Charge-recombination fluorescence from push-pull electronic systems constructed around amino-substituted styryl-BODIPY dyes

Article abstract of DOI:10.1002/chem.201301045

A small series of donor-acceptor molecular dyads has been synthesized and fully characterized. In each case, the acceptor is a dicyanovinyl unit and the donor is a boron dipyrromethene (BODIPY) dye equipped with a single styryl arm bearing a terminal amino group. In the absence of the acceptor, the BODIPY-based dyes are strongly fluorescent in the far-red region and the relaxed excited-singlet states possess significant charge-transfer character. As such, the emission maxima depend on both the solvent polarity and temperature. With the corresponding push-pull molecules, there is a low-energy charge-transfer state that can be observed by both absorption and emission spectroscopy. Here, charge-recombination fluorescence is weak and decays over a few hundred picoseconds or so to recover the ground state. Overall, these results permit evaluation of the factors affecting the probability of charge-recombination fluorescence in push-pull dyes. The photophysical studies are supported by cyclic voltammetry and DFT calculations. Poles apart: A series of push-pull dyes (see figure) with absorption and emission transitions in the far-red region has been subjected to detailed photophysical examination. These compounds exhibit large dipole moments under illumination and solvent-dependent emission. Copyright

Full text of DOI:10.1002/chem.201301045

DOI: 10.1002/chem.201301045
Charge-Recombination Fluorescence from Push–Pull Electronic Systems
Constructed around Amino-Substituted Styryl–BODIPY Dyes**
Adela Nano,[a] Raymond Ziessel,*[a] Patrycya Stachelek,[b] and Anthony Harriman*[b]
Abstract: A small series of donor–ac-
ceptor molecular dyads has been syn-
thesized and fully characterized. In
each case, the acceptor is a dicyanovin-
yl unit and the donor is a boron dipyr-
romethene (BODIPY) dye equipped
with a single styryl arm bearing a ter-
minal amino group. In the absence of
the acceptor, the BODIPY-based dyes
are strongly fluorescent in the far-red
region and the relaxed excited-singlet
states possess significant charge-trans-
fer character. As such, the emission
maxima depend on both the solvent
polarity and temperature. With the cor-
responding push–pull molecules, there
is a low-energy charge-transfer state
that can be observed by both absorp-
tion and emission spectroscopy. Here,
charge-recombination fluorescence is
weak and decays over a few hundred
picoseconds or so to recover the
ground state. Overall, these results
permit evaluation of the factors affect-
ing the probability of charge-recombi-
nation fluorescence in push–pull dyes.
The photophysical studies are support-
ed by cyclic voltammetry and DFT
calculations.
Keywords: donor–acceptor
tems · electrochemistry · electron
transfer fluorescence
photophysics
sys-
·
·
Introduction
original series of D–p–A compounds were based on a 4-ni-
trophenyl moiety as the electron acceptor with an N,N-di-
methylanilino group as the complementary electron donor.[8]
More recently, attention has turned to the use of stronger
organic donor and acceptor functions, including julolidine[9]
derivatives, pyran-containing compounds such as 2-(2-tert-
butyl-6-methyl-4H-pyran-4-ylidene)malonitrile,[10] 4,5-dicya-
noimidazole derivatives,[11] and the so-called “super-accept-
or” 7,7,8,8-tetracyanoquinodimethene.[12] The nature of the
terminals affects the optical and electronic properties of the
resultant materials, but so does the composition and length
of the p-conjugated backbone. Various conjugated spacers,
such as fluorene, stilbene, thiophene, polyenes, polyynes,
porphyrins, phthalocyanines, and merocyanine dyes, have
been incorporated into push–pull electronic systems.[13,14]
Along somewhat related lines, it is recognized that photo-
induced electron transfer remains one of the most intensive-
ly studied subjects in contemporary science and has particu-
lar relevance for understanding the workings of both natural
and artificial photosynthesis.[15] Innumerable studies have
addressed the various factors that promote light-induced
electron transfer in molecular dyads, triads, tetrads, and
higher-order analogues. The information gathered from such
studies, used in conjunction with our ever-growing aware-
ness of the natural process,[16] has led to new ideas about
how to achieve sustainable solar-fuel production.[17] It has
become clear that the efficacy of a light-induced electron-
transfer reaction depends critically on a range of parameters
that must be controlled to a supreme level of precision to
achieve the optimum output. The main variables include
both thermodynamic properties (excitation energy and elec-
trode potentials) and structural elements (separation dis-
tance, mutual orientation, and nature of the bridge).[18] The
Push–pull p-conjugated chromophores are of great topical
interest because of their putative applications in optoelec-
tronics,[1] organic light-emitting diodes,[2] and information-
storage devices.[3] Such compounds comprise an electron
donor (D) fitted with one or more electron-withdrawing (A)
groups connected through a p-conjugated spacer so as to
generate a substantive dipole moment across the molecule.
In addition, push–pull D–p–A chromophores have been in-
vestigated extensively for their nonlinear optical proper-
ties[4–6] and have proved to be useful materials for exploring
the intricacies of electron-transfer processes.[7] The field ben-
efits from the almost limitless variety of D–p–A systems
that can be isolated by combining different donor/acceptor
modules with known p-bridging spacer units. Many of the
[a] A. Nano, Dr. R. Ziessel
LCOSA, ICPEES, UMR7515
CNRS/Universitꢀ de Strasbourg/ECPM
25 rue Becquerel, 67087 Strasbourg
Cedex 02 (France)
Fax : (+33)3-68-85-27-61
E-mail: ziessel@unistra.fr
[b] P. Stachelek, Prof. Dr. A. Harriman
Molecular Photonics Laboratory
School of Chemistry, Bedson Building
Newcastle University
Newcastle upon Tyne, NE1 7RU (UK)
Fax: (+44) 191-222-8660
E-mail: anthony.harriman@ncl.ac.uk
[**] BODIPY: boron dipyrromethene (4,4-difluoro-8-(4-iodo)phenyl-
1,3,5,7-tetramethyl-4-bora-3a,4a-diaza-s-indacene).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201301045.
13528
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13528 – 13537

FULL PAPER
range of systems that has been developed to advance this
subject is incredibly far-flung and includes covalently linked
donor–acceptor pairs,[17,19] supramolecular assemblies based
on coordinative interactions[20] or hydrogen-bonding strat-
egies,[21] and self-associated donor–acceptor systems.[22]
Along these lines, boron dipyrromethene (BODIPY) deriva-
tives, featuring invaluable optical and electrochemical[23]
properties, have been less exploited as photoactive bridges
in D–p–A scaffolds and photoinduced electron transfer in-
volving BODIPY dyes is not so common.[24] The most popu-
lar reactants for such materials are bioinspired, such as por-
phyrins, carotenes, and quinones.
