Conjugated NDI-donor polymers: Exploration of donor size and electrostatic complementarity

Article abstract of DOI:10.1021/ma302340u

Conjugated donor-acceptor copolymers comprised of electron-deficient 1,4,5,8-naphthalenetetracarboxylic diimide (NDI) linked to a series of relatively electron-rich aromatics via ethynyl spacers were synthesized and characterized. While LUMO levels remained constant at -3.75 eV, HOMO levels were sensitive to the relatively electron-rich aromatic donors and systematically tuned from -5.68 to -5.17 eV. Regardless of the electron-rich comonomer, fluorescence and X-ray diffraction data were consistent with the polymer chains being assembled through the stacking of NDI moieties in an offset face-to-face fashion rather than alternating donor-acceptor stacks.

Full text of DOI:10.1021/ma302340u

Article
pubs.acs.org/Macromolecules
Conjugated NDI−Donor Polymers: Exploration of Donor Size and
Electrostatic Complementarity
Paul M. Alvey, Robert J. Ono, Christopher W. Bielawski, and Brent L. Iverson*
Department of Chemistry and Biochemistry, The University of Texas at Austin, Austin, Texas 78712, United States
S
* Supporting Information
ABSTRACT: Conjugated donor−acceptor copolymers com-
prised of electron-deficient 1,4,5,8-naphthalenetetracarboxylic
diimide (NDI) linked to a series of relatively electron-rich
aromatics via ethynyl spacers were synthesized and charac-
terized. While LUMO levels remained constant at −3.75 eV,
HOMO levels were sensitive to the relatively electron-rich
aromatic donors and systematically tuned from −5.68 to −5.17
eV. Regardless of the electron-rich comonomer, fluorescence
and X-ray diffraction data were consistent with the polymer chains being assembled through the stacking of NDI moieties in an
offset face-to-face fashion rather than alternating donor−acceptor stacks.
Previously, our group has exploited the complementary size
and electrostatics of the relatively electron-deficient NDI with
electron-rich 1,5-dialkoxynaphthalene (DAN) to affect the
macromolecular assembly of aqueous foldamers,14 self-assem-
bling polymers,15 and tunable liquid crystals.16 Although DAN
polymers have been reported,17 to the best of our knowledge
there has been no report of a conjugated, alternating polymer
incorporating both DAN and NDI. We reasoned that the
complementary nature of the DAN and NDI donor−acceptor
units could facilitate interchain assembly, thereby leading to
useful bulk properties. Alternatively, there are now theoretical
arguments asserting that electrostatic complementarity between
highly polarized groups on the periphery of aromatics, rather
than overall aromatic electron density, can provide the
dominant interactions between stacked aromatics.18 Such
considerations and a number of recent experimental examples
suggest that NDI units themselves may dominate the interchain
assembly of NDI-containing polymers, favoring an offset face-
centered stacking mode in which the strong NDI carbonyl
dipole moments complement each other from one unit to the
next.19 Herein, we report the synthesis and characterization of a
series of conjugated alternating NDI-based aromatic D−A
polymers linked by single alkynyl units and present evidence
that NDI self-stacking dominates the intermolecular polymer
chain assembly.
INTRODUCTION
■
The unique electronic properties of conjugated polymers have
rendered them desirable materials for applications that range
from photovoltaic (PV) cells1 to field effect transistors2 and
electrochromic devices.3 Conjugation along the polymer
backbone allows for extended π-delocalization that can result
in a relatively high HOMO level, a relatively low LUMO level,
or a small HOMO−LUMO band gap. As a result, certain
electronic characteristics are attainable such as p-type or n-type
behaviors, high charge carrier mobilities, and, for photovoltaic
purposes, a broad visible absorption spectrum.1−3
As a practical material, semiconducting polymers should have
tunable electronic and photovoltaic properties. One method to
fine-tune the electronic properties of these materials has been
to synthesize conjugated aromatic donor (electron rich)−
acceptor (electron poor) (D−A) polymers.1,4 Energy levels can
be tuned through careful selections of monomer units, as D−A
polymer HOMO−LUMO levels are well-approximated by the
individual monomeric donor HOMO and acceptor LUMO
orbitals.5 Hence, progress in this area has focused on
developing new monomers and implementing appropriate
combinations of these donor and acceptor moieties.6
1,4,5,8-Naphthalenetetracarboxylic diimide (NDI) has
emerged as a popular acceptor moiety due to its electron-
deficient aromatic core, high electron carrier mobility,7 and low-
lying LUMO.8 Recent synthetic advances have not only allowed
for substitution through the imide positions, but direct
substitution on the NDI aromatic core is now readily
accessible.9 In addition, solubility limitations have been
addressed and the construction of large polymers with NDI
is now possible.7,10 Numerous donor monomers have been
copolymerized with NDI in a conjugated fashion to alter
polymer electronic and photovoltaic properties. These groups
have included thiophene-based,7,10,11 thiazole-based,12 and
fluorene-based13 donors.
EXPERIMENTAL SECTION
■
Reagents. 1,4-Bis(ethynyl)benzene (1) was purchased from
Aldrich, and 2,6-dibromonaphthalene (3) was purchased from Alfa
Aesar. Pd(PPh3)2Cl2, Pd(Ph3)4, and CuI were purchased from
STREM. Dry toluene and TEA were obtained by distillation over
CaH. Dry NMP was purchased from Fisher Scientific and stored over
Received: November 13, 2012
Revised: January 18, 2013
Published: February 4, 2013
© 2013 American Chemical Society
718
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726

Macromolecules
Article
molecular sieves prior to use. All other chemicals were acquired from
Aldrich and used without further purification. All reactions were
carried out under argon.
