Exciton coupling effects in polymeric cis-indolenine squaraine dyes

Article abstract of DOI:10.1021/cm301109u

Dicyanovinylene-substituted cis-indolenine squaraines were polymerized by a Ni-mediated Yamamoto homocoupling reaction. Preparative recycling GPC was used to obtain polymer fractions with different molecular weight distributions in order to investigate the influence of molecular weight on the optical properties. In this attempt, conjugated cyclic trimers could also be isolated that are, to the best of our knowledge, the first of their kind. These trimers and the squaraine polymers were investigated by cyclic voltammetry, absorption spectroscopy, and static and time-resolved fluorescence spectroscopy. While the trimers show multiple fluorescence bands with, for squaraines, an exceptional huge Stokes' shift, the polymers show broad absorption in the red to near infrared (NIR) region with narrow fluorescence and only the common minor Stokes' shift. The spectral features of the squaraines were interpreted to be caused by excitonic coupling. By comparison of experimental and computed (CNDO/S2 CIS) spectra detailed, structural models of the polymers were derived that consist of helix (H-aggregate behavior) and zigzag motifs (J-aggregate behavior).

Full text of DOI:10.1021/cm301109u

Article
pubs.acs.org/cm
Exciton Coupling Effects in Polymeric cis-Indolenine Squaraine Dyes
Sebastian F. Volker and Christoph Lambert*
̈
Institut fur Organische Chemie, Universitat Wurzburg, Wilhelm-Conrad-Rontgen Research Center for Complex Material Systems and
̈
̈
̈
̈
Center for Nanosystems Chemistry, Am Hubland, 97074 Wurzburg, Germany
̈
ABSTRACT: Dicyanovinylene-substituted cis-indolenine
squaraines were polymerized by a Ni-mediated Yamamoto
homocoupling reaction. Preparative recycling GPC was used to
obtain polymer fractions with different molecular weight
distributions in order to investigate the influence of molecular
weight on the optical properties. In this attempt, conjugated
cyclic trimers could also be isolated that are, to the best of our
knowledge, the first of their kind. These trimers and the
squaraine polymers were investigated by cyclic voltammetry,
absorption spectroscopy, and static and time-resolved
fluorescence spectroscopy. While the trimers show multiple
fluorescence bands with, for squaraines, an exceptional huge
Stokes’ shift, the polymers show broad absorption in the red to
near infrared (NIR) region with narrow fluorescence and only the common minor Stokes’ shift. The spectral features of the
squaraines were interpreted to be caused by excitonic coupling. By comparison of experimental and computed (CNDO/S2 CIS)
spectra detailed, structural models of the polymers were derived that consist of helix (H-aggregate behavior) and zigzag motifs (J-
aggregate behavior).
KEYWORDS: squaraine dye, polymer, exciton coupling, time-resolved fluorescence spectroscopy
like absorption (ε > 105 M−1 cm−1) and fluorescence in the red
to near-infrared (NIR) region. These unique absorption and
INTRODUCTION
The goal of this work is to synthesize polysquaraine dyes that
are based on dicyanovinylene indolenine squaraine monomers
and to study their optical and electrochemical properties.
■
fluorescence properties make squaraines suitable for a variety of
applications in nonlinear optics,1−7 data imaging,8 organic
photovoltaics,9−17 or as fluorescence labels18−21 and sensors for
Squaraines (Scheme 1) are intensely colored dyes that are
anions,22 cations,23−27 and neutral molecules.28,29 There are
generated by 1,3-dicondensation reactions of nucleophilic
several highly recommended reviews that give more-detailed
compounds such as heterocycles or other electron-rich
aromatics with squaric acid. This results in a donor−
acceptor−donor structure with a strong and narrow cyanine-
insight into this remarkable class of dye.30−33
The absorption maximum of squaraine dyes can be
influenced by the nucleophilic component (e.g., anilines,
pyrroles, indolenines and its heteroatom analogues), by
replacing one oxygen of squaric carbonyl group by, e.g.,
Scheme 1. Different Types of Squaraine Dyes
sulfur34,35 or methylene active moieties, such as dicyano-
methylene9,34,36 or barbituric acid,34 and finally by adding
electron-donating substituents to the nucleophilic (hetero)cycle
component. We recently investigated the latter aspect, using
triarylamine substituents attached to a series of indolenine
squaraine dyes.37 These triarylamine−squaraine conjugates
display a bathochromic shift of the absorption maximum and
versatile redox behavior. A totally different way to shift the
absorption of squaraine dyes is to synthesize oligomers or
polymers that then show band shifts and broadenings due to
excitonic coupling of localized squaraine states. In this context,
we recently prepared polymers based on indolenine squaraine
dyes, and their optical and redox properties were inves-
Received: April 10, 2012
Revised: June 6, 2012
Published: June 6, 2012
© 2012 American Chemical Society
2541
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
Scheme 2. Yamamoto Polycondensation of trans-Dibromoindolenine Squaraine to Give a Polysquaraine
tigated.1,11 These polymers indeed show a bathochromic shift
of the absorption maximum but also a weaker band at higher
energy. This observation can be explained by a mixture of H-
and J-type alignment of chromophores, which then display
hypsochromic (H) and bathochromic (J) shifts.
moieties cannot always completely be ruled out. Another
synthetic approach to polysquaraines is to use metal-catalyzed
or metal-promoted C−C cross coupling reactions of squaraine
dye monomers in the final polycondensation step. Here, the
formation of unwanted 1,2-dicondensation products in the
polymer strand can be excluded by thorough purification of the
squaraine precursor monomer. This approach was successfully
used when dibrominated 1,4-dialkoxydivinylbenzene−pyrrole−
squaraine was coupled with either N-alkyl-2,7-diethynylcarba-
zole, dialkoxy-diethynylbenzene or 2,5-diethynyl-thiophene
derivatives using the Sonogashira cross coupling reaction.55,56
However, a major drawback of the Sonogashira reaction is the
risk of homocouplings of the terminal alkyne compounds.
In our recent work,11 we synthesized a squaraine
homopolymer without any bridges using a Ni-promoted
Yamamoto coupling from monomeric dibromoindolenine
squaraine (Scheme 2). In the present work, we use a cis-
indolenine squaraine in which one central oxygen atom of the
squaric ring is replaced by a dicyanovinylene moiety (see
Scheme 1). This results in a stronger central acceptor moiety
and induces a bathochromic shifted absorption as outlined
previously. Even more important, because of steric reasons, a cis
conformation, instead of the usually found trans conformation,
of the monomeric indolenine squaraine is induced.9,34,36,37,57
The goal of the present work is to use the highly bent cis isomer
for the Yamamoto polycondensation reaction, which should
lead to a pronounced zigzag structure or even helical structures
of the resulting polymer. Because of this zigzag motif, we expect
strongly modified optical properties, compared to the more
elongated structure of trans-indolenine squaraine polymers. In
order to modify solubility, we prepared two different squaraine
monomers: one with hexadecyl chains attached to the
indolenine nitrogen and one with 3,7-dimethyloctyl chains.
Concerning the synthesis of polysquaraines, there have been
various approaches since Treibs and Jacob discovered the
formation of insoluble colored compounds upon the reaction of
pyrrole derivatives with squaric acid.38−40 Low solubility and
the formation of 1,2-dicondensation products were the issues
that arose from early attempts of making carbazole−squaraine
polymers using either polyphosphoric acid or dimethylsulfoxide
(DMSO) as solvent.41 Later, it was Havinga et al. who
suggested that, by alternation of conjugated electron-donor and
electron-acceptor moieties, small-band-gap polymers could be
obtained.42−44 This idea was also supported by theoretical
considerations of Brocks et al., who proposed low band gaps in
polysquaraines.45 Havinga et al. proved this concept, using
benzo(1,2-4,5)-di(1-alkyl-2-methylene-3,3-dimethylpyrroline),
benzo(1,2−4,5)-di(3-alkyl-2-methylenethiazole), or terthio-
phene units as donors in the reaction with squaric acid to
yield polymers with band gaps of ∼1 eV.42−44 However,
especially in the field of pyrrole-derived squaraine dyes, there is
an ongoing effort to decrease the band gap, using variously
substituted pyrrole derivatives.46−48 Those studies have further
been extended, using electron-donating bridging units such as
2,5-divinylthiophene between the squaraine dyes, thus forming
copolymers.49 Even more popular is the use of 1,4-
dialkoxydivinylbenzene bridging units. Low-band-gap polymers
(∼1 eV) with a strong and broad absorption in the NIR region
near ∼1000 nm were obtained, showing intrinsic conductivities
of ∼10−5 S cm−1.50,51 Furthermore, dialkyl-substituted fluorene
was used as the bridge between pyrrole-squaraine, resulting in
polymers with high polydispersity and broad absorption.52
Instead of an electron donor as the bridge between squaraine
dyes, an electron acceptor (pyridopyrazine) was recently used,
giving a low-band-gap polymer that exhibits strong two-photon
absorption behaviors.3 Hecht et al. produced a series of
nonconjugated polysquaraines via a two-step one-pot synthesis
where isolated 2,4-dihydroxybenzene−squaraines are linked by
various amine bridges.53 These authors also attempted to
synthesize polysquaraines with different topologies, such as
helical, rodlike, or cyclic structures by polycondensation of
phenylenediamines with squaric acid. Unfortunately this
approach failed due to an exceeding low reactivity of the
intermediate semisquaraine.54 All those polymers mentioned so
far were obtained by polycondensation reaction of bifunctional
electron donors with squaric acid. This reaction usually is
performed in a solvent mixture of n-butanol and toluene (or
benzene) for almost no 1,2-dicondensation products are
formed in this mixture. However, traces of 1,2-dicondensation
EXPERIMENTAL SECTION
■
UV/vis/NIR. Ultraviolet−visible light−near-infrared (UV/vis/NIR)
spectra were measured in 1 cm quartz cuvettes, using a JASCO Model
V-670 spectrometer. The substances were dissolved in Uvasol solvents
from Merck, and the pure solvent was used as a reference. Absorption
spectra were recorded within a concentration range from 1 × 10−5
M
to 1 × 10−7 M.
