Synthesis, electrochemical, and optical properties of low band gap homo- and copolymers based on squaraine dyes

Article abstract of DOI:10.1002/pola.27073

A broad series of monomeric and polymeric squaraines was synthesized to investigate the impact of electron-donating bridges, such as unsaturated triarylamine, carbazole, and saturated piperazine groups, on the spectroscopic and redox properties. These bridges were attached to both standard trans-indolenine squaraines and dicyanomethylene-substituted cis-indolenine squaraines. The conjugates were investigated by absorption, steady-state, and time-resolved fluorescence spectroscopy and cyclic voltammetry. While addition of the donors resulted in significant redshift of the absorption of the model compounds, hardly any further shift or broadening was observed for the copolymers. Also the redox properties remained nearly unchanged compared with the model dyes. In contrast, immense broadening and redshift was observed for homopolymers. This behavior is explained by mostly excitonic coupling of localized squaraine transitions. The increasing distance of the chromophores determined by the bridges led to a decrease of the exciton coupling energy. We also performed semiempirical CNDO/S2 calculations on AM1 optimized structures.

Full text of DOI:10.1002/pola.27073

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Synthesis, Electrochemical, and Optical Properties of Low Band Gap
Homo- and Copolymers Based on Squaraine Dyes
€
Sebastian F. Volker, Theresa Dellermann, Harald Ceymann, Marco Holzapfel,
Christoph Lambert
€
€
€
€
Institut fur Organische Chemie, Center for Nanosystems Chemistry, Universitat Wurzburg, Am Hubland, 97074, Wurzburg,
Germany
Correspondence to: C. Lambert (E-mail: christoph.lambert@uni-wuerzburg.de)
Received 21 October 2013; accepted 5 December 2013; published online 11 January 2014
DOI: 10.1002/pola.27073
ABSTRACT: A broad series of monomeric and polymeric squara-
ines was synthesized to investigate the impact of electron-
donating bridges, such as unsaturated triarylamine, carbazole,
and saturated piperazine groups, on the spectroscopic and
redox properties. These bridges were attached to both stand-
ard trans-indolenine squaraines and dicyanomethylene-
substituted cis-indolenine squaraines. The conjugates were
investigated by absorption, steady-state, and time-resolved flu-
orescence spectroscopy and cyclic voltammetry. While addition
of the donors resulted in significant redshift of the absorption
of the model compounds, hardly any further shift or broaden-
ing was observed for the copolymers. Also the redox proper-
ties remained nearly unchanged compared with the model
dyes. In contrast, immense broadening and redshift was
observed for homopolymers. This behavior is explained by
mostly excitonic coupling of localized squaraine transitions.
The increasing distance of the chromophores determined by
the bridges led to a decrease of the exciton coupling energy.
We also performed semiempirical CNDO/S2 calculations on
C
V
AM1 optimized structures.
2014 Wiley Periodicals, Inc. J.
Polym. Sci., Part A: Polym. Chem. 2014, 52, 890–911
KEYWORDS: conjugated polymers; dyes/pigments; synthesis
and emission in the red to near-infrared (NIR) region, simi-
lar to cyanines. For years, their suitability for various appli-
cations has been investigated. The potential as
photoconductors in xerographic devices1 has been studied as
well as their application as materials for data imaging.2 Fur-
thermore squaraines can be used as fluorescence labels,3–6
in nonlinear optics,7–11 and in organic and hybrid photovol-
taics.12–19 Besides, their potential use as sensors for
anions20, cations,21–24 and some neutral compounds25,26
such as fructose27 has been studied. Several excellent
reviews give a deeper insight into chemistry and application
of this class of dyes.28–31
INTRODUCTION The use of functional dyes or p-conjugated
polymers for optoelectronic and other applications requires
fine-tuning of their optical and redox properties. In this
paper we focus on the synthesis and some basic electronic
(optical end electrochemical) properties of (co)polymers that
are based on squaraine dyes combined with other electron
donating p-conjugated building blocks such as carbazoles
and triarylamines along with their monomeric parent com-
pounds for comparison. This study shall provide a basis for
using these polymers in optoelectronic devices in the future.
Squaraines represent a class of organic dyes that is related
to cyanine dyes. The structure is similar, but, in contrast to
cyanines, there is an additional electron accepting 3-
oxocyclobut-1-enolate group centered in the middle. While
cyanine dyes are positively charged, squaraines are zwitter-
ionic and represent a donor-acceptor-donor structure. They
are usually synthesized by 1,3-dicondensation reactions of
squaric acid and nucleophilic compounds such as aniline
derivatives, pyrroles or indolenine methylene bases and its
heteroatom analogues. Squaraines are intensely colored and
There are various approaches to tune the properties of
squaraine dyes. These properties strongly depend on the
nucleophilic compound that is used in the squaraine synthe-
sis. Moreover, the size of the p-system as well as the addition
of further donor units at the periphery of the dyes shifts the
absorption and emission towards lower energy. Also, replace-
ment of one oxygen of the squaric carbonyl groups by dicya-
nomethylene,13,32–34 sulfur,33,35 or barbituric acid33 for
exhibit narrow and strong absorption (e > 105 M21 cm21
)
example leads to a significant change of the optical
Additional Supporting Information may be found in the online version of this article.
C
V
2014 Wiley Periodicals, Inc.
890
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
CHART 1 Target monomers M1-M12 and polymers P1-P7.
properties. The stronger the electron accepting unit in the
center of the squaraine dyes, the stronger is the induced
bathochromic shift.
Another way to tune the optical properties of squaraines is
the synthesis of oligomers and polymers where excitonic
coupling of localized states leads to significant broadening
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
891

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 1 Synthesis of functionalized dyes M1-M4.
and shifts of the absorption. We recently synthesized some
polymers where indeed the absorption spectra were consid-
erably broadened and red-shifted.14,34 Additionally there
were contributions in the absorption at higher energy which
could be explained by the concomitant existence of both H-
type and J-type alignments of the chromophores that lead to
hypsochromic and bathochromic shifts, respectively.
oligomers up to a pentamer. A significant redshift and broad-
ening was found upon stepwise increase of the oligomer
size.11
In this work here, we followed the idea of inserting addi-
tional very strong donor units into a polysquaraine system
and synthesized a series of indolenine-squaraine copolymers
with various electron rich bridging units. We used both satu-
rated (piperazine) and unsaturated rigid (3,6- and 2,7-carb-
azoles) and propeller like (triarylamine) bridges to study
their influence on the spectroscopic and electrochemical
properties. Furthermore we likewise bridged squaraine dyes
where an oxygen atom of the central carbonyl group was
replaced by a dicyanomethylene group. Besides this, we also
prepared some homopolymers and monomer conjugates for
comparison (see Chart 1).
Historically, polysquaraines were the first squaraines to be
reported about. Colored insoluble compounds were obtained
by Treibs and Jacob upon condensation of squaric acid with
pyrrole derivatives.36,37 Since then, there have been numer-
ous works on polysquaraines. Especially pyrrole-squaraine
polymer derivatives have been extensively studied.38,39 Fur-
thermore, additional donor units such as fluorene or 2,5-di-
alkoxyphenylene were inserted into polysquaraines to obtain
low-band-gap polymers with broad absorption in the NIR
region.40–42 Following this successful work, further donor
components were inserted into 2,5-dialkoxyphenylene-pyr-
role-squaraine polymers, such as 2,7-carbazole or thiophene
derivatives which also resulted in polymers with a broad
absorption in the NIR.43,44
EXPERIMENTAL
Synthesis
The synthetic protocols can be found in the Supporting
Information.
In the field of indolenine-squaraines, only few approaches
have been reported on the synthesis of donor-bridged squar-
aine copolymers. In search of polymers with broad absorp-
tion in the NIR region, Maeda et al. used both meta- and
para-linked phenylene bridges but the bathochromic shift of
the absorption was very small.45 A similar behavior was
observed by Kuster and Geiger where various bridges
between indolenine-squaraines were investigated. In this
work, a weak broadening and a small bathochromic shift of
the absorption compared with monomeric model compounds
was achieved even when 2,5-dialkoxyphenylene was used as
the bridge.46 However, Scherer et al. used thiophene groups
to link indolenine-squaraines and synthesized a series of
UV/Vis/NIR Spectroscopy
Ultraviolet/visible light/near-infrared spectra were measured
using either a JASCO V670 or a Carey 5000 spectrometer.
Compounds were dissolved in Uvasol solvents from Merck
and measured in a concentration range of ꢀ1025 to 1027
M
in 1 cm quartz cuvettes to exclude aggregate formation. The
pure solvent was used as reference.
Fluorescence Spectroscopy
Fluorescence measurements were performed with a Photon
Technology International (PTI) fluorescence spectrometer
QM-2000-4 with a 75 W xenon lamp and a cooled photomul-
tiplier (R928P) or a cooled InGaAs detector. The compounds
892
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
CHART 2 Previously published squaraine monomers M1334, M1450, M1550 and polymers Px14 and thiophene bridged pentamer
Py.11
were dissolved in Uvasol solvents from Merck, purged with
argon gas for 10 min and measured in 1 cm quartz cuvettes.
