Coupled Oscillators for Tuning Fluorescence Properties of Squaraine Dyes

Article abstract of DOI:10.1021/ja512338w

Combining a squaraine (S) and a BODIPY (B) chromophore in a heterodimer (SB) and two heterotrimers (BSB and SBS) by alkyne bridges leads to the formation of coupled oscillators whose fluorescence properties are superior compared to the parent squaraine ch

Full text of DOI:10.1021/ja512338w

Article
pubs.acs.org/JACS
Coupled Oscillators for Tuning Fluorescence Properties of Squaraine
Dyes
Christoph Lambert,* Thorsten Scherpf, Harald Ceymann, Alexander Schmiedel, and Marco Holzapfel
Institut fur Organische Chemie, Universitat Wurzburg, Wilhelm Conrad Rontgen Research Center for Complex Material Systems,
̈
̈
̈
̈
Center for Nanosystems Chemistry, Am Hubland, 97074 Wurzburg, Germany
̈
S
* Supporting Information
ABSTRACT: Combining a squaraine (S) and a BODIPY (B)
chromophore in a heterodimer (SB) and two heterotrimers (BSB and
SBS) by alkyne bridges leads to the formation of coupled oscillators
whose fluorescence properties are superior compared to the parent
squaraine chromophore. The lowest energy absorption and emission
properties of these superchromophores are mainly governed by the
squaraine part and are shifted by more than 1000 cm−1 to the red by
excitonic interaction between the squaraine and the BODIPY dye.
Employing polarization-dependent transient absorption and fluorescence
upconversion measurements, we could prove that the lowest energy
absorption in SB and BSB is caused by a single excitonic state but by two
for SBS. Despite the spectral red-shift of their lowest absorption band, the
fluorescence quantum yields increase for SB and BSB compared to the
parent squaraine chromophore SQA. This is caused by intensity
borrowing from the BODIPY states, which increases the squared transition moments of the lowest energy band dramatically
by 29% for SB and 63% for BSB compared to SQA. Thereby, exciton coupling leads to a substantial enhancement of fluorescence
quantum yield by 26% for SB and by 46% for BSB and shifts the emission from the red into the near-infrared. In this way, the
BODIPY-squaraine conjugates combine the best properties of each class of dye. Thus, exciton coupling in heterodimers and
-trimers is a valuable alternative to tuning fluorescence properties by, e.g., attaching substituents to chromophores.
only one type of chromophore is employed; i.e., homoag-
gregates are considered. Much less is known about hetero-
chromophore assemblies in which two or more different
chromophores are incorporated, although this kind of dye
assembly may open the chance to tune the SC properties
considerably.19,20
INTRODUCTION
■
Fluorescence dyes emitting in the red or near-infrared with high
quantum yield are highly sought after because of many
technological1−3 and medical applications.4,5 In this context,
tuning absorption and fluorescence energy characteristics of
organic chromophores by attaching small substituents to a
parent chromophore structure is a very popular but also limited
method.6,7 Often, broadening of absorption and/or decrease of
fluorescence quantum yield are unwanted side effects.8 Finding
completely new chromophore archetypes requires extensive
synthetic work and is also not easy to achieve.9 Therefore, we
follow an alternative route in which we combine different
chromophores to form covalently bound superchromophores
(SC).10 In these SCs the transition moments of individual
chromophores are coupled (exciton coupling) to produce new
coupled oscillators whose resonance energy may deviate
strongly from those of the individual chromophores.11 Nature
uses this concept in light harvesting employing huge complexes
of chlorophyll molecules in the chromosomes of green bacteria.
Artificial analogues are the J-aggregates of cyanine dyes whose
enhanced coherent excitation and emission properties have
been studied for decades.12−16 This pretty old concept17 has
recently experienced a revival because the formation of J-
aggregates may lead to what is nowadays frequently called
“aggregation induced emission” (AIE).18 In all these examples
In this paper we will investigate the photophysical properties
of heterodimer and heterotrimer SCs built up from an
indolenine squaraine dye and a BODIPY dye. The former
were chosen as they display sharp and intense absorption bands
in the red to near-infrared spectral region21−24 and they are
being used in numerous applications such as in biolabel-
ing,25−34 ion recognition,35−39 nonlinear optics40−47 and solar
energy conversion.48−72 BODIPY dyes are even more popular
as fluorescence dyes absorbing in the green spectral region.
They also show a relatively intense and narrow absorption
band.73−76 These are ideal conditions for investigating
chromophore interactions in SCs because coupling induced
shifts will easily be visible for strong and narrow absorption
bands. In particular, we focus on the squaraine-BODIPY
heterodimer SB, and the heterotrimers SBS and BSB (for their
synthesis, see Supporting Information (SI)). In these SCs, the
Received: December 3, 2014
Published: March 4, 2015
© 2015 American Chemical Society
3547
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
Figure 1. BODIPY and squaraine parent chromophores and heterodye conjugates along with their respective orientation of transition moments of
the lowest energy localized states.
individual dyes were held together by triple bonds, which
reduce flexibility and leads to fixed interchromophore distances
although different conformers may result from rotation around
the triple bonds. The steady-state and time-resolved optical
properties will be investigated and compared to those of the
parent chromophores SQA10 and the BODIPY dyes BODIPY-1
and BODIPY-277 (see Figure 1).
homodimer that gives two excitonic states, split by twice the
exciton coupling energy, J. The relative phase relations of the
localized transition moments make the lowest exciton state
being an allowed transition and the upper exciton state being
forbidden. This scenario refers to the well-known J-type
aggregate behavior.10,15,78
δEdimer = 2 ΔE2 + J2
(1)
RESULTS AND DISCUSSION
■
3hcε0 ln 10
ε(ν)
̃
9n
μ2 =
dν
̃
∫
The synthesis of the heterodimer SB and the two heterotrimers
BSB and SBS was accomplished by copper-free Sonogashira
reaction of appropriate bromine-substituted squaraines and
alkyne-terminated BODIPY dyes (Scheme 1, see SI for details).
The copper-free variant of Sonogashira reaction was used
because of less formation of homocoupling products, which are
difficult to remove from the target compounds.
The steady state absorption spectra of all SCs and their
parent compounds in toluene are given in Figure 2a−c. The
spectrum of SQA is typical of squaraine dyes and will serve as a
reference. The spectrum is quite narrow (fwhm = 560 cm−1)
and has a weak vibronic shoulder to the higher energy side. The
spectrum of the heterodimer SB displays a strong absorption at
15 000 cm−1 similar to SQA but red-shifted by 500 cm−1. A
second, much weaker band at 19 000 cm−1 is visible that is
shifted by 100 cm−1 compared to that of BODIPY-1, which we
consider to be the most suitable dye for comparison. This
drifting apart of electronic transitions is caused by exciton
coupling of localized transition moments at the squaraine and
BODIPY chromophore moieties, respectively. Figure 3 sketches
the electronic situation for the head-to-tail arrangement of a
2000π2N (n2 + 2)2
ν
̃
(2)
In heterodimers, the excitonic levels can be evaluated in the
usual manner and are split by δEdimer (eq 1), which is larger
than 2J of the homodimer. In this equation, 2ΔE is the energy
difference between the diabatic (noninteracting) energy levels
of the two chromophore units in the heterodimer. As zero
energy, we take the midpoint between these diabatic levels.