Spectra of films of the polymers were prepared by spin-casting of a
concentrated dye solution (2.5 mg mL−1 at 2000 rpm) in the
appropriate solvent onto a flat glass substrate (20 mm × 20 mm × 1
mm). UV/vis/NIR spectra of these samples were then measured using
a Carey Model 5000 UV/vis/NIR spectrometer with an integrating
sphere (ϕ = 15 cm).
Fluorescence. Fluorescence measurements were performed with a
Photon Technology International (PTI) QM fluorescence spectrom-
eter with a cooled photomultiplier (Model R928P) or an InGaAs
detector. Standard 1-cm cuvettes were used and spectra were recorded
in Uvasol solvents from Merck after purging the samples for 10 min
with argon gas. As a fluorescence standard, oxazine 1 in ethanol (Φfl =
2542
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
0.11) was used and the following equation was applied to determine
the quantum yields:58
7.05−7.01 (BB′, 2H), 4.73−4.62 (m, 2H, NCH2), 2.99 (s, 3H, CH3),
2.29 (s, 3H, O3SC6H4CH3), 1.85−1.77 (m, 1H, CH), 1.66−1.58 (8H,
CH3, CH2), 1.54−1.46 (m, 1H, CH), 1.38−1.28 (m, 2H, CH2), 1.25−
1.17 (m, 2H, CH2), 1.17−1.09 (m, 2H, CH2), 1.04 (d, 3J = 6.0 Hz, 3H,
CH3), 0.86 (d, 3J = 6.5 Hz, 3H, CH3), 0.85 (d, 3J = 6.6 Hz, 3H, CH3);
13C NMR (150 MHz, CDCl3): δ [ppm] = 195.7 (quart), 143.7
⎛
⎜
⎝
⎞
⎟
⎠
I(ν)
I(ν)
̃
× ODRef × (nD20)2
× OD × (nD20)2Ref
Φ = Φ
f
f,Ref
̃
Ref
(quart), 141.8 (quart), 141.2 (quart), 138.8 (quart), 129.7 (tert), 129.2
(tert), 128.4 (tert), 125.9 (tert), 123.1 (tert), 115.0 (tert), 54.4
(quart), 47.4 (sec), 39.0 (sec), 36.9 (sec), 35.0 (sec), 31.0 (tert), 27.9
(tert), 24.6 (sec), 23.0 (prim), 22.9 (prim), 22.6 (prim), 22.5 (prim),
21.2 (prim), 19.5 (prim), 14.9 (prim); ESI-MS pos (high resolution):
[M-OTos−]+ calcd: 300.26858 m/z, found: 300.26840 m/z, Δ = 0.58
ppm.
̃
where Φf is the quantum yield of the sample, I(ν) the integrated
emission band, OD the optical density of the absorption band at the
20
excitation wavelength, and nD the refraction index of the solvent.
Fluorescence Lifetimes. Fluorescence lifetimes were measured
with a PTI TM fluorescence lifetime spectrometer with a 650-nm
laser-diode for excitation. Colloidal silica in deionized water was used
as scatter solution to determine the instrument response. Lifetimes
were determined by fitting the decay curves with exponential decay
functions. Solvents and cuvette were used as in the steady-state
fluorescence experiments.
Cyclic Voltammetry. Cyclic voltammetry was performed for all
squaraine dyes using a BAS CV-50 W electrochemical workstation in
DCM and tetrabutylammonium hexafluorophosphate (TBAH ≈ 0.2
M) as supporting electrolyte. In our three electrode setup, a platinum
working electrode (1 mm in diameter), a helical platinum counter
electrode and a platinum pseudo-reference electrode were used in a
completely sealed and with argon flushed glass vessel. DCM was dried
over CaH2. Prior to use, the solvent was filtered over an activated
neutral aluminum oxide column under argon atmosphere for further
drying. All potentials were measured against the redox couple of
decamethylferrocene and its cation. The redox potential was then
calculated, relative to the ferrocene/ferrocenium (Fc/Fc+) redox
couple.59
NMR. Nuclear magnetic resonance (NMR) spectra were recorded
on either a Bruker Avance 400 or a Bruker Avance DMX 600 Fourier-
transform spectrometer. The chemical shifts are given in ppm, relative
to the internal standard tetramethylsilane (TMS). The coupling
constants are given in Hertz and the following abbreviations were used
for the spin multiplicities or for C-atom descriptions: s = singlet, d =
doublet, t = triplet, q = quartet, dd=doublet of doublet, m = multiplet;
prim=primary, sec=secondary, tert=tertiary, quart=quaternary.
Mass Spectrometry. Mass spectrometry was performed on a
Bruker Daltonik micrOTOF focus (ESI) system.
Gel Permeation Chromatography (GPC). Gel permeation
chromatography (GPC) was performed at 20 °C in chloroform with
a Shimadzu instrument (including a diode array detector (Model SPD-
M20A), system controller (Model CBM-20A), solvent delivery unit
(Model LC-20AD), and online degasser (Model DGU 20A9)).
Preparative chromatography was carried out in recycling mode on two
consecutive SDV columns (PSS SDV preparative 50 Å and 500 Å,
dimension: 20 × 600 mm, particle size: 10 μm) from PSS/Mainz,
Germany. Analytical chromatography was measured on a SDV column
(PSS SDV analytical linear S mixed bed, dimension: 8 × 300 mm,
particle size: 5 μm) from PSS/Mainz, Germany with polystyrene as a
standard.
5-Bromo-1-(3,7-dimethyloctyl)-2,3,3-trimethyl-3H-indol-1-
ium tosylate (3). 5-Bromo-2,3,3-trimethyl-3H-indole (1.01 g, 4.23
mmol) and 3,7-dimethyloctyl tosylate (1.78 g, 5.69 mmol) were
dissolved in MeNO2 under nitrogen atmosphere and refluxed for 17 h.
After cooling to room temperature, Et2O (40 mL) was added and the
product crystallized within 1 h. The precipitate was filtered off and
washed with Et2O to give a light brown solid (955 mg, 1.73 mmol,
41%). 1H NMR (600 MHz, CDCl3): δ [ppm] = 7.64 (dd, 3J = 8.5 Hz,
4J = 1.6 Hz, 1H, H-6), 7.62 (d, 4J = 1.6 Hz, 1H, H-4), 7.59−7.54 (AA′,
3
2H), 7.42 (d, J = 8.6 Hz, 1H, H-7), 7.04−7.00 (BB′, 2H), 4.70−4.55
(m, 2H, NCH2), 2.94 (s, 3H, CH3), 2.29 (s, 3H, O3SC6H4CH3),
1.81−1.74 (m, 1H, CH), 1.65−1.54 (8H, CH3, CH2), 1.54−1.45 (m,
1H, CH), 1.36−1.25 (m, 2H, CH2), 1.24−1.16 (m, 2H, CH2), 1.15−
3
3
1.07 (m, 2H, CH2), 1.01 (d, J = 6.2 Hz, 3H, CH3), 0.85 (d, J = 6.6
Hz, 6H, CH3); 13C NMR (150 MHz, CDCl3): δ [ppm] = 195.7
(quart), 143.6 (quart), 143.5 (quart), 140.3 (quart), 138.9 (quart),
132.5 (tert), 128.4 (tert), 126.6 (tert), 125.8 (tert), 124.1 (quart),
116.6 (tert), 54.5 (quart), 47.7 (sec), 39.0 (sec), 36.8 (sec), 34.9 (sec),
31.0 (tert), 27.9 (tert), 24.6 (sec), 22.90 (prim), 22.85 (prim), 22.6
(prim), 22.5 (prim), 21.3 (prim), 19.5 (prim), 15.0 (prim); ESI-MS
pos (high resolution): [M-OTos−]+ calcd: 378.17909 m/z, found:
378.17910 m/z, Δ = 0.03 ppm.
mSQ1. Compound 1 (2.10 g, 424 μmol), CN (64.5 mg, 202 μmol),
and pyridine (7.5 mL) were refluxed in a mixture of toluene and 1-
butanol (1:1, 16 mL) for 17 h. The solvent was evaporated under
reduced pressure and the residue purified by flash chromatography
(eluent: DCM) and dried in vacuo. The residue was crystallized from
hexane to give green crystals (100 mg, 138 μmol, 68%). 1H NMR (600
MHz, CDCl3): δ [ppm] = 7.38−7.32 (4H, CH), 7.22−7.18 (m, 2H,
CH), 7.03 (d, 3J = 7.9 Hz, 2H, CH), 6.49 (s, 2H, CH), 4.10−3.98 (m,
4H, NCH2), 1.82−1.74 (14H, CH3, CH), 1.70−1.58 (m, 4H, CH2),
1.55−1.48 (m, 2H, CH), 1.42−1.29 (m, 4H, CH2), 1.29−1.18 (m, 4H,
CH2), 1.18−1.11 (m, 4H, CH2), 1.03 (d, 3J = 6.5 Hz, 6H, CH3), 0.86
(d, 3J = 6.6 Hz, 12H, CH3); 13C NMR (150 MHz, CDCl3): δ [ppm] =
173.2 (quart), 171.7 (quart), 167.7 (quart), 166.6 (quart), 142.5
(quart), 141.9 (quart), 128.0 (tert), 124.5 (tert), 122.3 (tert), 118.9
(quart), 109.9 (tert), 89.0 (tert), 49.4 (quart), 42.9 (sec), 40.8 (quart),
39.1 (sec), 37.1 (sec), 34.1 (sec), 30.9 (tert), 28.0 (tert), 26.55 (prim),
26.53 (prim), 24.6 (sec), 22.7 (prim), 22.6 (prim), 19.7 (prim); ESI-
MS pos (high resolution): [M+] calcd: 724.50746 m/z, found:
724.50728 m/z, Δ = 0.25 ppm.