Our current interest in this field includes developing new
artificial systems for light-induced electron-transfer process-
es that incorporate BODIPY chromophores. Such dyes have
been used successfully in photovoltaic devices,[25] as ion sen-
sors,[26] and as fluorescent probes in biological analyses.[27]
Surprisingly, given the great popularity of these highly emis-
sive dyes, there have been few attempts to incorporate
BODIPY dyes into push–pull molecular architectures.[28]
Notably, D–p–A chromophores have been built from a
BODIPY-based core with electron-donor and electron-ac-
ceptor units attached at the 2- and 6-positions.[29] Such com-
pounds, although enabling facile tuning of the optical prop-
erties, do not promote light-induced charge separation. In
the present work, we report the synthesis of dipolar push–
pull chromophores having unusually strong donating (i.e.,
julolidine or triazatruxene) and withdrawing (i.e., dicyano-
vinyl) substituents attached to the central p-conjugated
backbone (Scheme 1). The photophysical properties of the
target molecules become understandable by virtue of sys-
tematic examination of model compounds lacking the dicya-
novinyl terminal.
Scheme 2. Synthesis of compound 5 and its precursors: i) p-TsOH cat., pi-
peridine, toluene, 1408C; ii) [PdACHTUNGRTNEUNG(PPh3)2Cl2] (10 mol%), HCOONa, DMF
under a flow of CO, 1008C; iii) AcOH/piperidine cat., methanol, 858C,
1 h.
ence of an excess amount of aldehyde. Formation of the
mono-functionalized analogue might be explained in terms
of julolidine being a strong electron donor that influences
the acidity of the methyl group located at the 5-position on
the BODIPY core. In any event, the result is the second
known example of selective mono-condensation of
a
BODIPY dye with an electrophilic aldehyde.[31] Transforma-
tion of JL into the carbaldehyde derivative 4 was realized
using a carboformylation reaction catalyzed by a palladi-
AHCTUNGTREGuNNNU m(II) precursor and with sodium formate as the reducing
reagent under a flow of CO at atmospheric pressure. The
target compound JLCV was synthesized in 50% yield by
condensation of 4 with malonitrile, using a Knoevenagel-
type reaction carried out according to a literature proce-
dure.[32,33] All compounds were fully characterized by
modern spectroscopic protocols, as outlined in the Experi-
mental Section and Supporting Information.
By way of analogy with the julolidine-based system, relat-
ed dyes were synthesized from the triazatruxene platform as
sketched in Scheme 3. This slightly ruffled scaffold has three
interconnecting amino groups and an anchor by which to
append secondary units. The iodo group on TX (compound
6 in the synthetic schemes) was converted to the corre-
sponding formyl derivative using a carboformylation reac-
tion, before the latter was transformed to the dicyanovinyl
derivative TXCV (compound 8 in the synthetic schemes)
using malonitrile as the reagent. It might be stressed that
some of these dipolar BODIPY-based chromophores are
difficult to isolate in high yield owing to their inherent insta-
bility during the purification procedures.
Scheme 1. Schematic representation of putative dipolar push–pull chro-
mophores built around a central BODIPY unit.
Results and Discussion
Synthetic strategy: The prototypic push–pull chromophore
JLCV (compound 5 in the synthetic schemes) was prepared
in three steps from the tetramethyl-BODIPY derivative 1 as
depicted in Scheme 2. Condensation of 1 with 9-formyl-10-
butoxy-1,1,7,7-tetramethyl-2,3,6,7-tetrahydro-1H,5H-pyrido-
Electrochemistry: The redox behavior of the new BODIPY
dyes was investigated by cyclic voltammetry in CH2Cl2 con-
taining 0.1m tetra-N-butylammonium hexafluorophosphate
ACHTUNGTRENNUNG
Chem. Eur. J. 2013, 19, 13528 – 13537
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13529

R. Ziessel, A. Harriman et al.
than for the isolated fragments. These potential shifts are a
clear consequence of the increased conjugation provided by
the connecting styryl linkage and to the occurrence of elec-
tronic coupling between the two units.
Turning attention to the triazatruxene-based scaffold, it is
notable that the parent compound, 9, exhibits two reversible
waves on oxidative scans but no observable peaks on reduc-
tion. These oxidative waves, which correspond to half-wave
potentials of 0.86 and 1.45 V versus SCE, respectively
(Table 1), are assigned to successive one-electron oxidation
of the aminocarbazole units. The difference between the ob-
served half-wave potentials is ascribed to electrostatic ef-
fects but the anticipated third oxidation step is not seen
within the available anodic window. Appending a monostyr-
yl BODIPY dye to the scaffold, thereby forming compounds
TX or 7, lowers the half-wave potential for one-electron oxi-
dation by approximately 200 mV owing to the increased
conjugation provided by the styryl arm. Comparing this
latter value with that derived for the analogous julolidine-
based dye, 4, indicates that oxidation is more difficult for 7
by approximately 280 mV. It is also notable that, although
the parent julolidine, 2, and triazatruxene, 9, compounds
have the same half-wave potential when isolated, these units
display quite disparate half-wave potentials when attached
to the BODIPY unit. This is an important point that illus-
trates the differing degrees to which the principal donor
couples to the extended BODIPY unit.
Scheme 3. i) [PdACHTUNGTRENNUNG(PPh3)2Cl2] (10 mol%), HCOONa, DMF under a flow of
CO, 1008C; ii) AcOH/piperidine cat., THF, 958C, 20–30 min.
(TBAPF6) as supporting electrolyte and the results are gath-
ered in Table 1. The starting material lacking the styryl arm,
1, displays a single reversible wave on both oxidative and re-
ductive scans owing to the formation of the p-radical cation
and p-radical anion, respectively, as has been observed pre-
viously for related compounds.[34] These waves are highly re-
versible (ipa/ipc ꢀ1) and exhibit the characteristic shape
(DEp =60–70 mV) of a Nernstian, one-electron process. The
monostyryl BODIPY dyes JL and 4 display two distinct
waves under oxidative scans, each of which is fully reversible
and corresponds to transfer of a single electron. The first
wave occurs with a half-wave potential of 0.38 V versus SCE
and this oxidation is considerably easier (about 480 mV)
For JL, DFT calculations indicate that both HOMO and
HOMO(À1) are spread over much of the molecule and in-
clude contributions from the BODIPY and julolidine resi-
dues (see the Supporting Information). As such, it is inaccu-
rate to ascribe the first oxidation step as being localized on
the julolidine unit. Identical calculations made with TX
show that the HOMO is also spread over much of the mole-
cule but incorporates only the amino group closest to the
styryl arm.