As a fluorescence standard, Oxazine 1 in ethanol (/fl 5 0.11)
was used and the following equation was applied to deter-
mine the quantum yields.47
nium and calculated against the ferrocene/ferrocenium (Fc/
Fc1) redox couple.49
Gel permeation Chromatography (GPC)
All measurements were performed at 20 ꢁC in chloroform
with a Shimadzu instrument (diode array detector SPD-
M20A, system controller CBM-20A, solvent delivery unit LC-
20AD, on-line degasser DGU 20A9). Preparative chromato-
graphy was carried out in recycling mode on two consecu-
tive SDV columns (50 Å and 500 Å, dimension: 20 3 600
mm, particle size: 10 mm) from PSS. Analytical chromatogra-
phy was carried out on a SDV column (analytical linear S
mixed bed, dimension: 8 3 300 mm, particle size: 5 mm)
from PSS with polystyrene as standard.
!
2
Ið~mÞ3ODRef3ðnD20
Þ
/fl5/fl;Ref
2
IðmÞRef3OD3ðn2D0ÞRef
~
where /fl is the quantum yield of the sample, /fl,Ref is the
~
quantum yield of the reference, I(m) the integrated emission
band, OD the optical density of the absorption band at the
20
excitation wavelength, and nD is the refraction index of the
solvent.
RESULTS AND DISCUSSION
Fluorescence Lifetimes
The samples, that were prepared likewise to the steady state
fluorescence measurements, were excited with a 650 nm
laser diode and measured with a PTI TM-2/2003 fluores-
cence lifetime spectrofluorometer. To determine the instru-
ment response (ꢀ1.5 ns) we used colloidal silica in de-
ionised water as scatter solution. The lifetimes were
obtained after fitting of the decay curves with exponential
decay functions.
Synthesis
Squaraine Dye Building Blocks
For a thorough investigation of the impact of the various
substitutions on the squaraine dyes, a completely nonfunc-
tionalized dye is necessary as well as functionalized dyes for
further syntheses. Therefore we synthesized M1 (Scheme 1)
Cyclic Voltammetry
Cyclic voltammetric measurements were performed in DCM,
THF or CDCl3 with tetrabutylammonium hexafluorophos-
phate (TBAHFP) (0.2 M) as supporting electrolyte. We used
a three electrode setup with a platinum working electrode
(ï 1 mm), a helical platinum counter electrode and either a
platinum pseudo reference electrode or an Ag/AgCl “leak
free” reference electrode48 in a completely sealed and with
argon flushed glass vessel. In case of the platinum pseudo
reference electrode, the potentials were measured against
the redox couple of decamethylferrocene/decamethylferroce-
SCHEME 2 Yamamoto coupling to the homopolymers P1a and
P1b.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
893

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Polymer Characterization: Number-Average Molecular
Weight (Mn), Number Average Degree of Polymerization (Xn),
Weight-Average Molecular Weight (Mw), and Polydispersity
(Mw/Mn)a
quaternary salt 1 in toluene and 1-butanol yielded the asym-
metric dye M2.
Squaraine Homopolymers
In analogy to our previous work, we synthesized squaraine
homopolymers based on monomer M1.14 Since poly-
indolenine squaraine Px (5 P1 with R 5 C16H33, Chart 2)
was only soluble in CHCl3 and solvent mixtures consisting of
both a chlorinated solvent and an alcohol, we replaced the
linear alkyl chain of polymer Px by a branched 3,7-dimethyl-
octyl chain hoping to increase the solubility.
Mn
Xn
Mw
Mw/Mn
P1a
P1b
P2
31,000
8,100
45.9
12.0
14.7
16.7
13.3
21.4
6.4
80,200
22,700
31,900
32,800
73,200
61,500
20,600
28,800
2.59
2.81
2.17
2.07
5.35
2.94
3.30
3.07
14,700
15,800
13,700
20,900
6,200
P3
P4
The Ni-mediated Yamamoto coupling reaction of M3 was car-
ried out under the same conditions as for Px (Scheme 2),
namely stirring at room temperature for six days in a solvent
mixture of DMF and toluene (1:1). The resulting polymer
P5
P6
P7
9,400
12.4
a
P1a, with
a
number average molecular weight of
Analysis was performed in CHCl3 with polystyrene as standard.
Mn 5 31,000 and a degree of polymerization of Xn 5 46, pre-
cipitated out of the solution during the reaction. This poly-
mer P1a is ꢀ30% longer than our previous polymer Px
(Xn 5 33), but solubility was again restricted to CHCl3 and
the aforementioned solvent mixtures.
which does not comprise any bromine substituents or other
functional groups at the periphery, the mono- and dibromi-
nated dyes M2 and M3, respectively, as well as the boronic
ester derivative M4.
Because P1a proved poorly soluble in pure solvents, we used
pure DMF as the solvent for the Yamamoto coupling of M3
under elsewise identical reaction conditions to provoke pre-
cipitation of the polymers at an earlier stage of polymeriza-
tion condensation reaction to get shorter and thus possibly
better soluble polymers. Indeed, this reaction resulted in
lower yield but also in the shorter polymer P1b (Mn 5 8100,
Xn 5 12) with increased solubility (soluble in pure DCM and
pure PhCN for example). All analytical GPC data for determin-
ing the polymer mass distribution can be found in Table 1.
We started with the quaternary salts 1 and 2, whose synthe-
sis is described in a previous publication.34 In situ deproto-
nation with pyridine to the highly nucleophilic methylene
bases and subsequent reaction with squaric acid in a solvent
mixture of 1-butanol and toluene (1:1) with azeotropic distil-
lation of water using a Dean-Stark trap, resulted in the non-
functionalized and dibrominated squaraine dyes M1 and M3,
respectively.
The bromine atoms in M3 were substituted by boronic ester
groups via the Pd-catalyzed Miyaura borylation reaction
using bis(pinacolato)diboron, Pd(PhCN)2Cl2 and 1,10-bis(di-
phenylphosphino)ferrocene (dppf), resulting in M4 in high
yields.
Triarylborane Substituted Squaraine Monomer
To investigate the impact of additional acceptor units at the
periphery of the dye molecule, we attached triarylboranes,
where the boron center is protected by six methyl groups
(M5, Scheme 3). These acceptor-squaraine-acceptor dyes are
topologically similar to the donor-squaraine-donor dyes
which we recently investigated.50 For the synthesis of M5,
overnight heating of M4 and triarylborane 5, which
was synthesized from 1,4-dibromo-2,3,5,6-tetramethylben-
zene and dimesitylfluoroborane according to literature
The monobrominated asymmetric dye M2 was obtained after
three steps from the quaternary salt 2 which was condensed
with 3,4-diethoxy-3-cyclobutene-1,2-dione in the presence of
NEt3 in boiling ethanol to the semisquaraine 3 which was
almost quantitatively hydrolyzed under acidic conditions to
the acidic form 4. Further condensation reaction with the
SCHEME 3 Synthesis of triarylborane substituted squaraine dye M5.
894
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 4 Synthesis of triarylamine precursor and the copolymers P2 and P3.
SCHEME 5 Synthesis of carbazole precursors.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
895

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 6 Synthesis of carbazole substituted squaraine dyes.
procedures,51 with Pd(PPh3)4 and Na2CO3 in a solvent mix-
ture of toluene and water resulted in M5 in high yield.
for all the physical characterizations throughout this work
was the part that remained insoluble during treatment in a
Soxhlet apparatus with DCM but dissolved when replacing
DCM by CHCl3.
Donor Substituted Squaraines
Following our previous work on donor substituted squaraine
dyes,50 we synthesized various copolymers using triaryl-
amine (Scheme 4), carbazoles (Scheme 7) and a non-
conjugating piperazine (Scheme 8) as bridging units as out-
lined below:
Carbazole Substituted Squaraines
Triarylamines possess a propeller-like structure which shows
some flexibility.52 In order to increase conjugation, we
replaced the triarylamine by the more rigid carbazole bridge
which was connected either via its 2- or 3- position to the
squaraines.
Triarylamine Squaraine Copolymers
The triarylamine precursor 7 was synthesized in two steps
applying the copper catalyzed Ullmann coupling of bis(4-
bromophenyl)amine and 1-iodo-4-methoxybenzene to 6 in
the first step (Scheme 4). Halogen-metal exchange with
nBuLi and subsequent addition of the borolane eventually
gave the diboronic ester 7 in low yield. Copolymerizations
of 7 with either M3 or the previously published M634 were
carried out by continuous heating in THF/water for five
days to give the copolymers P3 and P2, respectively. Both
polymers are of similar size with a degree of polymeriza-
tion Xn ꢀ 15. However, in case of P3 solubility turned out
to be poor. A major part of the polymer was insoluble in
common organic solvents. Therefore the fraction we used
The synthesis of the carbazole precursors is straight forward
(Scheme 5). All non-alkylated brominated carbazoles were
synthesized according to literature procedures. Deprotona-
tion of the mono-(853, 1154) or dibrominated (1455, 1756
)
carbazoles with either NaH or K2CO3 and subsequent alkyla-
tion with 3,7-dimethyloctyl bromide or tosylate in DMF gave
the alkylated carbazoles 9, 12, 15, 18 in moderate to very
good yields. The boronic ester was attached via halogen-
lithium exchange in THF, using nBuLi for the monobromi-
nated and tBuLi for the dibrominated species, and subse-
quent addition of the borane to give the boronic esters 10,
13, 16 and 19.
896
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 7 Synthesis of carbazole-squaraine copolymers.
SCHEME 8 Synthesis of piperazine-squaraine monomer and polymer.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
897

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
3,6-dichloro-9H-carbazole gave substituted squaraine M11. In
this reaction, the temperature plays a particularly crucial role.
Under refluxing conditions the product could not be isolated.
We always found mixtures with substances where additional
carbazole units were coupled to the one that was already
attached at the squaraine core. This could be avoided by reduc-
ing the temperature to 80 ꢁC for a reaction time of 18 h.