Synthetic Procedures. Synthetic procedures were performed in
standard glassware.
Materials. The chemicals used were purchased from commercial
suppliers and used without further purification. Reactions under
nitrogen (dried over Sicapent from Merck; oxygen was removed by
copper catalyst R3−11 from BASF) were performed in flame-dried
glass ware with solvents that were dried according to literature
procedures and stored under nitrogen. For flash chromatography,
Merck silica gel 32−63 μm was used. 5-Brom-2,3,3-trimethyl-3H-
indole,11 triethylammonium 2-butoxy-3-(dicyanomethylene)-4-oxocy-
clobut-1-enolate (CN),60 3,7-dimethyloctyl tosylate,61 2,11 and
mSQ237 were synthesized according to the given literature.
1-(3,7-Dimethyloctyl)-2,3,3-trimethyl-3H-indol-1-ium tosy-
late (1). 2,3,3-Trimethyl-3H-indole (2.00 g, 12.6 mmol) and 3,7-
dimethyloctyl tosylate (2.75 g, 8.79 mmol) were refluxed in MeNO2
for 17 h. The solvent and remaining indole were removed under
reduced pressure. The residue was digested with Et2O several times to
mSQ3. To a solution of 3 (1.12 g, 2.03 mmol) in pyridine (7.5
mL), CN (325 mg, 1.02 mmol) dissolved in a mixture of toluene and
1-butanol (1:1, 40 mL) was added. The mixture was refluxed for 17 h
using a Dean−Stark trap. The solvent was evaporated under reduced
pressure, and the residue was purified by flash chromatography
(eluent: DCM → DCM/EA 40:1 → 30:1 → 25:1) and dried in vacuo
1
to give a red violet shining solid (737 mg, 835 μmol, 82%). H NMR
4
(600 MHz, CD2Cl2): δ [ppm] = 7.50 (d, J = 1.9 Hz, 2H, H-4), 7.48
(dd, 3J = 8.3 Hz, 4J = 1.9 Hz, 2H, H-6), 6.96 (d, 3J = 8.4 Hz, 2H, H-7),
6.46 (s, 2H, CH), 4.06−3.93 (m, 4H, NCH2), 1.80−1.71 (14H, CH3,
CH), 1.67−1.55 (m, 4H, CH2), 1.55−1.47 (m, 2H, CH), 1.41−1.30
(m, 4H, CH2), 1.29−1.10 (8H, CH2), 1.01 (d, 3J = 6.5 Hz, 6H, CH3),
3
0.86 (d, J = 6.6 Hz, 12H, CH3); 13C NMR (150 MHz, CD2Cl2): δ
[ppm] = 173.2 (quart), 171.6 (quart), 167.9 (quart), 167.2 (quart),
144.8 (quart), 141.4 (quart), 131.2 (tert), 126.0 (tert), 118.9 (quart),
117.7 (quart), 111.8 (tert), 89.7 (tert), 49.8 (quart), 43.4 (sec), 40.9
1
obtain a wine red solid (2.76 g, 5.80 mmol, 66%). H NMR (600
MHz, CDCl3): δ [ppm] = 7.64−7.60 (AA′, 2H), 7.56−7.49 (4 H),
2543
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
Scheme 3. Synthetic Pathway to the Polysquaraines and the Cyclic Trimers
(quart), 39.4 (sec), 37.4 (sec), 34.2 (sec), 31.20 (tert), 28.3 (tert),
26.61 (prim), 26.57 (prim), 25.0 (sec), 22.8 (prim), 22.7 (prim), 19.7
(prim); ESI-MS pos (high resolution): [M+] calcd: 880.32849 m/z,
found: 880.32851 m/z, Δ = −0.02 ppm.
(tert), 118.9 (quart), 110.6 (tert), 89.7 (tert), 49.2 (quart), 44.6 (sec),
41.2 (quart), 31.9 (sec), 29.69 (sec), 29.684 (sec), 29.681 (sec), 29.66
(sec), 29.65 (sec), 29.62 (sec), 29.57 (sec), 29.45 (sec), 29.42 (sec),
29.3 (sec), 27.5 (sec), 26.8 (sec), 26.7 (prim), 22.7 (sec), 14.1 (prim);
ESI-MS pos (high resolution): [M+2] calcd: 1336.01970 m/z, found:
1336.01980 m/z, Δ = 0.08 ppm.
pSQ2. A mixture of Ni(COD)2 (152 mg, 571 μmol), 2,2′-bipyridine
(89.0 mg, 571 μmol), 1,5-cyclooctadiene (62.0 mg, 571 μmol),
degassed toluene (1 mL) and degassed DMF (2 mL) was stirred at
room temperature (rt) under a nitrogen atmosphere for 30 min. A
solution of mSQ2 (250 mg, 238 μmol) in degassed toluene (4.5 mL)
and degassed DMF (3.5 mL) was added and the resulting mixture was
stirred at rt for 6 d. The reaction mixture was poured into MeOH/HCl
(20%) (4:1) (500 mL) and stirred for 90 min. The purple precipitate
was filtered off and washed consecutively with hexane and MeOH
using a Soxhlet extractor. Each washing ran overnight. The remaining
solid was dissolved in CHCl3 (30 mL) and poured into MeOH/HCl
(20%) (4:1) (500 mL). The resulting precipitate was filtered off and
dried under high vacuum. The product was obtained as purple solid
(167 mg, 78%). Preparative GPC was used to separate the product
into two batches of different molecular weight and to isolate the cyclic
trimer:
pSQ3. A mixture of Ni(COD)2 (224 mg, 816 μmol), 2,2′-bipyridine
(127 mg, 816 μmol), 1,5-cyclooctadiene (88.0 mg, 816 μmol),
degassed toluene (2 mL), and degassed DMF (2 mL) was stirred at 65
°C under nitrogen atmosphere for 30 min. A solution of mSQ3 (300
mg, 340 μmol) in degassed toluene (4 mL) and degassed DMF (4
mL) was added and the resulting mixture was stirred at 65 °C for 6 d.
The reaction mixture was poured into MeOH/HCl (20%) (4:1) (500
mL) and stirred for 5 h. The purple precipitate was filtered off and
washed consecutively with hexane, MeOH and acetone using a Soxhlet
extractor. Each washing ran overnight. The major part remained in the
acetone fraction (192 mg, 77%) with very little solid that remained
insoluble in acetone. Preparative GPC was used to separate the
acetone fraction into two batches of different molecular weight and to
isolate the cyclic trimer: pSQ3-1: 58 mg, pSQ3-2: 53 mg, tSQ3: 14
mg (6.80 μmol, 6%). pSQ3-1, pSQ3-2: 1H NMR (400 MHz, CDCl3):
δ [ppm] = 7.64−7.44 (CH), 7.18−7.04 (CH), 6.62−6.44 (CH), 4.23−
3.87 (CH2), 1.92−1.58 (CH, CH2, CH3), 1.58−1.46 (CH), 1.46−1.10
pSQ2-1: 61 mg, pSQ2-2: 40 mg, tSQ2: 13 mg (4.86 μmol, 6%).
1
pSQ2-1, pSQ2-2: H NMR (400 MHz, CDCl3): δ [ppm] = 7.66−
7.44 (CH), 7.20−7.06 (CH), 6.66−6.46 (CH), 4.20−3.80 (CH2),
1.92−1.74 (CH3), 1.54−1.42 (CH2), 1.42−1.33 (CH2), 1.33−1.12
1
(CH2), 1.10−0.98 (CH3), 0.92−0.76 (CH3). tSQ3: H NMR (600
MHz, CDCl3): δ [ppm] = 7.74 (dd, 3J = 8.3 Hz, 4J = 1.6 Hz, 6 H, H-
6), 7.56 (d, 4J = 1.4 Hz, 6H, H-4), 7.15 (d, 3J = 8.3 Hz, 6H, H-7), 6.64
(s, 6H, CH), 4.17−4.04 (m, 12H, NCH2), 1.90−1.81 (42H, CH3,
CH), 1.74−1.63 (m, 12H, CH2), 1.56−1.50 (m, 6H, CH), 1.45−1.33
(12H, CH2), 1.32−1.22 (12H, CH2), 1.20−1.15 (12H, CH2), 1.07 (d,
3J = 6.3 Hz, 18H, CH3), 0.89−0.86 (36H, CH3); 13C NMR (150 MHz,
CDCl3): δ [ppm] = 174.3 (quart), 171.2 (quart), 167.8 (quart), 166.3
(quart), 143.8 (quart), 141.4 (quart), 135.9 (quart), 125.8 (tert), 119.7
1
(CH2), 0.90−0.81 (CH3). tSQ2: H NMR (600 MHz, CDCl3): δ
[ppm] = 7.75 (dd, 3J = 8.3 Hz, 4J = 1.4 Hz, 6H, H-6), 7.57 (d, 4J = 1.1
Hz, 6H, H-4), 7.17 (d, 3J = 8.3 Hz, 6H, H-7), 6.66 (s, 6H, CH), 4.12−
4.04 (m, 12H, NCH2), 1.93−1.85 (m, 12H, CH2), 1.83 (s, 36H, CH3),
1.54−1.45 (m, 12H, CH2), 1.42−1.34 (m, 12H, CH2), 1.34−1.19
(132H, CH2), 0.87 (t, 3J = 6.9 Hz, 18H, CH3); 13C NMR (150 MHz,
CDCl3): δ [ppm] = 173.0 (quart), 171.3 (quart), 167.9 (quart), 166.4
(quart), 143.8 (quart), 141.5 (quart), 135.8 (quart), 125.8 (tert), 119.7
2544
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
Figure 1. (Left) Chromatograms of three preparative recycling GPC runs of pSQ3 with separation into three fractions indicated by vertical dotted
lines. (Right) Analytical GPC run (detection at 400 nm) of the separated fractions with different molecular weight of pSQ3 in CHCl3.