A
similar distribution is observed for
HOMO(À1) but the second amino group (this being sepa-
rated from the styryl unit by 3 bonds compared with only 1
bond for the proximal amino group) makes an important
contribution to HOMO(À2). The distal amino group (which
is removed by 5 bonds from the styryl unit) is the major
contributor to HOMO(À3) (see the Supporting Informa-
tion). The energy-minimized structures indicate that the ju-
lolidine-based N atom present in JL is out of alignment with
the styryl unit by only 2.58 but the proximal N atom on the
triazatruxene unit present in TX is twisted out of alignment
by approximately 108. Such structural facets must contribute
significantly towards the differing extents of electronic cou-
pling found for these two electronic systems.
Table 1. Electrochemical data for the BODIPY, judolidine, and triaza-
truxene derivatives and relevant control compounds.[a]
Compound E0’(ox, soln) [V] (DE [mV]) E0’(red, soln) [V] (DE [mV])
ACHTUNGTRENNUNG
1
2
+1.15 (70)
+0.86 (60)
À1.24 (60)
–
JL (3)
4
JLCV (5)
TX (6)
+0.38 (70), +0.79 (60)
+0.38 (60), +0.81 (60)
+0.38 (70), +0.80 (60)
+0.67 (70), +0.90 (70)
+0.66 (70), +0.92 (70)
+0.72 (60), +0.97 (60)
+0.86 (70), +1.45 (70)
À1.17 (70)
À1.16 (70)
À1.04 (irr.), À1.24 (70)
À1.09 (70)
7
À1.06 (70)
TXCV (8)
9
À0.98 (irr.), À1.10 (60)
–
Further confirmation for the disparate electronic charac-
ter of JL and TX comes from comparison of their absorp-
tion spectra recorded in 2-methyltetrahydrofuran (MTHF).
[a] Potentials determined by cyclic voltammetry in deoxygenated di-
chloromethane, containing 0.1m TBAPF6 [electrochemical window from
+1.6 to À2.2 V], at a solute concentration of approximately 1.5 mm and
at RT. Potentials were standardized versus ferrocene (Fc) as internal ref-
Thus, the spectrum recorded for TX (lMAX =607 nm, eMAX
=
103055mÀ1 cmÀ1) resembles those reported for related styryl-
based BODIPY dyes, although the half-width for the
lowest-energy transition is somewhat elevated, and the total
oscillator strength[35] (f) is 0.86. This latter transition can be
erence and converted to the SCE scale assuming that E1/2ACTHNUTRGNEUNG
(Fc/Fc+)=
+0.38 V (DEp =60 mV) versus SCE. The error in half-wave potentials is
Æ10 mV. For irreversible processes the peak potentials (Eap) are quoted.
All reversible redox steps result from one-electron processes.
13530
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13528 – 13537

Charge-Recombination Fluorescence
FULL PAPER
accounted for in terms of three vibrational modes of
1200 cmÀ1, each with a half-width of 1135 cmÀ1. For JL
under the same conditions, the corresponding absorption
transition (lMAX =631 nm, eMAX =89630mÀ1 cmÀ1) is consid-
erably broader, with f=0.94, and redshifted. Spectral decon-
struction into Gaussian-shaped bands requires three vibra-
tional modes of 1200 cmÀ1 with half-widths of 1500 cmÀ1.
One-electron reduction of the BODIPY core is easier by
approximately 80 mV in both JL and 4 relative to the origi-
nal dye, 1. In fact, this observation is consistent with the op-
tical properties of the monostyryl BODIPY dyes, as dis-
cussed later, and is a consequence of the extended p-delo-
above and are reproduced in Figure 1. Fluorescence is readi-
ly apparent from both compounds in MTHF at room tem-
perature and occurs in the far-red region (Figure 1). For TX,
the emission spectrum is broad, but partially resolved, and
has a maximum at 672 nm in MTHF. The entire spectral en-
velope can be deconstructed into Gaussian-shaped compo-
nents with a common half-width of 1050 cmÀ1. The accompa-
nying vibronic modes coupled with nonradiative decay are
estimated from spectral curve fitting to correspond to a mix-
ture of torsional modes (n=750 cmÀ1, S=0.96, in which S
refers to the Huang–Rhys factor) and skeletal bending
modes (n=1400 cmÀ1, S=0.79).[36] There is good agreement
between the excitation and absorption spectra across the
entire visible and near-UV regions. In MTHF, the fluores-
cence quantum yield is 0.56, and the excited-singlet-state
lifetime, obtained from monoexponential decay curves, is
3.5 ns (Table 2). The spectrum has the appearance of a
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
somewhat easier to reduce (Table 1). This generic behavior
follows from the situation outlined above for the corre-
sponding oxidative processes. Quantum chemical
calculations performed at the DFT level (B3LYP
Table 2. Summary of the photophysical properties recorded for the various amino-
substituted styryl–BODIPY compounds in MTHF at room temperature.
6-31G*) are suggestive of the LUMO being distrib-
uted over the styryl–BODIPY without significant
penetration onto the amino-based donor (see the
Supporting Information).
Property
JL
TX
JLCV
TXCV
lABS [nm]
eMAX [mÀ1 cmÀ1
f[a]
631
103055
0.94
607
89630
0.86
638
88900
1.01
619
114720
1.02
]
In the case of the prototypic push–pull dye,
JLCV, reductive scans point to an irreversible
wave attributable to one-electron reduction of the
dicyanovinyl unit and a reversible wave due to re-
duction of the BODIPY core. Again, electrostatic
repulsion perturbs the latter value. Similar proper-
ties are apparent for TXCV, although there are im-
portant disparities in the derived half-wave poten-
tials for reduction of both units (Table 1), as might
be anticipated in light of the above discussion com-
paring the electronic properties of the two sets of
donors. It is interesting to note that this electronic
effect carries over to reduction of the dicyanovinyl
unit.
dGS [D][b]
lFLU [nm]
S[c]
2.5
3.0
9.3
8.0
720
672
725
685
1.7:1.15
0.30
2.8
0.96:0.79
0.56
3.5
1.05:0.70
0.0020
0.13
0.82:0.65
0.0023
0.26
FF
tS [ns]
kRAD [107 sÀ1
lCR [eV][d]
]
10.7
15.6
1.5
0.9
0.10
0.06
0.075
0.068
[a] Oscillator strength calculated from the integrated absorption spectrum. [b] Com-
puted dipole moment for the ground state. [c] Huang–Rhys factor calculated from the
deconstructed emission spectrum, with the values referring to the medium- and low-
frequency modes, respectively. [d] Reorganisation energy accompanying charge-re-
combination emission extracted from spectral curve fitting.
single component. On cooling the solution, the emission
maximum undergoes a substantive redshift until reaching
Photophysics of the BODIPY–julolidine and BODIPY–tri-
ACHTUNGTRENNUNGazatruxene dyes: The absorption spectral features of the two
amino-substituted styryl-based BODIPY dyes were outlined
727 nm at 160 K from whence further lowering of the tem-
perature causes the maximum to revert to higher energy.