Carbazole Squaraine Copolymers
The reaction of squaraine M6 with carbazole 16 in toluene/
water resulted in polymer P4 with reasonable yield and
mass distribution (Mn 5 13,700, Xn 5 13), whereas the same
reaction conditions for dibromosquaraine M3 and carbazole
diboronic ester 16 only gave low yield and much shorter
polymers P5 (Mn 5 7090, Xn 5 7.2) (see Scheme 7). This is
most likely due to the precipitation of shorter polymer
strands during reaction in this particular solvent mixture
after a couple of hours. However, the reaction of squaraine
diboronic ester M4 with dibromocarbazole 15 in THF/water
proceeded very well to give longer polymers P5
(Mn 5 20,900, Xn 5 21) in higher yield. This load of the lon-
ger polymers P5 was used for the characterization through-
out this work. The copolymerization reaction for M3 and
3,6-disubstituted carbazole 19 in toluene/water with Na2CO3
gave good yields but short polymers P6 (Mn 5 6200,
Xn 5 6.4). The synthetic approach in THF/water, analogous to
the synthesis of P5, was not pursued in this case, because
no major differences neither of the optical properties
between the 2,7-carbazole or 3,6-carbazole bridged copoly-
mers nor of the optical and electrochemical properties of the
monomeric model compounds were observed.
FIGURE 1 Complete IR spectra of M1, M12 and P7 and magni-
fied spectra of M12 and P7 (inset).
Suzuki coupling of squaraine M6 with carbazole 10 in tolu-
ene/water (Scheme 6) gave both mono- and disubstituted
squaraines M7 and M8, respectively, after careful separation
by preparative recycling GPC. Squaraine M3 was coupled
with both 10 and 13 to yield the substituted squaraines M9
and M10, respectively, to investigate the impact of the differ-
ent linking positions of the carbazoles.
We also connected a carbazole unit via its N-position to
a squaraine dye which yields an asymmetric dye conjugate
for potential future purposes. Buchwald-Hartwig coupling
reaction
of
monobrominated
squaraine
M2
with
TABLE 2 Redox Potentials (vs. Ferrocene/Ferrocenium) and Band Gaps of Monomers Determined by Both Cyclic Voltammetry
(E(gap)CV) and Absorption Spectroscopy (E(gap)opt
)
3rd red
2nd red
1st red
1st ox
2nd ox
3rd ox
4th ox
HOMOa
LUMOa
(eV)
E(gap)CVa
E(gap)optb
Solvent E1/2 (mV) E1/2 (mV) E1/2 (mV) E1/2 (mV) E1/2 (mV) E1/2 (mV) E1/2 (mV) (eV)
(eV)
(eV)
M1
THFc
THFe
THFe
DCMe
DCMc
DCMc
DCMc
DCMc
DCMc
DCMe
22631d
22312d
22487d
21660
21320
21655
21678
21502
21468
21670
21653
21701
21711
21664
21717
64
442
25.32
25.51
25.32
25.23
25.31
25.26
25.16
25.20
25.25
24.95
25.16
24.98
23.60
23.94
23.61
23.48
23.66
23.69
23.49
23.51
23.46
23.45
23.51
23.45
1.72
1.57
1.71
1.75
1.65
1.57
1.67
1.69
1.79
1.50
1.65
1.53
1.88
1.71
1.84
1.85
1.69
1.65
1.77
1.79
1.86
1.73
1.75
1.65
M6
245
64
M5
22647d
454
524
570
542
523
544
557
231
277
221
M4
71
M7
146
96
802d
M8
M9
1
M10
M11
M12
35
88
1151
994d
277
2214
26
2190
M1450 DCMc
M1550 DCMc
587
833
491
a
b
Obtained from the half wave potentials of cyclic voltammetry measure-
ments. All redox potentials are given versus Fc/Fc1 which has a potential
of 0.46 eV in TBAHFP/DCM and in TBAHFP/THF 0.56 eV versus saturated
calomel electrode (SCE).61 The potential of SCE is 0.244 V versus the nor-
mal hydrogen electrode (NHE),62 which has an absolute potential of 4.46
eV versus vacuum.63 Therefore, in DCM E(HOMO/LUMO) 5 25.16 eV –
Determined in DCM by the intersection point of a tangent on the
inflexion point of the low energy side of the absorption band and the x-
axis.
c
Platinum pseudo reference electrode.
Potentials extracted from DPV.
d
e
Ag/AgCl “leak free” reference electrode.
E
1/2(1st ox/1st red) and the electrochemically derived band gap is
E(gap)CV 5 E(LUMO) 2 E(HOMO). In THF, the absolute potential of Fc/
Fc1 has to be replaced by 25.26 eV.
898
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 3 Redox Potentials (vs. Ferrocene/Ferrocenium) and Band Gaps of Monomers Determined by Both Cyclic Voltammetry
(E(gap)CV) and Absorption Spectroscopy (E(gap)opt
)
1st red
1st ox
2nd ox
3rd ox
4th ox
HOMOa
(eV)
LUMOa
(eV)
E(gap)CVa
E(gap)optb
Solvent
E1/2 (mV)
E1/2 (mV)
E1/2 (mV)
E1/2 (mV)
E1/2 (mV)
(eV)
(eV)
P1b
P2
P3
P4
P5
P7
DCMc
DCMd
21,671
21,466
–
62
570
407
325e
563
530
219
25.22
25.27
–
23.49
23.69
–
1.73
1.58
–
1.63
1.62
112
25e
147
60
607
465e
c
CDCl3
843e
(1.68)/1.74
1.63
DCMd
DCMd
DCMd
21,454
21,656
21,744
25.31
25.22
24.98
23.71
23.50
23.42
1.60
1.72
1.56
1.76
2183
(1.54)/1.72
a
b
Obtained from the half wave potentials of cyclic voltammetry meas-
urements. All redox potentials are given versus Fc/Fc1 which has a
potential of 0.46 eV in TBAHFP/DCM and in TBAHFP/THF 0.56 eV versus
saturated calomel electrode (SCE).61 The potential of SCE is 0.244 V ver-
Determined in CHCl3 by the intersection point of a tangent on the
inflexion point of the low energy side of the absorption band and the x-
axis, the values in brackets were obtained when the tangent was placed
on the low energy shoulder of the main absorption band.
sus the normal hydrogen electrode (NHE)62
potential of 4.46 eV versus vacuum.63 Therefore, in DCM E(HOMO/
,
which has an absolute
Platinum pseudo reference electrode.
c
d
Ag/AgCl “leak free” reference electrode.
e
LUMO) 5 –5.16 eV
–
E
1/2(1st ox/1st red) and the electrochemically
Potentials extracted from DPV.
derived band gap is E(gap)CV 5 E(LUMO) – E(HOMO). In THF, the abso-
lute potential of Fc/Fc1 has to be replaced by 25.26 eV.
Piperazine-Squaraine Monomer and Copolymer
of polycondensation reactions which should ideally show a
Flory distribution with PDI 5 2. The deviation from this val-
ues can be understood by the work up extraction procedures
which remove both very short (5very soluble) and very long
(5insoluble) polymer fractions from the crude polymer
product.
Piperazine was chosen as a nonconjugating electron donor
that may be used as a bridge to connect squaraine dyes.
Because the Buchwald-Hartwig coupling reaction did not
prove successful for the polymerization step for a similar
system in our lab, a different synthetic route was followed
where the piperazine was attached to the methylene base
before the dye condensation synthesis (Scheme 8).
IR Spectroscopy
It is well known that in the condensation reaction of a squaric
acid derivative with nucleophilic compounds the central ring
can be disubstituted both in the 2,4-positions to the squaraine
dye and in 3,4-positions of a cyclobutenedione compound.57–59
The solvent that is used in this condensation reaction step
plays a crucial role, e.g. the use of a toluene/1-butanol solvent
mixture usually gives 2,4-substituted products only. In the case
of monomeric squaraine dyes, even traces of 3,4-substituted
products can be removed via flash chromatography to give
pure 2,4-isomeric squaraine products. A subsequent transition
metal catalyzed or mediated polycondensation reaction there-
fore will give polymers with only isomerically pure squaraine
subunits where the central ring is disubstituted at the 2,4-posi-
tions. However, this may be different in cases where the
squaraine forming reaction is the polycondensation step itself
as for the bifunctional methylene base 22 and squaric acid to
form P7. Therefore, even though it is unlikely, the formation of
3,4-substituted subunits cannot be completely ruled out. To
prove their absence, we used IR spectroscopy. In case of 3,4-
substitution, a diketone structure of the cyclobutene ring
In the first step, the quaternary indolenine 2 salt was quan-
titatively deprotonated to give the highly nucleophilic meth-
ylene base 20. Thereafter, analogous reaction sequences
were used to synthesize both the model compound and the
copolymer. Buchwald-Hartwig reaction of the methylene
base 20 with N-methylpiperazine gave 21 which was
treated with squaric acid in a condensation reaction to yield
the model compound M12 in moderate yield. For the poly-
mer precursor, 20 was coupled twice with piperazine to
give 22. Here, immediate recrystallization of the crude
product from ethanol proved essential to prevent the prod-
uct from further side reactions. Polymerization was then
carried out analogous to the monomer squaraine condensa-
tion synthesis in toluene/1-butanol (1:1) using a Dean-
Stark trap. However, the polycondensation reaction was car-
ried out under nitrogen atmosphere and the reaction time
was extended to six days to obtain polymer P7 (Mn 5 9400,
Xn 5 12).
shows a C5O vibration at around 1680 to 1800 cm21 60
Due
.