Because the transition moments are in general unequal for
heterodimers, they do not cancel for the upper excitonic level in
the head-to-tail arrangement, thus both states, the lower and
the upper are allowed. This is exactly what we observe for SB.
Interestingly, while the sum of the squared transition moments
μ2 (calculated by integration of the spectra, see eq 2) for the
squaraine and the BODIPY band are roughly that of the sum of
the parent chromophores (189 D2 vs 160 D2), the one of the
squaraine band in SB is larger (159 D2 vs 123 D2) and the one
of the BODIPY band in SB is smaller (30.4 D2 vs 36.5 D2) than
those of the respective parent compounds (see Table 1). This
intensity borrowing is caused by the phase relations of the
transition moments, which intensify the lower state on the
3548
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
Scheme 1. Synthesis of SB, BSB and BSB by Copper-Free Sonogashira Reaction
expense of the upper state in case of a head-to-tail
arrangement.79,80 The fact that SB shows an angle of ca. 120°
between the chromophore long axes leads to some modification
of intensity and orientation of resulting transition moments but
we consider the head-to-tail model as qualitatively correct for
our purposes. From the absorption maxima of SQA and
state diagram in Figure 3 with three exciton states and a state
splitting of δEtrimer (see eq 4) for the lower and upper state,
which is in case of identical ΔE and J values larger than that of
the heterodimer. In the same way as for SB we estimate |J| =
1100 cm−1 by eq 4 for the exciton coupling energy in BSB,
which is in excellent agreement with that in SB. Exciton
coupling theory also predicts that the lowest exciton level
carries most of the oscillator strength, the middle state is
forbidden and the upper states carries less oscillator strength
(for a compilation of all eigenvalues and eigenvectors, see SI).
At this point we would like to address an issue that has
frequently led to misunderstandings: The exciton coupling
energy is a quantity that enters in, e.g., eq 3, and which leads to
a certain splitting/shifting of excited energy levels. With
increasing energy difference 2ΔE between the diabatic states,
the influence of J on this splitting becomes weaker but not the
coupling energy itself, as is obvious when looking at eq 4.
Therefore, the fact that many heterochromophore aggregates
show only little energy displacements compared to their parent
chromophores does not allow concluding on weak coupling, it
may also be caused by a large 2ΔE. However, comparing eq 1
with eq 4 shows that the effect of J on the splitting is larger for
the trimer than for the dimer because of the 2J2 term in the
root. For a heteropolymer this transforms into a 4J2 term.10
BODIPY-1 we estimate ΔE = (vmax(BODIPY-1) −
̃
v
max(SQA))/2 = 1700 cm−1 and with the experimental state
̃
splitting of SB = 4000 cm−1 we evaluate an exciton coupling
energy |J| = 1050 cm−1 by eq 1 (the coupling is negative for a
head-to-tail arrangement).
In case of BSB the absorption spectra show an even more
intense squaraine-type band at 14 400 cm−1, which is strongly
red-shifted compared to SQA by 1100 cm−1, and a broad
BODIPY-type band at 19 000 cm−1. Again, the sum of squared
transition moments of BSB is larger than that of the parent
chromophores (271 D2 vs 169 D2) and the squaraine band is
more intense (201 D2 vs 123 D2) and the BODIPY band is
weaker (70.1 D2 vs 2 × 36.5 D2 = 73 D2) than in the parent
chromophores.
For BSB the exciton interaction matrix reads as in eq 3 where
we adopt the nearest-neighbor approximation and neglect
dispersion interactions between the chromophores in the
ground state. Solving the secular determinant eq 3 yields the
3549
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
Figure 2. (a−c) Steady-state absorption spectra in toluene. In each panel the spectra of SB, BSB, and SBS are compared with their parent
compounds SQA and BODIPY-1 or BODIPY-2, respectively. In panel (c) the pump wavenumbers for the FLUC and TA measurement are
indicated. (d) Fluorescence spectra in toluene.
Figure 3. Exciton state diagram for dimers and trimers. The blue and orange vectors indicate the phase relations of transition moments and are not
to scale.
However, the BODIPY band is much weaker and hardly visible
ΔE − ε
J
0
at 19 000 cm−1. For this reason we refrain from evaluating the
separated squared transition moments and give that of the sum
only, which is 339 D2, compared to 291 D2 for the sum of
individual parent compounds. As reference compound for SBS
we chose BODIPY-2 as this chromophore carries two acetylene
groups that shift the absorption maximum by 900 cm−1 to
lower energy compared to BODIPY-1.
J
−ΔE − ε
J
= 0
0
J
ΔE − ε
(3)
(4)
δEtrimer = 2 ΔE2 + 2J2
In contrast to BSB and SB, which both show narrow
squaraine bands similar to that of SQA, the heterotrimer SBS
exhibits a much broader squaraine band at 14 600 cm−1.
The exciton state diagram for SBS (Figure 3) is similar to
that of BSB with the middle state now being closer to the lower
3550
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
Table 1. Optical Data of Squaraine and BODIPY Dyes in Toluene at RT
v
abs,1/cm−1 (ε/M−1 cm−1
)
v
abs,2/cm−1 (ε/M−1 cm−1
)
μ21 +
v /cm−1
̃
̃
̃
fl
a
b
c
d
e
f
g
[fwhm/cm−1
]
μ21/D2
[fwhm/cm−1
]
μ22/D2
μ22/D2
[fwhm/cm−1
]
ϕf
J/cm−1
SQA
15 500
123
123
36.5
44.9
189
271
339
15 400
[590]
0.57 0.007
0.78 0.010
0.92 0.013
0.72 0.014
0.83 0.025
0.53 0.035
(340 000)
[560]
BODIPY-1
BODIPY-2
SB
18 900
36.5
44.9
30.4
70.1
18 300
[1090]
(70 600)
[1090]
18 000
17 500
[920]
(86 900)
[970]
15 000
159
201
19 000
14 600
[600]
1050
1100
(348 000)
[700]
(29 600)
BSB
14 400
19 000
14 100
[610]
(435 000)
[710]
(65 200)
h
SBS
14 600
19 000
14 100
990
i
(484 000)
(19 600)
[750]
1250
a
b
Absorption energy (molar extinction coefficient) [full width at half-maximum] of the squaraine-type band. Squared transition moment of the
c
d
squaraine-type band. Absorption energy (molar extinction coefficient) [full width at half-maximum] of the BODIPY-type band. Squared transition
moment of the BODIPY-type band. Fluorescence wavenumber [full width at half-maximum]. Absolute fluorescence quantum yield. Exciton
coupling energy. Evaluated using BODIPY-2 as the reference chromophore. Evaluated using BODIPY-1 as the reference chomophore.
e
f
g
h
i
Figure 4. Vectorial representation of phase relations of transition moments of SBS for different orientations (DFT optimized geometries of model
compounds without solubilizing alkyl chains at the bottom, see SI): orange vector = squaraine, blue vector = BODIPY, black vector = vector sum.
The relative length of the orange and the blue vectors are arbitrary as they depend on the coefficients of eigenvectors (see SI).
state. As above, we evaluate |J| = 990 cm−1, again in very good
agreement with the coupling in SB and BSB. Using BODIPY-1
as reference compound would lead to |J| = 1250 cm−1.
Compared to the BODIPY parent chromophores, attaching
two trimethylsilylacetylene groups to SQA only leads to a 400
middle state in case of the cisoid arrangement of chromophores.