The emission intensity and lifetime decrease slightly during
the redshift, in line with the Englman–Jortner energy-gap
law,[37] but recover during the subsequent blueshift (see the
Supporting Information). These changes are modest but
there is a larger effect as the solvent starts to freeze at
around 140 K. Here, the emission maximum shifts rapidly
with falling temperature until settling at 640 nm at approxi-
mately 115 K.
At <110 K, at which temperatures MTHF forms a glassy
matrix,[38] the emission profile resembles that expected for a
monostyryl BODIPY dye lacking significant charge-transfer
interactions (Figure 2).[39] The emission peak lies at 640 nm
(i.e., 15635 cmÀ1) and the profile can be deconstructed into
Gaussian-shaped bands with
a common half-width of
680 cmÀ1 and accompanying vibronic components of 1300
and 580 cmÀ1. The Huang–Rhys factors (S), which are 0.96
(n=750 cmÀ1) and 0.79 (n=1600 cmÀ1) at room temperature,
Figure 1. Absorption and fluorescence spectra recorded for compounds
JL (black curves) and TX (grey curves) in MTHF at room temperature.
Chem. Eur. J. 2013, 19, 13528 – 13537
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13531

R. Ziessel, A. Harriman et al.
(i.e., n=1300 cmÀ1, S=0.21) and low-frequency (i.e., n=
600 cmÀ1, S=0.53) vibronic peaks with a half-width of
815 cmÀ1. The Huang–Rhys factors[36] derived at 80 K, al-
though much smaller than those found at room temperature,
are still somewhat too high for a pure p–p* transition.
These spectral shifts can be explained in terms of solvation
of a polar species, taking into account that the dielectric
constant of fluid MTHF increases with decreasing tempera-
ture.[41] Again, it is proposed that freezing of the solvent hin-
ders some important structural change that is essential for
effective charge transfer along the molecular axis.
Aided by the quantum chemical studies outlined above,
our understanding of the photophysical properties of TX
can be summarized as follows (Table 2): Illumination creates
an excited-singlet state possessing significant charge-transfer
character, with the resultant transition dipole moment being
spread over most of the molecule. The radiative rate con-
stant (kRAD) decreases with increasing polarity of the sol-
vent, due to a corresponding change in geometry,[42] while
the competing nonradiative rate constant (knr) increases
under the same conditions, due to the energy-gap law.[37]
The net result is that fluorescence is reasonably pronounced
in nonpolar media but is extinguished in strongly polar envi-
ronments. It appears that formation of the excited state de-
mands a change in geometry, as is evident by the unusually
large Stokesꢂ shift of 1550 cmÀ1 in MTHF. This structural
change, presumed to involve mutual alignment of the
p cloud and the N lone pair, is restricted in rigid media in
which the excited state more closely resembles a locally ex-
cited p–p* species. The ground state does not appear to be
a polar species, in agreement with the DFT calculations
(mGS =3.0 D), since the position of the absorption maximum
varies with the polarizability (PONS) of the solvent[43] but not
with the dielectric constant. Indeed, there is a linear rela-
tionship between the wavenumber (nABS) at the maximum of
the lowest-energy absorption transition and PONS (Figure 3),
as has been reported for many classes of nonpolar chromo-
phores,[44] including BODIPY dyes.[45] This linearity permits
estimation of the radius of the solvent cavity housing the
dye as being 4.3 ꢃ and predicts the absorption maximum
under vacuum to be at 560 nm, similar to that of the mono-
styryl dye lacking the amino group.[39] Now the change in
dipole moment accompanying excitation can be estimated
from the Lippert–Mataga relationship[46] as being 15.7 D for
TX (Figure 2). Similar behavior is observed for JL
(Table 2), where the change in dipole moment on excitation
is calculated as approximately 17 D and the Stokesꢂ shift in
MTHF is 1960 cmÀ1.
Figure 2. Effect of temperature on the fluorescence spectral profile re-
corded for TX in MTHF. The arrow indicates the effect of warming the
solution from 80 to 290 K in equal increments.
have values of 0.30 (n=735 cmÀ1) and 0.11 (n=1570 cmÀ1)
in the glassy matrix and are in better accord with those ex-
pected[40] for a p–p* transition. The implication, therefore, is
that the glassy matrix restricts internal charge transfer by
virtue of its rigid nature; it might be noted that the polarity
of glassy MTHF has been likened to that of N,N-dimethyl-
formamide at room temperature.[41] In other solvents at
room temperature, the fluorescence maximum recorded for
TX shows a marked dependence on the static dielectric con-
stant (eS) and undergoes a redshift as the polarity increases.
Thus, the emission maximum transforms from 650 nm in di-
butyl ether (eS =3.1), to 668 nm in ethyl acetate (eS =6.02),
to 720 nm in dichloromethane (eS =8.93), and to approxi-
mately 740 nm in acetonitrile (eS =37.5). The quantum yield
and lifetime fall steadily as the emission peak moves to
lower energy and fluorescence is difficult to resolve in
strongly polar solvents; for details see the Supporting
Information.
For the corresponding julolidine-based fluorophore, JL,
fluorescence is redshifted relative to that of TX and has a
maximum at 720 nm in MTHF at room temperature
(Figure 1). Spectral curve fitting again requires the involve-
ment of three Gaussian-shaped components with a common
half-width of 950 cmÀ1, and with underlying vibronic compo-
nents of 755 and 1310 cmÀ1. Excitation and absorption spec-
tra are in accord across the relevant spectral window. At
room temperature in MTHF solution, FF has a value of 0.30
and tS is 2.8 ns; again, decay curves are well explained in
terms of monoexponential fits. Cooling the solution results
in a redshift, the maximum reaching 755 nm at 160 K and
with an accompanying decrease in emission yield, until ap-
proaching the freezing point of the solvent (see the Support-
ing Information). On lowering the temperature further,
there is a substantial blueshift and a marked escalation in
emission. At 80 K in glassy MTHF, FF reaches a limiting
value of 0.58, and the emission maximum is found at
685 nm. The spectral profile remains fairly broad even in
the glassy matrix but can be analyzed in terms of medium-
In glassy MTHF at 80 K, emission spectra recorded for
the two compounds do not coincide and the julolidine-based
system JL is subjected to a redshift of approximately 45 nm
relative to TX. The same is true for excitation spectra. The
parent monostyryl BODIPY dye[39] that lacks charge-trans-
fer interactions shows absorption and emission maxima at
573 and 583 nm, respectively, thereby giving a Stokesꢂ shift
of approximately 300 cmÀ1. Extrapolation of the Lippert–
Mataga plot recorded for TX also indicates that the Stokesꢂ
13532
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13528 – 13537

Charge-Recombination Fluorescence
FULL PAPER
MTHF shifts the absorption and fluorescence peaks to 575
and 585 nm, respectively, and FF increases to 0.95.