Physical Characterization
Molecular Weight Distribution
to the delocalized character of 2,4-substituted units, this vibra-
tion is not observed in this case, but a vibration of the delocal-
ized central C5C double bonds at around 1600 cm21 should
be seen. Thus, IR spectra of polymer P7 and, for comparison,
of monomers M12 and M1 were recorded. Figure 1 shows the
complete IR spectra and a magnified part of M12 and P7
(inset). As can be seen for the monomers, there is absolutely
no signal at around 1680 to 1800 cm21. For the polymer very
Molecular weight distributions of all isolated polymers were
determined by analytical GPC using polystyrene standards
(see Table 1). In general the degree of polymerization Xn
depends on the solubility of the polymer under the given
reaction conditions and varies from about 7 up to 45. The
polydispersity PDI 5 Mw/Mn varies between 2 and 5 typical
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
899

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
ubility issues, P3 was measured in CDCl3. In cases in which
the reduction is of particular interest (such as M5), CVs were
measured in THF. All data of the monomers and polymers can
be found in Tables 2 and 3 and selected CVs in Figure 2 to 7.
In addition to the reduction and two oxidation processes
that are common for indolenine-squaraine dyes (M1, M6),
FIGURE 2 Cyclic voltammograms of M1, M6 and M5 in
TBAHFP/THF (0.2 M) at a scan rate of 250 mV s21
.
weak bands in this region indicate indeed negligible formation
of 3,4-substituted squaraine moieties.
Cyclic Voltammetry
Monomers
FIGURE 3 Cyclic voltammograms of carbazole-squaraines M7-
M11 and piperazine-squaraine M12 in TBAHFP/DCM (0.2 M) at
Cyclic voltammetry (CV) measurements were performed in
DCM for most squaraine monomers and polymers. Due to sol-
a scan rate of 250 mV s21
.
900
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
carbazole decreases the oxidation potentials by around 40
mV. Scanning over a third, chemically irreversible, oxidation
process leads to a different back-reduction (dotted lines).
The irreversibility of the third process is likely to be due to
the unprotected 3-position of the carbazoles where dimeriza-
tion could occur after oxidation of the carbazole moiety.65,66
This interpretation is supported by the third clean reversible
oxidation of M11, where the 3,6-positions are protected by
chlorine atoms and no further dimerization of the carbazole
moiety is possible.
The piperazine substituted dye M12 exhibited two clean oxi-
dation waves (Fig. 3, bottom black line) when the scan did
not include the third oxidation (red line). The third process is
chemically irreversible and no clean back-reduction is
detected for this process. The half wave potentials of the first
two oxidation waves (E1/2 5 2214 mV, 231 mV) and of the
reduction (E1/2 5 21.71 V) are very similar to those of our
previously published dianisylaminosquaraine dye M15 (Chart
2) (E1/2 5 2190 mV, 221 mV and 21.72 V respectively) that
comprises two bisphenylamine substituents instead of two 4-
methylpiperazines. The comparison indicates that the first oxi-
dation is localized at the squaraine moiety and not at either
of the substituents. In analogy to the absorption spectroscopy,
this observation demonstrates the strong electron donating
character of the piperazine moieties.
Homopolymers and Copolymer
The reduction and the two oxidation processes of the
squaraine homopolymer P1b are about 30 mV at lower
potential than those of our previously published homopoly-
mer that has a higher molecular weight but is otherwise
identical.14 As mentioned in our previous work, the redox
potentials were rather similar to those of the dibrominated
monomeric model compound and polymerization thus has
no major impact on the redox potentials.
FIGURE 4 Cyclic voltammograms of P2 and P5 in TBAHFP/
DCM (0.2 M) at a scan rate of 250 mV s21
.
for the borane-squaraine conjugate M5 we found two reduc-
tion processes far below 22 V. Differential pulse voltammetry
(DPV) shows that the second of these processes covers two
electron reduction. Both, M1 and M6 also showed an (chemi-
cally irreversible) reduction process at 22.63 V and 22.31 V,
respectively. These chemically irreversible reductions are
accompanied by belated back oxidation processes starting at
ꢀ2700 mV and 2400 mV, respectively (see Fig. 2). Thus, in
analogy we assign the process at 21.66 and 22.49 V in M5
to reductions of the squaraine moiety and the process at
22.65 V to the reductions of the two triarylboranes. The lat-
ter values are similar to those found in other triarylboranes.64
All donor-squaraine copolymers show one reduction wave
when measured in DCM. Only P3, which was measured in
CDCl3, does not show a clean reduction wave within the
operating window of the solvent. Up to four oxidation waves
were observed for the polymers. In case of the triarylamine-
squaraine copolymers P2 (Fig. 4, top) and P3 (not shown),
the back-reduction peak becomes distinctly sharper when a
potential high enough to cover the fourth oxidation was
applied. This sharp wave is a sign for adsorbed polymers on
the electrode that might have precipitated at the fourfold
oxidized state. In contrast, when only the third oxidation
process is covered, the back-reduction shows the normal
reversible signal. The voltammograms of the carbazole-
squaraine copolymers P4 (not shown) and P5 (Fig. 4, bot-
tom) display a very symmetric reduction wave with a small
peak separation of only 20 to 30 mV. The same is true for
the first two oxidation processes of P5 (8 mV and 32 mV).
We assume that these polymers adsorb to the electrode
which then might lead to the observed behavior typical of
thin layers where diffusion processes do not lead to the usu-
ally observed hysteresis behavior of cyclic voltammograms
Donor Substituted Monomers
The voltammograms of the carbazole substituted dyes are
depicted in Figure 3 (top: M9, M10, M11; middle: M7, M8).
Concerning the first two oxidation steps and the reduction,
only weak differences are found for squaraines linked to
carbazole in either the 2- or 3-position (M9 vs. M10). The
N-linked carbazole conjugate M11 shows both a somewhat
lower reduction potential and higher oxidation potentials.
Comparison of the mono- and dicarbazole-substituted dyes
M7 and M8, respectively, displays that the additional
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
901

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 4 Absorption and Fluorescence Spectroscopic Data of Monomers (cy 5 Cyclohexane)
vmax (cm21
)
emax (M21 cm21
)
kfl (nm)
vfl (cm21
)
/
fl
sfl (ns)
~
~
Solvent
k (nm)
M1
M2
M3
DCM
639
644
642
647
646
651
646
650
656
652
655
697
711
720
703
715
720
719
731
733
671
674
667
669
673
666
645
649
644
679
682
685
700
677
681
672
703
706
697
15,700
15,500
15,600
15,500
15,500
15,400
15,500
15,400
15,300
15,300
15,300
14,300
14,100
13,900
14,200
14,000
13,900
13,900
13,700
13,600
14,900
14,800
15,000
14,900
14,800
15,000
15,500
15,400
15,500
14,700
14,700
14,600
14,300
14,800
14,700
14,900
14,200
14,200
14,300
336,000
340,000
334,000
327,000
316,000
354,000
244,000
310,000
278,000
432,000
308,000
217,000
230,000
217,000
200,000
179,000
197,000
233,000
236,000
245,000
288,000
298,000
328,000
332,000
355,000
294,000
318,000
312,000
248,000
188,000
202,000
207,000
236,000
292,000
341,000
351,000
160,000
193,000
189,000
649
649
655
658
658
660
657
660
665
665
666
719
726
743
729
732
742
742
747
758
691
687
681
684
684
681
658
660
655
710
706
706
716
704
695
684
756
730
716
15,400
15,400
15,300
15,200
15,200
15,200
15,200
15,200
15,000
15,000
15,000
13,900
13,800
13,500
13,700
13,700
13,500
13,500
13,400
13,200
14,500
14,600
14,700
14,600
14,600
14,700
15,200
15,200
15,300
14,100
14,200
14,200
14,000
14,200
14,400
14,600
13,200
13,700
14,000
0.19
0.35
0.18
0.46
0.33.
0.45
0.41
0.24
0.39
0.30
0.42
0.40
0.56
0.50
0.16
0.51
0.47
0.14
0.47
0.54
0.44
0.49
0.60
0.51
0.55
0.58
0.30
0.40
0.47
0.03
0.24
0.25
0.41
0.24
0.48
0.72
0.42
0.30
0.52
1.0
Toluene
DCM
1.9
1.4
Toluene
DCM
1.8
1.3
Toluene
cy
–
1.7
M4
DCM
1.6
Toluene
DCM
2.0
M5
1.7
Toluene
DCM
2.1
M634
2.5
Toluene
cy
4.0
3.9
M7
DCM
3.7
Toluene
cy
3.7
–
M8
DCM
3.7
Toluene
cy
3.7
–
M9
DCM
2.6
Toluene
cy
2.7
–
M10
M11
DCM
2.5
Toluene
cy
2.5
–
DCM
1.3
Toluene
cy
2.0
–
M12
DCM
0.19 (91.6)a, 1.1 (8.1)a, 6.2 (0.3)a
Toluene
DCM
1.7
–
M13b
M14c
Toluene
DCM
4.5
0.83
2.5
2.3
0.29
2.2
2.4
Toluene
cy
M15c
DCM
Toluene
cy
a
b
The values in brackets are relative amplitudes of the corresponding
lifetime.
Ref. 34.
Ref. 50.
c
under semiinfinite boundary conditions in solution. The poly-
merization does not have a strong impact on the potentials
of P5 as both reduction and oxidation values are in the same
range as those of the model monomer M10. Also the poten-
tials of P7 do not differ very much to those of the parent
monomer M12.