This transition is now allowed and refers to the intense
shoulder in the absorption spectra of SBS at ca. 15 000 cm−1.
The steady-state fluorescence spectra (see Figure 2d) of all
squaraine-containing compounds show very small Stokes shifts
(on the order of 100−500 cm−1) and similar band shapes as
that of the parent squaraine SQA, which displays mirror image
to its absorption spectrum. The fact that the fluorescence
spectrum of SBS does not display mirror image to its
absorption spectrum supports the assumption that the broad
squaraine absorption of SBS stems from two different excitonic
states as only the lowest one should lead to fluorescence. The
fluorescence spectra of the BODIPY dyes are also mirror
images of their respective absorption spectra. Comparison of all
these spectra shows that the lowest excitonic states of the
heterodimer and the heterotrimers have essentially squaraine
character. As a consequence of exciton coupling, the
fluorescence of these SCs is strongly red-shifted (800−1300
cm−1) compared to that of SQA and extends into the NIR.
Fluorescence quantum yields were determined for all dyes with
81
cm−1 shift (15 100 cm−1 vs 15 500 cm−1 (SQA)). We stress
that the choice of the proper reference chromophore is crucial
for the evaluation of the electronic coupling, which thereby
introduces a major source of error. The above evaluated
coupling energies thus should be considered as a function of
the choice of diabatic states (= noninteracting parent
chromophores).
The broadness of the squaraine band in SBS might be caused
by different conformers, which may adopt a cisoid and a transoid
structure concerning the relative orientation of squaraine dyes
to each other. This fact requires modifying the phase relations
of transition moments as given in Figure 4. Now we have to
take into account the ca. 120° angle between the chromophores
(for DFT computed molecular structure, see SI), which leads to
a nonvanishing vector sum of transition moments for the
3551
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
Table 2. Time-Resolved Optical Data of Squaraine and BODIPY Dyes in Toluene at RT
a
b
c
d
e
f
τf/ns (TCSPC) kf = ϕf/τf/s−1
τi/ps (ai) (FLUC) τi/ps (TA) τa/ps (r(t = 0)) anisotropy (FLUC) τa/ps (r(t = 0)) anisotropy (TA)
SQA
1.7
3.35 × 108
0.31 (−0.091)
16 (−0.027)
1400 (1)
1.0
200 (0.4)
200 (0.36)
740
[16 900/15 300]
[15 500/14 200]
1900
[15 500]
[16 900/15 300]
180 (0.35)
[15 500/15 400]
BODIPY-1
BODIPY-2
SB
4.7
4.5
1.9
3.79 × 108
4.61 × 108
170 (0.115)
1900 (1)
1.4
440 (0.36)
50
[16 400/14 500]
[16 400/14 500]
490
2000
[14 900]
1.0
BSB
1.8
0.51 (−0.104)
35 (−0.176)
1500 (1)
980 (0.38)
50
[15 400/13 900]
320
[15 400/13 900]
1800
[14 500]
1090 (0.025)
[19 000/13 700]
0.70 (−0.070)
34 (−0.123)
1700 (1)
[19 000/[13 700]
0.29 (−0.246)
70 (−0.197)
450 (0.150)
1200 (1)
SBS
1.3
4.08 × 108
0.13
960 (0.15)
1400 (0.36)
5.0
[15 600/14 000]
[14 600/14 500]
88
1200
[14 600]
[15 600/14 000]
a
Fluorescence lifetime measured TCSPC, excitation at 15 200 cm−1 (squaraine dyes) and 23 900 cm−1 (BODIPY dyes), emission at the emission
b
c
d
maximum. Radiative rate constant. Fluorescence lifetime (amplitudes) [excitation/fluorescence wavenumber] measured by FLUC. Globally fitted
lifetimes of SADS measured by TA [pump wavenumber]. Lifetime of anisotropy decay (initial anisotropy) [excitation/fluorescence wavenumber]
measured by FLUC. Lifetime of anisotropy decay (initial anisotropy) [excitation/probe wavenumber] measured by TA.
e
f
an integration sphere. The amazing point is that for both SB (ϕf
= 0.72) and BSB (ϕf = 0.83) the fluorescence quantum yield
increases compared to SQA (ϕf = 0.57) due to the interaction
with the BODIPY dye. This contradicts the gap rule, which
suggests increasing rates of nonradiative processes while going
to lower fluorescence energies.82,83 In contrast, the quantum
yield of SBS is much smaller than that of SB and BSB and even
SQA, which is likely caused by the allowed middle state in the
cisoid conformer.
The BSB and SBS superchromophores behave also much
differently from a recently described squaraine-fluorene-
squaraine heterotrimer,84 which displays very similar absorption
and emission properties to its parent squaraine chromophore
presumably because the excited state of the fluorene moiety is
too high in energy and excitonic couplings exert only little
changes of state energies. On the other hand, Therien et al.
compared alkyne-connected Zn-porphyrin oligomers consisting
of identical chromophores with oligomers of alternating slightly
different Zn-porphyrins.85,86 While impressive red-shifts of
absorption and fluorescence were found for both series, the
spectral width of the absorption band also increases strongly
with the number of porphyrin units.
To cover and compare the whole temporal evolution of all
chromophores by fluorescence spectroscopy, we measured the
fluorescence decays of all chromophores by time-correlated
single-photon counting (TCSPC, IRF ca. 200 ps) in toluene.
From the results in Table 2 it is obvious that while the BODIPY
dyes show monoexponential decays with lifetimes of ca. 4−5 ns
the squaraine dyes have lifetimes of ca. 1.7−1.9 ns, with
exception of SBS, which has a significantly reduced lifetime of
1.3 ns. Using these lifetimes and the fluorescence quantum
yield we calculated the fluorescence rate kf = ϕf/τf (see Table
2), which is larger for SB and BSB compared to SQA and
explains the effect of intensity borrowing on the basis of rate
constants.
For the squaraine dyes better time resolution was obtained
by fluorescence upconversion (FLUC) measurements with fs-
time resolution (see Table 2 and SI). These measurements
show multiexponential decays even for the SQA parent
chromophore but the longest lifetime is in good agreement
with the TCSPC results in all cases. For SQA, small negative
amplitudes for the 0.31 and 16 ps component are indicative of
dynamic changes of the fluorescence spectral band shape, which
is probably caused by intramolecular vibrational relaxation
(IVR) and vibrational cooling (VC)87−89 because solvent
relaxation effects are supposed to be small considering the
nonpolar excited state and the nonpolar toluene.90 Similar
observations were made for BSB but not for SB, which shows
only two decay components. In any case, the longest
component dominates by far the decay to the ground state.
We stress that for technical reasons it was not possible to excite
The above-described steady-state optical properties of SB,
BSB, and SBS can quantitatively be understood by exciton
coupling theory. In order to probe the influence of the
excitonically coupled chromophores on the dynamic optical
properties we performed time-resolved fluorescence and
transient absorption experiments.
3552
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
the chromophores at the absorption maxima of the spectra and
to measure the fluorescence decay at the maxima of the
respective fluorescence spectra. This has to do with the
excitation and emission wavenumber being very close and also
close to the laser fundamental, which leads to stray-light
artifacts even when using a type-II BBO crystal for sum-
frequency generation in the FLUC experiments. The observed
rise-time and decay-time components may then depend on the
specific excitation/emission wavenumber combination. There-
fore, we refrain from discussing the minor components in more
detail.