C16H12O2, 236.0837; found, 236.0834. IR: 739, 831, 858, 963, 1041,
1174, 1203, 1229, 1332, 1376, 1449, 1484, 1590, 3284 cm−1.
9,10-Diethynylanthracene (10). Monomer 10 was synthesized in
the same manner as that given for 4, but with 9 as starting material
with some modification explained below. Column 1: 100% Hex
(further purified by recrystallization from ethanol to afford
disubstituted material); purified product was stored as the silyl-
protected intermediate. Because of the instability of 10, this material was
deprotected and immediately stannylated and polymerized. The material
was protected from light during deprotection, and after a reaction
period of 1 h, the reaction mixture was quickly passed through a silica
plug. The solvent was removed by rotary evaporation, and 10 was
collected as a yellow solid (0.1571 g, 0.69 mmol, 61% yield) and
immediately used in the next step. 13C NMR (400 MHz, CDCl3) δ:
132.90, 127.49, 127.46, 118.23, 90.34, 80.64 ppm. 1H NMR (400
MHz, CDCl3) δ: 8.61 (m, 4H), 7.62 (m, 4H), 4.07 (s, 2H) ppm. IR:
663, 776, 859, 980, 1029, 1170, 1167, 1222, 1370, 1435, 1623, 3280
cm−1.
Polymer P2. A solution of 1 (0.0216 g, 0.17 mmol) and n-BuLi
(0.43 mL of 1.6 M in Hex, 0.69 mmol) was allowed to stir in 2 mL of
dry THF at −78 °C for 0.5 h. To this solution was added tributyltin
iodide (0.19 mL, 0.66 mmol), and the mixture temperature was
allowed to reach room temperature while stirring for 2 h. The solution
was poured over water and extracted with DCM. The DCM extracts
were combined, dried over NaSO4, and the solvent was removed by
rotary evaporation. Proton NMR showed satisfactory conversion to 2,
and this material was dissolved with 12 (0.1675 g, 0.17 mmol) in 2 mL
of dry toluene. The mixture was subjected to three freeze, pump, and
thaw cycles. Pd(PPh3)4 (0.0100 g, 5 mol %) was added, and the
solution was heated at 90 °C. The polymerization progress was
monitored by UV−vis spectroscopy. Once the solution absorbance
bathochromically shifted to its absorbance maximum, iodobenzene
(0.19 mL, 1.7 mmol) was added to the solution, and the reaction
proceeded for an additional 2 h. The toluene was concentrated to a
minimal volume by rotary evaporation, and the crude material was
precipitated with MeOH/2N aqueous HCl (10/1 (v/v)). The solid
was purified with a Soxhlet extraction apparatus using methanol,
acetone, and DCM in succession. The polymers were collected from
the DCM fraction and concentrated to a minimum volume by rotary
evaporation. The polymers were precipitated two more times in
MeOH/2N aqueous HCl (10/1 (v/v)) to yield a bright red solid
(0.1037 g, 0.11 mmol, 64% yield). IR: 724, 767, 793, 839, 930, 1020,
1199, 1226, 1316, 1393, 1395, 1447, 1513, 1578, 1666, 1708, 2206,
2856, 2927, 2960 cm−1. GPC: Mn 18.7, Mw 60.9, PDI 3.3.
Instrumentation. NMR spectra were taken on a Varian Unity 400
spectrometer. Melting points were detected using a MEL-TEMP
apparatus. GPC analyses were performed on polymer solutions in
THF using a Waters Model 510 HPLC pump, two fluorinated
polystyrene columns (IMBHW-3078 and I-MBLMW-03078) arranged
in series, and a Waters 486 tunable absorbance detector (λ = 450 nm).
Calibration was based on polystyrene standards in THF. Absorption
spectra were obtained on an Agilent 8453 UV−vis spectrometer.
Fluorescence measurements were made on a PTI fluorimeter (4 nm
slits) with an 814 photomultiplier detection system using a 75 W
xenon short arc lamp. IR spectra were obtained using polymer solids
on a PerkinElmer Spectrum 100 FT-IR equipped with a universal ATR
(UATR) accessory. Electrochemical cyclic voltammetry was performed
under a nitrogen atmosphere. The cell was equipped with platinum
working, tungsten counter, and silver electrodes. Thin films were
measured in a 0.1 M tetrabutylammonium hexafluorophosphate
(TBAP) MeCN solution at a scan rate of 50 mV s−1 and referenced
to Fc/Fc+ by shifting (Fc*)0/+ to 0.0 V.20 X-ray powder diffraction
(XRD) patterns were obtained with a Scintag X1 theta−theta
diffractometer equipped with a Cu X-ray tube and a solid-state X-ray
detector set to count Cu Kα radiation. Samples were prepared by
smearing a small amount of polymer onto a zero background quartz
plate sample holder. Suitable samples were obtained by methanol
precipitation or slow evaporation from CH2Cl2 to yield a powder and
film, respectively. Both methods gave the same XRD pattern.
2,6-Dibromo-1,5-bis(methoxy)naphthalene (6).21 To a sol-
ution of 2,6-dibromonaphthalene-1,5-diol22 (5.90 g, 19.0 mmol) in
dry, degassed NMP (85 mL) at 0 °C was added NaH (60% mineral oil
dispersion, 1.70 g, 42.0 mmol). After allowing the solution to stir at 0
°C for 5 min, MeI (6.20 g, 44.0 mmol) was added and the solution
warmed to room temperature while stirring overnight. The solution
was poured onto ice (400 g), extracted with diethyl ether, and the
organic fractions were poured through a short neutral alumina plug.
