiv B(C6F5)3 (i.e., one boron center per BT unit). The
absorption spectra of this mixture shown in Figure 3b displays a
slight decrease of the intensity in the bands at 378 and 568 nm,
and the appearance of a weak band at λmax = 810 nm, which is
likely due to Lewis adduct formation.44 Since it is reasonable that
the basicity of the BT nitrogens in 2 and 3 should not greatly
differ, the difference in the spectral characteristics in Figure 3a,b
suggests that steric interference is pronounced in the case of 3;
thus, the equilibrium favors free borane and chromophore, as
shown in Scheme 2. Increasing the concentration of B(C6F5)3 in
solution to upward of 5 mol equiv results in near complete
disappearance of the UV-vis spectra features of 3 and an
increase in the intensity of the long wavelength transition, see
Supporting Information. This result implies that B-N adduct
formation is possible, but not as favored as observed with 2. We
thus examined the addition of BBr3, which leads to an immediate
solution color change from violet to yellow-green. The absorp-
tion spectrum provides more detail; most notably the change-
transfer absorption band is red-shifted by 207 nm, as shown in
Figure 3b, and is considerably broadened. More detailed char-
acterization of the 3-BBr3 interaction could not be achieved due
to the instability of the product. Thus, we initially conclude that
while the absorption spectra of BT containing chromophores can
be altered upon Lewis acid addition, steric constraints are more
In contrast to the observations for 2, there is no observable
color change by visual inspection of 3 in o-DCB upon addition of
4634
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Scheme 2. Reactions of 2 and 3 with B(C6F5)3 at Room Temperature in o-DCBa
a B(C6F5)3 forms a strong adduct with the external N-atoms of the BT unit in 2. Less pronounced adduct formation is observed with the internal BT
N-atom of 3. R1 = hexyl.
Scheme 3. Incorporation of a Pyridal Nitrogen into the Core
D/A Backbone of the Organic Materials via Incorporation of a
PT Fragmenta
[2,1,3]thiadiazole (9) are related to 10 and 11, respectively, and
provide well-defined species to enable more detailed character-
ization via NMR spectroscopy. All four novel materials exhibit
unique properties in their own right and for this reason are
described in some detail below.
The synthesis of small molecules 8 and 9 is outlined in
Scheme 4. The commercially available precursor 4,7-dibromo-
pyridal[2,1,3]thiadiazole (6) was used as received, while 4,7-
bis(5-bromo-2-thienyl)[1,2,5]thiadiazolo[3,4-c]pyridine (7) and
the stannylated compounds 5-(trimethylstannyl)-4,4-bis(hexyl)-
4H-cyclopenta[2,1-b;3,4-b0]dithiophene (4) and 4,4-bis(2-ethyl-
hexyl)-2,6-bis(trimethylstannyl)-4H-cyclopenta[2,1-b:3,4-b0]-
dithiophene (5) were synthesized and purified according to
literature procedures.3,26 Full experimental details and spectro-
scopic characterization can be found in the Supporting Informa-
tion. The targets 8 and 9 were obtained as purple solids via
reaction of 2 equiv of 4 with 1 equiv of 6 or 7, respectively, in
yields greater than 70%. These compounds exhibit good solubi-
lity in most organic solvents and were structurally characterized
by elemental analysis, multinuclear NMR solution spectroscopy,
and mass spectrometry. In comparison to the related BT small
molecules 2 and 3 described above, species 8 and 9 exhibit
asymmetric π-conjugated organic structures, resulting in chemi-
cally distinct NMR spectra for the aromatic component. For
example, compound 8 exhibits seven aromatic resonances in the
1H NMR spectrum, indicating all aromatic protons exist in
chemically different environments. The asymmetry is also ob-
served in the 13C NMR spectrum, where two inequivalent CDT
bridging carbon atoms can be identified. Neither 8 nor 9 showed
obvious thermal transitions in the 0-250 °C range in the solid
state, as determined using differential scanning calorimetery
(DSC).
Polymers 10 and 11 were synthesized via a microwave assisted
Stille polymerization procedure, as shown in Scheme 4.52 Both
polymers incorporate 2-ethylhexyl side chains on the donor
CDT unit to promote good solubility in organic media, particu-
larly chlorobenzene (CB) and o-DCB (∼10 mg per mL). The
average number molecular weights were determined by gel
permeation chromatography (GPC) at 150 °C in 1,2,5-trichlor-
obenzene, and were found to be 16 and 18 kg mol-1 for 10 and
11, respectively. The polydispersity index (PDI) of each polymer
was approximately 2. Due to the nature of the step-growth
polymerization, it is likely that each polymer exhibits a regioirre-
gular structure.26 No thermal transitions in the solid state were
detected by DSC in the 0-250 °C range.
a It is hypothesized that the pyridal N-atom has a higher basicity and is
more sterically accessible than the azole N-atom, allowing for stronger
interactions with organoboranes.
pronounced when BT is flanked by two CDT units, as in the
situation with 1. Subsequent efforts were therefore focused on
the development of new π-conjugated small molecules and
polymers that would more strongly bind B(C6F5)3.
Design and Synthesis of Conjugated Polymers and Small
Molecules Containing the Pyridalthiadiazole (PT) Acceptor
Unit. On the basis of the known affinity of B(C6F5)3 to bind
to pyridine13,56,57 we turned our attention to the [1,2,5]-
thiadiazolo[3,4-c]pyridine (PT) acceptor subunit, as shown in
Scheme 3.26,58 The PT unit is a strong electron acceptor26 which
can lead to charge-transfer characteristics when coupled with
complementary donor fragments. Very recently, thiophene-
based copolymers incorporating the PT unit have been reported
as effective narrow band gap materials for high performance BHJ
solar cells.59 Furthermore, PT offers a pyridal N-atom for
possible Lewis acid binding, which is more basic and accessible
than the BT counterpart (Scheme 3). These considerations led
us to the design of four novel polymers and model compounds as
shown in Scheme 4. The polymer poly[(4,4-bis(2-ethylhexyl)-
cyclopenta-[2,1-b:3,4-b0]dithiophene)-2,6-diyl-alt-([1,2,5]-
thiadiazolo[3,4-c]pyridine)-4,7-diyl] (10) incorporates the CDT
unit as the donor moiety and the PT unit as the acceptor moiety.