UV/Vis/NIR Absorption and Fluorescence Spectroscopy
UV/Vis/NIR-absorption and fluorescence spectra (if possible)
were recorded for all compounds in two solvents. The mono-
mers were measured in DCM and toluene, only the
carbazole-series was measured in DCM and cyclohexane. If
not stated elsewise, all the spectra in toluene are very
902
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 5 Absorption and Fluorescence Spectroscopic Data of Polymers
vmax (cm21
)
emax (M21 cm21
)
kfl (nm)
vfl (cm21
)
/
sfl (ns)
a
b
~
~
Solvent
k (nm)
fl
P1a
P1b
P2
CHCl3
739
x
13,500
x
183,000
x
749
x
13,400
x
0.28
x
1.0 (97.1), 8.1 (2.9)
x
Toluene
CHCl3
733
x
13,600
x
142,000
x
744
x
13,400
x
0.25
x
1.1 (97.0), 7.7 (3.0)
x
Toluene
CHCl3
729
738
682
682
726
735
685
x
13,700
13,500
14,700
14,700
13,800
13,600
14,600
x
160,000
122,000
181,000
x
759
766
705
701
748
755
698
x
13,200
13,100
14,200
14,300
13,400
13,200
14,300
x
0.07
0.10
0.14
0.22
0.09
0.20
0.28
x
0.88 (98.7), 10 (1.3)
2.0 (87.3), 6.2 (12.7)
1.3 (97.4), 5.0 (2.6)
2.1 (98.6), 8.7 (1.4)
1.7 (93.1), 6.4 (6.9)
0.84 (31.1), 3.5 (68.9)
1.3 (97.7), 5.0 (2.3)
x
Toluene
CHCl3
P3
Toluene
CHCl3
P4
151,000
90,000
247,000
x
Toluene
CHCl3
P5
Toluene
CHCl3
P6
676
677
691
693
14,800
14,800
14,500
14,400
179,000
152,000
151,000
x
694
692
711
709
14,400
14,500
14,100
14,100
0.22
0.39
0.02
0.16
1.6 (94.3), 5.5 (5.7),
2.1 (91.6), 6.0 (8.4)
0.58
Toluene
CHCl3
P7
Toluene
1.2 (93.4), 4.1 (6.6)
a
b
The values of the extinction coefficient are per monomer unit.
The values in brackets are relative amplitudes of the corresponding
x 5 insoluble or not completely soluble.
lifetime.
similar to those in DCM, only slightly shifted to lower energy.
Generally the fluorescence spectra are mirror images to the
main absorption bands and fluorescence quantum yields are
higher in toluene than in DCM, if the compounds fluoresce at
all. The polymers were measured in CHCl3 and, if possible,
toluene. All data of the monomers and polymers are found
in Tables 4 and 5, respectively. In some cases, we also deter-
mined the fluorescence lifetime.
acceptor. For the boron containing substituents conformational
issues certainly play a role which may inhibit efficient p-orbital
overlap. As has been demonstrated before for M13, replacing
one oxygen of the squaric acid moiety by a dicyanomethylene
group shifts the absorption maximum of M1 strongly to lower
energy. This is also visible in the squaraine derivatives M6-M8.
This effect is most apparent for the carbazolyl derivatives M8
versus M9 with a shift of 1000 cm21. The dicyanomethylene
group also leads to a cis-configuration of the indolenine groups
relative to the central squaric acid moiety while the usually
most stable configuration is trans.13,70
Squaraine Monomers
Magnified parts of the spectra of the nonfunctionalized (M1),
functionalized (M2-M4) and triarylborane substituted (M5)
squaraines in DCM are shown in Figure 5. All compounds in
common is the typical cyanine like structure of the absorption
band. A steep rise at the low energy side of the main absorp-
tion band with a huge extinction coefficient of e ꢀ 300,000
M21 cm21 and a small shoulder caused by vibronic progres-
sion at the high energy side are the dominant features of the
spectra. Nonfunctionalized squaraine M1 has the absorption
maximum at 15,700 cm21 in DCM. It appears that any substit-
uent (bromine, boronic ester, triarylborane, carbazolyl, pipera-
zinyl) shifts the absorption maximum to lower energy. For
piperazinyl this shift is strongest (1000 cm21 between M12
and M1 in DCM), followed by the carbazolyl substituents (800
cm21). Interestingly, the position of connection of the carba-
zolyl substituents (M9 vs. M10) does hardly influence the
absorption maximum. The shift of the piperazinyl substituent
in M12 surpasses that of the recently investigated squaraine
M14 with triarylamine substituents but does not reach that of
M15 with dianisylamino substituents. Even the formal p-elec-
tron acceptor/r-donor67–69 substituents triarylborane (M5)
and boronic ester (M4) shift the absorption maximum to lower
energies as well as bromine which acts as a p-donor and r-
The fluorescence quantum yield of monomeric squaraine
dyes are generally higher in nonpolar toluene or cyclohexane
than in DCM and range between about 20 and 60% for the
series investigated here, with both M9 and M10 having the
highest quantum yield (both ca. 60%) which again highlights
that the specific position of carbazolyl attachment is unim-
portant for the optical properties.
The fluorescence lifetimes were recorded in the ns time
regime. The samples were excited with a laser diode with a
fixed wavelength at 650 nm. Fluorescence lifetimes of all
monomeric squaraines are in the lower ns time regime and
are all monoexponential. They are slightly larger in toluene
and cyclohexane compared with DCM. Only M12 shows three
lifetimes in DCM. However, due to the low quantum yield of
0.03, the signal to noise ratio is not very good which limits
the accuracy in this case.
Polymers
For the polymers we measured absorption and fluorescence
spectra in CHCl3 because of better solubility than in DCM,
and in toluene.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
903

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 5 Fluorescence (left) and absorption spectra (right) of monomeric squaraine dyes M1-M5, M9-M12 (top) and M6-M8
(bottom).
FIGURE 6 Fluorescence (left) and absorption spectra (right) of homopolymers P1a/b and model dye M1.
904
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 7 Fluorescence (left) and absorption spectra (right) of carbazole-squaraine copolymers.
Homopolymers
any impact on the absorption (Fig. 7, right) or fluorescence
spectra (Fig. 7, left) compared with the model compounds
M9 and M10, respectively. The redshift of the absorption
maxima is only 100 to 300 cm21 to 14,800 cm21 (P6) and
14,600 cm21 (P5) and there is only little broadening of the
absorption band. This indicates only little electronic coupling
between the squaraine chromophores over the carbazole
bridges.
The homopolymers P1a/b show a broad main absorption
band, covering a range of ꢀ4000 cm21 with a distinct
intense absorption feature at the low energy side (Fig. 6,
right). The broadening and the shift of the absorption com-
pared with the monomer M1 (ꢀ 2100 cm21) can be
explained by simple excitonic coupling of localized transition
moments within the elongated polymer chains which show
mainly J-type behavior (5red shifted) with some H-type con-
tributions (5blue shifted).11,14,34 The maximum absorption
band of P1b is at 13,600 cm21, at only slightly lower energy
than P1a (13,500 cm21) which shows that the effective con-
jugation length is reached at a degree of polymerization
Xn 5 12. However, we will show below that this simple exci-
ton coupling model does not explain all optical features of
squaraine polymers.
The absorption band of the carbazole-dicyanomethylene
squaraine copolymer P4 is also rather narrow but at lower
energy (13,800 cm21) as expected for a dicyanomethylene
squaraine dye.
For the triarylamine-squaraine copolymers (P2, P3) we do
also find absorption spectra that are essentially similar to
the one of the monomeric dye conjugate (M1450 vs. P3) con-
cerning the peak with highest intensity (see Fig. 8). However,
for P3 in CHCl3 there is a small but significant band at lower
energy (13,800 cm21) which is absent in toluene. Polymer
P2 also shows a similar but even weaker absorption peak at
lower energy (12,700 cm21) than the most intense signal. In
addition, the signal at about 15,000 cm21 which usually
appears as a shoulder in all squaraine dyes is now more pro-
nounced and shows a distinct peak maximum.
In contrast to the monomeric compounds, the fluorescence
spectra of the homopolymers (Fig. 6, left) are no mirror
images to the absorption spectra. The shape is narrower and
similar to the monomeric fluorescence spectra but distinctly
shifted. The Stokes shift is as small as for the monomers
(100–300 cm21) and fluorescence obviously only occurs
from the lowest energy excitonic state with a quantum yield
of around 0.25. For the homopolymers we find biexponential
fluorescence decays with both lifetimes in the low ns regime.
The shorter lifetime of 1 ns is in the same range as those of
the monomers and contributes with 97%. The slightly longer
lifetime is around 8 ns.
The copolymer of squaraine and piperazine P7 also display
an absorption spectrum that is very similar to its monomer
M12 (Fig. 9, left). But again, a low energy band of low inten-
sity is found at 13,100 cm21 both in CHCl3 and in toluene.
These low energy bands of P2, P3, and P7 are plotted
together in Figure 9 (right) for better comparison. The origin
of these features is not fully understood but as they only
appear for copolymers of squaraines with amine donors we
suppose that they are caused by charge transfer interactions
which only occur in some conformers, possibly in those in
which the donor is coplanar to the squaraine dye. Semiem-
pirical computations indeed suggest a lowering of excitation
energy in such fully coplanar conformers, see below.