(tert), 118.8 (quart), 110.4 (tert), 89.6 (tert), 49.3 (quart), 43.1 (sec),
41.2 (quart), 39.2 (sec), 37.2 (sec), 34.3 (sec), 31.0 (tert), 28.0 (tert),
26.70 (prim), 26.67 (prim), 24.6 (sec), 22.7 (prim), 22.6 (prim), 19.7
(prim); ESI-MS pos (high resolution): [M+2] calcd: 1083.73800 m/z
found: 1083.73987 m/z, Δ = 1.73 ppm.
the preparative splitting. The Mw values are 27 801 (pSQ2-1)
and 46 659 (pSQ3-1) for the heavier molecular weight
fractions and 8 264 (pSQ2-2) and 11 941 (pSQ3-2) for the
lighter ones (see Table 1). Because the optical properties of the
a
Table 1. Polymer Characterization
RESULTS
■
number-
average
molecular
weight, Mn
number-average
degree of
polymerization,
Xn
weight-
average
molecular
weight, Mw
Synthesis. The synthesis of polymeric squaraines followed a
straightforward route, analogous to the one recently published
by our group.11 5-Bromo-2,3,3-trimethyl-3H-indole, which was
synthesized via Fischer−Indole synthesis from p-bromophe-
nylhydrazine hydrochloride, according to literature procedures,
and commercially available 2,3,3-trimethyl-3H-indole were
alkylated with either 1-iodohexadecane or 3,7-dimethyloctyl
tosylate in MeNO2 to yield the quaternary salts 1−3. The
dicondensation reactions to the dicyanovinylene-substituted
squaraines mSQ1−mSQ3 were carried out with 1−3,
respectively, and triethylammonium 2-butoxy-3-(dicyano-
methylene)-4-oxocyclobut-1-enolate (CN) in a solvent mixture
of toluene and 1-butanol (1:1) and pyridine as a base at high
temperatures. The Ni-mediated Yamamoto homocoupling
reaction was chosen for the final polycondensation step of
the bromine substituted squaraines mSQ2 and mSQ3. This
reaction was performed under inert gas atmosphere with
Ni(COD)2, 1,5-cyclooctadiene, and 2,2′-bipyridine in a solvent
mixture of deaerated toluene and DMF (1:1) at either room
temperature (pSQ2) or 65 °C (pSQ3) for several days.
Various solvents were used for the Soxhlet extractions in
order to purify the polymers. The weight-average molecular
weights (Mw) of the total crude polycondensation products of
pSQ2 and pSQ3 are 13 609 and 27 258, the polydispersities
(PDI) are 2.23 and 3.29, respectively. Since the main part of the
polymers was soluble in acetone, this fraction was separated on
a preparative recycling GPC to yield batches of polymers of
different molecular weight distributions (Figure 1, left). This
allows investigation of the influence of molecular weight on the
optical properties in those systems. In a first run on the
preparative recycling GPC, the fractions pSQ3-1 (pSQ2-1) and
pSQ3-2 (pSQ2-2) were separated as indicated by the dotted
lines. In this process, an eye-catching peak was observed in the
chromatograms of both polycondensation reactions at a similar
retention time, which could be isolated in this process. This
peak refers to a cyclic trimer whose formation is facilitated by
the cis structure of the squaraine monomer was confirmed by
mass spectrometry and NMR spectra. These cyclic trimers
(tSQ3, tSQ2) could be isolated in 6% yield. The polymeric
fractions show relatively small polydispersities (1.2−1.8) due to
polydispersity,
Mw/Mn
pSQ2-
total
6103
6.8
13609
2.230
pSQ2-1
pSQ2-2
17543
6621
8284
19.7
7.4
27801
8264
1.585
1.248
3.290
pSQ3-
total
11.4
27258
pSQ3-1
pSQ3-2
25639
8567
35.5
11.8
46659
11941
1.820
1.394
a
Analytical GPC was performed in CHCl3 with polystyrene as a
standard. “pSQx-total” assigns polymer batches after Soxhlet
extractions but before preparative splitting.
hexadecyl- and dimethyloctyl-substituted squaraines do not
differ, we thus can view the four polymers as being a series with
an increasing number-average degree of polymerization (Xn):
7.4 (pSQ2-2), 11.8 (pSQ3-2), 19.7 (pSQ2-1), and 35.5
(pSQ3-1). The polydispersity (PDI) increases as the degree of
polymerization increases. All those polymers are quite soluble
in solvents such as CHCl3, DCM, toluene, 1,2-dichlorobenzene,
or chlorobenzene.
Cyclic Voltammetry. Electrochemical characterization of
the squaraines was done by cyclic voltammetry (CV) in DCM/
0.2 M TBAH of a monomer, a cyclic trimer and both the high-
and low-molecular-weight batches of one polymer. The
voltammograms are shown in Figure 2, and the redox potentials
are listed in Table 2. Since the alkyl side groups only affect the
solubility but not the redox potentials, we only give data of the
squaraines with dimethyloctyl chains and the monomer with
the hexadecyl chain. Each substance shows one reduction wave
and two oxidation waves. Compared to the monomer, the
waves of the reduction and the first oxidation are broader in
case of the cyclic trimer, whereas the second oxidation wave
remains remarkably sharp. However, in case of the polymers,
this wave also broadens slightly. The first oxidation potentials
of the cyclic trimer and the polymers are at 30−40 mV and the
reduction potentials at 50−60 mV lower than those of the
monomer (207 mV and −1405 mV, respectively) which can be
2545
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
In toluene, the polymers display a very broad absorption in
the red-to-NIR region, which covers all the absorption of the
trimer (12 000−17 500 cm−1) but with three almost equally
intense maxima at ∼12 900 cm−1, 13 600 cm−1, and 14 500
cm−1. The spectra of the polymers are almost identical,
irrespective of the molecular weight and alkyl chain substituent.
In particular, the transition moments as determined from
integration (see eq 1) of the main absorption bands in toluene
ranges for all compounds between 8.9 (pSQ3-1, pSQ3-2), and
10.7 D (tSQ3).62
⎡
⎢
⎤
⎥
3hcε0 ln 10
2000π N ⎣(n + 2) ⎦
9n
ε
ν
μ2 =
dν
̃
eg
2
2
2
∫
̃
(1)
Figure 2. Cyclic voltammograms of a monomer, a trimer, and two
polymers of different molecular weight in DCM/TBAH (∼0.2 M) at a
scan rate of 250 mV s−1.
Since we noticed a distinct difference of the proportions of
the local maxima in the absorption spectra of the polymers in
toluene and DCM, we measured the absorption of pSQ3-1 in a
series of solvents that differ in polarity, aromatic character, and
hydrogen bond donor property (see Figure 4). When we first
compare the spectra in CHCl3, DCM, and toluene, the local
maxima are at the same energy but the bands at higher energies
increase in relative intensity from CHCl3 to toluene (Figure 4,
left). In PhCN, the three local maxima have almost the same
intensity but a weak shoulder, which is due to a fourth band at
even higher energy, is visible. In DMF, this shoulder is more
pronounced and shifts the maximum of the highest energy band
to even higher energies, while the maxima of the other bands at
lower energy decrease in intensity. In acetone, this trend
increases further: now, the former shoulder at highest energy is
the maximum absorption and all other bands show decreasing
relative intensities on going to lower energy. The absorption
maximum at highest energy shifts hypsochromic, from 14 800
cm−1 (PhCN) to 15 100 cm−1 (DMF) and to 15 400 cm−1
(acetone). Again, the measured transition moments of the main
transitions range between 8.9 (PhCN, toluene) and 9.6
(CHCl3, DCM).
explained by the lack of the bromine substituents. In summary,
the redox potentials do not differ greatly in this series, which
demonstrates weak electronic interaction between the squar-
aine chromophores in the ground state.
Optical Spectroscopy: UV/vis/NIR Absorption and
Fluorescence. Steady-state absorption and fluorescence
measurements were carried out in DCM and toluene solutions.
The data are summarized in Table 3, and the spectra in toluene
are shown in Figure 3.
All squaraine dyes show an intense absorption at ca. 14 000
cm−1. The extinction coefficients of this lowest-energy band
range from 51 000 to 61 000 M−1 cm−1 (monomer unit)−1 in
toluene and 65 000−74 000 M−1 cm−1 (monomer unit)−1 in
DCM for the polymers and from 194 000 M−1 cm−1 (monomer
unit)−1 to 236 000 M−1 cm−1 (monomer unit)−1 in toluene and
220 000−245 000 M−1 cm−1 (monomer unit)−1 in DCM for
the monomers and cyclic trimers. The absorption maxima are
at somewhat lower energy in toluene than in DCM. For the
polymers, this difference is negligible (<100 cm−1), whereas for
the cyclic trimers and the monomers, the energy difference is
300 cm−1. The different alkyl chains do not have any significant
impact neither on the energy of the absorption nor on the
fluorescence. The monomers show the typical absorption
properties of squaraine dyes with a steep rise of the main
absorption band at the low energy side with the maximum at
14 300 cm−1 (mSQ1) or 14 100 cm−1 (mSQ2, mSQ3) and a
small shoulder on the high energy side. The cyclic trimers again
possess this very prominent and sharp absorption at 14 300
cm−1 but also a weaker band at 13 600 cm−1 (ε = 50 000 M−1
cm−1 (monomer unit)−1) and a very weak shoulder at 12 400
cm−1 (ε = 4000 M−1 cm−1 (monomer unit)−1).