A polymer structure with thiophene spacers between the CDT
and PT units, namely poly[(4,4-bis(2-ethylhexyl)cyclopenta-
[2,1-b:3,4-b0]dithiophene)-2,6-diyl-alt-(40,70-bis(2-thienyl)-[1,2,5]-
thiadiazolo[3,4-c]pyridine)-5,5-diyl] (11), was targeted to ex-
plore more subtle steric and electronic effects. The two small
molecules bis(4,4-bis(hexyl)-4H-cyclopenta[2,1-b;3,4-b0]dithiophene)-
4,7-pyridal[2,1,3]thiadiazole (8) and bis{2-thienyl-(4,4-bis-
(hexyl)-4H-cyclopenta[2,1-b;3,4-b0]dithiophene)}-4,7-pyridal-
4635
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Scheme 4. Synthesis of Small Molecules 8 and 9 and Polymers 10 and 11a
a (i) Microwave irradiation, 170 °C, 5 mol % Pd(PPh3)4, toluene; (ii) microwave irradiation, 200 °C, 5 mol % Pd(PPh3)4, xylenes. R1 = hexyl, R2 =
2-ethylhexyl.
Figure 4. Normalized UV-vis-NIR absorption spectra for small molecules 8 (a) and 9 (b). Solution spectra at 25 °C (black) and 110 °C (red)
recorded in o-DCB. As-cast film spectra on quartz (blue) obtained at 25 °C.
Optical Absorption Properties of 8 and 9. Absorption
characteristics of 8 and 9 were determined to provide a baseline
measure when we examine the effect of adding Lewis acids to
them and their polymeric counterparts. The normalized UV-
vis-NIR absorption spectra of 8 and 9 are shown in Figure 4, and
all the relevant data are summarized in Table 1. The λmax values
in o-DCB at 25 °C for the low energy transitions of 8 and 9 occur
at 618 and 626 nm, respectively. Heating these solutions
to 110 °C using a temperature controlled heating stage
(see Supporting Information) has little effect on the spectral
characteristics, and implies that there is negligible interchromo-
phore aggregation under these conditions.52 Transitioning from
solution to thin film, the λmax of 8 red shifts 28 nm while that of 9
remains mostly unchanged. However, the λonset values for 8 and
9 are red-shifted 26 and 28 nm, respectively. The optical band
gaps of 8 and 9, determined from the onset of the thin film
absorption, were estimated to be 1.57 and 1.61 eV, respectively.
These similar values suggest the increased π-conjugated length
afforded by the two additional thiophene units in 9 offsets any
significant LUMO level destabilization caused by their ability to
4636
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Table 1. Summary of Optical Data of Small Molecules and Polymers
compound
λmaxa (nm) (εmax [M-1 cm-1])
λonseta (nm)
λmaxb (nm)
λonsetb (nm)
band gapc (eV)
1.57
8
618 (35 440)
386 (30 670)
626 (37 500)
418 (30 520)
805 (44 980)d
420 (15 410)d
680 (38 120)d
450 (22 450)d
766
646
390
626
416
830
412
690
452
790
9
742
890
855
768
944
875
1.61
1.31
1.42
10
11
a Solution (o-DCB). b Film (quartz). c Estimated from onset of film absorption. d Approximate polymer extinction coefficient determined using
molecular weight of repeat unit.
Figure 5. Normalized UV-vis-NIR absorption spectra for polymers 10 (a) and 11 (b). Solution spectra at 25 °C (black) and 110 °C (red) recorded in
o-DCB. As-cast film spectra on quartz (blue) obtained at 25 °C.
inductively donate electron density.60 It is also worth pointing
out that 8 has a smaller band gap than the corresponding BT-
containing analogues (1.87-1.95 eV),3 but is comparable to
DAD small molecules based upon the thienopyrazine acceptor
unit (1.55-1.55 eV),3 thus demonstrating the high electron
affinity of the PT unit.
Optical Absorption Properties of 10 and 11. The absorp-
tion spectra of polymers 10 and 11 in o-DCB solution at 25 and
110 °C, and as thin films, are shown in Figure 5, with full details
listed in Table 1. The λmax values of 10 and 11, at 25 °C in o-
DCB, are 805 and 680 nm, respectively. Polymer 10 exhibits a
broad absorption spectrum with an estimated λonset in solution of
and 11 have absorption spectra that are significantly red-shifted
due to the higher electron affinity of the PT unit.
Reaction of B(C6F5)3 with 8 or 9. A slight color change from
blue to blue-green is observed upon addition of 1 equiv B(C6F5)3
to 8 in o-DCB. Analysis by UV-vis spectroscopy at 25 °C reveals
a large red shift in λmax (Δ = 136 nm) and λonset (Δ = 26 nm), as
shown in Figure 6a. These are hallmarks of B(C6F5)3 adduct
formation, and based on the lack of reactivity with 3 discussed
above, we presume that binding occurs via the pyridal N-atom, as
depicted in Scheme 5.3,62 Figure 6a also exhibits a shoulder at
approximately 615 nm, due to the presence of unreacted 8. This
observation implies that equilibrium exists between free and
coordinated 8 at 25 °C in solution. Transitioning from solution
to the solid state, equilibrium is driven toward adduct formation,
and the absorption spectrum only exhibits features from the 8-
B(C6F5)3 adduct. Comparatively, addition of greater than 1
equiv of Lewis acid to 8 in solution shifts the equilibrium in
favor of adduct formation, as determined by UV-vis spectros-
copy, see Supporting Information. Upon addition of Py, the
borane is displaced from 8. Parent 8 can be recovered by solvent
removal and separation from the (C6F5)3B-Py adduct by flash
column chromatography with hexanes. Adduct formation is
therefore chemically reversible.