Photophysics of the push–pull compounds JLCV and
TXCV: The push–pull compounds differ from the parent
amino-substituted styryl–BODIPY dyes by virtue of having
an electron-affinic dicyanovinyl unit appended through the
meso position. It might be noted that this type of electron
acceptor has a long history in the generic field of molecular
photophysics.[50] Absorption spectra recorded for these new
push–pull molecules, JLCV and TXCV, in MTHF at room
temperature are similar to those described for the corre-
sponding BODIPY–amine compounds, JL and TX, but are
redshifted by approximately 10 nm and slightly broadened
(Figure 4). Oscillator strengths, which are 1.02 for both com-
Figure 3. Effect of solvent on a) absorption and b) fluorescence spectral
properties recorded for TX at room temperature: panel (a) represents
the effect of solvent polarizability on the absorption maximum for the
lowest-energy transition whereas panel (b) refers to a Lippert–Mataga
plot for the Stokesꢂ shift versus the solvent Pekkar function (f(e)). The
points are experimental values and the solid lines refer to fits to the cor-
responding theoretical model.
shift is 300 cmÀ1 under vacuum, where charge-transfer inter-
actions are likely to be minimal. As such, the amino-substi-
tuted derivatives can be considered to possess increased
conjugation lengths because of the N lone pair, which ac-
counts for the redshifted absorption maximum, in addition
to their inherent charge-transfer character. An important
point to bear in mind is that this charge-transfer effect does
not lead to light-induced charge separation.
Overall, the photophysical properties recorded in fluid
solution appear consistent with those reported by Rurack[47]
and Baruah[48] and their co-workers for the corresponding
monostyryl BODIPY dye bearing a terminal N,N-dimethyl-
Figure 4. a) Absorption and b) fluorescence spectra recorded for JLCV in
MTHF at room temperature (see the Supporting Information for corre-
sponding plots for TXCV). In part (a), the solid black curve is the experi-
mental spectrum, the light grey curve is the deconstructed charge-transfer
transition, and the dark grey curves are Gaussian components for the
normal p–p* transition. In part (b), the solid black curve is the experi-
mental spectrum, the light grey curve is the deconstructed spectrum for
the emissive state and the dark grey curve is the proposed impurity.
ACHTUNGTRENNUNGanilino unit. Here, the absorption spectrum is hardly affect-
ed by changes in solvent but the emission maximum moves
to lower energy as the solvent polarity increases. A particu-
lar emphasis in the work outlined by Baruah et al.[48] is
placed on establishing the most appropriate descriptor for
solvent polarity. We are less concerned with this point since
our main interest lies with the push–pull dyes. Nonetheless,
it is important to highlight the fact that kRAD for the N,N-di-
methylanilino-based dye[47,48] is approximately 1.9ꢄ108 sÀ1,
which is slightly higher than that found in this work, but
consistent with the general observation that kRAD decreases
with increasing charge-transfer character. As reported by
others for related compounds,[47–49] the amino N atom is
readily protonated in acidic solution and this causes a
marked spectral shift for both absorption and emission
maxima. In the case of JL, for example, protonation in
pounds, are increased relative to the parent compounds and,
most importantly, there are indications for the presence of a
charge-transfer transition lying under the lowest-energy
p–p* absorption band. The latter situation is most evident
for JLCV but spectral curve fitting confirms the presence of
a charge-transfer contribution to the total absorption spec-
tral profile of TXCV. The maxima of these transitions are
found at 660 and 724 nm for TXCV and JLCV in MTHF, re-
spectively. The origin of both transitions is considered to in-
volve charge transfer from the styryl–BODIPY to the dicya-
novinyl unit since DFT calculations show the HOMO and
LUMO(1) to be localized on the styryl-based BODIPY resi-
Chem. Eur. J. 2013, 19, 13528 – 13537
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13533

R. Ziessel, A. Harriman et al.
due whereas the LUMO is centered on the dicyanovinyl ap-
pendage. Interestingly, the DFT calculations indicate sub-
stantially increased dipole moments for both push–pull com-
pounds relative to the original amino derivative. Thus, the
computed ground-state dipole moments are 9.3 and 8.0 D
for JLCV and TXCV, respectively.
Fluorescence from the target push–pull systems is difficult
to resolve in liquid MTHF and it is clear that the emission is
extensively quenched relative to the analogues lacking the
dicyanovinyl unit. For TXCV, fluorescence is decreased by a
factor of approximately 300-fold relative to TX while the
emission maximum is shifted to 695 nm (Table 2). In glassy
MTHF at temperatures <115 K, the emission spectrum re-
sembles that recorded for the control compound TX, and at
80 K FF has a value of 0.32, which remains less than that
measured for TX (FF =0.85) under the same conditions
(Figure 5). As the glass melts, there is a dramatic fall in FF
Figure 6. Time-resolved fluorescence decay profile recorded for TXCV in
MTHF following excitation at 635 nm and with detection at 710 nm. The
instrument response function is shown as a black solid curve and the
experimental data points appear as open circles. The analytical fit is
shown as a red curve superimposed over the data points. For this
analysis, c2 =1.09.
Figure 5. Effect of temperature on the fluorescence spectral profile re-
corded for TXCV in MTHF. The arrow indicates the effect of warming
the solution from 80 to 290 K in equal increments. The inset shows the
effect of temperature on the relative emission quantum yield.
Figure 7. Decay-associated emission spectra derived from analysis of the
time-resolved fluorescence spectra recorded for TXCV in MTHF follow-
ing excitation at 590 nm. The grey curve corresponds to the longer-lived
species and the black curve is for the shorter-lived component. See the
Supporting Information for the analogous spectra derived for JLCV.
and a significant redshift, which reaches a maximum at
170 K where the fluorescence peak lies at 740 nm. Under
critical examination of the emission spectral profiles it be-
comes clear that there are important differences between
room-temperature spectra recorded for the push–pull
system and the simpler amino derivative. Indeed, the total
emission spectrum recorded for TXCV appears as a mixture
of a fluorescence profile somewhat similar to that of TX
and a new band lying at lower energy (Figure 5). This
second component is broad and essentially featureless, with
the maximum being strongly temperature dependent.