We also measured the solid-state absorption spectra of
pSQ3-1 in a film (see Figure 5) that was formed by spin-
coating a solution of pSQ3-1 (2.5 mg mL−1 at 2000 rpm) onto
a glass substrate. Taking into account the differences of the
absorption spectra in different solvents, we were interested if
these properties are maintained in the film when prepared from
solutions in different solvents. Thus, we chose three solvents to
prepare concentrated solutions for spin-coating: CHCl3, DMF,
and toluene. The spectra of the films were then measured with
an integrating sphere. All of these films share the fact that the
main absorption band loses its structure with the three local
maxima and we observe a slight broadening and bathochromic
shift. For films spin-coated from CHCl3 and toluene, the
spectra are almost identical and quite similar to the spectra
a
b
c
Table 2. Oxidation (ox) and Reduction (red) Potentials , Electrochemical Band Gaps , Optical Band Gaps, and Band Gaps
d
Derived from the Intersection of Absorption and Fluorescence of Selected Squaraines
opt
g,onset
opt
g,inter
E1/2(ox1) (mV)
E1/2(ox2) (mV)
E1/2(red1) (mV)
ECg V (eV)
E
(eV)
E
(eV)
mSQ2
tSQ3
207
167
175
576
600
573
−1405
−1456
−1465
1.61
1.62
1.64
pSQ3-1
1.52 (solution)
1.56
1.56
1.45 (film)
pSQ3-2
165
574
−1463
1.63
1.52 (solution)
1.45 (film)
a
b
c
opt
In DCM/TBAH (∼0.2 M) at a scan rate of 250 mV s−1 vs Fc/Fc+. EgCV = E1/2(ox1) − E1/2(red1). E
g,inter
from the onset of the absorption band.
g,onset
d
opt
E
= energy at the intersection of the normalized absorption and fluorescence spectra
2546
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
a
Table 3. Optical Properties of Squaraine Monomers, Trimers, and Polymers
b
Xn
νmax
̃
(cm−1
)
εmax (M−1cm−1
)
μ (D)
ν
̃
(cm−1
)
Φf
τf (ns)
τ (ns) kf
̅
(× 108 s−1
)
knr
̅
(× 108 s−1
)
fl
̅
mSQ1
mSQ2
14300
14100
(14400)
236000
10.10
9.58
14000
13800
0.41
0.57
4.5
4.1
0.91
1.39
1.31
1.05
194000
(220000)
mSQ3
tSQ2
14100
230000
10.37
9.86
13800
0.56
0.11
4.0
6.8
1.39
0.16
1.09
1.31
14300
(14600)
198000
(245000)
10800/12100
10800/12100
12500
tSQ3
14300
(14600)
222000
(221000)
10.67
9.17
8.95
9.28
8.92
0.08
0.19
0.13
0.10
0.06
7.2
0.11
0.52
0.38
0.31
0.19
1.27
2.20
2.53
2.78
3.03
pSQ2-2
7.4
11.8
19.7
35.5
13000
(13100)
61000
(67000)
5.4 [0.26]
1.6 [0.74]
3.7
3.4
3.2
3.1
pSQ3-2
13000
(13000)
56000
(74000)
12400
5.7 [0.18]
1.8 [0.82]
pSQ2-1
12900
(12900)
59000
(71000)
12400
5.7 [0.16]
1.6 [0.84]
pSQ3-1 toluene
12900
51000
12300
5.6 [0.14]
1.6 [0.86]
DCM
CHCl3
PhCN
DMF
(13000)
12900
13000
15100
(65000)
71000
48000
49000
9.57
9.57
8.94
9.02
a
Xn = average degree of polymerization, νmax
and polymers), μ = transition moment, νfl = fluorescence maxima, Φf = quantum yields, τf = fluorescence lifetimes, τ = average fluorescence lifetime,
kf = rate constant of the fluorescent decay in toluene, and knr
̃
= absorption maxima, εmax = extinction coefficients (the values are per monomer unit for the trimers
̃
̅
̅
̅ = rate constant of nonradiative decay in toluene. The values shown in parentheses are
b
obtained from measurements in DCM. The values given in square brackets are relative amplitudes of the different lifetimes.
Figure 3. Complete absorption spectra (top left) and absorption (solid lines) and fluorescence (dotted lines) spectra of a monomer (top right), a
trimer (bottom left), and the polymers (bottom right) in toluene. Excitation of the trimer was at 700 nm, of the polymers at 680 nm, and of the
monomer at 640 nm.
measured in CHCl3 solution with the maximum at
12 800 cm−1. However, in the case of DMF, the situation is
different. Similar to the spectrum in solution, we find the
absorption maximum of the film at the high energy side of the
main absorption band with a pronounced bathochromic shift at
14 500 cm−1. Since annealing can lead to a structural
reorganization in the films, we recorded absorption spectra
after annealing the substrates for 10 and 30 min at 130 °C.
2547
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
Figure 4. (Left) Normalized absorption spectra and (right) complete spectra of pSQ3-1 in various solvents of different polarity. Because of low
solubility, the extinction coefficient ε was not determined in acetone.
Figure 5. Absorption spectra of pSQ3-1 in solution and films spin-coated from solutions of different solvents.
While no changes were observed in the films prepared from
CHCl3 and toluene, that from DMF shows significant
differences upon annealing. As mentioned previously, the
maximum is at the high energy side of the main absorption
band immediately after film preparation. After 10 min at 130
°C, we find a plateau from 13 500 cm−1 to 14 500 cm−1 and,
after 30 min of annealing, the maximum moves to the low-
energy side of the absorption band at 13 400 cm−1. Thus, it
appears that there is a thermodynamically favored structure
which is already present in solutions and in films of CHCl3 and
toluene but not in DMF solution and its pristine film but which
is gradually formed upon annealing of the films.
fluorescence is generally very weak for the chromophores
presented here. The fluorescence of mSQ2 in DCM is
described in our previous work.37 In toluene, the fluorescence
spectra of the monomers are mirror images to the absorption
spectra with a small Stokes’ shift of 300 cm−1 and the maxima at
14 000 cm−1 (mSQ1) (see Figure 3) or 13 800 cm−1 (mSQ2,
mSQ3). The fluorescence spectra of the polymers are similar in
shape to the monomers but at significantly lower energy
(∼12 000 cm−1). Therefore, they are no mirror images to the
broad absorption spectra and there exists a Stokes’ shift of
500−600 cm−1. As do the absorption properties of the cyclic
trimers, the fluorescence properties completely differ from
those of the monomers and polymers. Two rather intense
fluorescence bands can be observed at 12 100 cm−1 and 10 800
Fluorescence. Fluorescence measurements were carried
out in toluene solutions, because, in other solvents (see above),
2548
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
cm−1, with one shoulder at the low-energy side of the latter and
a shoulder at ∼9500 cm−1. It is noteworthy that, of all
compounds investigated here, the absorption maximum of the
cyclic trimers is at the highest energy, whereas the fluorescence
maximum is, by far, at the lowest energy (apparent Stokes’ shift
of maxima = 2200 cm−1).
The fluorescence quantum yields of both bromine-
substituted monomers are basically the same, with values of
∼0.56, while mSQ1 has a quantum yield of 0.41. The lower
quantum yield of the unsubstituted squaraine is in agreement
2
α1τ12 + α2τ2
α1τ1 + α2τ2
τ =
̅
(4)
While the excitation wavelength for both static and time-
resolved fluorescence measurements was chosen for technical
reasons (fixed laser diode wavelength, small Stokes’ shift) to be
at the high-energy side of the absorption bands, it is unlikely
that the emission spectra or the kinetics are influenced by this
choice, since we assume fast energy transfer dynamics beyond
the time resolution of our setup (ca. 200 ps by deconvolution
with IRF).
with a recently published work by Wurthner et al., where a
̈
halogen effect in squaraines was reported, describing the rise of
quantum yield in a series of squaraines substituted with H, Cl,
Br, and I.63 The quantum yields of the cyclic trimers are 0.11
(tSQ2) and 0.08 (tSQ3), which may be considered to be equal,
within experimental error. For the polymers, the fluorescence
quantum yields decrease as the degree of polymerization
increases, from ϕf = 0.19 (Xn = 7.4) to 0.06 (Xn = 35.5) (see
Table 3).
Fluorescence lifetimes were measured in the nanosecond
time regime with a gated detection method at 650 nm
excitation with a laser diode. For the monomers and the cyclic
trimers, monoexponential decays were observed. The lifetime is
4.5 ns for the monomer mSQ1 and 7.0 ns for the cyclic trimers.
In the case of the polymers, biexponential decays were
monitored, with lifetimes of ∼5.6 and 1.7 ns (see Figure 6).