890 nm. Polymer 11 is blue-shifted relative to 10 with an λonset
=
855 nm in solution at 25 °C. The introduction of thiophene
linkages61 between the CDT donor unit and the PT acceptor unit
has the primary effect of destabilizing the LUMO energy level
(vide infra), resulting in a widening of the band gap, and thus the
blue-shifted absorption spectra of 11 versus 10. The optical
absorption of polymers 10 and 11 shows a dependence on
solution temperature, as heating to 110 °C results in a slight
blue shift of the low energy absorption band. This result is
consistent with aggregation of the polymer chains in solution.52
A
transition from solution to film results in red-shifted absorption
spectra for both 10 and 11. The optical band gaps of 10 and 11 in
the solid state were estimated to be 1.31 and 1.42 eV, respec-
tively. Compared to the related BT derivatives,4,52 polymers 10
In an analogous fashion to 8, small molecule 9 was subjected to
the addition of 1 equiv of B(C6F5)3. The resulting solution and
film absorption spectra are shown in Figure 6b. Red shifts of
4637
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Figure 6. Normalized UV-vis-NIR absorption spectra of (a) 8 as-cast thin film (blue), 8 þ 1 equiv B(C6F5)3 in o-DCB solution (orange) and as-cast
thin film (green); (b) 9 as-cast thin film (blue), 9 þ 1 equiv B(C6F5)3 in o-DCB solution (orange) and as-cast thin film (green).
a
Scheme 5. Proposed Adduct Formation between 8 and B(C6F5)3
a Adduct formation is fully reversible in the presence of Py. R1 = hexyl.
Table 2. Summary of Optical Data of Small Molecule and Polymer Lewis Adducts with B(C6F5)3
compound
λmaxa (nm)
λonseta (nm)
λmaxb (nm)
λonsetb (nm)
band gapc (eV)
1.31
8-B(C6F5)3
752
892
764
950
562, 640, 384
820
558, 386
868
9-B(C6F5)3
10-B(C6F5)3
11-B(C6F5)3
1022
1162
1236
1152
1290
1395
1.08
0.96
0.89
632, 416
998
568, 444
1050
440
410
986
1090, 480
450
a Solution (o-DCB). b Film (quartz). c Determined from λonset
.
194 and 280 nm for λmax and λonset, respectively, were observed
in solution. Again, in solution, an equilibrium exists between free
and bound borane, and upon transitioning to the solid state, only
the 9-B(C6F5)3 adduct is observed. The optical band gap in the
solid state of the adduct 9-B(C6F5)3 was determined to be 1.08
eV (Table 2).
Further insight into the reaction of 8 and 9 with B(C6F5)3 was
obtained by using multinuclear NMR solution spectroscopy. The
1H NMR spectrum of an equal molar solution of 8 and B(C6F5)3
in CD2Cl2 revealed significant broadening of the aromatic proton
resonances at 300 K (Figure 7). This observation indicates that
an exchange mechanism exists between 8 and possibly the 8-
B(C6F5)3 adduct that is commensurate with the NMR time scale.
Upon cooling to 280 K the aromatic resonances sharpen,
revealing resonances attributable to 8 and a new species, assigned
to 8-B(C6F5)3. Further cooling to 230 K shifts the equilibrium
toward adduct formation, and thus, only signals from 8-
B(C6F5)3 can be observed. The equilibrium constant at 280 K was
determined to be 1.2 ꢀ 102 M-1. For reference, no significant
change in the 1H NMR spectrum of 8 was observed upon cooling
from 300 to 230 K, see Supporting Information. A similar
equilibrium between B(C6F5)3 and lutidine has recently been
observed and characterized.63,64 The 19F NMR spectrum at
300 K shown in Figure 8 displays 3 major resonances at -128.2,
-143.8, and -160.9, for the ortho, meta, and para resonances of
free B(C6F5)3, respectively, consistent with an equilibrium
4638
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Figure 8. 19F NMR spectra of B(C6F5)3 (top) and B(C6F5)3 þ 1 equiv
of 8 at various temperatures from 300 to 230 K. All spectra were
recorded in CD2Cl2. Upon cooling the reaction mixture from 300 to 230
K, adduct formation is observed as indicated by the disappearance of the
resonances for free B(C6F5)3 and the appearance of 15 inequivalent
fluorine resonances for four coordinate boron with restricted motion.
Figure 7. Aromatic region of the 1H NMR spectra of 8 (top) and 8 þ 1
equiv B(C6F5)3 at various temperatures from 300 to 230 K. All spectra
were recorded in CD2Cl2. Upon addition of B(C6F5)3 the aromatic
resonances of 8 are broadened due to rapid exchange of bound and
unbound Lewis acid. Upon cooling to 280 K, resonances for 8 and 8-
B(C6F5)3 are observed. Further cooling to 230 K drives the equilibrium
fully toward adduct formation and resonances for 8-B(C6F5)3 are
exclusively observed.
B(C6F5)3 has a stronger propensity to form adducts with harder
N-containing bases.42 As an aside, we investigated the related and
sterically less hindered molecule 2,20-bithiophene, which con-
tains no N-atoms, and observed no evidence of adduct formation
with B(C6F5)3 via 1H and 19F NMR solution spectroscopy at 300
K (see Supporting Information). We therefore conclude that any
interaction of B(C6F5)3 with the S atoms of 8 or 9 is highly
unlikely.
mixture that favors the unbound starting materials. Upon cooling
to 230 K, the three ortho, meta, and para resonances of free
B(C6F5)3 disappear, and 15 new resonances from -125 to -166
ppm are observed which indicates that all the fluorines in
B(C6F5)3 are inequivalent and is consistent with formation of
8-B(C6F5)3, in which the boron center exists in a 4-coordinate
pseudotetrahedral orientation.42,44,57 Steric interactions between
the C6F5 rings and either the hexyl alkyl chains or the thiophene
rings restrict rotation about the B-N and B-C bonds, thus
rendering all 15 fluorine atoms inequivalent. Addition of an equal
molar amount of pyridine to this solution results in regeneration
of 8 and the Py-B(C6F5)3 adduct. A similar equilibrium was
Reaction of B(C6F5)3 with Polymers 10 and 11. In o-DCB
solution, 10 was subjected to the addition of varying equivalents
of B(C6F5)3, and the resulting absorption spectra are provided in
Figure 9a. Upon addition of 0.1 equiv B(C6F5)3 the absorption
peak at ∼805 nm decreases, while a new red-shifted absorption
peak appears at ∼980 nm. Addition of 0.5 equiv of B(C6F5)3
results in almost complete disappearance of the absorption band
for 10, and an increase in the strength of the low energy band.