Time-resolved emission studies carried out for TXCV in
MTHF show that decay profiles require analysis in terms of
dual-exponential kinetics (Figure 6). One component, with a
lifetime of 2.4 ns, is a minor species and is tentatively as-
signed to a trace impurity on account of its spectral profile
being similar to that of TX. The second component, which
has a lifetime of 260 ps, is the dominant species and corre-
sponds to the redshifted species (Figure 7). From the spec-
tral curve fitting analysis, FF for this latter component is es-
timated to be approximately 0.002, hence kRAD must be ap-
proximately 107 sÀ1 at room temperature. The same analysis
can be made for JLCV, where the shorter lifetime is 130 ps
and the residual impurity has a lifetime of 2.8 ns. In this
case, the emission spectrum shows only a minor contribution
from the “impurity” whilst the shorter-lived species has an
emission maximum at 725 nm. The emission quantum yield
for this latter species is very low in MTHF at room tempera-
ture, where kRAD is estimated to be approximately 2ꢄ107 sÀ1.
Temperature-dependent emission studies again show that
the blueshifted emission characteristic of JL is apparent in
the glassy matrix at <110 K but disappears on melting. As
for TXCV, the peak position for the redshifted species is
strongly temperature dependent. On the basis that this emis-
sion arises from a charge-transfer state, the reorganization
energy for charge recombination is approximately 0.1 eV
(Table 2).
The photophysical properties of these push–pull dyes can
be interpreted as follows: The lowest-lying singlet-excited
13534
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13528 – 13537

Charge-Recombination Fluorescence
FULL PAPER
state in both compounds is accessed through direct excita-
tion into the charge-transfer transition identified on the red
edge of the strong absorption band associated with the
amino-substituted styryl–BODIPY dye. This state is weakly
emissive. The excited-state lifetimes are fairly short, owing
to the energy-gap law;[37] the corresponding radiative rate
constants are relatively small because of the low oscillator
strengths for the charge-transfer absorption transition.[42] Il-
lumination into the BODIPY dye forms the corresponding
localized state possessing considerable charge-transfer char-
acter but this species undergoes rapid internal conversion to
populate the lower-energy state. The polar solvent stabilizes
this latter state and thereby gives rise to the observed red-
shift as the temperature falls. In the glassy matrix, such sta-
bilization is not possible,[51] because the solvent cannot re-
Adding the dicyanovinyl acceptor, by means of an orthog-
onal phenylene ring, provides a route for increasing the mo-
lecular dipole moment and is accompanied by a large de-
crease in fluorescence. Electronic coupling is so strong, de-
spite the bridge, that the absorption spectrum shows evi-
dence of a low-lying charge-transfer state. On vitrification,
the emission spectrum moves towards that of the amino–
BODIPY derivative. This effect is explained in terms of the
glass being unable to adequately solvate the charge-transfer
state and preventing the essential geometry change. There is
considerable interest in monitoring the properties of optical
glasses formed by freezing an organic solvent, especially
with regard to the local polarity.[41] In this respect we note
that compounds JLCV and TXCV function as exquisite mir-
rors for the vitrification of the solvent and clearly distin-
guish between a rigid glass and an amorphous glassy regime.
These effects are due to changes in viscosity, rather than po-
larity, but could be adapted to develop improved optical
sensors for glass transition temperatures.
ACHTUNGTRENNUNGorient around the emergent dipole, and the energy of the
charge-transfer state increases dramatically. At <115 K, this
state lies at higher energy than the singlet-excited state asso-
ciated with the amino-substituted styryl–BODIPY dye. The
net result is complete recovery of blueshifted fluorescence
in the glass. In fact, the fluorescence properties (both inte-
grated area and peak position) provide a convenient means
by which to monitor global changes in the nature of the sol-
vent as a function of temperature. Although other dyes have
been used[41,52] as optical probes for the temperature-
induced change in the dielectric constant of MTHF, these
push–pull molecules are especially useful for following
vitrification.
Experimental Section
Synthesis and characterization
General procedure A: In a round-bottomed flask, piperidine and a crystal
of p-toluenesulfonic acid (p-TsOH) were added to a solution of com-
pound
tetrahydro-1H,5H-pyridoAHCTNUGTRENNNUG
1
(1 equiv) and 9-formyl-10-butoxy-1,1,7,7-tetramethyl-2,3,6,7-
[3,2,1-ij]quinoline (1.5 equiv) in toluene. The
solution was heated at 1408C until it had evaporated to dryness. The
colored solids were purified by silica gel chromatography before being
recrystallized.
Conclusion
General procedure B: [PdACTHNUTRGNEUNG(PPh3)2Cl2] (10 mmol%) and sodium formate
(1.2 equiv) were added to a solution of 3 (1 equiv) in anhydrous DMF
(8 mL) in a two-necked flask equipped with a reflux condenser, a gas
bubbler, and a magnetic stirring bar. The reaction mixture was degassed
under a continuous flow of CO at atmospheric pressure and stirred at
908C for 3 h. After cooling to room temperature, the mixture was ex-
tracted with dichloromethane and washed several times with water. The
organic phase was dried over hygroscopic cotton wool and evaporated.
The crude residue was then purified by flash chromatography.
The two amino-substituted monostyryl BODIPY dyes un-
dergo a significant increase in dipole moment on promotion
to the first-allowed excited state. Both dyes are strongly flu-
orescent, without the anticipated quenching by the amino
group, with the emission maximum depending on the local
polarity. The relevant excited states, which possess strong
charge-transfer character, are accessed by means of a key
geometry change that is inhibited in a frozen glass at low
temperature. Consequently, in the case of TX in MTHF, vit-
rification of the solvent is accompanied by a 90 nm blueshift,
which occurs over a temperature range of approximately
30 K. These compounds are characterized by relatively high
radiative rates and it is interesting to note how kRAD de-
pends on the strength of the donor. In fact, it can be argued
that kRAD provides an indirect measure of the internal
charge-transfer characteristics for this class of styryl–
BODIPY dyes. Data collected in different solvents for JL,
TX, and the N,N-dimethylanilino[47,48] derivative show rea-
sonable linearity between kRAD and the emission peak maxi-
mum, which suggests comparable coupling elements for all
three dyes. In the same solvent, the measured Stokesꢂ shifts
are 1960, 1550, and 1420 cmÀ1 for JL, TX, and the N,N-di-
methylanilino compound, respectively, which could be taken
as a relative measure of the donor ability.