DISCUSSION
■
When we compare the pSQ polymers with those of our
previous work (trans-indolenine squaraine without dicyanovi-
nylene moiety in the center; see Scheme 2), we see that,
although the polymer weights are similar, the lowest-energy
absorption band in toluene comprises three almost equally
intense features within the 12 000−17 500 cm−1 region, while
that of the trans-indolenine-squaraine polymer is highly
asymmetric with a very intense band at the lowest-energy
side of the band manifold between 13 000 cm−1 and 17 300
cm−1. The spectral features of trans-indolenine-squaraine
polymer were discussed in terms of exciton coupling of
localized squaraine transitions rather than by conjugational
effects. This seems also to be reasonable for the tSQ trimers
and the pSQ polymers, because the electrochemical character-
ization indicates weak electronic interactions of the squaraine
chromophores. The strong and sharp absorption bands of
squaraine dyes should indeed lead to strong exciton coupling
due to Coulomb interactions of transition densities. In order to
assess these interactions, structural models of the squaraine
trimer and the polymers are necessary. Thus, we performed
semiempirical AM1 calculations69 on the cyclic trimer and
some hexamer superstructures for mimicking the polymer: an
open loop structure, a helical hexamer, an elongated hexamer
with a zigzag structure, and a mixed hexamer in which both
features are present. In case of the cyclic trimer, we also
optimized the geometry at the density functional theory (DFT)
level, using the B3LYP/6-31G* functional/basis set combina-
tion.70−73 The AM1-optimized structures are depicted in Figure
7.
Figure 6. Fluorescence decay of pSQ3-1 in toluene. Excitation energy:
ν
̃
= 15 400 cm−1 (λ = 650 nm).
Because of the bent structure of the cisoid dicyanovinylene
squaraine chromophores and the stiff but conformationally
flexible biaryl axis, there are different macromolecular super-
structures possible for the polymers. The most rigid structure is
that of the trimer, which proved to be asymmetric caused by
some strain in the ring (see Figure 7). While one squaraine
moiety is planar, the other two adopt a bent structure, which
deviates from planarity in order to close the cycle. The AM1-
and DFT-optimized geometries agree very well, which gives
confidence that the AM1 computed structures of the polymers
are also reasonable. In contrast to the cyclic trimer, the helical
hexamer is free of strain and has ca. 2.7 monomer units per
helix turn (Figure 7). In the helix motif, the squaraine
chromophores possess a tilt and shift π-stack arrangement
where the alkyl chains extend radially from the tube formed by
the helix (not shown in Figure 7, because the alkyl chains were
replaced by methyl groups in the calculations). Alternatively to
the helix, an open loop structure can be formed. Furthermore,
an elongated structure is conceivable which adopts a zigzag
motif with a ca. 114° angle between the centers of the squaric
acid ring. In reality, mixtures of the zigzag (or loop) and the
However, a closer inspection of the biexponential decays
revealed an increase of relative amplitude of the shorter
component with increasing Xn. The lifetimes are in the same
range as those of other molecular squaraine compounds that
have been reported in the literature.36,37,64−68
The fluorescence quantum yield (Φf) and the fluorescence
lifetime (τf) were used to calculate the average rate constant of
the fluorescent decay (kf̅ ) and the nonradiative decay (kn̅ r),
using eqs 2 and 3. For the polymers, in these equations, we
used the average fluorescence lifetime (τ), which was
̅
determined by eq 4, to calculate the rate constants.
Φ
τf
f
̅
kf =
(2)
(3)
1 − Φ
f
̅
knr =
τf
2549
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
Figure 7. AM1-optimized structures (for simplicity, the long alkyl chains are replaced by methyl groups) of tSQ, open loop hexamer, helical hexamer
(top view and side view), elongated zigzag hexamer, and mixed zigzag/helical hexamer. The orientation of localized transition moments of the lowest
energy squaraine transition is indicated by orange arrows. The phase relation of transition moments, which leads to the lowest energy absorption of
the aggregate, is indicated by the direction of the arrows.
Table 4. CNDO/S2 CIS Calculations on AM1 Optimized Cyclic Trimer and Four Different Hexameric Structures
states/cm−1 (osc. strength)
cycl. trimer
helical hexamer
elongated zigzag hexamer
open-loop hexamer
mixed hexamer
1
2
3
4
5
6
14800 (0.2)
16500 (2.4)
17000 (2.0)
15300 (0.2)
16000 (0.3)
16500 (0.1)
17200 (0.5)
17500 (3.6)
18000 (3.9)
14700 (7.1)
15300 (0.4)
15900 (1.2)
16400 (0.1)
16800 (0.3)
16900 (0.4)
14700 (1.2)
15200 (0.8)
15700 (1.9)
16500 (5.5)
16800 (0.2)
17200 (0.8)
14700 (2.2)
15100 (2.6)
15900 (0.6)
16700 (0.3)
16900 (1.9)
17000 (2.2)
helical motif may be present in a polymer strand, as indicated
by the mixed zigzag/helical structure in Figure 7.
For the helix-like arrangement, exciton coupling theory
predicts two different excitonic interactions: one H-type
behavior (parallel orientation of transition moments) due to
the stacked π-systems (see Figure 7) and another interaction
that is the consequence of two consecutive chromophores
which are arranged in a ca. 70° angle and is analogous to the
situation in the trimer. Both excitonic interactions together lead
to two allowed transitions into higher lying states, followed by
four states that are forbidden (see Table 4). The CNDO/S2
calculations indeed give two strong transitions at 17 800 and
17 500 cm−1 and four very weak lower energy transitions
(between 17 200 and 15 300 cm−1). Much in contrast, exciton
coupling theory predicts a single intense transition in the case
of the elongated zigzag hexamer at low energy and five
transitions at higher energy (J-type behavior). This is in
agreement with that found by the CNDO/S2 calculations: a
strong transition at 14 700 cm−1 and five almost-negligible
transitions between 15 300 cm−1 and 16 900 cm−1. The open-
loop hexamer also shows a single very prominent band but at
intermediate energy (16 500 cm−1). Interestingly, the individual
energies of the six transitions of these three hexamers are very
similar and only their associated transition moments vary.
Finally, in a mixed zigzag/helical structure, we observed four
allowed transitions: two at low energy and two at high energy.
For the interpretation of the absorption spectra by exciton
coupling theory, we must keep in mind that the transition
moment of the lowest energy transition of a monomeric
squaraine chromophore is polarized along the long molecular
axis, i.e., along the N−N vector of the indolenine nitrogen
atoms. Because of the C2v symmetry of the monomeric unit,
there is also a short-axis polarized transition which is at a
significantly higher energy and does not play a role in the
discussion here.37 If we assume ideal C3 symmetry of the tSQ
trimer, simple exciton theory predicts an A-symmetric state into
which electronic transition is forbidden and which is below two
degenerate E states, which are optically allowed.74 Because of
the asymmetry of the trimer, the degeneracy will be lifted and
the selection rules will not strictly apply. Semiempirical
CNDO/S2-CIS computations75 indeed support this interpre-
tation (see Table 4): although the absolute transition energies
are estimated to be too high by ca. 3000 cm−1, these
calculations yield two allowed bands (at 17 000 and 16 500
cm−1) and a very weak one (at 14 800 cm−1). In fact, we
observe two bands at different energy (14 300 and 13 600
cm−1) which are likely the “E-type” bands and a very weak one
at lower energy (12 400 cm−1) which is supposedly the
forbidden A-type band.
2550
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
We assume that the polymers pSQ display in solution both
helix-like sequences or loops, as well as elongated structures.
Thus, the experimentally observed absorption spectra will show
a superposition of the spectral features that are associated with
these superstructures, that is, two bands at the high-energy side
for the helix sections and one single band at the low-energy side
for the elongated sections and one in the middle of the
excitonic manifold as being present in the open-loop form. In
fact, we observe three bands (see Figure 3) in toluene but four
bands in the more-polar solvents (see Figure 4). The fact that
these spectral features are clearly visible suggests that, indeed,
only a limited number of structures/conformers is present in
In the following, we interpret the relative amplitudes of the
biexponential decay as the mole fraction x and 1 − x,
respectively, of two different superstructures present within one
polymer strand. As we will argue below, the shorter lifetime
(mole fraction 1 − x) is likely to be associated with the helix or
open-loop structures while the long lifetime (mole fraction x)
refers to the elongated zigzag chain. A plot of the mole fraction
x versus the degree of polymerization Xn (see Figure 8) shows
̃
solution. From the bandwidth Δν of the excitonic manifold, we
̃
estimate the electronic exciton coupling as V = Δν/4 ≈ 620
cm−1, which agrees very well with that of the cis-indolenine
squaraine polymer in Scheme 2 (640 cm−1).76
Following Kasha’s rule, fluorescence is expected to be
emitted from the lowest excited state of the excitonic manifold,
irrespective of the energy of excitation into this manifold.
Although, for the monomer squaraines, a mirror image
fluorescence with very small Stokes’ shift is observed, the
cyclic trimer displays a very structured fluorescence spectrum
with a large Stokes’ shift and which extends into the NIR
(>12 000 cm−1). This somewhat unexpected behavior may be
interpreted in two different ways: one possibility is that the
strong distortion of two of the three squaraine chromophores
leads to a strong vibronic progression in the lowest-energy
emitting state. The observation of an almost unperturbed and
unshifted narrow absorption band at 14 300 cm−1 for the
unperturbed squaraine chromophore but a much broader and
weaker band at lower energy (13 600 cm−1) support this
interpretation. The other interpretation is that two of the three
chromophores behave different and almost independent, which
would lead to two different fluorescence emission bands. A
similar observation has recently been made for strained small
oligothiophene macrocycles,77 while Law et al. observed dual or
triple fluorescence in squaraine dyes, which these authors
interpreted to be caused by conformationally different
structures being present in solution or by solute−solvent
complexes.78−81 However, dual (or triple) fluorescence from a
squaraine trimer would require negligible energy transfer within
the trimer; otherwise, only a single emission from the lowest of
the three excitonic states would be observed. Since this seems
unlikely in such a small chromophore arrangement, we prefer
the first interpretation. A clarification could be made by
measuring the excitation spectra and/or the lifetime of the
emission at all wavelengths, neither of which can be done in our
laboratory for technical reasons within this particular wave-
length region.