The solvent was removed by rotary evaporation and the crude material
was purified by recrystallization from acetone to yield a light brown
solid (5.24 g, 15.0 mmol, 79% yield); mp 149−156 °C. 13C NMR (400
MHz, CDCl3) δ: 153.68, 131.32, 129.87, 119.68, 113.83, 61.76 ppm.
1H NMR (400 MHz, CDCl3) δ: 7.78 (d, J = 8.0 Hz, 2H), 7.63 (d, J =
8.0 Hz, 2H), 3.99 (s, 6H) ppm. CI-HRMS (positive ion) calculated for
C12H10Br2O2, 343.9048; found, 343.9048.
2,6-Diethynylnaphthalene (4). 3 (0.3258 g, 1.14 mmol) was
dissolved in 11 mL of a 50/50 toluene/TEA solution. The solution
was degassed with argon. Pd(PPh3)2Cl2 (0.1145 g, 15 mol %), CuI
(0.0105 g, 5 mol %), and (tert-butyldimethylsilyl)acetylene (0.3229 g,
2.30 mmol) were subsequently added, and the solution was heated at
reflux for 24 h. The solvent was removed by rotary evaporation, and
the material was passed through a short silica column (column 1: Hex)
to yield a mixture of the disubstituted and monosubstituted protected
alkyne intermediates. This mixture was taken up in 11 mL of THF
with TBAF (2.3 mL of 1.0 M in THF, 2.3 mmol) and stirred at room
temperature for 1 h. The solvent was removed by rotary evaporation,
and the crude material was purified by column chromatography
(column 2: Hex) to afford the desired product as a white solid (0.1508
g, 0.86 mmol, 75% yield); mp 144−148 °C. 13C NMR (400 MHz,
CDCl3) δ: 132.37, 132.03, 129.33, 127.86, 120.45, 83.67, 78.23 ppm.
1H NMR (400 MHz, CDCl3) δ: 7.99 (s, 2H), 7.74 (d, J = 12.0 Hz,
2H), 7.54 (dd, J = 1.3 Hz, J = 8.0 Hz, 2H), 3.18 (s, 2H) ppm. CI-
HRMS (positive ion) calculated for C14H8, 176.0626; found, 176.0626.
IR: 678, 706, 819, 884, 1255, 1364, 1494, 1597, 3268 cm−1.
2,6-Diethynyl-1,5-bis(methoxy)naphthalene (7). Monomer 7
was synthesized in the same manner as that given for 4, but with 6 as
starting material. Column 1: 5% DCM/Hex; column 2: 30% DCM/
Hex; collected as an off-white solid (0.1936 g, 0.82 mmol, 72% yield);
mp 142−147 °C. 13C NMR (400 MHz, CDCl3) δ: 159.04, 130.27,
129.14, 117.85, 111.77, 83.18, 80.31, 61.94 ppm. 1H NMR (400 MHz,
CDCl3) δ: 7.86 (d, J = 8.0 Hz, 2H), 7.49 (d, J = 8.0 Hz, 2H), 4.13 (s,
6H), 3.45 (s, 2H) ppm. CI-HRMS (positive ion) calculated for
Polymer P5. P5 was synthesized using the same procedure and
scale as that given for P2, but with 4 and 12 as starting materials. P5
was collected as a red solid (0.1376 g, 0.14 mmol, 82% yield). IR: 693,
725, 776, 836, 931, 1201, 1222, 1316, 1382, 1447, 1514, 1577, 1666,
1709, 2207, 2855, 2925, 2963 cm−1. GPC: Mn 8.2, Mw 23.0, PDI 2.8.
Polymer P8. P8 was synthesized using the same procedure and
scale as that given for P2, but with 7 and 12 as starting materials. P8
was collected as a purple solid (0.1619 g, 0.15 mmol, 88% yield). IR:
693, 725, 775, 822, 926, 1062, 1200, 1222, 1314, 1352, 1381, 1451,
1573, 1666, 1709, 2191, 2856, 2926, 2960 cm−1. GPC: Mn 14.9, Mw
58.4, PDI 3.9.
Polymer P11. P11 was synthesized using the same procedure and
scale as that given for P2, but with 10 and 12 as starting materials. P11
was collected as a purple solid (0.0698 g, 0.07 mmol, 41% yield). IR:
694, 725, 775, 923, 1212, 1221, 1252, 1312, 1380, 1451, 1575, 1661,
1709, 2179, 2856, 2926, 2961 cm−1. GPC: Mn 3.3, Mw 10.0, PDI 3.0.
RESULTS
■
Design. DAN, benzene, naphthalene, and anthracene were
selected to explore how the physical size and donating
properties influence the properties of conjugated NDI−donor
polymers. DAN and NDI are known to stack in an alternating
fashion14a,16a due to electrostatic complementarity that drives
desolvation effects in polar media.23 Similar self-assembly has
also been observed in a 1:1 complex of an anthracene-based
719
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726

Macromolecules
Article
Figure 1. Electrostatic potential maps of the individual monomers (top) and the monomers connected by alkyne linkages (bottom). Models were
generated with the same electrostatic potential color scale in Spartan using the DFT B3LYP G-31* method.
host and NDI guest.24 On the contrary, benzene and
naphthalene are not expected to stack as well with NDI due
to a lesser degree of size and electrostatic complementarity,
respectively. Although the differences are small, electrostatic
potential maps (Figure 1) show that the area of electron density
on the aromatic cores follows the following trend: anthracene >
DAN > naphthalene > benzene. The electrostatic potential
maps also show that the oxygen atoms on DAN strongly
contribute to the increased electron density on the aromatic
core when compared to naphthalene which lacks these oxygen
atoms. In general, acene-based molecules form polycrystalline
assemblies in which the aromatics exhibit some sort of stacked
geometry facilitated by the rigid and planar acene structure.