In order to circumvent the above-mentioned problems we
measured transient absorption (TA) pump−probe spectra
where we excited the samples in toluene at the maximum of
their absorption with 140 fs laser pulses and probed the spectral
difference with a white-light continuum between 12 500 and
25 000 cm−1 (400−800 nm).90 The original chirp- and stray-
light corrected data can be found in the SI. Here we focus on
global target fits to the transient wavenumber × time map. In
these target fits we assume a consecutive model (A→ B → C
etc. and A → S0, B → S0 etc.) where all spectral components
are associated with species and, thus, yield “species associated
difference spectra” (SADS), which are given in Figure 5. Each
component is also allowed to decay directly to the ground state.
The efficiencies for each reaction pathway are given in the
reaction scheme in Figure 5 and were determined under the
assumption that all SADS at the maximum bleaching signal
show the same optical density because they should possess the
same extinction coefficient. In general the SADS consist of
three contributions: (negative) ground state bleaching (GSB)
signals where the steady-state spectra show prominent
absorption bands; a (negative) stimulated emission (SE) signal
where the steady-state spectra show fluorescence (intensity
divided by ν
̃
2), and (positive) excited state absorption (ESA) to
higher lying states.
For SQA the SADS show strong bleaching associated with
the strong S1 ← S0 GSB superimposed by SE S1 → S0
contribution to the low energy side. These two contributions
merge to a strong negative signal with a minimum at 15 500
cm−1. At 18 000−23 000 cm−1 there is a broad and featureless
ESA band (see SI). Between the first (τ = 1.0 ps) and the
second (τ = 740 ps) SADS there is a slight red-shift of the SE
flank of the 15 500 cm−1 signal and between the second and the
third (τ = 1.9 ns) SADS there is a slight blue-shift of both the
GSB and the SE flanks. The shorter time constant (1.0 ps) is
probably associated with an IVR process (see fluorescence
upconversion results above). The slower process (τ = 740 ps) is
difficult to assign to a specific process. One might speculate
about vibrational cooling, which is so slow that in parallel
relaxation to the ground state occurs. In fact this would be a
continuous process and the representation by a single time
constant is inadequate. Similarly long decay components were
observed before for analogous squaraine dyes by other
authors.91,92 There, the hundred ps component was argued to
be caused by an isomerization reaction. However, the long
lifetime speaks against isomerization via a conical intersection
as do the almost identical transient spectra for all three
components (see Figure 5 and a magnification of the SADS in
the SI). Furthermore, fluorescence anisotropy measurements
(see below) also show no indication of an isomerization
process.93 Thus, an unspecified relaxation processes that goes
along with little geometric/electronic changes appear to be the
most likely explanation.
In SB the situation is similar to SQA, however, the global
analysis procedure requires four components (see Table 2)
whose SADS exhibit an increasing red-shift of the SE flank as
does the low-energy side of the ESA at ca. 17 000−18 000
cm−1. The SADS of BSB show a very similar scenario with very
similar lifetimes of the components but with more pronounced
red-shifts. Again, for SQA, SB and BSB the slowest component
is in good agreement with the time-resolved emission
measurements. The shorter components cannot be compared
with the FLUC data because dynamic shifts of bands result in
different lifetimes and amplitudes in a single color measurement
as FLUC than in a globally fitted broad-band measurement
such as the TA experiments. The dynamic properties of SB and
BSB obviously are governed by the squaraine part of the SC.
Much in contrast to SQA, SB and BSB, the SADS of SBS
show some additional features. These are, besides the
superimposed GSB and SE at ca. 14 500 cm−1, weak ESA
signals at 13 800 and 15 200 cm−1 for the component with the
shortest lifetime (τ = 0.13 ps) and somewhat stronger ESA for
all longer-lived components at the latter wavenumber. In
addition, the successive red-shifts of the four SADS are even
stronger than those in BSB and the associated lifetimes for the
three shortest SADS are about 1 order of magnitude shorter
compared to those of SB and BSB. This might be caused by the
presence of cisoid and transoid conformers in SBS (see Figure
4) that are absent in BSB. The ESA at 15 200 cm−1 may be
Figure 5. Species associated difference spectra of a global target fit to
the transient absorption measurements in toluene. The inset gives
state diagrams with the lifetimes and efficiencies of the corresponding
excited states. S0 and the initially pumped state are represented by
black bars. The efficiency differences to 100% refer to the direct
pathways to the ground state.
3553
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
Figure 6. (a) Fluorescence upconversion measurements (circles) of BSB with pump pulse at magic angle (MA), parallel (∥) and perpendicular (⊥)
orientation relative to the gate pulse (pump at 15 400 cm−1, fluorescence at 13 900 cm−1). The MA data are not to scale to the parallel and
perpendicular data. Global convoluted fit curves (red lines). (b) Experimental fluorescence anisotropy and convoluted anisotropy r.
caused by a two-exciton state, that is, transitions into a state
where both squaraine chromophores are excited.94 The more
pronounced red-shift points toward stronger molecular
reorientation processes in the excited state.
the middle state and the lowest exciton state have an angle of
ca. 90°, which theoretically refers to an initial anisotropy r(t =
0) = −0.2. In case of the transoid isomer we can only excite the
lowest exciton state, which should result in r(t = 0) = 0.4. If we
assume a mixture of cisoid and transoid conformers with, e.g.,
equal amounts this would result in an anisotropy of (r(90°) +
r(0°))/2 = (−0.2 + 0.4)/2 = 0.1 in reasonable agreement with
experiment.
As indicated above, the deviations of the optical properties of
SBS from, e.g., BSB may be traced back to excitation to the
middle state being allowed caused by the presence of
conformers. Such geometric changes should be visible in the
time-resolved anisotropy of fluorescence, which is sensitive to
changes of the orientation of the transition moment of the
emitting state vs the transition moment of the primarily excited
state.95 Therefore, we determined the time-resolved fluores-
cence anisotropy r(t) (see eq 5) by FLUC for all squaraine
containing chromophores. In this equation, θ refers to the angle
between the excitation and the emission transition moment.
The anisotropy was determined by measuring the time-
dependent fluorescence at parallel, perpendicular and magic
angle orientation of the pump pulse relative to the gate pulse
(see Figure 6). These time traces were deconvoluted by the
instrument response (assumed to be Gaussian-shaped) and
globally fitted, which yields the isotropic decay times and
amplitudes and the anisotropic decay time together with its
amplitude.96 In all cases, r(t) shows a monoexponential decay
(Table 2) whose lifetime is related to the rotational diffusion
and, thus reflect the increasing hydrodynamic radius of
chromophores in the series τa(SQA) = 200 ps, τa(SB) = 440
ps, τa(BSB) = 980 ps, and τa(SBS) = 960 ps.97
In order to support the assumption that we excite a mixture
of conformers of SBS we performed polarization dependent TA
measurements. In general, unlike fluorescence measurements,
where the anisotropy is always between −0.2 and +0.4, for TA
measurements the anisotropy can adopt all values between −∞
and +∞ because several signals (GSB, SE, ESA) can contribute
at a single wavenumber.98 However, at the GSB/SE band of the
squaraine dyes (around 14 000−16 000 cm−1), there is hardly
any ESA signal contribution. Because GSB of the lowest excited
state and the ESA signal are expected to possess the same
direction of transition moments this should simplify the
analysis. This assumption is confirmed by measuring the TA
anisotropy of SQA by pumping at 15 500 cm−1 and probing at
14 200 cm−1, which indeed gives r(t = 0) = 0.36 as expected for
a lowest energy pumped and probed state with minimal overlap
with other states. For SBS we pumped at 14 600 cm−1, which is
at the lowest exciton level and probed at 14 500 cm−1 yielding
r(t = 0) = 0.36. Again, this high anisotropy shows that we
pumped and probed the identical lowest energy state. Thus,
this TA experiment at 14 600 cm−1 pump wavenumber
together with the FLUC anisotropy measurement, which
shows a much lower anisotropy at 15 600 cm−1 pump
wavenumber, proves that the squaraine absorption band
between 14 000 and 17 000 cm−1 of SBS consists of the
lower and the middle excitonic states.