Considering the better solubility of the shorter polymer and
the nearly size-independent electrochemical properties (see
above) of squaraine homopolymers,14,34 the increase in
length of P1b may only affect morphological but not elec-
tronic properties.
Copolymers
Copolymerization of a squaraine dye with either a 3,6- or
2,7-carbazole to polymers P6 or P5, respectively, has hardly
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
905

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 8 Fluorescence (left) and absorption spectra (right) of triarylamine-squaraine M14 and copolymers P2 and P3 in toluene
and CHCl3.
The fluorescence spectra of all copolymers show essentially
mirror image behavior to the absorption spectra with excep-
tion of the weak low energy band of P2, P3, and P7 that do
not show any equivalent signal in the fluorescence spectra.
the amplitude whereas the shorter lifetime contributes with
one third.
All in common is the fact that the fluorescence lifetimes of
the related monomers are right between the two lifetimes of
the polymers. The lifetime found for P7 in CHCl3 is unreli-
able because of the low signal to noise ratio caused by the
low quantum yield.
The fluorescence quantum yield for P7 (0.02 in CHCl3 and
0.16 in toluene) is only slightly smaller than for the analo-
gous monomer M12 which is due to the saturated pipera-
zine bridge. All other polymers however show a much
smaller fluorescence quantum yield than their monomers,
e.g. for carbazole copolymer P5 the quantum yield is 0.21 in
CHCl3 while it is 0.51 in DCM for M10.
To sum up, the optical behavior of all the copolymers is
rather similar, no matter if the bridge is saturated (pipera-
zine) or unsaturated, rigid (carbazole) or propeller like (tri-
arylamine) and does not differ very much from those of the
analogous monomers. These kinds of bridges obviously do
not provide any significant coupling of individual squaraine
chromophores by p-conjugation and the distance of chromo-
phores is widened in such a way that exciton coupling
decreases to an extent which is not noticeable.
For all copolymers we find biexponential fluorescence
decays. The shorter lifetimes are between 0.8 and 2.1 ns
and dominate the decay with amplitudes exceeding 90%.
The larger lifetimes are between 3.5 and 10 ns. However,
for P4 in toluene the larger lifetime makes two thirds of
FIGURE 9 Fluorescence (dotted line) and absorption spectra (solid line) of piperazine-squaraine monomer M12 and copolymer P7
(left) and complete absorption spectra of donor-squaraine copolymers P2, P3 and P7 (right).
906
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
A similar behavior was reported by Kuster and Geiger for
indolenine-squaraines.46 They synthesized a series of linear
squaraine dimers and oligomers and investigated the impact
of various bridges. They also found only little broadening
and small bathochromic shifts of the absorption bands. The
induced red-shift of the bridges was in the order
tetrafluorophenylene< 2,5-dialkyloxyphenylene< phenylene <
fluorene. Maeda et al. bridged indolenine-squaraines via
phenylene units.45 These dyes were linked in para- or meta-
positions also leading to little broadening and red-shift which
was more pronounced for the para-linked system.
of 4.46 eV versus vacuum.63 Therefore, in DCM E(HOMO/
LUMO) 5 25.16 eV – E1/2(1st ox/1st red) and the electro-
chemically derived band gap is E(gap)CV 5 E(LUMO) 2
E(HOMO). In THF, 25.16 eV is replaced by 25.26 eV.
To obtain the optical band gap E(gap)opt we used the onset
of the lowest energy transition on the low energy side in the
absorption spectra. Therefore we calculated the first deriva-
tion of the absorption spectra and placed a tangent at the
inflexion point. The intersection point of the tangent with
the x-axis in eV gave the optical band gap E(gap)opt
.
The values of the HOMOs all are between 24.95 eV for
squaraines with strong electron donors and 25.32 eV. The
LUMO values are between 23.35 eV and 23.71 eV for
squaraines with the strong electron accepting dicyanomethy-
lene group in the center. The differing values for M6 can be
attributed to the THF solvent in the cyclic voltammetry
measurements. The band gaps of all squaraine conjugates
are between 1.52 eV and 1.88 eV. Comparison of the band
gaps determined by optical and electrochemical methods
shows that the latter values are 0.05 to 0.13 eV lower than
the optical derived values. Only for P1b the optical band gap
is 0.1 eV smaller than the electrochemical band gap. For M1,
M5 and M6 different solvents were used for the cyclic vol-
tammetry (THF) and the absorption spectroscopy (DCM), for
this reason a direct comparison should be taken with care in
these cases. Also for the polymers different solvents were
used for cyclic voltammetry (DCM) and absorption spectros-
copy (CHCl3). However, the optical properties are rather sim-
ilar for our systems in DCM and CHCl3 and for this reason
the band gaps are also in good agreement. In the cyclic
voltammetry and absorption section above, we already dis-
cussed that no major impact on the optical or electrochemi-
cal properties is observed upon polymerization. This is also
reflected in the optical and electrochemical band gaps of
polymers versus monomers.
Also thiophenes were used as bridges in squaraine oligo-
mers.11 Much in contrast to the observations stated above,
both a broadening as well as a significant red-shift was
observed. The absorption maximum of a pentamer was
found at 13,800 cm21 in CHCl3 compared with 14,900 cm21
for a dithienyl substituted indolenine squaraine monomer. In
contrast to phenylene bridges in the work of Kuster and Gei-
ger (and the carbazole and triphenylamine bridges used in
this study) which show a computed torsional angle of 36 to
41ꢁ, thiophene bridges may adopt a more coplanar confor-
mation with the squaraine p-system thus enhancing p-conju-
gation. Semiempirical AM1 computations for model
hexamers indeed gave dihedral angles of 22 to 32ꢁ, see
below.
However, for N-alkylpyrrole squaraine dyes, significant
broadening and red-shift of the absorption was reported
upon copolymerization with an electron rich bridge. Ajaya-
ghosh and Eldo used 2,5-dialkoxydivinylbenzene to bridge N-
alkylpyrrole squaraines to form conjugated polymers with
intense and broad absorption in the NIR region and low
band gaps.41,42 Into that system, Zhang et al. inserted addi-
tional donors such as thiophene derivatives, 2,7-carbazole or
another 2,5-dialkoxyphenylene via the Sonogashira coupling
and also obtained low band gap polymers with broad
absorption in the NIR region.43,44 In all these cases the con-
jugation pathway is provided by vinylene groups (and
ethynyl groups in the latter cases) which allow for a better
coplanar orientation of aromatic and heterocyclic ring
moieties.
Semiempirical Calculations
In order to gain insight into the electronic structure of the
polymers, the interaction of monomers and in particular the
interaction with bridge molecules in the copolymers we per-
formed semiempirical calculations. The structures of model
hexamers were first optimized using the AM1 hamiltonian71
which is known to give reliable geometries for the squara-
ines.34 In all cases we used the most elongated structure
that we were able to model and other conformers may have
more strongly bent zig-zag structures. Although the effective
conjugation length may not be reached with hexamers, it is
pretty close to. In order to probe the effect of direct conjuga-
tion, we modified the dihedral angle between the squaraine
monomers in the homo-hexamer and between the squaraine
monomers and the bridge molecules in the copolymers.
Thus, a dihedral angle of 0ꢁ will enhance conjugation, an
angle of 90ꢁ will inhibit conjugation and will leave excitonic
coupling—besides some hyperconjugative effects—to be the
only mechanism at work for interaction. Using the thus
obtained geometries, we performed CNDO/S2 calculations72
So far, as described above, the broadening and red-shift is
stronger in our homopolymers P1a/b, where no bridges
are used and the squaraine chromophores are directly
linked via a C-C bond between the indolenine rings. Thus,
taken together the information given above, the exciton
coupling model to explain the optical properties should be
modified to include conjugational effects at least in some
cases.
Band Gaps
HOMO and LUMO values were obtained from the half wave
potentials of cyclic voltammetry measurements. The poten-
tial of Fc/Fc1 in TBAHFP/DCM is 0.46 eV and in TBAHFP/
THF 0.56 eV versus the saturated calomel electrode (SCE).61
Furthermore, the potential of SCE is 0.244 versus the normal
hydrogen electrode (NHE), which has an absolute potential
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
907

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 6 CNDO/S2 Excited State Energies and Oscillator Strengths Based on AM1 Optimized Geometries as well as Electronic Cou-
pling V in cm21
Polymer
optimizedc
Polymer dihedral
Polymer dihedral
Monomera
Dimerb
angle 5 0ꢁ
angle 5 90ꢁ
21
21
21
21
21
~
~
~
~
~
v (cm
)
f
v (cm
)
f
v (cm
)
f
v (cm
)
f
v (cm
)
f
Without bridge
16,678
1.79
16,500
3.92
16,437
9.79
15,997
10.39
16,817
6.67
P1 (M1)
18,065
0.19
17,019
17,682
18,346
18,886
19,122
671
0.07
0.74
0.01
0.32
0.80
16,655
17,429
18,199
18,823
19,095
774
0.04
0.81
0.00
0.35
0.00
17,341
17,922
18,435
18,831
18,906
522
0.10
0.78
0.24
0.91
3.08
V !
783
Thiophene
16,659
2.28
16,365
4.24
0.29
16,774
10.23
16,449
11.22
17,544
5.04
bridge Py
17,412
17,093
17,594
18,002
18,358
18,541
442
0.23
0.44
0.03
0.16
1.88
16,873
17,380
17,876
18,274
18,475
507
0.04
0.61
0.04
0.23
0.82
17,785
18,050
18,316
18,568
18,723
295
0.25
1.55
0.20
0.42
4.82
V !