Several such derivatives have found use in devices such as
OFETs.25 Note the strong local electron density on the DAN
oxygen atoms. It is likely that these areas of localized high
electron density on the DAN periphery can interact with the
relative lack of electron density around the carbon atoms of the
NDI carbonyl groups, an interaction that is not possible with
the other donors.
the mode of conjugation in order to alleviate steric hindrance
that would otherwise prevent coplanarity of aromatic donor
and acceptor units. Thus, we anticipate that all of the aromatic
units in the polymers P2, P5, P8, and P11 (Scheme 1) can
adopt a highly conjugated and coplanar conformation. The
planar structure is expected to facilitate interchain assembly
through some type of stacked arrangement. The minimized
structures in Figure 1 indicate that, consistent with our design
criteria, the donor moieties should be able to adopt a relatively
planar arrangement when connected to two NDI moieties
through alkynyl linkages. The alkynyl linkages and subsequent
conjugation are expected to result in strong delocalization and
diminished electrostatic separation between the donor moieties
and NDI, although some level of donor and acceptor properties
should remain with the acene and NDI units, respectively.
Synthesis and Characterization. NDI monomer 1210b,12
(Scheme 1) was prepared from dibrominated 1,4,5,8-
naphthalenetetracarboxylic dianhydride9b and 2-octyldodecyl-
amine26 using literature procedures. Synthesis of the donor
monomers 2, 5, 8, and 11 is shown in Scheme 2. Dibrominated
naphthalene derivative 6 was synthesized by the deprotonation
of 2,6-dibromonaphthalene-1,5-diol and its reaction with
Instead of directly connecting together the aromatic cores of
NDI and the donors, an alkynyl linkage was selected to serve as
720
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726

Macromolecules
Article
occurred. To prevent loss of the stannyl groups on the silica
column, 2, 5, 8, and 11 were used without purification prior to
polymerization.
Scheme 1. Copolymerization of NDI 12 with Donor
Monomers
D−A polymers (P2, P5, P8, and P11) were synthesized
through the copolymerization of dibrominated NDI 12 with
diethynyl donor monomers 2, 5, 8, and 11 using the Stille
protocol29 (Scheme 1). An advantage of this route is that it
permits precise control over polymer architecture. An
analogous Sonogashira approach was successfully utilized by
Guo and Watson to copolymerize substituted diethynyl
benzene monomers with pyromellitic diimide (PMDI).30
The polymerization progress was monitored by UV−vis
spectroscopy, and reactions were stopped upon reaching the
maximum bathochromic shift to provide solubility in common
organic solvents such as tetrahydrofuran, methylene chloride,
chloroform, and toluene. Excess iodobenzene was added to the
reaction mixture to cap the polymers. The reaction proceeded
another 2 h before the solvent was removed, and crude
polymers were collected by precipitation from methanol.
Purification was achieved by successive Soxhlet extractions
with methanol, acetone, and methylene chloride. The filtrate
collected from the extraction with methylene chloride was then
concentrated, and the polymers were isolated following two
more precipitations from methanol. Reactions resulted in
satisfactory yields with the exception of P11, which was isolated
in 41% yield. Even with the short reaction period, a significant
amount of insoluble material remained in the Soxhlet thimble.
Therefore, the reported yields only represent the readily soluble
fraction of each sample.
Number-average (Mn) and weight-average (Mw) molecular
weights were determined by gel permeation chromatography
(GPC). Mn values that ranged from 3.3 to 18.7 kDa were
observed with polydispersity indices (PDI) between 2.8 and
3.9. These values may be overestimated in analogy to previously
reported GPC data for other rigid-rod polymers.31
Scheme 2. General Synthesis of Monomers 2, 5, 8, and 11
1
The presence of aggregated polymers was supported by H
NMR spectroscopy, as resolved spectra of these polymers were
unattainable even at 130 °C. Similar characterization problems
were encountered with the conjugated PMDI−benzene
polymers.30 In light of these NMR characterization complica-
tions, infrared spectroscopy (IR) was used to structurally
characterize the polymers as shown in Figure 2 for P2, 12, and
1. The IR spectrum of 12 exhibits strong signals from 2828 to
methyl iodide. Naphthalene 3 was commercially available, and
9 was prepared by the direct bromination of anthracene
according to literature procedures.27
Next, (tert-butyldimethylsilyl)acetylene (TBDMSA) moieties
were appended onto the aromatic cores of 3, 6, and 9 using
standard Sonogashira conditions, followed by TBAF depro-
tection to afford terminal alkynes 4, 7, and 10 (Scheme 2).
These intermediates, along with commercially available 1, were
subsequently stannylated with tributyltin iodide to yield
suitable electron-rich coupling partners (2, 5, 8, and 11).
Because of its instability,28 10 was stannylated immediately after
deprotection and polymerized before significant decomposition
Figure 2. IR spectrum of 12 (green), 1 (pink), and P2 (black). Red
arrows indicate the presence of an internal alkyne signal at 2207 cm−1
and absence of a terminal alkyne signal at 3262 cm−1 for P2.