Further addition of B(C6F5)3, up to 10 equiv, results in a
progressive red shift of λmax (980 to 1000 nm), while the λonset
is only slightly red-shifted. This observation can likely be
attributed to equilibrium existing between free and bound 10
in solution. Increasing the concentration of B(C6F5)3 in solution
shifts the equilibrium in favor of adduct formation. These
observations are in stark contrast to the reaction of 1 and
B(C6F5)3, where complete quenching of the absorption of 1 is
not observed upon addition of excess Lewis acid (Figure 2a).
Given the electronic communication along the backbone of a
conjugated polymer, it seemed reasonable to us that B(C6F5)3
binding could lead to an amplification effect, where interaction
with a single PT site would impact the optical properties of
several PT-CDT repeat units. To examine this possibility we re-
examined in more detail the interaction between oligomer 8 and
1
observed in CD2Cl2 by H and 19F NMR spectroscopy for an
equal molar mixture of 9 and B(C6F5)3. Here, the equilibrium
constant at 280 K was determined to be 2.8 ꢀ 102 M-1. The
slightly stronger affinity of 9 compared to 8 for binding of
B(C6F5)3 is likely attributed to (i) a more accessible pyridal
N-atom due to a diminished steric impact from the CDT alkyl
substituents, and (ii) an increased basicity at the pyridal N-atom
due to the inductively donating effect of the additional thiophene
linkages. The B(C6F5)3 adducts of 8 and 9 are sensitive to H2O
and readily hydrolyze to yield pyridinum borate salts, a known
decomposition pathway for nitrogen-based B(C6F5)3
adducts.65,66 It is also worth mentioning that while the S atoms
of 8 and 9 have available lone pairs of electrons for potentially
binding of Lewis bases, previous work has demonstrated that
4639
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Figure 9. UV-vis-NIR absorption spectra of polymer 10 (a) and oligomer 8 (b) with varying equivalents of B(C6F5)3 in o-DCB at room temperature
under N2.
Figure 10. Normalized UV-vis-NIR absorption spectra of (a) 10 with 1 equiv B(C6F5)3 in o-DCB solution at room temperature under N2 (orange),
with 1 equiv B(C6F5)3 as thin films (green). (b) Absorption spectra of 10 plus varying equivalents of B(C6F5)3 as thin films on quartz cast from CB at
1500 rpm under an N2 atmosphere.
B(C6F5)3, see Figure 9b. By looking at the change in absorbance
upon addition of 0.5 equiv of B(C6F5)3, one can estimate that
approximately 25% of 8 is bound, while increasing the concen-
tration to 2.5 equiv shifts the equilibrium toward nearly quanti-
tative adduct formation. Further addition, up to 5 equiv, has
minimal impact on the optical absorption spectrum. If one
assumes a similar binding constant for 10, then one-quarter of
the PT fragments in the polymer should be bound under similar
conditions. This assumption is reasonable given the similar
basicity and local steric environment of the PT units in 8 and
10. In fact, it would not be surprising if the binding constant for
10 is smaller given the more restricted environment within the
macromolecular framework and the fact that attaching one
B(C6F5)3 unit leads to a net decrease in electron density on
the backbone. We can thus conclude that only a fractional
number of PT units (25% or less) in 10 are occupied upon
addition of 0.5 equiv of B(C6F5)3, and that this structural
modification completely modifies the intrinsic absorption fea-
tures of the entire polymer chain. Beyond this point, when
additional Lewis acid is introduced, there is a further increase
in the number of bound PT units, leading to the further red shift
in absorbance shown in Figure 9a. Therefore, there is an amplifica-
tion effect of Lewis acid binding to a conjugated polymer backbone
that is not readily achieved with small molecule counterparts.
Transitioning from solution to thin film, a red shift of 52 and
128 nm is observed for the λmax and λonset of 10-B(C6F5)3
(Figure 10a, Table 2), a possible result of closer interpolymer
chain interactions and the anticipated shift toward B-N adduct
formation. It is noteworthy that, upon addition of B(C6F5)3 to
10, a red shift of ∼340 nm in the thin film absorption onset,
compared to parent spectrum of 10, can be achieved. Investiga-
tion of the binding of B(C6F5)3 to 10 in the solid state
(Figure 10b) revealed no further changes after addition of
approximately 0.6 equiv of Lewis acid, which suggests that
binding saturates after (on average) every second PT unit has
been coordinated (Scheme 6). Furthermore, we suggest that the
additional encumbrance by the borane should suppress inter-
chain S N interactions that can mediate solid state
3 3 3
organization.67,68
In a similar fashion, the 11-B(C6F5)3 adduct can be readily
obtained upon addition of B(C6F5)3 to 11. The optical absorp-
tion spectra in solution and as thin films are shown in Figure 11
4640
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
a
Scheme 6. Proposed Interaction between Polymer 10 and B(C6F5)3
a On the basis of optical absorption saturation studies in the solid state, only half of the PT acceptor units can interact with B(C6F5)3. Adduct formation is
fully reversible in the presence of Py. R2 = 2-ethylhexyl.