Figure 8. Plot of the mole fraction of helix structure and of the
fluorescence quantum yield versus the degree of polymerization.
that, for increasing chain lengths (Xn), the fraction of the helix
structure also increases and approaches a limiting value of ca.
0.86. This appears to be reasonable as longer chains possess a
higher probability to form loops and helix structures than
shorter chains. In a reversed trend, the fluorescence quantum
yield decreases with the chain length and approaches ca. 0.06.
Surprisingly, the absorption spectra in toluene do not reflect
this change of mole fraction with increasing degree of
polymerization. From these observations and eq 2−4 we derive
the following characteristics for purely zigzag (τf = 5.6 ns, ϕf =
0.41, kf = 1.1 × 108 s−1, knr = 7.3 × 107 s−1) and helix structure
(τf = 1.6 ns, ϕf = 0, kf = 0 s−1, knr = 6.3 × 108 s−1), which are
also given in Figure 9. The assignment of the helix-type
structure to the short-lifetime component is made because in
the limiting case of infinitely long chains this structure must
have a vanishing fluorescence rate constant, as required for an
H-type superstructure. This model requires that energy transfer
between the two different superstructures is slow, relative to the
decay rates.
Much in contrast to the trimer, the polymers obviously emit
from the lowest excited state of the exciton manifold, which can
easily be seen from the shape of the fluorescence band, which is
a mirror image of the monomer absorption but distinctly
shifted to lower energy. The two different time constants of the
fluorescence decay may be interpreted as being caused by the
two different superstructures (e.g., helical vs elongated zigzag).
The lowest-energy state of the helical structure should be a dark
state with a small fluorescence rate constant kf (H-type states
with vanishing transition moment should have small
fluorescence rate constant kf according to the Strickler−Berg
equation)82 while the one with the zigzag structure (J-type
band) should be allowed and thus should be associated with a
significant kf value.
Figure 9. Relaxation pathways depending on helix or zigzag
superstructures.
2551
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
Taking the above outlined structural model into account, we
can also explain the solvatochromic behavior of the excitonic
manifold (see Figure 4) by a fractional increase of helix or
open-loop structures relative to the zigzag chain. While solvents
that may form C−H hydrogen bonds to the squaric acid
oxygen, such as DCM and chloroform, favor a zigzag chain,
more polar solvents (such as DMF or acetone, which are unable
to solvate the negative charge of the squaric acid core) induce a
folding to yield a more helix-like arrangement.
An interesting point concerns the optical properties of the
films: While the films spin-coated from CHCl3 or toluene
clearly show a J-type behavior and, consequently, a large mole
fraction of zigzag structures, the pristine film formed from
DMF clearly displays a more pronounced H-type in the
absorption spectra which indicates large fractions of helix or
open-loop structures. Obviously, these structural motifs are
maintained when spin-coating from solutions in DMF.
Annealing then leads to a breakup of these structures to yield
more zigzag fractions. This aspect highlights the possibility to
induce supramolecular structures by choosing the appropriate
solvent for film-forming applications.
Concerning the band gap of the squaraine polymers, we
estimate 1.63 eV from the differences of the redox potentials for
oxidation and reduction in solution. This is approximately the
same as that of the monomer mSQ2. The optical band gap
from the onset of absorption spectra is dependent neither on
the polymer chain length nor on the solvent and is 1.52 eV (in
the films prepared from CHCl3 it is 1.45 eV). From the
intersection of absorption and fluorescence, we obtain 1.56 eV.
These values are slightly smaller than those from electro-
chemistry which reflects the exciton binding energy to be on
the order of ca. 0.1 eV. The band gap of the polysquaraines
derived from trans-indolenine is about the same.11
ACKNOWLEDGMENTS
■
This work was supported by the Deutsche Forschungsgemein-
schaft within the Research Group Fo 1809.
REFERENCES
■
(1) Scherer, D.; Dorfler, R.; Feldner, A.; Vogtmann, T.; Schwoerer,
̈
M.; Lawrentz, U.; Grahn, W.; Lambert, C. Chem. Phys. 2002, 279,
179−207.
(2) Ashwell, G. J.; Ewington, J.; Moczko, K. J. Mater. Chem. 2005, 15,
1154−1159.
(3) Shi, Q. Q.; Chen, W. Q.; Xiang, J. F.; Duan, X. M.; Zhan, X. W.
Macromolecules 2011, 44, 3759−3765.
(4) Collini, E.; Carlotto, S.; Ferrante, C.; Bozio, R.; Polimeno, A.;
Bloino, J.; Barone, V.; Ronchi, E.; Beverina, L.; Pagani, G. A. Phys.
Chem. Chem. Phys. 2011, 13, 12087−12094.
(5) Webster, S.; Fu, J.; Padilha, L. A.; Przhonska, O. V.; Hagan, D. J.;
Van Stryland, E. W.; Bondar, M. V.; Slominsky, Y. L.; Kachkovski, A.
D. Chem. Phys. 2008, 348, 143−151.
(6) Chen, C. T.; Marder, S. R.; Cheng, L. T. J. Am. Chem. Soc. 1994,
116, 3117−3118.
(7) Chen, C. T.; Marder, S. R.; Cheng, L. T. J. Chem. Soc.Chem.
Commun. 1994, 259−260.
(8) Emmelius, M.; Pawlowski, G.; Vollmann, H. W. Angew. Chem.,
Int. Ed. 1989, 28, 1445−1471.
(9) Mayerhoffer, U.; Deing, K.; Gruß, K.; Braunschweig, H.;
̈
Meerholz, K.; Wurthner, F. Angew. Chem., Int. Ed. 2009, 48, 8776−
̈
8779.
(10) Wei, G.; Wang, S.; Renshaw, K.; Thompson, M. E.; Forrest, S.
R. ACS Nano 2010, 4, 1927−1934.
(11) Volker, S. F.; Uemura, S.; Limpinsel, M.; Mingebach, M.;
̈
Deibel, C.; Dyakonov, V.; Lambert, C. Macromol. Chem. Phys. 2010,
211, 1098−1108.
(12) Wei, G. D.; Lunt, R. R.; Sun, K.; Wang, S. Y.; Thompson, M. E.;
Forrest, S. R. Nano Lett. 2010, 10, 3555−3559.
(13) Wei, G.; Xiao, X.; Wang, S.; Zimmerman, J. D.; Sun, K.; Diev, V.
V.; Thompson, M. E.; Forrest, S. R. Nano Lett. 2011, 11, 4261−4264.
(14) Maeda, T.; Nakao, H.; Kito, H.; Ichinose, H.; Yagi, S.;
Nakazumi, H. Dyes Pigm. 2011, 90, 275−283.
CONCLUSIONS
■
(15) Wang, S. Y.; Hall, L.; Diev, V. V.; Haiges, R.; Wei, G. D.; Xiao,
X.; Djurovich, P. I.; Forrest, S. R.; Thompson, M. E. Chem. Mater.
2011, 23, 4789−4798.
(16) Silvestri, F.; Irwin, M. D.; Beverina, L.; Facchetti, A.; Pagani, G.
A.; Marks, T. J. J. Am. Chem. Soc. 2008, 130, 17640−17641.
(17) Merritt, V. Y.; Hovel, H. J. Appl. Phys. Lett. 1976, 29, 414−415.
(18) Gassensmith, J. J.; Arunkumar, E.; Barr, L.; Baumes, J. M.;
DiVittorio, K. M.; Johnson, J. R.; Noll, B. C.; Smith, B. D. J. Am. Chem.
Soc. 2007, 129, 15054−15059.
(19) Sameiro, M.; Goncalves, T. Chem. Rev. 2009, 109, 190−212.
(20) Arunkumar, E.; Fu, N.; Smith, B. D. Chem.Eur. J. 2006, 12,
4684−4690.
(21) Xiang, Z. M.; Nesterov, E. E.; Skoch, J.; Lin, T.; Hyman, B. T.;
Swager, T. M.; Bacskai, B. J.; Reeves, S. A. J. Histochem. Cytochem.
2005, 53, 1511−1516.
In conclusion, we successfully synthesized the first dicyanovi-
nylene-substituted polysquaraines by applying the Ni-mediated
Yamamoto homocoupling reaction. In this attempt, we were
able to isolate conjugated cyclic trimers that, to the best of our
knowledge, are the first of their kind and exhibit multiple
fluorescence bands with, for squaraines, unusually large Stokes’
shift. The polymers show spectroscopic features such as a rather
broad absorption in the red-to-NIR region with distinct maxima
but a narrow fluorescence, red-shifted relative to the monomer
emission. The fluorescence quantum yield decreases as the
degree of polymerization increases. These features can best be
described by the formation of an excitonic manifold due to at
least two different superstructures: a zigzag chain and a helix-
like arrangement. While the former shows typical J-type
behavior, the latter may be conceived as an H-type aggregate.
The fractions to which these structures are formed depend on
the chain length and the solvent. Controlling the superstructure
will allow one to tune the optical properties, which, in a first
attempt, was successfully done in films.
(22) Ros-Lis, J. V.; Martínez-Man
2002, 2248−2249.
(23) Ros-Lis, J. V.; Marcos, M. D.; Martínez-Man
Angew. Chem., Int. Ed. 2005, 44, 4405−4407.