721
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726

Macromolecules
Article
3000 cm−1 (C−H stretching vibrations of the alkyl chains) and
signals at 1656 and 1710 cm−1 (carbonyl stretching vibrations)
while the spectrum of 1 shows a signal at 3262 cm−1 (C−H
stretching vibrations of the terminal alkyne). The IR spectrum
of P2 maintains both the alkyl C−H and carbonyl stretching
signals of 12 and lacks the C−H stretching vibration of a
terminal alkyne as seen in 1. The asymmetric, internal alkynes
of P2 result in a new stretching vibration at 2207 cm−1.
Optical and Electrochemical Properties. The absorption
spectra for all polymers in chloroform and as a thin film are
shown in Figure 3. The polymer absorption maximum (λmax
)
Figure 4. Normalized (top) and non-normalized (bottom) polymer
fluorescence spectra in chloroform (4 × 10−6 M) of P2 (black), P5
(red), P8 (purple), and P11 (teal). Concentration is based on the
molecular weight of the polymer repeat unit.
of the dramatic fluorescence decrease seen with P8 and P11
could be due to interchain assembly through intermolecular
donor−acceptor interactions. Alternatively, increased intra-
molecular charge transfer could be playing a significant role
in the fluorescence quenching of P8 and P11.
Electrochemical data were collected by thin film cyclic
voltammetry (CV) using a standard three-electrode cell with
platinum working, tungsten counter, and silver reference
electrodes. No oxidation peaks were observed within the
solvent operating window during the anodic sweep, but two
clear reduction peaks were observed for each polymer during
the cathodic sweep (Table 1), results that are characteristic of
n-type materials. Cyclic voltammograms can be found in the
Supporting Information.
LUMO energy levels (Table 1) were determined from the
E1/2 values of the first reduction potentials. As previously
reported,10a the E1/2 from the first reduction potentials are
around −1.0 V and resemble that of the parent NDI monomer8
(see Supporting Information). LUMO levels are relatively
consistent among the four polymers with values between −3.73
and −3.79 eV, reflecting the dominant NDI contribution to the
overall polymer LUMO energy level. A likely explanation is that
the LUMO is predominately localized on the NDI units, even
in the conjugated systems. The HOMO energy levels were
estimated from the calculated LUMO energies and the Egopt. As
expected, the HOMO energies exhibited a strong dependence
on the donor monomer used. P2 and P5, possessing
comparable visible absorbance onsets, have calculated
HOMO levels of −5.68 and −5.65 eV, respectively. The
acene units, which were expected to be stronger donor
monomers, exhibited an increase of the HOMO level; the
Figure 3. Normalized absorption spectra in chloroform (top) and as a
thin film (bottom) of P2 (black), P5 (red), P8 (purple), and P11
(teal). The inset picture of the polymers dissolved in chloroform
shows the bright red color of P2 and P5 and the purple and teal color
of P8 and P11, respectively.
and the corresponding optical band gap (Egopt) energies are
listed in Table 1. Absorption maxima in solution were 558, 534,
576, and 678 nm for P2, P5, P8, and P11, respectively, and
little change from these numbers was observed in the thin film
measurements. In the thin films, peak broadening was clearly
visible for P11, and P5 appears to develop a shoulder much like
the one present in P2. The maximum shift (blue or red)
observed between the solution and thin film absorption maxima
was only 7 nm, suggesting that the polymers adopt a similar
structure in both solution and solid state.
The normalized and non-normalized fluorescence spectra of
the polymers in chloroform are shown in Figure 4. There was a
dramatic decrease in fluorescence in P8 and P11, amounting to
approximately a 10-fold and 50-fold decrease of emission
intensity in chloroform, respectively, when compared to that of
P2. Two possible explanations could account for such a
dramatic decrease in fluorescence. Since DAN and anthracene
can stack with NDI, it is reasonable to assume that at least some
Table 1. A Summary of Optical and Electrochemical Properties
a
a
b
c
soln abs λmax (nm)
film abs λmax (nm)
soln flu λmax (nm)
Egopt (eV)
E1/2 red1 (V)
E1/2 red2 (V)
LUMO (eV)
HOMO (eV)
P2
558
534
576
678
561
527
583
673
646
579
677
791
1.89
1.90
1.82
1.44
−1.01
−1.05
−1.06
−1.07
−1.31
−1.34
−1.42
−1.31
−3.79
−3.75
−3.74
−3.73
−5.68
−5.65
−5.56
−5.17
P5
P8
P11
a
b
c
Values were estimated using the vacuum ferrocene reference value of −4.8 eV at 0.0 V. LUMO = −(E1/2 red1 + 4.8) eV. HOMO = LUMO −
Egopt
.
722
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726

Macromolecules
Article
DAN-containing P8 exhibited a HOMO of −5.56 eV, a 0.09 eV
increase over the HOMO of P5. Incorporation of anthracene
resulted in the highest HOMO energy level, as P11 exhibited a
HOMO energy of −5.17 eV.
Using the maximum thin film absorption edge for each
polymer, the Egopt values for P2, P5, and P8 were calculated to
be 1.89, 1.90, and 1.82 eV, respectively. The additional electron
donation from the oxygen atoms on the naphthalene rings in
P8 resulted in a minimum band gap change (0.08 eV) when
compared to P5, which lacks oxygen atoms on the naphthalene
rings. Replacing naphthalene with anthracene, however,
resulted in a significant lowering of the band gap; P11
exhibited the lowest band gap of 1.44 eV.