Finally it is instructive to see what happens if the BODIPY
state of SB and BSB is excited at 18 900 cm−1 (the BODIPY
band of SBS is too weak for such an experiment). While the TA
spectra (SADS) and their associated lifetimes are similar to the
ones pumped at the maximum of the squaraine band (see SI),
the initial anisotropy of the FLUC measurement of BSB is
0.025 (τa(BSB) = 1090 ps in excellent agreement with the
experiment at 15 400 cm−1 pump wavenumber, see Table 2 and
SI). This surprisingly low initial value indicates energy transfer
from the upper excitonic state to the emitting lowest state being
much faster than our time resolution (<200 fs).
2 3 cos2 θ − 1
I (t) − I⊥(t)
I (t) + 2I⊥(t)
r(t) =
=
·
5
2
(5)
Even more interesting is the anisotropy (amplitude) at t = 0,
which can be related to the angle θ between the excitation and
the emission transition moment before rotational diffusion or
other processes such as energy transfer or energy relaxation
leads to a decay of anisotropy with time (see eq 5). The initial
anisotropy r(t = 0) is almost 0.4 for SQA, SB, and BSB (see
Table 2) referring to 0° angle; that is, the transition moment
for absorption and emission are parallel, which is what we
expect for the emitting state being the same as the excited state.
However, SBS shows r(t = 0) = 0.15, which refers to 40°. As
stated above, because of technical reasons we excited the
samples at the high energy flank of the squaraine band at
16 400 cm−1 for SB, 15 400 cm−1 for BSB and 15 600 cm−1 for
SBS. In case of SBS this means that we excited predominantly
the middle exciton level of the cisoid conformer. Internal
conversion to the lowest exciton state is a prerequisite to emit
fluorescence. As sketched in Figure 4 the transition moment of
3554
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
second wire grid the probe beam was then adjusted to parallel,
perpendicular and magic angle relative to the probe beam. This
ensures that the sample was excited with the same intensity in the
parallel and perpendicular measurements. The pump pulse (Ø ca. 0.5
mm) and probe pulse (Ø ca. 0.1 mm) met at ca. 6° vertical angle in
the sample cuvette The probe light was measured by a spectrograph
equipped with a CMOS sensor (Ultrafast Systems, Helios) in the
range between 11 900 and 25 000 cm−1 with an intrinsic resolution of
1.5 nm. Every second pump pulse was blocked by a mechanical
chopper (working at 500 Hz) to measure I and I0. In order to
compensate intensity fluctuations, a reference beam was split off and
also detected with an identical spectrograph.
By means of a computer-controlled linear stage (retro reflector in
double pass setup) the relative temporal delay between pump and
probe pulse was varied in 20 fs steps from 0 fs to 4 ps and from 4 ps to
8 ns in logarithmic steps with a maximum step size of 200 ps. Steady
state absorption spectra were recorded before and after the transient
absorption experiment to exclude degradation of the sample. The raw
data were corrected for stray light prior to data analysis of the
difference spectra map (time × wavelength).
The maps recorded under magic angle conditions were analyzed
with GLOTARAN99,100 including the correction for the white light
dispersion (chirp) and modeling the instrument response function and
the coherent artifact. The anisotropic data were analyzed by a
simultaneous reconvolution fit of the parallel, perpendicular and magic
angle traces at selected wavelengths with a self-written implementation
in MatLab. A detailed description of this procedure can be found in
the literature.101 Briefly, the measured signal intensity I is the
convolution of the instrument response IRF (taken as Gaussian
shaped) with the product of the population decay function S and
respective anisotropy function, which depends on r.
CONCLUSIONS
■
Combination of two different chromophores, squaraine and
BODIPY, in a heterodimer and two heterotrimers allowed
tuning the superchomophore properties. The lowest energy
absorption and emission properties are mainly governed by the
squaraine part of the SCs, which can be shifted by more than
1000 cm−1 to the red by the excitonic interaction with the
BODIPY dye. The exciton coupling energy is ca. 1000 cm−1 for
all squaraine-containing SCs, which is substantial compared to,
10,49
e.g., squaraine polymers with |J| ∼ 400−700 cm−1
but
similar to thiophene-bridged squaraine dyes.40 Employing
polarization-dependent TA and FLUC measurements we
could prove that the squaraine absorption in SB and BSB is
caused by a single excitonic state but they are two for SBS. In
this way, the dynamic properties of SBS differ from those of all
other squaraine compounds in terms of dynamic Stokes shift,
lifetimes and, most importantly, fluorescence quantum yield,
which is decreased compared to SQA. For SB and BSB the
situation is quite different: despite the spectral red-shift of the
lowest absorption band, the fluorescence quantum yields
increase for SB and BSB compared to SQA, which contradicts
Siebrand’s gap rule. This is caused by intensity borrowing from
the BODIPY states, which increases the squared transition
moments of the squaraine band dramatically by 29% for SB and
63% for BSB compared to SQA. In this way, exciton coupling
can lead to a substantial enhancement of fluorescence quantum
yield by 26% for SB and 46% for BSB and shift the emission
from the red to the near-infrared while keeping the bandwidth
almost constant. Thus, exciton coupling in heterochromophore
systems represents an alternative way to tune and improve
fluorescence properties of dyes provided that the excited state
energies of the two chromophore units are not too far away. In
conclusion, the here investigated superchomophores combine
the best of two worlds, the high fluorescence quantum yield of
BODIPY dyes with the low energy fluorescence of squaraine
chromophores.
⎡
⎢
⎢
⎢
⎤
⎥
⎥
⎥
⎛
⎞
I (t)
I⊥(t)
⎡
⎢
⎢
⎤
⎥
⎥
1 + 2r(t)
1 − r(t)
1
⎜
⎟
= ⎜S(t)
⎟ ⊗ IRF
⎜
⎟
⎢
⎣
⎥
⎦
⎢
⎣
⎥
⎦
⎝
⎠
I
(t)
MA
(6)
Emission Spectroscopy. Steady-state fluorescence measurements
were performed with an Edinburgh Instruments FLS980 spectrometer.
The compounds were dissolved in toluene (Uvasol from Merck) and
purged with argon gas for 30 min prior to each measurement.