Table 3. Summary of HOMO-LUMO Energy Levels Ob-
tained via UPS and Absorption Spectroscopy
compounda HOMOb LUMOc Egd compound HOMOb LUMOc Egd
8
-4.86 -3.29 1.57 10
8-B(C6F5)3 -5.37 -4.06 1.31 10-B(C6F5)3 -5.24 -4.28 0.96
-4.79 -3.18 1.61 11 -4.91 -3.49 1.42
-5.01 -3.70 1.31
9
9-B(C6F5)3 -5.08 -4.00 1.08 11-B(C6F5)3 -5.23 -4.34 0.89
a Thin films on Au substrates. b In electron volts. Obtained from UPS
c
measurements. In electron volts. Estimated by using the UPS-deter-
mined HOMO value and the optical band gap (Eg). d Estimated from the
onset of film absorption.
electrolytes. Table 3 summarizes the UPS results; spectra can
be found in the Supporting Information. It is worth pointing out
that the HOMO levels determined via UPS for 10-11 match
those obtained by CV to within ∼0.1 eV, see Supporting
Information. The HOMO energy levels for polymers 10 and
11 were determined to be -5.01 and -4.91 eV, respectively. On
the basis of the onset of optical absorption, we estimated the
LUMO energy levels at -3.70 and -3.49, for 10 and 11,
respectively. The destabilization of both the HOMO and LUMO
of 11 compared to 10 is a result of the increased electron density
across the π-conjugated system afforded by the additional two
electron rich thiophene moieties per repeat unit. A similar but
less pronounced trend can be observed with 8 (ELUMO = -3.29
eV) and 9 (ELUMO = -3.18 eV).
Interaction of the small molecules and/or polymers with
B(C6F5)3 results in a synergetic lowering of both the HOMO
and LUMO energy levels, with the LUMO exhibiting the greatest
change. For example, the HOMO and LUMO energies of
polymer 10 were determined to be -5.01 and -3.70 eV,
respectively (Table 3). Binding B(C6F5)3 to the polymeric
backbone via a B-N dative interaction yields the adduct 10-
B(C6F5)3 vide supra, which was determined to have HOMO and
LUMO energies of -5.24 and -4.28 eV, respectively (Table 3).
Here, interaction of the polymer with B(C6F5)3 causes a low-
ering of the HOMO by 0.23 eV while the LUMO is lowered by
0.58 eV. The greater change in the LUMO energy by 0.35 eV
results in the narrowing of the optical band gap. These data imply
that coordination of B(C6F5)3 to the PT acceptor unit not only
removes electron density away from this unit, but also from the
entire π-conjugated system. Quite remarkably, for each material,
the HOMO is lowered in energy by ∼0.2-0.5 eV, upon B-
(C6F5)3 coordination, and thus this method of adduct formation
presents itself as a facile route toward significantly modifying
Figure 11. Normalized UV-vis-NIR absorption spectra of 11 as thin
films (blue), with 1 equiv B(C6F5)3 in o-DCB solution at room
temperature under N2 (orange), with 1 equiv B(C6F5)3 as thin films
(green).
and are summarized in Table 2. The 11-B(C6F5)3 adduct has
significant absorption into the NIR-region with λonset values in
solution and the solid state of 1236 and 1395 nm, respectively.
Compared to 10-B(C6F5)3, 11-B(C6F5)3 has a smaller optical
band gap, which we attribute to a stronger B-N interaction as a
result of the additional thiophene linkages, which likely provide a
more basic and sterically accessible coordination site for binding
B(C6F5)3. It is also worth mentioning that, for both polymers,
addition of B(C6F5)3 only affects the low energy charge transfer
energy band, while the high energy band at ∼400-450 nm
remains virtually unchanged. While the complexation of halo-
genated main group Lewis acids to rigid-rod poly(p-phenylene-
benzobisthiazole) type polymers has been reported,69-72 no
significant changes in the optical properties of the parent
polymers were observed. Thus, these findings therefore present
a facile method for red-shifting the optical absorption spectra of
narrow band gap polymers.
Determination of the Frontier Molecular Orbital Energy
Levels and the Impact of B(C6F5)3 Coordination. To gain
insight into the nature of the electronic states we estimated the
HOMO energy levels of each material as thin films using
ultraviolet photoemission spectroscopy (UPS). The more typical
approach involving electrochemical determination via cyclic
voltammetry (CV) experiments was not feasible due to the
instability of the B(C6F5)3 adducts in typical supporting
4641
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Table 4. Summary of HOMO and LUMO Energy Level
Values Determined by DFT Calculationsa
compound
HOMO (eV)
LUMO (eV)
Eg (calcd)
8
-4.73
-5.17
-4.65
-4.95
-2.80
-3.46
-2.86
-3.48
1.93
1.71
1.79
1.47
8-BCl3
9
9-BCl3
a B3LYP/6-31G(d,p).
C6F5 aryl rings, were used to enable convergence to a global
minimum. Relative energies determined from the optimized
structures show that adduct formation at the pyridal N-atom is
more favorable by a minimum of 27 kJ/mol over adduct
formation at the azole N-atom (Figure 12). The optimized
structures also reveal that binding of the borane to one of the
azole N-atoms results in a distortion from planarity of a CDT unit
by nearly 60°, whereas binding of B to the pyridal N-atom only
forces one CDT unit 25° out of plane (Figure 13). These results
are consistent with our previous proposal that the more pro-
nounced affinity of the Lewis acidic borane to bind the pyridal
N-atom is a consequence of a combination of increased basicity
and steric accessibility. Furthermore, it is noteworthy to mention
that the distortion from planarity along the π-conjugated back-
bone upon adduct formation would be anticipated to decrease π-
electron conjugation.71 The observed band gap narrowing im-
plies that the increased electron redistribution afforded by the
increase of the electron affinity of the PT acceptor overcompen-
sates for loss in delocalization.