Structural and Organization Properties. The aggrega-
tion properties of the different polymers were investigated to
probe the nature of interchain interactions. In particular, the
dramatic loss of fluorescence seen with P8 and P11 could in
part be the result of aromatic D−A stacking interactions
between chains. Unfortunately, the strong absorbance
associated with intramolecular charge transfer deterred
detection of possible intermolecular donor−acceptor inter-
actions with UV−vis spectroscopy. Serial dilution fluorescence
was instead conducted to investigate the interchain stacking
behavior of the polymers (Figure 5). Although each polymer
XRD and Modeling. Bulk polymers collected from
methanol precipitation (slow evaporation from dichloro-
methane to form a film gave analogous results) were analyzed
by X-ray diffraction (XRD) to determine the actual stacking
mode of the polymer chains. Four peaks were consistent in all
of the XRD patterns (Figure 6). These peaks corresponded to
d-spacings of 23, 12−13, 4.6, and 3.5 Å. In addition, the
variation among peak intensities at 2θ ≈ 3.8° indicates that P8
and P11 possess a higher relative crystallinity than P2 and P5.
The d-spacing of 23 Å is representative of a lamellar structure
with interdigitated side chains11b as this distance is shorter than
the extended length of the NDI moiety (∼35 Å). A scale model
of P2 and the proposed packing mode (representative of P5,
P8, and P11 as well) is shown in Figure 7. The viewpoint of a
single layer (Figure 7B) along the Y−X axis shows the 23 Å
distance between two parallel polymer chains. The distance of
3.5 Å matches the characteristic distance between stacked
aromatic units seen in numerous XRD studies of NDI-based or
other aromatic systems.19a−c The polymer chains were then
separated by a distance of 3.5 Å along the Z-axis (Figure 7C) in
the scale model. Importantly, the observed d-spacing of 4.6 Å
closely resembles the reported centroid-to-centroid distance
between adjacent NDI moieties stacked in an offset face-to-face
fashion,19d and the polymer chains were adjusted as such in the
model. It was then possible to draw parallel planes along the Y-
axis that matched the 4.6 Å d-spacing (Figure 7C). In addition,
another set of parallel planes could be drawn along the Y-axis
that have a distance of ∼12.5 Å (Figure 7D) which resembles
the d-spacing of 13.3 Å. The difference may be due to the error
associated with a broad XRD peak. It also seems plausible that
the distance (12.5 Å) between these particular parallel planes
would be sensitive to the donor size and could explain the small
difference among the polymers at 2θ ≈ 7° (11.7, 12.2, 13.3, and
13.5 Å).
For the sake of completeness, two other models were
generated: one that set the centroid-to-centroid distance
between NDI and benzene at 4.6 Å and another one that
assumed a complete alternating donor−acceptor assembly
along the Z-axis. Neither of these models accounted for all of
the XRD peaks (see Supporting Information).
DISCUSSION
■
Four new conjugated aromatic D−A polymers incorporating
NDI as the acceptor moiety were synthesized and charac-
terized. A Stille coupling reaction between diethynyl aryl
monomers and dibromo NDI yielded high molecular weight
polymers that were readily soluble in organic solvents. As with
previously reported conjugated NDI polymers, the NDI−aryl
polymers reported herein also exhibited interesting electro-
chemical properties. LUMO levels remained around −3.75 eV
for all polymers examined, a higher value than that of the
commonly used PV acceptor PCBM (−4.3 eV), while the
HOMO levels were sensitive to the aromatic donor groups and
resembled those of commonly employed thiophene or
fluorene-based monomers (HOMO approximately −5.8 to
−4.8 eV). Most notably, polymer P11 possesses one of the
smallest conjugated NDI−donor polymer band gaps to date as
well as HOMO/LUMO levels that may be suitable for use in
PV devices.32
Figure 5. Normalized serial dilution fluorescence pattern of P2
(black), P5 (red), P8 (purple), and P11 (teal) in chloroform.
Concentration is based on the molecular weight of the polymer repeat
unit. The solutions were 2-fold diluted in serial fashion, and each point
represents the maximum fluorescence at that particular concentration.
exhibited different fluorescence intensities (Figure 4), their
serial dilution plots are presented as a normalized series to
compare the overall fluorescence behavior of the polymers at
various concentrations.
All of the polymers exhibited a similar dramatic decrease in
fluorescence at high concentrations in chloroform. The
fluorescence intensity for each polymer increased with
decreasing concentration, and a maximum signal was attained
in the range of 125−62 μM before the fluorescence intensity
again decreased with decreasing concentration.
All of the polymers experienced a dramatically quenched
fluorescence at high concentrations which is characteristic of an
aggregated material. In addition, polymers P8 and P11
displayed dramatically quenched fluorescence at all concen-
723
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726

Macromolecules
Article
Figure 6. XRD patterns of P2 (black), P5 (red), P8 (purple), and P11 (teal). Peaks are labeled with their corresponding d-spacing value, and the
inset plots represent an enlarged portion of the pattern.
Figure 7. Scale representation (A) of P2 and the proposed stacking (B−D) to account for the d-spacing values obtained by XRD. NDI = blue and
benzene = red. Side chains were removed for clarity (C, D) and marked by a black dot on the blue NDI.
trations examined compared to P2 and P5. NDI is known to
quench fluorescence when stacked with a fluorophore.33,19c
inference, the concentration-dependent fluorescence quench-
ing, is the result of interactions between the NDI units
common to all the polymers. To the extent that this is true, the
diminished fluorescence of P8 and P11 is most likely the result
of intramolecular charge transfer and not interchain aromatic
D−A interactions.