General procedure C: In a Schlenk tube equipped with a magnetic stir-
ring bar, malonitrile (1.2 equiv) and several drops of piperidine/acetic
acid were added to a suspension of the aldehyde precursor (1 equiv) in
methanol (2.5 mL). The mixture was stirred at 85–958C for 1 h.
Details regarding the spectroscopic investigations are also provided in
the Supporting Information.
Compound JL (3): Compound 3 was synthesized according to general
procedure A, starting from 4,4-difluoro-8-(4-iodo)phenyl-1,3,5,7-tetra-
methyl-4-bora-3a,4a-diaza-s-indacene (BODIPY; 226 mg, 0.50 mmol).
Column chromatography (silica gel, CH2Cl2/petroleum ether 4:6, v/v) and
recrystallization from ethanol/CH2Cl2 gave 3 (220 mg, 58%) as a dark
blue solid. Rf =0.30 (CH2Cl2/petroleum ether 4:6); 1H NMR (CDCl3,
300 MHz): d=0.99 (t, 3J=7.3 Hz, 3H), 1.33 (s, 6H), 1.41–1.42 (m, 9H),
1.46 (s, 3H), 1.52–1.57 (m, 2H), 1.70–1.74 (m, 4H), 1.80–1.90 (m, 2H),
2.58 (s, 3H), 3.16 (t, 3J=5.7 Hz, 2H), 3.23 (t, 3J=5.7 Hz, 2H), 3.83 (t,
3J=6.6 Hz, 2H), 5.95 (s, 1H), 6.58 (s, 1H), 7.08 (d, 3J=8.2 Hz, 2H), 7.32
3
3
(d, J=16.2 Hz, 1H), 7.37 (s, 1H), 7.42 (d, J=16.2 Hz, 1H), 7.83 ppm (d,
3J=8.2 Hz, 2H); 13C NMR (CDCl3, 50 MHz): d=14.3, 14.1, 14.7, 14.7,
15.2, 19.6, 30.2, 31.0, 32.4, 32.4, 32.8, 36.6, 40.1, 47.0, 47.6, 94.6, 113.2,
113.3, 113.3, 113.3, 117.4, 118.3, 120.3, 120.3, 122.0, 123.6, 126.9, 130.7,
135.3, 136.4, 138.3, 139.5, 142.7, 144.7, 152.2, 156.6, 157.6 ppm; 11B NMR
(CDCl3, 128 MHz): d=1.07 ppm (t, J(B,F)=32.6 Hz); EIMS (neat
matter): m/z calcd for C40H47BF2IN3O: 761.2; found: 761.1 (100) [M+],
Chem. Eur. J. 2013, 19, 13528 – 13537
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13535

R. Ziessel, A. Harriman et al.
688.2 (30) [MÀ
N
8.3 Hz, 2H), 8.26 (d, 3J=8.5, 1H), 8.29 ppm (d, 3J=7.9 Hz, 2H);
13C NMR (CDCl3, 50 MHz): d=13.9, 13.9, 14.1, 14.6, 14.9, 22.4, 22.5,
22.7, 26.3, 26.4, 29.3, 29.7, 31.4, 31.4, 31.4, 31.9, 47.1, 84.2, 110.3, 110.6,
119.3, 119.8, 119.8, 119.8, 121.4, 121.5, 121.6, 122.7, 123.3, 123.4, 124.6,
129.7, 130.2, 130.9, 131.3, 138.5, 138.8, 139.4, 140.9, 141.0, 141.2, 142.0,
154.5, 155.1, 158.7 ppm; 11B NMR (CDCl3, 128 MHz): d=1.07 ppm (t,
J(B,F)=32.0 Hz); IR: n˜ =710, 727, 800, 984, 1018, 1039, 1077, 1158, 1194,
Compound 4: Compound 4 was synthesized according to general proce-
dure B, starting from compound 3 (155 mg, 0.203 mmol). The crude resi-
due so obtained was purified by column chromatography (silica gel), elut-
ing with toluene/CH2Cl2 (8:2, v/v) to give the desired compound as a
dark blue solid (95 mg, 71%). Rf =0.46 (toluene/CH2Cl2 8:2); 1H NMR
(CDCl3, 300 MHz): d=0.98 (t, 3J=7.2 Hz, 3H), 1.33–1.42 (m, 18H),
1.51–1.57 (m, 2H), 1.73 (t, 3J=6.0 Hz, 4H), 1.81–1.88 (m, 2H), 2.59 (s,
3H), 3.16–3.24 (m, 4H), 3.83 (t, 3J=6.7 Hz, 2H), 5.96 (s, 1H), 6.59 (s,
1H), 7.46–7.48 (m, 3H), 7.54 (d, 3J=8.1 Hz, 2H), 8.02 (d, 3J=8.1 Hz,
2H), 10.12 ppm (s, 1H); 13C NMR (CDCl3, 50 MHz): d=14.2, 14.4, 14.6,
15.0, 19.5, 29.74, 30.1, 30.8, 30.9, 32.3, 32.7, 36.4, 40.0, 46.9, 47.5, 113.1,
117.3, 118.4, 120.3, 121.9, 123.5, 125.3, 126.9, 128.3, 129.1, 129.8, 130.2,
132.8, 135.6, 135.8, 136.5, 137.9, 139.1, 142.2, 142.3, 142.4, 144.7, 152.3,
156.8, 157.6, 191.7 ppm; 11B NMR (CDCl3, 128 MHz): d=1.07 ppm (t,
J(B,F)=33.2 Hz); IR: n˜ =701, 763, 976, 1061, 1116, 1156, 1194, 1254,
À
1299, 1373, 1409, 1463, 1500, 1540, 1589, 2229 ( CN), 2853, 2923,
2954 cmÀ1
;
EIMS (neat matter): m/z calcd for C66H68BF2N7: 1007.5;
found: 1007.3 (100) [M]+, 988.3 (20) [MÀF]+; elemental analysis calcd
(%) for C66H68BF2N7: C 78.63, H 6.80, N 9.73; found: C 78.42, H 6.62, N
9.42.
Acknowledgements
1290, 1317, 1464, 1497, 1532, 1581, 1702 (CHO), 2869, 2927, 2957 cmÀ1
We thank the CNRS, EPSRC (EP/G04094X/1), ECPM-Strasbourg, and
Newcastle University for financial support of this work. We gratefully ac-
knowledge IMRA Europe S.A.S. (Sophia Antipolis, France) for awarding
a PhD fellowship to A.N. Drs. Stꢀphane Jacob and Gilles Dennler
(IMRA Europe) are thanked for many helpful and fruitful discussions.