524
Carbazole bridge
17,105
2.35
17,211
4.44
0.13
17,828
11.30
17,532
11.74
18,331
10.45
P5 (M10)
17,663
17,978
18,156
18,306
18,452
18,507
170
0.36
0.49
0.01
0.22
0.36
17,730
17,939
18,145
18,316
18,387
214
0.11
0.58
0.02
0.20
0.55
18,407
18,490
18,568
18,630
18,670
85
0.46
0.62
0.00
0.12
0.21
V !
226
a
c
The monomer consists of the sequence bridge-squaraine-bridge.
The dimer consists of the sequence bridge-squaraine-bridge-
The polymer consists of the sequence bridge-[squaraine-bridge]6.
b
squaraine-bridge whose structure was taken from the optimization of
the polymer.
to get the excited states. For infinitely long polymers, the
nearest-neighbor approximation leads to an exciton band-
width that is four times the electronic interaction energy V.73
Thus, we estimated V by (E2 2 E1)/4 where E1 and E2 are
the energies of the lowest and highest excited state of the
excitonic manifold, respectively. Furthermore, we assume the
relative orientation of the squaraine chromophores in the
model hexamers has only little influence on the electronic
interaction because we investigate stretched conformers
only. In these conformers the angle between the chromo-
phores approaches 180ꢁ leading to a J-type exciton behavior
for which only transitions into the lowest excited state of the
exciton manifold is optically allowed. As representative
examples we chose the homopolymer P1, the carbazole
bridged copolymer P5 and the thiophene bridged copolymer
Py that has been investigated in form of its monodisperse
pentamer by Scherer et al.11 The AM1 optimizations gave, as
expected, planar squaraines and dihedral angles between the
squaraine dyes of 39 to 41ꢁ for P1, and between the squar-
aine dyes and the bridging group of 22 to 32ꢁ for Py and 40
to 46ꢁ for P5. The CNDO/S2 calculated transition energies
and oscillator strengths of these model hexamers together
with those of the monomers (squaraines, end-capped with
two bridge units) and of model dimers are given in Table 6.
The CNDO/S2 calculations of the monomeric model com-
pounds gave lowest energy transitions around 17,000 cm21
which is about 2000 cm21 too high in energy compared with
experiment. This is obviously a systematic problem of the
CNDO/S2 parameterization. In the dimeric model compounds,
exciton coupling yields two excited states, the one being lower
in energy is an allowed transition with the twofold oscillator
strength, the other one at higher energy is forbidden and
shows almost no oscillator strength. Half of the splitting of the
two states refers to the exciton coupling V. For the model hexa-
mers the CNDO/S2 computations gave six lowest energy tran-
sitions in a close energy range which form the exciton
manifold. The lowest energy transition carries almost all of the
908
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
oscillator strength which is thus about sixfold of the one of the
respective monomer. This situation thus resembles a typical J-
aggregate behavior. For the homopolymer P1 the CNDO/S2 cal-
culation predicts an only slight decrease of lowest energy com-
pared with the monomer squaraine (see Table 6). This
contrasts the experimental behavior where we found a quite
remarkable red-shift of 2100 cm21. However, the calculated
exciton coupling energy V 5 671 cm21 is in good agreement
with one fourth of the experimentally observed band width
(640 cm21 from the peak to peak width of the highest and
lowest energy peak in Fig. 6). The electronic coupling derived
from the dimer (V 5 783 cm21) is somewhat larger which indi-
cates that the model hexamer does not quite reach the full
polymer properties concerning the full exciton band width of
4 3 V. Artificial planarization of the polymer leads to an
increase of apparent exciton coupling energy (774 cm21) com-
pared to the one of the optimized hexamer (671 cm21) and
90ꢁ orientation between the squaraine dyes to a decrease to
522 cm21. From these values it is apparent that about 78% of
the electronic coupling of the optimized polymer stems from
exciton coupling interactions and the remainder from
conjugation.
might even be confined to only the monomer. However, even
in the case of the 90ꢁ-twisted polymer where conjugation is
blocked, the CNDO/S2 calculations indicate a significant exci-
ton coupling interaction. Particularly in this case, the semi-
empirical computations predict a hypsochromic shift of the
allowed lowest energy transition upon polymer formation.
Thus, from a missing bathochromic shift, one cannot con-
clude about the absence of exciton coupling interactions
because the whole excitonic manifold can be shifted to
higher energies, for what reasons so ever. Thus, it remains to
be elucidated how the lowest energy transition within the
exciton manifold depends on the degree of excitonic interac-
tion and the degree of polymerization.
CONCLUSIONS
We followed the idea that the insertion of a strong electron
donor into a polymeric squaraine system would lead to an
intense and broad absorption in the NIR region. Therefore,
we synthesized a series of donor-squaraine copolymers and
monomeric model compounds using diverse synthetic
strategies. No matter if we used transition metal catalyzed/
mediated or polycondensation reactions, polymers could
always be obtained in sufficient yield and molecular weight.
The situation in the thiophene-bridged Py is quite similar,
however with a smaller electronic coupling which is about
2/3 of P1. The part which is caused by exciton coupling
interactions drops to about 66%. In the carbazole-bridged
We investigated the impact of various bridging units on the
spectroscopic and electrochemical properties. For this pur-
pose, we chose saturated (piperazine), unsaturated rigid (3,6-
and 2,7-carbazole) and propeller like (triarylamine) groups.
While the absorption maxima of the model compounds gener-
ally shift bathochromically upon extension of the p-system at
the periphery or the addition of donors (both r and p),
copolymerization of a squaraine dye and a donor did not
have a major impact on the absorption spectra. These were
only marginally broadened and red-shifted compared with the
model compounds. Furthermore, the electrochemical proper-
ties remained rather similar to their monomeric model conju-
gates. This behavior was observed for all bridges we used. We
explain this behavior to be caused by the decrease of exciton
coupling energy with the power-3-dependence on the chro-
mophore distance provoked by insertion of the bridging
groups. However, conjugative effects also play a role for the
interchromophore interactions and may reach about 50% for
longer squaraine separations as in P5.
1
P5 the electronic coupling is distinctly smaller (ca. = of P1)
4
and again, the exciton contribution drops to 50%.
1
jlDlA
V
dd5
(1)
hc4pe0 rD3A
The decreasing exciton contribution from P1 > Py > P5 can
be explained by the power-3-dependence of V on rDA in the
point-dipole approximation according to eq. 174 where
j5eDeA23ðeDeDAÞðeAeDAÞ is an orientation factor which can
be evaluated by eD and eA, which are the unit vectors (orien-
tation) of the transition moments lD and lA being localized
at the chromophores D and A, respectively, and the vector
eDA connecting the two. lD 5 lA is the transition moment of
a squaraine monomer.
Thus, the CNDO/S2 calculations can explain the small
changes in the carbazole-squaraine copolymer compared
with the monomer model compound and the distinctly stron-
ger coupling in Py and P1. However, the fact that both P1
and Py (but not P5) show distinctly lower energies of their
lowest transitions than the respective monomer is still unex-
plained. However, for all polymers, the lowest transition
energy is somewhat lower for the 0ꢁ conformer than in the
optimized geometry which supports the assumption that
conformational issues are responsible for the small low
energy peaks seen in the spectra of Figure 9, right.
ACKNOWLEDGMENTS
This work was supported by the Deutsche Forschungsgemein-
schaft within the Forschergruppe 1809 and by the the
Bavarian State Ministry for Science, Research and the Arts
within the collaborative research project “Solar Technologies
Go Hybrid.”
REFERENCES AND NOTES
It is tempting to argue that the experimentally observed
missing shift of lowest energy absorption of e.g. P5 versus
M10 is caused by a small effective conjugation length which
1 K. -Y. Law, Chem. Rev. 1993, 93, 449–486.
2 M. Emmelius, G. Pawlowski, H. W. Vollmann, Angew. Chem.
Int. Ed. Engl. 1989, 28, 1445–1471.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
909

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
3 J. J. Gassensmith, E. Arunkumar, L. Barr, J. M. Baumes, K.
M. DiVittorio, J. R. Johnson, B. C. Noll, B. D. Smith, J. Am.
Chem. Soc. 2007, 129, 15054–15059.
30 A. Ajayaghosh, Acc. Chem. Res. 2005, 38, 449–459.
31 S. Sreejith, P. Carol, P. Chithra, A. Ajayaghosh, J. Mater.
Chem. 2008, 18, 264–274.
4 E. Arunkumar, N. Fu, B. D. Smith, Chem. Eur. J. 2006, 12,
4684–4690.
32 A. L. Tatarets, I. A. Fedyunyayeva, T. S. Dyubko, Y. A.
Povrozin, A. O. Doroshenko, E. A. Terpetschnig, L. D.
Patsenker, Anal. Chim. Acta 2006, 570, 214–223.
5 M. Sameiro, T. Goncalves, Chem. Rev. 2009, 109, 190–212.
33 A. L. Tatarets, I. A. Fedyunyaeva, E. Terpetschnig, L. D.
Patsenker, Dyes Pigm. 2005, 64, 125–134.
6 Z. M. Xiang, E. E. Nesterov, J. Skoch, T. Lin, B. T. Hyman, T.
M. Swager, B. J. Bacskai, S. A. Reeves, J. Histochem. Cyto-
chem. 2005, 53, 1511–1516.
€
34 S. F. Volker, C. Lambert, Chem. Mater. 2012, 24, 2541–2553.
7 G. J. Ashwell, J. Ewington, K. Moczko, J. Mater. Chem. 2005,
15, 1154–1159.