The XRD data were strikingly similar among all four
polymers and combined with models to further refine the mode
of interchain aggregation. The only model that was able to
explain the observed XRD data for all of the polymers
contained interdigitated side chains with the NDI moieties
stacked in an offset face-to-face fashion (Figure 7). Although
DAN and anthracene are known to form aromatic D−A
interactions with NDI, none of the aromatic donor units
investigated in this study were able to interrupt NDI self-
A
noticeable difference was expected in the fluorescence behavior
of P2, P5, P8, and P11 during the dilution experiments if
interchain assembly was driven by intermolecular D−A
interactions because these units, representing a range of
different sizes and donor capabilities, are expected to display
different levels of interaction with NDI. In contrast to this
expectation, the strong similarity among the concentration
dependencies of the fluorescence patterns in Figure 5 can most
easily be interpreted to indicate that all four polymers aggregate
in a similar fashion. In other words, aggregation is apparently
not dependent on the nature of the donor monomer, ruling out
D−A interactions as the primary driving force for aggregation.
A reasonable explanation is that interchain aggregation and, by
724
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726

Macromolecules
Article
stacking in these conjugated polymer chains. NDI is well-
known to have a propensity for self-association in an offset face-
centered stacking mode, presumably because such a stacking
geometry provides for maximum complementary electrostatic
interactions between the carbon and oxygen atoms of the
highly polarized NDI carbonyl groups on adjacent NDI units.
Taken together, our results are therefore consistent with more
recent discussions of aromatic stacking dominated by
interactions between highly polarized groups on the periphery
of aromatic units rather than overall polarization of the
aromatic ring itself (i.e., D−A interactions). Interestingly, the
XRD patterns additionally suggest that P8 and P11 have a
higher degree of crystallinity than P2 or P5, possibly the result
of larger acene donor moieties.
It is worth pointing out that the branched side chains, which
are necessary to solubilize the polymers, might also drive the
observed polymer assembly geometry as evidenced by an
interdigitated side chain arrangement. We have previously
shown that side chains can have a great impact on the solid-
state packing of donor−acceptor assemblies,16b and this
influence should not be ruled out.
REFERENCES
■
(1) Cheng, Y.-J.; Yang, S.-H.; Hsu, C.-S. Chem. Rev. 2009, 109, 5868.
(2) (a) Murphy, A. R.; Frechet, J. M. J. Chem. Rev. 2007, 107, 1066.
(b) Anthony, J. E.; Facchetti, A.; Heeney, M.; Marder, S. R.; Zhan, X.
́
Adv. Mater. 2010, 22, 3876.
(3) Beaujuge, P. M.; Reynolds, J. R. Chem. Rev. 2010, 110, 268.
(4) (a) Chen, J.; Cao, Y. Acc. Chem. Res. 2009, 42, 1709. (b) Havinga,
E. E.; Hoeve, W.; Wynberg, H. Polym. Bull. 1992, 29, 119.
(5) (a) Thompson, B. C.; Kim, Y.-G.; McCarley, T. D.; Reynolds, J.
R. J. Am. Chem. Soc. 2006, 128, 12714. (b) 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. 2007, 130, 732.
(6) (a) Abbotto, A.; Calderon, E. H.; Dangate, M. S.; De Angelis, F.;
Manfredi, N.; Mari, C. M.; Marinzi, C.; Mosconi, E.; Muccini, M.;
Ruffo, R.; Seri, M. Macromolecules 2010, 43, 9698. (b) Roncali, J.
Chem. Rev. 1997, 97, 173.
(7) (a) Yan, H.; Chen, Z.; Zheng, Y.; Newman, C.; Quinn, J. R.;
Dotz, F.; Kastler, M.; Facchetti, A. Nature 2009, 457, 679. (b) Katz, H.
E.; Lovinger, A. J.; Johnson, J.; Kloc, C.; Siegrist, T.; Li, W.; Lin, Y. Y.;
Dodabalapur, A. Nature 2000, 404, 478.
(8) Chopin, S.; Chaignon, F.; Blart, E.; Odobel, F. J. Mater. Chem.
2007, 17, 4139.
(9) (a) Chaignon, F.; Falkenstrom, M.; Karlsson, S.; Blart, E.;
Odobel, F.; Hammarstrom, L. Chem. Commun. 2007, 42, 64.
(b) Thalacker, C.; Roger, C.; Wurthner, F. J. Org. Chem. 2006, 71,
CONCLUSION
̈
̈
■
8098.
The electronic properties of conjugated D−A polymers
containing NDI have been thoroughly investigated over the
past few years. A common feature of these systems is the ability
to tune the HOMO/LUMO energy levels, making these
polymers popular candidates for electronic materials inves-
tigations. However, a thorough understanding of the polymer
chain organization is necessary to take these systems past the
investigation level and on to application. The ideal material will
exhibit both suitable electronic properties and predictable self-
assembly into well-organized architectures. The results reported
here have shed considerable light on the interchain stacking
behavior of a set of NDI D−A polymer systems and have
brought into sharp focus the apparent dominance of NDI−NDI
interactions. We are currently developing next-generation
polymers based on these findings.
(10) (a) Guo, X.; Watson, M. D. Org. Lett. 2008, 10, 5333. (b) Chen,
Z.; Zheng, Y.; Yan, H.; Facchetti, A. J. Am. Chem. Soc. 2008, 131, 8.
(11) (a) Chen, J.; Shi, M.-M.; Hu, X.-L.; Wang, M.; Chen, H.-Z.
Polymer 2010, 51, 2897. (b) Durban, M. M.; Kazarinoff, P. D.;
Luscombe, C. K. Macromolecules 2010, 43, 6348. (c) Durban, M. M.;
Kazarinoff, P. D.; Segawa, Y.; Luscombe, C. K. Macromolecules 2011,
44, 4721. (d) Mohamad, K. A.; Yousuke, K.; Uesugi, K.; Fukuda, H.