Fluorescence quantum yields were measured with an integrating
sphere and the FLS980 spectrometer applying the method of Bardeen
et al.102 to correct for self-absorption. Fluorescence lifetimes were
determined by time-correlated single-photon counting (TCSPC) with
the FLS980 spectrometer by exciting the samples with pulsed laser
diodes at 15 200 cm−1 (SQA, SB, BSB, SBS) or at 23 900 cm−1
(BODIPY-1, BODIPY-2) under magic angle conditions and using a
fast PMT detector (H10720) for fluorescence detection. Deconvolu-
tion of the data (4096 channels) was done by measuring the
instrument response function with a scatterer (LUDOX).
Femtosecond Fluorescence-Upconversion Spectroscopy.
We used a commercial fluorescence upconversion setup (Halcyone
from Ultrafsat Systems). The laser system was the same as for the fs-
TA-experiments. The output beam was again divided in two parts. One
part seeded the optical parametric amplifier (TOPAS from Newport
Spectra Physics) to generate the pump pulse with a pulse length of 140
fs and an excitation wavenumber of 16 900 cm−1 for SQA, 16 400 cm−1
for SB, 15 600 cm−1 for SBS and 15 400 cm−1 for BSB. The other part
of the output beam was used as the gate pulse (12 500 cm−1), which
was delayed over a maximum of 3 ns in 20 fs steps from 0 fs to 4 ps
and in logarithmic steps from 4 ps to 3 ns with a maximum step size of
80 ps with a computer-controlled linear stage. The samples were
prepared as for the fs-TA-experiments in 2 mm fused silica cuvettes
(Spectrocell Inc.) and were stirred during the experiment. The pump
pulse was focused onto the cuvette and the fluorescence light was
collected and focused on a 0.5 mm BBO type II crystal for frequency
upconversion with the gate pulse whose focus was adjusted to lay ca.
10 mm behind the BBO crystal. All lenses in the setup had a focal
length of 100 mm and a thickness of 1.85 mm. The upconverted light
EXPERIMENTAL SECTION
■
Steady-State Absorption Spectroscopy. All dyes were
dissolved in toluene (Uvasol from Merck) and the UV/vis/NIR-
absorption spectra were measured in 1 cm quartz cuvettes from
Hellma using a Cary 5000 spectrometer. The pure solvent was used as
reference.
Femtosecond Transient-Absorption Spectroscopy. The
samples were dissolved in toluene (Uvasol from Merck), degassed
for 30 min, filtered and stirred throughout the measurement. The
experiments were performed in 2 mm fused silica cuvettes (Spectrocell
Inc.) at RT and the optical density was adjusted to ca. 0.3 at the
corresponding excitation wavenumber. The pump−probe measure-
ments were performed with a Helios transient spectrometer from
Ultrafast Systems and an amplified Ti:sapphire oscillator (Solstice)
from Newport Spectra Physics (pulse length of 100 fs) with a
repetition rate of 1 kHz and a fundamental wavenumber of 12 500
cm−1 (800 nm). The output beam from the Solstice amplifier was split
into two parts. A small part was focused onto a vertically oscillating
CaF2 crystal to generate a white light continuum between 11 900 cm−1
(840 nm) and 25 000 cm−1 (400 nm), which was polarized
horizontally (the polarization was adjusted by using a wire grid
(Thorlabs)) and used as the probe pulse. The main part was used to
pump an optical parametric amplifier (TOPAS-C) from Light
Conversion to generate the pump pulse with a pulse length of 140
fs at the corresponding excitation wavenumbers (SQA: 15 500 cm−1,
SB: 14 900 cm−1, BSB: 14 500 cm−1, SBS: 14 600 cm−1). By means of
a λ/2-plate and a wire grid (Thorlabs) the polarization axis of the
pump pulse was set to 45° relative to that of the probe beam. With a
3555
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
was focused on the entrance slit of a double monochromator and
measured by a PMT detector. For polarization dependent measure-
ments the pump beam was set to 45° by a λ/2 plate relative to the
horizontally oriented gate beam. The polarization of both beams was
purified by wire grids (Moxtek). In front of the cuvette, the pump
beam was finally adjusted to parallel, perpendicular and magic angle
relative to the gate beam with a wire grid for reasons already explained
in the section above. The analyses of the anisotropic fluorescence
upconversion data was done exactly as for the anisotropic transient
absorption data.
(23) Sreejith, S.; Carol, P.; Chithra, P.; Ajayaghosh, A. J. Mater. Chem.
2008, 18, 264.
(24) Yagi, S.; Nakazumi, H. Top. Heterocycl. Chem. 2008, 14, 133.
(25) Volkova, K. D.; Kovalska, V. B.; Tatarets, A. L.; Patsenker, L. D.;
Kryvorotenko, D. V.; Yarmoluk, S. M. Dyes Pigm. 2006, 72, 285.
(26) 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.
(27) Terpetschnig, E.; Szmacinski, H.; Ozinskas, A.; Lakowicz, J. R.
Anal. Biochem. 1994, 217, 197.
(28) Thomas, J.; Sherman, D. B.; Amiss, T. J.; Andaluz, S. A.; Pitner,
J. B. Bioconjugate Chem. 2007, 18, 1841.
(29) Renard, B.-L.; Aubert, Y.; Asseline, U. Tetrahedron Lett. 2009,
ASSOCIATED CONTENT
* Supporting Information
Synthetic protocols, eigenvalues and eigenvectors of exciton
coupling theory, raw data of time-resolved measurements. This
material is available free of charge via the Internet at http://
pubs.acs.org.
■
S
50, 1897.
(30) Gao, F.-P.; Lin, Y.-X.; Li, L.-L.; Liu, Y.; Mayerhoffer, U.; Spenst,
̈
P.; Su, J.-G.; Li, J.-Y.; Wurthner, F.; Wang, H. Biomaterials 2014, 35,
1004.
̈
(31) Terpetschnig, E.; Szmacinski, H.; Lakowicz, J. R. Anal. Chim.
Acta 1993, 282, 633.
(32) Arunkumar, E.; Fu, N.; Smith, B. D. Chem.Eur. J. 2006, 12,
4684.
(33) 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.
AUTHOR INFORMATION
Corresponding Author
christoph.lambert@uni-wuerzburg.de
■
Notes
(34) Xiang, Z.; 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.
(35) Ros-Lis, J. V.; Martinez-Manez, R.; Sancenon, F.; Soto, J.;
Spieles, M.; Rurack, K. Chem.Eur. J. 2008, 14, 10101.
(36) Ros-Lis, J. V.; Martinez-Manez, R.; Soto, J. Chem. Commun.
2002, 2248.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the DFG for funding this work within the Research
Group FOR 1809.
REFERENCES
(37) Ajayaghosh, A.; Arunkumar, E.; Daub, J. Angew. Chem., Int. Ed.
2002, 41, 1766.
(38) Radaram, B.; Mako, T.; Levine, M. Dalton Trans. 2013, 42,
16276.
(39) Ananda Rao, B.; Kim, H.; Son, Y.-A. Sens. Actuators, B 2013,
■
(1) Stender, B.; Voelker, S. F.; Lambert, C.; Pflaum, J. Adv. Mater.
2013, 25, 2943.
(2) Tessler, N.; Medvedev, V.; Kazes, M.; Kan, S.; Banin, U. Science
2002, 295, 1506.
188, 847.
(3) Yao, L.; Zhang, S.; Wang, R.; Li, W.; Shen, F.; Yang, B.; Ma, Y.
Angew. Chem., Int. Ed. 2014, 53, 2119.