(24) Ros-Lis, J. V.; Martínez-Manez, R.; Sancenon
́
ez, R.; Soto, J. Chem. Commun.
́
ez, R.; Soto, J.
́
́
, F.; Soto, J.;
Spieles, M.; Rurack, K. Chem.Eur. J. 2008, 14, 10101−10114.
(25) Chenthamarakshan, C. R.; Eldo, J.; Ajayaghosh, A. Macro-
molecules 1999, 32, 5846−5851.
(26) Ajayaghosh, A.; Arunkumar, E.; Daub, J. Angew. Chem., Int. Ed.
2002, 41, 1766−1769.
AUTHOR INFORMATION
■
Corresponding Author
(27) Chenthamarakshan, C. R.; Ajayaghosh, A. Tetrahedron Lett.
1998, 39, 1795−1798.
*E-mail: christoph.lambert@uni-wuerzburg.de.
(28) Ros-Lis, J. V.; Garcia, B.; Jimen
Sancenon
Soc. 2004, 126, 4064−4065.
́ ́
ez, D.; Martínez-Manez, R.;
Notes
́
, F.; Soto, J.; Gonzalvo, F.; Valldecabres, M. C. J. Am. Chem.
The authors declare no competing financial interest.
2552
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553

Chemistry of Materials
Article
(29) Sreejith, S.; Divya, K. P.; Ajayaghosh, A. Angew. Chem., Int. Ed.
2008, 47, 7883−7887.
(66) Padilha, L. A.; Webster, S.; Hu, H. H.; Przhonska, O. V.; Hagan,
D. J.; Van Stryland, E. W.; Bondar, M. V.; Davydenko, I. G.;
Slominsky, Y. L.; Kachkovski, A. D. Chem. Phys. 2008, 352, 97−105.
(30) Ajayaghosh, A. Acc. Chem. Res. 2005, 38, 449−459.
(31) Sreejith, S.; Carol, P.; Chithra, P.; Ajayaghosh, A. J. Mater. Chem.
2008, 18, 264−274.
́
(67) Silvestri, F.; Lopez-Duarte, I.; Seitz, W.; Beverina, L.; Martínez-
Díaz, M. V.; Marks, T. J.; Guldi, D. M.; Pagani, G. A.; Torres, T. Chem.
Commun. 2009, 4500−4502.
(32) Yagi, S.; Nakazumi, H. Topics Heterocyc. Chem. 2008, 14, 133−
181.
(68) Toro, C.; De Boni, L.; Yao, S.; Ritchie, J. P.; Masunov, A. E.;
Belfield, K. D.; Hernandez, F. E. J. Chem. Phys. 2009, 130, 214504.
(69) Stewart, J. J. P. MOPAC2009; Stewart Computational
Chemistry: Colorado Springs, CO, http://openMOPAC.net, 2008.
(70) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(71) Frisch, M. J., et al. Gaussian 03, revision E.01; Gaussian, Inc.:
Wallingford, CT, 2004.
(33) Beverina, L.; Salice, P. Eur. J. Org. Chem. 2010, 1207−1225.
(34) Tatarets, A. L.; Fedyunyaeva, I. A.; Terpetschnig, E.; Patsenker,
L. D. Dyes Pigm. 2005, 64, 125−134.
(35) Kim, S. H.; Han, S. K.; Kim, J. J.; Hwang, S. H.; Yoon, C. M.;
Keum, S. R. Dyes Pigm. 1998, 39, 77−87.
(36) Tatarets, A. L.; Fedyunyayeva, I. A.; Dyubko, T. S.; Povrozin, Y.
A.; Doroshenko, A. O.; Terpetschnig, E. A.; Patsenker, L. D. Anal.
Chim. Acta 2006, 570, 214−223.
(72) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785−
789.
(73) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213−
222.
(74) Kasha, M.; Rawls, H. R.; Ashraf El-Bayoumi, M. Pure Appl.
Chem. 1965, 11, 371−392.
(37) Volker, S. F.; Renz, M.; Kaupp, M.; Lambert, C. Chem.Eur. J.
̈
2011, 17, 14147−14163.
(38) Treibs, A.; Jacob, K. Angew. Chem., Int. Ed. 1965, 4, 694−695.
(39) Treibs, A.; Jacobs, K. Liebigs Ann. Chem. 1966, 699, 153−167.
(40) Ajayaghosh, A. Chem. Soc. Rev. 2003, 32, 181−191.
(41) Chen, Y.-Y.; H.K. Hall, J. Polym. Bull. 1986, 16, 419−425.
(42) Havinga, E. E.; Pomp, A.; Tenhoeve, W.; Wynberg, H. Synth.
Met. 1995, 69, 581−582.
(75) WinMOPAC, Version 2.0; Fujitsu, Ltd.; 1997−1998.
(76) McRae, E. G.; Kasha, M. The Molecular Exciton Model, Vol. 3;
Academic Press: New York, 1964; pp 23−42.
(77) Mena-Osteritz, E.; Zhang, F.; Gotz, G.; Reineker, P.; Bauerle, P.
̈
̈
Beilstein J. Nanotechnol. 2011, 2, 720−726.
(43) Havinga, E. E.; Tenhoeve, W.; Wynberg, H. Synth. Met. 1993,
55, 299−306.
(44) Havinga, E. E.; Hoeve, W. t.; Wynberg, H. Polym. Bull. 1992, 29,
119−126.
(45) Brocks, G.; Tol, A. J. Phys. Chem. 1996, 100, 1838−1846.
(46) Lynch, D. E.; Geissler, U.; Peterson, I. R.; Floersheimer, M.;
Terbrack, R.; Chi, L. F.; Fuchs, H.; Calos, N. J.; Wood, B.; Kennard, C.
H. L.; Langley, G. J. J. Chem. Soc., Perkin Trans. 1997, 2, 827−832.
(47) Chenthamarakshan, C. R.; Ajayaghosh, A. Chem. Mater. 1998,
10, 1657−1663.
(48) Lu, H. C.; Whang, W. T.; Cheng, B. M. Chem. Phys. Lett. 2010,
500, 267−271.
(49) Huo, E. F.; Wu, J. Y.; Song, B. F.; Li, Y.; Jiang, Q.; Xie, M. G.
Chin. Chem. Lett. 2007, 18, 1531−1534.
(78) Mang, C. Y.; Zhou, J.; Liu, C. P.; Wu, K. C. Chin. J. Chem. 2009,
27, 2323−2328.
(79) Law, K. Y. J. Phys. Chem. 1987, 91, 5184−5193.
(80) Law, K. Y. J. Phys. Chem. 1989, 93, 5925−5930.
(81) Law, K. Y. Chem. Phys. Lett. 1992, 200, 121−124.
(82) Strickler, S. J.; Berg, R. A. J. Chem. Phys. 1962, 37, 814−822.
(50) Eldo, J.; Ajayaghosh, A. Chem. Mater. 2002, 14, 410−418.
(51) Ajayaghosh, A.; Eldo, J. Org. Lett. 2001, 3, 2595−2598.
(52) Wu, J. Y.; Huo, E. F.; Wu , Z. B.; Lu, Z. Y.; XieM. G.; Jiang, Q. e-
Polym. 2007, Paper 77.
(53) Block, M. A. B.; Hecht, S. Macromolecules 2004, 37, 4761−4769.
(54) Block, M. A. B.; Khan, A.; Hecht, S. J. Org. Chem. 2004, 69,
184−187.
(55) Zhang, W.; Tao, F.; Meng, K. G.; Wang, Z.; Xi, L. Y.; Li, Y.;
Jiang, Q. J. Mater. Sci. 2011, 46, 5363−5370.
(56) Zhang, W.; Tao, F.; Meng, K. G.; Xi, L. Y.; Wang, Z.; Li, Y.;
Jiang, Q. Polym. Bull. 2012, 68, 349−360.
(57) Volkova, K. D.; Kovalska, V. B.; Tatarets, A. L.; Patsenker, L. D.;
Kryvorotenko, D. V.; Yarmoluk, S. M. Dyes Pigm. 2007, 72, 285−292.
(58) Sens, R.; Drexhage, K. H. J. Lumin. 1981, 24−5, 709−712.
(59) Noviandri, I.; Brown, K. N.; Fleming, D. S.; Gulyas, P. T.; Lay,
P. A.; Masters, A. F.; Phillips, L. J. Phys. Chem. B 1999, 103, 6713−
6722.
(60) Zubatyuk, R. I.; Baumer, V. N.; Tatarets, A. L.; Patsenker, L. D.;
Shishkin, O. V. Acta Crystallogr. Sect. E.: Struct. Rep. Online 2004, 60,
O2252−O2254.
(61) Sheepwash, E. E.; Rowntree, P. A.; Schwan, A. L. J. Labelled
Compd. Radiopharm. 2008, 51, 391−398.
(62) Lewis, J. E.; Maroncelli, M. Chem. Phys. Lett. 1998, 282, 197−
203.
(63) Mayerhoffer, U.; Fimmel, B.; Wurthner, F. Angew. Chem., Int.
̈
̈
Ed. 2012, 51, 164−167.
(64) Fu, J.; Padilha, L. A.; Hagan, D. J.; Van Stryland, E. W.;
Przhonska, O. V.; Bondar, M. V.; Slominsky, Y. L.; Kachkovski, A. D. J.
Opt. Soc. Am. BOpt. Phys. 2007, 24, 67−76.
(65) Liang, K. N.; Farahat, M. S.; Perlstein, J.; Law, K. Y.; Whitten, D.
G. J. Am. Chem. Soc. 1997, 119, 830−831.
2553
dx.doi.org/10.1021/cm301109u | Chem. Mater. 2012, 24, 2541−2553