Figure 12. Ground state geometry optimizations of (a) 8 and (b) 9
and their corresponding adducts with BCl3. Methyl groups were
used in place of the hexyl side chains on the dithiophene bridging
carbon, while chlorine atoms were used instead of C6F5 on boron.
Color scheme: carbon, gray; nitrogen, blue; sulfur, orange; boron,
pink; chlorine, green. Optimized structures calculated using DFT at
the B3LYP/6-31G(d,p) level of theory. Relative energies are shown
under each diagram.
Lewis adduct formation significantly modifies the molecular
orbital energy levels. Focusing on 8, the HOMO and LUMO
energy levels were calculated to be -4.73 and -2.80 eV,
respectively. With coordinated BCl3, these energy levels decrease
to -5.17 and -3.46, respectively. The changes in absolute
energies determined by UPS measurements are in good agree-
ment with these trends. Specifically, both the HOMO and
LUMO energies decrease in energy upon Lewis acid complexa-
tion. However, the LUMO energy level is more affected, con-
sistent with the experimental results in Table 3. Examination of
the electron distribution of the HOMO also reveals a slightly
greater localization on the CDT units upon Lewis acid binding
(Supporting Information).
Figure 13. Perspectives of (a) 8-BCl3 binding through azole N-atom
(left) and (b) 8-BCl3 binding through pyridal N-atom (right) that
highlight the torsion angle between the CDT and PT units.
both of the HOMO/LUMO energy levels of π-conjugated
materials. Comparing polymers 10 to 11, adduct formation with
B(C6F5)3 results in a lowering of the HOMO energy levels by
0.23 and 0.32 eV, respectively. It is plausible that the slightly
larger change for 11 is due to the increased strength of the acid-
base adduct, which is anticipated to be more effective at depleting
electron density from the π-conjugated system. Comparing 8 to
9, adduct formation results in a decrease in the HOMO energy
levels by 0.51 and 0.29 eV, respectively. Interestingly, all B-
(C6F5)3 adducts have LUMO energy levels below -4 eV, which
may lend themselves useful as n-type materials in the fabrication
of organic solar cells.73
Computational Investigation of Regioselective Lewis Acid
Coordination and Band Gap Narrowing. Density functional
theory (DFT) methods were used to gain insight into the nature
of the chromophore-borane interaction. Analogues of 8 and 9
were optimized at the B3LYP/6-31G** level of theory with the
simplification of replacing the hexyl chains with methyl groups.
Adducts with BCl3 bound independently to each N-atom of the
PT unit were subsequently optimized, see Supporting Informa-
tion. The Cl atoms, which have a similar electron affinity as the
’ CONCLUSIONS
To summarize, we have shown how to modulate the absorp-
tion properties of donor/acceptor conjugated polymers by
coordination of Lewis acids to heteroatoms on the acceptor
fragments. When examining the widely used polymer 1, one finds
that the driving force for binding B(C6F5)3 to the BT azole
N-atom is considerably reduced relative to previously examined
small molecule analogues. From the reactivity of model com-
pounds 2 (which binds strongly) and 3 (which binds weakly), we
surmise that the flanking CDT units adjacent to each BT
fragment in 1 provide steric interference that shifts the equilib-
rium for B(C6F5)3 adduct formation toward the unbound
species. It is possible to circumvent these steric limitations by
using the smaller BBr3; however, the ensuing adducts are
unstable, most likely as a result of propensity for B-Br bond
cleavage and bromide generation, and are therefore difficult to
properly characterize.
4642
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
Successful formation of polymer-B(C6F5)3 adducts required
the introduction of the PT subunit as the acceptor segment. Our
original thoughts for structural design focused on the more basic
and sterically accessible pyridal N-atom, relative to the azole
N-atom in BT, as shown in Scheme 3. Indeed, we find that
polymer 10 has a greater affinity for B(C6F5)3 than 1, despite
their structural similarities. Similar tendencies are observed with
the small molecule analogues 8 and 3. Examination of the
differences by which the absorption characteristics of 8 and 10
change with a given quantity of borane shows that there is a
previously not observed amplification effect, whereby the bind-
ing of B(C6F5)3 to a fractional number of basic sites can
change the optical properties of the entire backbone. Introduc-
tion of electron donating thiophene units adjacent the PT
acceptor, specifically the generation of 11 and 9, has the effect
of increasing both the distance between the PT unit and the alkyl
side chains and the electron density across the π-conjugated
system. The increased electron density, and more open coordi-
nation site of the PT unit, results in stronger B-N interactions
upon Lewis acid adduct formation. One therefore observes larger
changes in the optical band gaps of 9 and 11, compared to 8 and
10, upon B(C6F5)3 coordination. From a materials perspective, it
is worth pointing out that, through the binding of B(C6F5)3, new
NIR-absorbing polymers can be readily generated with band gaps
of 0.96 and 0.89 eV, for 10- B(C6F5) and 11- B(C6F5),
respectively.
The combination of experimental estimates of HOMO-
LUMO levels and the DFT results provides additional insights
on the role of Lewis acid binding on the preferences for site
attachment and the electronic structure of the chromophores. By
using BCl3 as the model acid, we find that there is a thermo-
dynamic preference for attaching to the pyridal N-atom over the
azole N-atom, consistent with the composite body of experi-
mental results. Coordination of the chromophore to the electron
deficient boron results in a lowering of both HOMO and LUMO
levels. However, the effect is more pronounced for the LUMO,
which resides predominantly near the acceptor fragment. This
differential influence over the HOMO and the LUMO is
responsible for the observed decrease in band gap energy.
’ REFERENCES
(1) Forrest, S. R.; Thompson, M. E. Chem. Rev. 2007, 107, 923–925.
(2) Meier, H. Angew. Chem., Int. Ed. 2005, 44, 2482–2506.
(3) Karsten, B. P.; Bijleveld, J. C.; Viani, L.; Cornil, J.; Gierschner, J.;
Janssen, R. A. J. J. Mater. Chem. 2009, 19, 5343–5350.