(40) Scherer, D.; Dorfler, R.; Feldner, A.; Vogtmann, T.; Schwoerer,
̈
M.; Lawrentz, U.; Grahn, W.; Lambert, C. Chem. Phys. 2002, 279, 179.
(41) Odom, S. A.; Webster, S.; Padilha, L. A.; Peceli, D.; Hu, H.;
Nootz, G.; Chung, S. J.; Ohira, S.; Matichak, J. D.; Przhonska, O. V.;
Kachkovski, A. D.; Barlow, S.; Bredas, J. L.; Anderson, H. L.; Hagan, D.
J.; Van Stryland, E. W.; Marder, S. R. J. Am. Chem. Soc. 2009, 131,
7510.
(4) Guo, Z.; Park, S.; Yoon, J.; Shin, I. Chem. Soc. Rev. 2014, 43, 16.
(5) Pansare, V. J.; Hejazi, S.; Faenza, W. J.; Prud’homme, R. K. Chem.
Mater. 2012, 24, 812.
(6) Mayerhoffer, U.; Fimmel, B.; Wurthner, F. Angew. Chem., Int. Ed.
̈
̈
2012, 51, 164.
(7) Heckmann, A.; Dummler, S.; Pauli, J.; Margraf, M.; Kohler, J.;
̈
̈
(42) Ohira, S.; Rudra, I.; Schmidt, K.; Barlow, S.; Chung, S.-J.; Zhang,
Stich, D.; Lambert, C.; Fischer, I.; Resch-Genger, U. J. Phys. Chem. C
Q.; Matichak, J.; Marder, S. R.; Bredas, J.-L. Chem.Eur. J. 2008, 14,
́
2009, 113, 20958.
11082.
(8) Volker, S. F.; Renz, M.; Kaupp, M.; Lambert, C. Chem.Eur. J.
̈
(43) Toro, C.; De Boni, L.; Yao, S.; Ritchie, J. P.; Masunov, A. E.;
Belfield, K. D.; Hernandez, F. E. J. Chem. Phys. 2009, 130, 214504/1.
(44) 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.
(45) Webster, S.; Odom, S. A.; Padilha, L. A.; Przhonska, O. V.;
Peceli, D.; Hu, H.; Nootz, G.; Kachkovski, A. D.; Matichak, J.; Barlow,
S.; Anderson, H. L.; Marder, S. R.; Hagan, D. J.; Van Stryland, E. W. J.
Phys. Chem. B 2009, 113, 14854.
(46) Webster, S.; Peceli, D.; Hu, H.; Padilha, L. A.; Przhonska, O. V.;
Masunov, A. E.; Gerasov, A. O.; Kachkovski, A. D.; Slominsky, Y. L.;
Tolmachev, A. I.; Kurdyukov, V. V.; Viniychuk, O. O.; Barrasso, E.;
Lepkowicz, R.; Hagan, D. J.; Van Stryland, E. W. J. Phys. Chem. Lett.
2010, 1, 2354.
2011, 17, 14147.
(9) Muller, T. J. J.; Bunz, U. H. F. Functional Organic Materials:
̈
Syntheses, Strategies and Applications; Wiley-VCH: Weinheim, 2007.
(10) Volker, S. F.; Schmiedel, A.; Holzapfel, M.; Renziehausen, K.;
̈
Engel, V.; Lambert, C. J. Phys. Chem. C 2014, 118, 17467.
(11) Kuster, S.; Geiger, T. Dyes Pigm. 2015, 113, 110.
(12) Wurthner, F.; Kaiser, T. E.; Saha-Moeller, C. R. Angew. Chem.,
̈
Int. Ed. 2011, 50, 3376.
(13) Mobius, D. Adv. Mater. 1995, 7, 437.
̈
(14) Kobayashi, T. J-Aggregates; World Scientific: Singapore, 1996.
(15) Van Amerongen, H.; Valkunas, L.; Van Grondelle, R.
Photosynthetic Excitations; World Scientific: Singapore, 2000.
(16) Kirstein, S.; Daehne, S. Int. J. Photoenergy 2006, 2006, 1.
(17) Scheibe, G. Kolloid-Z. 1938, 82, 1.
(47) Belfield, K. D.; Bondar, M. V.; Haniff, H. S.; Mikhailov, I. A.;
Luchita, G.; Przhonska, O. V. ChemPhysChem 2013, 14, 3532.
(48) Silvestri, F.; Irwin, M. D.; Beverina, L.; Facchetti, A.; Pagani, G.
A.; Marks, T. J. J. Am. Chem. Soc. 2008, 130, 17640.
(18) Hong, Y.; Lam, J. W. Y.; Tang, B. Z. Chem. Soc. Rev. 2011, 40,
5361.
(19) Czikkely, V.; Dreizler, G.; Foersterling, H. D.; Kuhn, H.;
Sondermann, J.; Tillmann, P.; Wiegand, J. Z. Naturforsch., A:
Astrophys., Phys. Phys. Chem. 1969, 24, 1821.
(20) Kiprianov, A. I. Russ. Chem. Rev. 1971, 40, 594.
(21) Ajayaghosh, A. Acc. Chem. Res. 2005, 38, 449.
(22) Beverina, L.; Salice, P. Eur. J. Org. Chem. 2010, 1207.
(49) Volker, S. F.; Uemura, S.; Limpinsel, M.; Mingebach, M.;
̈
Deibel, C.; Dyakonov, V.; Lambert, C. Macromol. Chem. Phys. 2010,
211, 1098.
(50) Merritt, V. Y.; Hovel, H. J. Appl. Phys. Lett. 1976, 29, 414.
3556
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557

Journal of the American Chemical Society
Article
(51) Morel, D. L.; Ghosh, A. K.; Feng, T.; Stogryn, E. L.; Purwin, P.
E.; Shaw, R. F.; Fishman, C. Appl. Phys. Lett. 1978, 32, 495.
(52) Wang, S.; Mayo, E. I.; Perez, M. D.; Griffe, L.; Wei, G.;
Djurovich, P. I.; Forrest, S. R.; Thompson, M. E. Appl. Phys. Lett. 2009,
94, 233304.
(87) Glasbeek, M.; Zhang, H. Chem. Rev. 2004, 104, 1929.
(88) Elsaesser, T.; Kaiser, W. Annu. Rev. Phys. Chem. 1991, 42, 83.
(89) Pigliucci, A.; Duvanel, G.; Daku, L. M. L.; Vauthey, E. J. Phys.
Chem. A 2007, 111, 6135.
(90) Reynolds, L.; Gardecki, J. A.; Frankland, S. J. V.; Horng, M. L.;
Maroncelli, M. J. Phys. Chem. 1996, 100, 10337.
(53) Mayerhoffer, U.; Deing, K.; Gruss, K.; Braunschweig, H.;
̈
(91) de, M. G.; Marchena, M.; Zitnan, M.; Pandey, S. S.; Hayase, S.;
Douhal, A. Phys. Chem. Chem. Phys. 2012, 14, 1796.
Meerholz, K.; Wurthner, F. Angew. Chem., Int. Ed. 2009, 48, 8776.
̈
(54) Fan, B.; Maniglio, Y.; Simeunovic, M.; Kuster, S.; Geiger, T.;
(92) Gude, C.; Rettig, W. J. Phys. Chem. A 2000, 104, 8050.