35 S. H. Kim, S. K. Han, J. J. Kim, S. H. Hwang, C. M. Yoon, S.
R. Keum, Dyes Pigm. 1998, 39, 77–87.
8 E. Collini, S. Carlotto, C. Ferrante, R. Bozio, A. Polimeno, J.
Bloino, V. Barone, E. Ronchi, L. Beverina, G. A. Pagani, Phys.
Chem. Chem. Phys. 2011, 13, 12087–12094.
36 A. Treibs, K. Jacob, Angew. Chem. Int. Ed. Engl. 1965, 4,
694–695.
37 A. Treibs, K. Jacobs, Liebigs Ann. Chem. 1966, 699, 153–
9 Q. Q. Shi, W. Q. Chen, J. F. Xiang, X. M. Duan, X. W. Zhan,
Macromolecules 2011, 44, 3759–3765.
167.
38 H. -C. Lu, W. -T. Whang, B. -M. Cheng, Synth. Met. 2010,
160, 1002–1007.
10 S. Webster, J. Fu, L. A. Padilha, O. V. Przhonska, D. J.
Hagan, E. W. Van Stryland, M. V. Bondar, Y. L. Slominsky, A.
D. Kachkovski, Chem. Phys. 2008, 348, 143–151.
39 H. C. Lu, W. T. Whang, B. M. Cheng, Chem. Phys. Lett. 2010,
500, 267–271.
€
11 D. Scherer, R. Dorfler, A. Feldner, T. Vogtmann, M.
40 J. Y. Wu, E. F. Huo, Z. B. Wu, Z. Y. Lu, M. G. Xie, Q. Jiang,
e-Polymers 2007, 077.
Schwoerer, U. Lawrentz, W. Grahn, C. Lambert, Chem. Phys.
2002, 279, 179–207.
41 A. Ajayaghosh, J. Eldo, Org. Lett. 2001, 3, 2595–2598.
42 J. Eldo, A. Ajayaghosh, Chem. Mater. 2002, 14, 410–418.
12 T. Maeda, H. Nakao, H. Kito, H. Ichinose, S. Yagi, H.
Nakazumi, Dyes Pigm. 2011, 90, 275–283.
43 W. Zhang, F. Tao, K. G. Meng, Z. Wang, L. Y. Xi, Y. Li, Q.
Jiang, J. Mater. Sci. 2011, 46, 5363–5370.
€
13 U. Mayerhoffer, K. Deing, K. Gruß, H. Braunschweig, K.
Meerholz, F. Wurthner, Angew. Chem. Int. Ed. Engl. 2009, 48,
€
44 W. Zhang, F. Tao, K. G. Meng, L. Y. Xi, Z. Wang, Y. Li, Q.
Jiang, Polym. Bull. 2012, 68, 349–360.
8776–8779.
€
14 S. F. Volker, S. Uemura, M. Limpinsel, M. Mingebach, C.
Deibel, V. Dyakonov, C. Lambert, Macromol. Chem. Phys. 2010,
211, 1098–1108.
45 T. Maeda, T. Tsukamoto, A. Seto, S. Yagi, H. Nakazumi,
Macromol. Chem. Phys. 2012, 213, 2590–2597.
46 S. Kuster, T. Geiger, Dyes Pigm. 2012, 95, 657–670.
47 R. Sens, K. H. Drexhage, J. Lumin. 1981, 24–5, 709–712.
48 Innovative Instruments, Inc. Tampa, USA.
15 S. Y. Wang, L. Hall, V. V. Diev, R. Haiges, G. D. Wei, X. Xiao,
P. I. Djurovich, S. R. Forrest, M. E. Thompson, Chem. Mater.
2011, 23, 4789–4798.
16 G. Wei, S. Wang, K. Renshaw, M. E. Thompson, S. R.
Forrest, ACS Nano 2010, 4, 1927–1934.
49 I. Noviandri, K. N. Brown, D. S. Fleming, P. T. Gulyas, P. A.
Lay, A. F. Masters, L. Phillips, J. Phys. Chem. B 1999, 103,
6713–6722.
17 G. Wei, X. Xiao, S. Wang, K. Sun, K. J. Bergemann, M. E.
Thompson, S. R. Forrest, ACS Nano 2012, 6, 972–978.
€
50 S. F. Volker, M. Renz, M. Kaupp, C. Lambert, Chem. Eur. J.
2011, 17, 14147–14163.
18 G. Wei, X. Xiao, S. Wang, J. D. Zimmerman, K. Sun, V. V. Diev,
M. E. Thompson, S. R. Forrest, Nano Lett. 2011, 11, 4261–4264.
51 S. Yamaguchi, T. Shirasaka, K. Tamao, Org. Lett. 2000, 2,
4129–4132.
19 G. D. Wei, R. R. Lunt, K. Sun, S. Y. Wang, M. E. Thompson,
S. R. Forrest, Nano Lett. 2010, 10, 3555–3559.
52 G. R. Freeman, H. A. Levy, G. M. Brown, Acta Crysallogr. A
1969, 25, S145–S146.
ꢀ
ꢀ
20 J. V. Ros-Lis, R. Martınez-Manez, J. Soto, Chem. Commun.
53 A. Midya, Z. Xie, J.-X. Yang, Z.-K. Chen, D. J. Blackwood, J.
Wang, S. Adams, K. P. Loh, Chem Commun (Camb) 2010, 46,
2091–2093.
2002, 2248–2249.
ꢀ
ꢀ
21 J. V. Ros-Lis, M. D. Marcos, R. Martınez-Manez, J. Soto,
Angew. Chem. Int. Ed. Engl. 2005, 44, 4405–4407.
54 M. Tavasli, S. Bettington, M. R. Bryce, A. S. Batsanov, A. P.
Monkman, Synthesis 2005, 1619–1624.
ꢀ
ꢀ
ꢀ
22 J. V. Ros-Lis, R. Martınez-Manez, F. Sancenon, J. Soto, M.
Spieles, K. Rurack, Chem. Eur. J. 2008, 14, 10101–10114.
55 D. Patra, D. Sahu, H. Padhy, D. Kekuda, C. W. Chu, H. C. Lin,
J. Polym. Sci. Part A: Polym. Chem. 2010, 48, 5479–5489.
23 A. Ajayaghosh, E. Arunkumar, J. Daub, Angew. Chem. Int.
Ed. Engl. 2002, 41, 1766–1769.
56 N. Berton, I. Fabre-Francke, D. Bourrat, F. Chandezon, S.
Sadki, J. Phys. Chem. B 2009, 113, 14087–14093.
24 C. R. Chenthamarakshan, A. Ajayaghosh, Tetrahedron Lett.
1998, 39, 1795–1798.
57 C. R. Chenthamarakshan, J. Eldo, A. Ajayaghosh, Macromo-
lecules 1999, 32, 251–257.
25 S. Sreejith, K. P. Divya, A. Ajayaghosh, Angew. Chem. Int.
Ed. Engl. 2008, 47, 7883–7887.
58 Y. -Y. Chen, J. H. K. Hall, Polym. Bull. 1986, 16, 419–425.
59 A. H. Schmidt, Synthesis 1980, 961–994.
ꢀ
ꢀ
ꢀ
26 J. V. Ros-Lis, B. Garcia, D. Jimenez, R. Martınez-Manez, F.
ꢀ
Sancenon, J. Soto, F. Gonzalvo, M. C. Valldecabres, J. Am.
Chem. Soc. 2004, 126, 4064–4065.
60 A. Treibs, K. Jacob, Liebigs Ann. Chem. 1968, 712, 123–137.
61 N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910.
27 S. Saito, T. L. Massie, T. Maeda, H. Nakazumi, C. L. Colyer,
Sensors 2012, 12, 5420–5431.
62 V. V. Pavlishchuk, A. W. Addison, Inorg. Chim. Acta 2000,
298, 97–102.
28 L. Beverina, P. Salice, Eur. J. Org. Chem. 2010, 1207–1225.
29 S. Yagi, H. Nakazumi, Topics Heterocyclic Chem. 2008, 14,
133–181.
63 D. Tsiplakides, D. Archonta, C. G. Vayenas, Top. Catal. 2007,
44, 469–479.
910
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
64 R. Stahl, C. Lambert, C. Kaiser, R. Wortmann, R. Jakober,
Chem. Eur. J. 2006, 12, 2358–2370.
69 M. E. Glogowski, J. L. R. Williams, J. Organomet. Chem.
1981, 218, 137–146.
65 J. F. Morin, M. Leclerc, D. Ades, A. Siove, Macromol. Rapid
Commun. 2005, 26, 761–778.
70 T. Lin, B. X. Peng, Dyes Pigm. 1999, 43, 73–76.
71 J. J. P. Stewart, MOPAC2009, Stewart Computational Chem-
istry, Colorado Springs, CO, USA, HTTP://openMOPAC.net,
2008.
66 C. J. Xia, R. C. Advincula, Chem. Mater. 2001, 13, 1682–
1691.
67 C. D. Entwistle, T. B. Marder, Angew. Chem. Int. Ed. Engl.
2002, 41, 2927–2931.
72 WinMOPAC, Fujitsu Ltd., 1997–1998.
73 M. Kasha, E. G. McRae, Phys. Processes Radiat. Biol. 1964,
68 C. D. Entwistle, T. B. Marder, Chem. Mater. 2004, 16, 4574–
23–42.
4585.
€
74 T. Forster, Annalen Der Physik 1948, 2, 55–75.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 890–911
911