(4) Bijleveld, J. C.; Shahid, M.; Gilot, J.; Wienk, M. M.; Janssen,
R. A. J. Adv. Funct. Mater. 2009, 19, 3262–3270.
(5) Luo, M.; Shadnia, H.; Qian, G.; Du, X. B.; Yu, D. B.; Ma, D. G.;
Wright, J. S.; Wang, Z. Y. Chem.—Eur. J. 2009, 15, 8902–8908.
(6) Hellstrom, S.; Zhang, F. L.; Inganas, O.; Andersson, M. R. Dalton
Trans. 2009, 10032–10039.
(7) Zhang, F. L.; Bijleveld, J.; Perzon, E.; Tvingstedt, K.; Barrau, S.;
Inganas, O.; Andersson, M. R. J. Mater. Chem. 2008, 18, 5468–5474.
(8) Beaujuge, P. M.; Ellinger, S.; Reynolds, J. R. Nat. Mater. 2008,
7, 795–799.
(9) Son, H. J.; Han, W. S.; You, D. H.; Min, K. T.; Kwon, S. N.; Ko, J.;
Kang, S. O. J. Org. Chem. 2009, 74, 3175–3178.
(10) van Mullekom, H. A. M.; Vekemans, J. A. J. M.; Meijer, E. W.
Chem.—Eur. J. 1998, 4, 1235–1243.
(11) Albota, M.;; et al. Science 1998, 281, 1653–1656.
(12) Entwistle, C. D.; Marder, T. B. Chem. Mater. 2004,
16, 4574–4585.
(13) Lesley, M. J. G.; Woodward, A.; Taylor, N. J.; Marder, T. B.;
Thornton, A.; Bruce, D. W.; Kakkar, A. K. Chem. Mater. 1998,
10, 1355–1365.
(14) Haldi, A.; Kimyonok, A.; Domercq, B.; Hayden, L. E.; Jones,
S. C.; Marder, S. R.; Weck, M.; Kippelen, B. Adv. Funct. Mater. 2008,
18, 3056–3062.
(15) Chen, C. T. Chem. Mater. 2004, 16, 4389–4400.
(16) Zeng, W. D.; Cao, Y. M.; Bai, Y.; Wang, Y. H.; Shi, Y. S.; Zhang,
M.; Wang, F. F.; Pan, C. Y.; Wang, P. Chem. Mater. 2010, 22, 1915–1925.
(17) Mishra, A.; Fischer, M. K. R.; Bauerle, P. Angew. Chem., Int. Ed.
2009, 48, 2474–2499.
(18) Choi, H.; Baik, C.; Kang, S. O.; Ko, J.; Kang, M. S.; Nazeer-
uddin, M. K.; Gratzel, M. Angew. Chem., Int. Ed. 2008, 47, 327–330.
(19) Dennler, G.; Scharber, M. C.; Brabec, C. J. Adv. Mater. 2009,
21, 1323–1338.
(20) Cheng, Y. J.; Yang, S. H.; Hsu, C. S. Chem. Rev. 2009,
109, 5868–5923.
(21) Kroon, R.; Lenes, M.; Hummelen, J. C.; Blom, P. W. M.; De
Boer, B. Polym. Rev. 2008, 48, 531–582.
(22) Chen, J. W.; Cao, Y. Acc. Chem. Res. 2009, 42, 1709–1718.
(23) Qian, G.; Wang, Z. Y. Chem.—Asian J. 2010, 5, 1006–1029.
(24) Gong, X.; Tong, M. H.; Xia, Y. J.; Cai, W. Z.; Moon, J. S.; Cao,
Y.; Yu, G.; Shieh, C. L.; Nilsson, B.; Heeger, A. J. Science 2009,
325, 1665–1667.
’ ASSOCIATED CONTENT
(25) Qian, G.; Wang, Z. Y. Can. J. Chem. 2010, 88, 192–201.
(26) Blouin, N.; Michaud, A.; Gendron, D.; Wakim, S.; Blair, E.;
Neagu-Plesu, R.; Belletete, M.; Durocher, G.; Tao, Y.; Leclerc, M. J. Am.
Chem. Soc. 2008, 130, 732–742.
(27) Roncali, J. Chem. Rev. 1997, 97, 173–205.
(28) Romero-Nieto, C.; Durben, S.; Kormos, I. M.; Baumgartner, T.
Adv. Funct. Mater. 2009, 19, 3625–3631.
(29) Hancock, J. M.; Jenekhe, S. A. Macromolecules 2008,
41, 6864–6867.
(30) Kappaun, S.; Horner, S.; Kelterer, A. M.; Waich, K.; Grasse, F.;
Graf, M.; Romaner, L.; Niedermair, F.; Mullen, K.; Grimsdale, A. C.; Saf,
R.; List, E. J. W.; Zojer, E.; Slugovc, C. Macromol. Chem. Phys. 2008,
209, 2122–2134.
S
Supporting Information. Full details on materials synth-
b
esis, spectroscopic characterization, and all computational re-
sults. Complete ref 11. This material is available free of charge via
the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
bazan@chem.ucsb.edu
(31) Terashima, T.; Nakashima, T.; Kawai, T. Org. Lett. 2007,
9, 4195–4198.
’ ACKNOWLEDGMENT
Financial support through the Center for Energy Efficient
Materials (DOE) and the National Science Foundation (DMR
Program) is gratefully acknowledged. G.C.W. is grateful for a
NSERC PDF scholarship. Drs. Robert Coffin, Junghwa Seo, and
Asit Parta are acknowledged for contributions to synthesis, UPS/
XPS, and CV measurements, respectively.
(32) Sun, M. T. J. Chem. Phys. 2006, 124, 054903.
(33) Yasuda, T.; Yamamoto, T. Macromolecules 2003,
36, 7513–7519.
(34) Monkman, A. P.; Palsson, L. O.; Higgins, R. W. T.; Wang, C. S.;
Bryce, M. R.; Batsanov, A. S.; Howard, J. A. K. J. Am. Chem. Soc. 2002,
124, 6049–6055.