(93) For none of the squaraine dyes do we have any indication for
the presence of different isomers/conformers in solution by, e.g., NMR
spectroscopy. While rotation about the triple bond should be faster
than the NMR time scale and, thus, be invisible, isomerization within
the squaraine dyes around the double bonds should be slow and visible
as two sets of NMR signals.
(94) Knoester, J. Proc. Int. Sch. Phys. “Enrico Fermi” 2002, 149, 149.
(95) Vogel, S. S.; Thaler, C.; Blank, P. S.; Koushik, S. V. In FLIM
Microscopy in Biology and Medicine; Periasamy, A., Clegg, R. M., Eds.;
Chapman and Hall/CRC Press: Boca Raton, FL, 2009.
(96) Cross, A. J.; Fleming, G. R. Biophys. J. 1984, 46, 45.
(97) Fleming, G. R. In Chemical Applications of Ultrafast Spectroscopy;
Oxford University Press: Oxford, U.K., 1986.
(98) Jonas, D. M.; Lang, M. J.; Nagasawa, Y.; Joo, T.; Fleming, G. R.
J. Phys. Chem. 1996, 100, 12660.
(99) Van Stokkum, I. H. M.; Larsen, D. S.; Van Grondelle, R.
Biochim. Biophys. Acta, Bioenerg. 2004, 1657, 82.
(100) Snellenburg, J. J.; Laptenok, S. P.; Seger, R.; Mullen, K. M.;
Stokkum, I. H. M. v. J. Stat. Software 2012, 49, 1.
Hany, R.; Nuesch, F. Int. J. Photoenergy 2009, 1.
(55) Wei, G.; Wang, S.; Renshaw, K.; Thompson, M. E.; Forrest, S.
R. ACS Nano 2010, 4, 1927.
(56) Wei, G.; Lunt, R. R.; Sun, K.; Wang, S.; Thompson, M. E.;
Forrest, S. R. Nano Lett. 2010, 10, 3555.
(57) Beverina, L.; Drees, M.; Facchetti, A.; Salamone, M.; Ruffo, R.;
Pagani, G. A. Eur. J. Org. Chem. 2011, 2011, 5555.
(58) Wei, G.-D.; Xiao, X.; Wang, S.-Y.; Zimmerman, J. D.; Sun, K.;
Diev, V. V.; Thompson, M. E.; Forrest, S. R. Nano Lett. 2011, 11,
4261.
(59) Wang, S.; Hall, L.; Diev, V. V.; Haiges, R.; Wei, G.; Xiao, X.;
Djurovich, P. I.; Forrest, S. R.; Thompson, M. E. Chem. Mater. 2011,
23, 4789.
(60) Wei, G.; Xiao, X.; Wang, S.; Sun, K.; Bergemann, K. J.;
Thompson, M. E.; Forrest, S. R. ACS Nano 2012, 6, 972.
(61) Chen, C.-H.; Cheng, W.-T.; Tsai, M.-L.; Huang, K.-T. Ind. Eng.
Chem. Res. 2012, 51, 3630.
(62) Xiao, X.; Wei, G.; Wang, S.; Zimmerman, J. D.; Renshaw, C. K.;
Thompson, M. E.; Forrest, S. R. Adv. Mater. 2012, 24, 1956.
(63) Deing, K. C.; Mayerhoffer, U.; Wurthner, F.; Meerholz, K. Phys.
̈
̈
(101) Steeger, M. Ph.D. Thesis, University of Wurzburg, 2014.
̈
Chem. Chem. Phys. 2012, 14, 8328.
(102) Ahn, T.-S.; Al-Kaysi, R. O.; Muller, A. M.; Wentz, K. M.;
(64) Kylberg, W.; Zhang, Y.; Aebersold, A.; Araujo de Castro, F.;
Geiger, T.; Heier, J.; Kuster, S.; Ma, C.-Q.; Bauerle, P.; Nuesch, F.;
Tisserant, J.-N.; Hany, R. Org. Electron. 2012, 13, 1204.
(65) Maeda, T.; Tsukamoto, T.; Seto, A.; Yagi, S.; Nakazumi, H.
Macromol. Chem. Phys. 2012, 213, 2590.
Bardeen, C. J. Rev. Sci. Instrum. 2007, 78, 086105.
(66) Bagnis, D.; Beverina, L.; Huang, H.; Silvestri, F.; Yao, Y.; Yan,
H.; Pagani, G. A.; Marks, T. J.; Facchetti, A. J. Am. Chem. Soc. 2010,
132, 4074.
(67) Choi, H.; Kamat, P. V. J. Phys. Chem. Lett. 2013, 4, 3983.
(68) Maeda, T.; Arikawa, S.; Nakao, H.; Yagi, S.; Nakazumi, H. New J.
Chem. 2013, 37, 701.
(69) Maeda, T.; Hamamura, Y.; Miyanaga, K.; Shima, N.; Yagi, S.;
Nakazumi, H. Org. Lett. 2011, 13, 5994.
(70) Maeda, T.; Nakao, H.; Kito, H.; Ichinose, H.; Yagi, S.;
Nakazumi, H. Dyes Pigm. 2011, 90, 275.
(71) Maeda, T.; Shima, N.; Tsukamoto, T.; Yagi, S.; Nakazumi, H.
Synth. Met. 2011, 161, 2481.
(72) Jiang, J.-Q.; Sun, C.-L.; Shi, Z.-F.; Zhang, H.-L. RSC Adv. 2014,
4, 32987.
(73) Bessette, A.; Hanan, G. S. Chem. Soc. Rev. 2014, 43, 3342.
(74) Ulrich, G.; Ziessel, R.; Harriman, A. Angew. Chem., Int. Ed. 2008,
47, 1184.
(75) Boens, N.; Leen, V.; Dehaen, W. Chem. Soc. Rev. 2012, 41, 1130.
(76) Loudet, A.; Burgess, K. Chem. Rev. 2007, 107, 4891.
(77) Popere, B. C.; Della Pelle, A. M.; Thayumanavan, S.
Macromolecules 2011, 44, 4767.
(78) McRae, E. G.; Kasha, M. Phys. Processes Radiat. Biol., Proc. Int.
Symp. 1964, 23.
(79) Guelen, D. Photosynth. Res. 2006, 87, 205.
(80) Shibata, Y.; Tateishi, S.; Nakabayashi, S.; Itoh, S.; Tamiaki, H.
Biochemistry 2010, 49, 7504.
(81) Chen, J.; Winter, R. F. Chem.Eur. J. 2012, 18, 10733.
(82) Siebrand, W. J. Chem. Phys. 1967, 46, 440.
(83) Siebrand, W. J. Chem. Phys. 1967, 47, 2411.
(84) Moreshead, W. V.; Przhonska, O. V.; Bondar, M. V.;
Kachkovski, A. D.; Nayyar, I. H.; Masunov, A. E.; Woodward, A.
W.; Belfield, K. D. J. Phys. Chem. C 2013, 117, 23133.
(85) Duncan, T. V.; Susumu, K.; Sinks, L. E.; Therien, M. J. J. Am.
Chem. Soc. 2006, 128, 9000.
(86) Susumu, K.; Therien, M. J. J. Am. Chem. Soc. 2002, 124, 8550.
3557
DOI: 10.1021/ja512338w
J. Am. Chem. Soc. 2015, 137, 3547−3557