Jpn. J. Appl. Phys. 2011, 50, 091603. (e) Piyakulawat, P.; Keawprajak,
A.; Chindaduang, A.; Hanusch, M.; Asawapirom, U. Synth. Met. 2009,
159, 467. (f) Polander, L. E.; Tiwari, S. P.; Pandey, L.; Seifried, B. M.;
Zhang, Q.; Barlow, S.; Risko, C.; Bredas, J.-L.; Kippelen, B.; Marder, S.
́
R. Chem. Mater. 2011, 23, 3408. (g) Wei, Y.; Zhang, Q.; Jiang, Y.; Yu,
J. Macromol. Chem. Phys. 2009, 210, 769.
(12) Kudla, C. J.; Dolfen, D.; Schottler, K. J.; Koenen, J.-M.; Breusov,
D.; Allard, S.; Scherf, U. Macromolecules 2010, 43, 7864.
(13) Piyakulawat, P.; Keawprajak, A.; Wlosnewski, J.; Forster, M.;
Asawapirom, U. Synth. Met. 2011, 161, 1238.
(14) (a) Lokey, S. R.; Iverson, B. L. Nature 1995, 375, 303.
(b) Gabriel, G. J.; Sorey, S.; Iverson, B. L. J. Am. Chem. Soc. 2005, 127,
2637. (c) Bradford, V. J.; Iverson, B. L. J. Am. Chem. Soc. 2008, 130,
1517.
(15) Reczek, J. J.; Iverson, B. L. Macromolecules 2006, 39, 5601.
(16) (a) Reczek, J. J.; Villazor, K. R.; Lynch, V.; Swager, T. M.;
Iverson, B. L. J. Am. Chem. Soc. 2006, 128, 7995. (b) Alvey, P. M.;
Reczek, J. J.; Lynch, V.; Iverson, B. L. J. Org. Chem. 2010, 75, 7682.
(17) Mori, T.; Kijima, M. Opt. Mater. 2007, 30, 545.
(18) Wheeler, S. E. J. Am. Chem. Soc. 2011, 133, 10262.
(19) (a) Shao, H.; Gao, M.; Kim, S.-H.; Jaroniec, C. P.; Parquette, J.
R. Chem.Eur. J. 2011, 17, 12882. (b) Liu, K.; Yao, Y.; Wang, C.; Liu,
Y.; Li, Z.; Zhang, X. Chem.Eur. J. 2012, 18, 8622. (c) Shao, H.;
Parquette, J. R. Chem. Commun. 2010, 46, 4285. (d) Tomasulo, M.;
Naistat, D. M.; White, A. J. P.; Williams, D. J.; Raymo, F. M.
Tetrahedron Lett. 2005, 46, 5695. (e) Avinash, M. B.; Govindaraju, T.
Adv. Funct. Mater. 2011, 21, 3875.
ASSOCIATED CONTENT
■
S
* Supporting Information
NMR spectra, IR spectra, cyclic voltammograms, GPC traces,
and additional models. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail biverson@mail.utexas.edu.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
(20) Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C.
Adv. Mater. 2011, 23, 2367.
The authors declare no competing financial interest.
(21) Thomas, H.; Stuhr-Hansen, N.; Westerlund, F.; Laursen, B. W.;
Magnussen, M.; Sørensen, H. O.; Bjørnholm, T.; Christensen, J. B.
Tetrahedron Lett. 2009, 50, 7374.
(22) Wheeler, A. S.; Ergle, D. R. J. Am. Chem. Soc. 1930, 52, 4872.
(23) Martinez, C. R.; Iverson, B. L. Chem. Sci. 2012, 3, 2191.
(24) Jiang, W.; Han, M.; Zhang, H.-Y.; Zhang, Z.-J.; Liu, Y. Chem.
Eur. J. 2009, 15, 9938.
ACKNOWLEDGMENTS
■
This work was supported by the Robert A. Welch Foundation,
grant F1188 to B.L.I. We thank Eric Anslyn and his group for
use of their fluorometer. We also thank Steve Swinnea for
collecting the XRD patterns.
725
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726

Macromolecules
Article
(25) Kymissis, I. Organic Field Effect Transistors: Theory, Fabrication
and Characterization; Springer Science+Business Media, LLC: New
York, 2009.
(26) Letizia, J. A.; Salata, M. R.; Tribout, C. M.; Facchetti, A.; Ratner,
M. A.; Marks, T. J. J. Am. Chem. Soc. 2008, 130, 9679.
(27) Jones, S.; Atherton, J. C. C. Synth. Commun. 2001, 31, 1799.
(28) Zhao, Z.; Yu, S.; Xu, L.; Wang, H.; Lu, P. Tetrahedron 2007, 63,
7809.
(29) Wurthner, F.; Suraru, S.-L. Synthesis 2009, 2009, 1841.
̈
(30) Guo, X.; Watson, M. D. Macromolecules 2011, 44, 6711.
(31) Avila-Ortega, A.; Vaz
Polym. Chem. 2007, 45, 1993.
́
quez-Torres, H. J. Polym. Sci., Part A:
(32) Ahmed, E.; Ren, G.; Kim, F. S.; Hollenbeck, E. C.; Jenekhe, S. A.
Chem. Mater. 2011, 23, 4563.
(33) Alp, S.; Erten, S.; Karapire, C.; Koz, B.; Doroshenko, A. O.; Icli,
̈
S. J. Photochem. Photobiol., A 2000, 135, 103.
726
dx.doi.org/10.1021/ma302340u | Macromolecules 2013, 46, 718−726