4643
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644

Journal of the American Chemical Society
ARTICLE
(35) Liaw, D. J.; Wang, K. L.; Chang, F. C. Macromolecules 2007,
40, 3568–3574.
(72) Roberts, M. F.; Jenekhe, S. A. Chem. Mater. 1993,
5, 1744–1754.
(36) Yamamoto, T.; Fukumoto, H.; Koizumi, T. J. Inorg. Organomet.
Polym. Mater. 2009, 19, 3–11.
(73) Anthony, J. E.; Facchetti, A.; Heeney, M.; Marder, S. R.; Zhan,
X. W. Adv. Mater. 2010, 22, 3876–3892.
(37) Tolosa, J.; Zucchero, A. J.; Bunz, U. H. F. J. Am. Chem. Soc.
2008, 130, 6498–6506.
(38) Zucchero, A. J.; Wilson, J. N.; Bunz, U. H. F. J. Am. Chem. Soc.
2006, 128, 11872–11881.
(39) Wilson, J. N.; Bunz, U. H. F. J. Am. Chem. Soc. 2005,
127, 4124–4125.
(40) Moliton, A.; Hiorns, R. C. Polym. Int. 2004, 53, 1397–1412.
(41) Patil, A. O.; Heeger, A. J.; Wudl, F. Chem. Rev. 1988,
88, 183–200.
(42) Welch, G. C.; Coffin, R.; Peet, J.; Bazan, G. C. J. Am. Chem. Soc.
2009, 131, 10802–10803.
(43) Job, A.; Wakamiya, A.; Kehr, G.; Erker, G.; Yamaguchi, S. Org.
Lett. 2011, 12, 5470–5473.
(44) Piers, W. E. Adv. Organomet. Chem. 2005, 52, 1-76.
(45) Erker, G. Dalton Trans. 2005, 1883–1890.
(46) Piers, W. E.; Chivers, T. Chem. Soc. Rev. 1997, 26, 345–354.
(47) Chen, E. Y. X.; Marks, T. J. Chem. Rev. 2000, 100, 1391–1434.
(48) Parks, D. J.; Piers, W. E. J. Am. Chem. Soc. 1996,
118, 9440–9441.
(49) Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed. 2010,
49, 46–76.
(50) Chase, P. A.; Welch, G. C.; Jurca, T.; Stephan, D. W. Angew.
Chem., Int. Ed. 2007, 46, 8050–8053.
(51) Welch, G. C.; Juan, R. R. S.; Masuda, J. D.; Stephan, D. W.
Science 2006, 314, 1124–1126.
(52) Coffin, R. C.; Peet, J.; Rogers, J.; Bazan, G. C. Nat. Chem. 2009,
1, 657–661.
(53) Zhu, Z.; Waller, D.; Gaudiana, R.; Morana, M.; Muhlbacher, D.;
Scharber, M.; Brabec, C. Macromolecules 2007, 40, 1981–1986.
(54) Muhlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z. G.; Waller,
D.; Gaudiana, R.; Brabec, C. Adv. Mater. 2006, 18, 2884.
(55) Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger,
A. J.; Bazan, G. C. Nat. Mater. 2007, 6, 497–500.
(56) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1966, 5, 218–&.
(57) Focante, F.; Mercandelli, P.; Sironi, A.; Resconi, L. Coord.
Chem. Rev. 2006, 250, 170–188.
(58) Fukumoto, H.; Yamamoto, T. J. Polym. Sci., Part A: Polym.
Chem. 2008, 46, 2975–2982.
(59) Huaxing, Z.; Yang, L.; Price, S. C.; Knight, K. J.; You, W. Angew.
Chem., Int. Ed. 2010, 49, 7922–7995.
(60) Melucci, M.; Favaretto, L.; Zanelli, A.; Cavallini, M.; Bongini,
A.; Maccagnani, P.; Ostoja, P.; Derue, G.; Lazzaroni, R.; Barbarella, G.
Adv. Funct. Mater. 2010, 20, 445–452.
(61) Moule, A. J.; Tsami, A.; Buennagel, T. W.; Forster, M.;
Kronenberg, N. M.; Scharber, M.; Koppe, M.; Morana, M.; Brabec,
C. J.; Meerholz, K.; Scherf, U. Chem. Mater. 2008, 20, 4045–4050.
(62) Karsten, B. P.; Janssen, R. A. J. Org. Lett. 2008, 10, 3513–3516.
(63) Geier, S. J.; Stephan, D. W. J. Am. Chem. Soc. 2009,
131, 3476–3477.
(64) Geier, S. J.; Gille, A. L.; Gilbert, T. M.; Stephan, D. W. Inorg.
Chem. 2009, 48, 10466–10474.
(65) Di Saverio, A.; Focante, F.; Camurati, I.; Resconi, L.; Berin-
ghelli, T.; D’Alfonso, G.; Donghi, D.; Maggioni, D.; Mercandelli, P.;
Sironi, A. Inorg. Chem. 2005, 44, 5030–5041.
(66) Beringhelli, T.; Maggioni, D.; D’Alfonso, G. Organometallics
2001, 20, 4927–4938.
(67) Naraso; Wudl, F. Macromolecules 2008, 41, 3169–3174.
(68) Karikomi, M.; Kitamura, C.; Tanaka, S.; Yamashita, Y. J. Am.
Chem. Soc. 1995, 117, 6791–6792.
(69) Roberts, M. F.; Jenekhe, S. A. Chem. Mater. 1994, 6, 135–145.
(70) Roberts, M. F.; Jenekhe, S. A. Polymer 1994, 35, 4313–4325.
(71) Roberts, M. F.; Jenekhe, S. A.; Cameron, A.; Mcmillan, M.;
Perlstein, J. Chem. Mater. 1994, 6, 658–670.
4644
dx.doi.org/10.1021/ja110968m |J. Am. Chem. Soc. 2011, 133, 4632–4644