Dr. Takashi Mitsumoto and Dr. Hiroyuki Katsuda are also acknowledged
for their continuous encouragement.
EIMS (neat matter): m/z calcd for C41H48BF2N3O2: 663.3; found: 663.2
(100) [M]+, 590.2 (35) [MÀ
C41H48BF2N3O2: C 74.20, H 7.29, N 6.33; found: C 73.92, H 6.98, N 6.12.
Compound JLCV (5): Compound 5 was synthesized according to general
procedure C, starting from compound 4 (39 mg, 0.056 mmol). After cool-
ing the reaction to room temperature, the precipitate formed was isolated
by centrifugation and washed several times with methanol, giving the
product as
a
dark blue powder (20 mg, 50%). 1H NMR (CDCl3,
300 MHz): d=0.98 (t, 3J=7.3 Hz, 3H), 1.33–1.42 (s, 18H), 1.50–1.55 (m,
2H), 1.71–1.75 (m, 4H), 1.80–1.90 (m, 2H), 2.59 (s, 3H), 3.17 (t, 3J=
5.8 Hz, 2H), 3.25 (t, 3J=5.8 Hz, 2H), 3.82 (t, 3J=6.7 Hz, 2H), 5.96 (s,
1H), 6.60 (s, 1H), 7.73–7.48 (m, 3H), 7.55 (d, 3J=8.2 Hz, 2H), 7.84 (s,
1H), 8.05 ppm (d, 3J=8.3 Hz, 2H); 13C NMR (CDCl3, 100 MHz): d=
14.3, 14.5, 14.6, 15.1, 19.5, 29.4, 29.7, 30.0, 30.8, 32.3, 32.3, 32.7, 36.4, 39.9,
46.9, 47.5, 75.9, 83.9, 112.5, 112.9, 113.5, 117.2, 118.7, 120.4, 121.9, 123.5,
126.9, 129.8, 130.0, 130.5, 131.1, 131.2, 132.7, 134.7, 135.9, 138.7, 142.2,
142.6, 144.8, 152.3, 157.1, 157.7, 158.9, 165.5 ppm; 11B NMR (CDCl3,
[1] M. C. Ruiz Delgado, J. Casado, V. Hernꢅndez, J. T. Lꢆpez Navarrete,
J. Orduna, B. Villacampa, R. Alicante, J.-M. Raimundo, P. Blan-
chard, J. Roncali, J. Phys. Chem. C 2008, 112, 3109–3120.
[2] J. Li, D. Liu, Z. Hong, S. Tong, P. Wang, C. Ma, O. Lengyel, C.-S.
Lee, H.-L. Kwong, S. Lee, Chem. Mater. 2003, 15, 1486–1490.
[3] B. A. Reinhardt, L. L. Brott, S. J. Clarson, A. G. Dillard, J. C. Bhatt,
R. Kannan, L. Yuan, G. S. He, P. N. Prasad, Chem. Mater. 1998, 10,
1863–1874.
[4] D. R. Kanis, M. A. Ratner, T. J. Marks, Chem. Rev. 1994, 94, 195–
242.
128 MHz): d=1.04 (t, JACHTUNGTRENNUNG(B,F)=32.9 Hz); IR: n˜ =702, 795, 854, 984, 1014,
[5] M. Blanchard-Desce, V. Alain, P. V. Bedworth, S. R. Marder, A.
Fort, C. Runser, M. Barzoukas, S. Lebus, R. Wortmann, Chem. Eur.
J. 1997, 3, 1091–1104.
[6] a) F. Tancini, Y.-L. Wu, B. Schweizer, J.-P. Gisselbrecht, C. Boudon,
P. D. Jarowski, M. T. Beels, I. Biaggo, F. Diederich, Eur. J. Org.
Chem. 2012, 2756–2771; b) M. Kivala, F. Diedrich, Acc. Chem. Res.
2009, 42, 235–248.
[7] A. Myers Kelley, J. Phys. Chem. A 1999, 103, 6891–6903.
[8] a) S. Priyadarshy, M. J. Terien, D. N. Beratan, J. Am. Chem. Soc.
1996, 118, 1504–1510; b) R. A. Rijkenberg, D. Bebelaar, W. J.
Buma, J. W. Hofstraat, J. Phys. Chem. A 2002, 106, 2446–2456; c) B.
Schmidt, C. Sobotta, S. Malkmus, S. Laimgruber, M. Braun, W.
Zinth, P. Gilch, J. Phys. Chem. A 2004, 108, 4399.
À
1156, 1196, 1257, 1294, 1320, 1465, 1496, 1533, 1583, 2228 ( CN), 2963,
2961, 2926 cmÀ1; EIMS (neat matter): m/z calcd for C44H48BF2N5O:
711.4; found: 711.2 (100) [M]+, 692.2 (30) [MÀF]+; elemental analysis
calcd (%) for C44H48BF2N5O: C 74.26, H 6.80, N 9.84; found: C 73.94, H
6.62, N 9.62.
Compound 7: Compound 7 was synthesized according to general proce-
dure B, starting from compound 6a (101 mg, 0.077 mmol). Flash chroma-
tography from silica gel (toluene/petroleum ether 7:3, v/v) gave the prod-
uct as a dark blue solid (78%, 73 mg). Rf =0.48 (toluene/petroleum ether
7:3); 1H NMR (CDCl3, 300 MHz): d=0.78–0.85 (m, 9H), 1.19–1.43 (m,
25H), 1.93–2.04 (m, 6H), 2.66 (s, 3H), 4.88–4.96 (m, 6H), 6.04 (s, 1H),
6.73 (s, 1H), 7.35 (t, 3J=7.4 Hz, 2H), 7.43–7.50 (m, 3H), 7.55 (d, 3J=
8.0 Hz, 2H), 7.62–7.70 (m, 3H), 7.71 (s, 1H), 7.85 (d, 3J=16.0 Hz, 1H),
[9] C.-C. Lee, A. T. Hu, Dyes Pigm. 2003, 59, 63–69.
8.04 (d, J=8.0 Hz, 2H), 8.22–8.29 (m, 3H), 10.12 ppm (s, 1H); 11B NMR
(CDCl3, 128 MHz): d=1.10 ppm (t, JACTHNUTRGNEUGN(B,F)=32.6 Hz); IR: n˜ =706, 727,
3
[10] Y. J. Chang, T. J. Chow, J. Mater. Chem. 2011, 21, 3091–3099.
[11] J. Kulhanek, F. Bures, A. Wojciechowski, M. Makowska-Janusik, E.
Gondek, I. V. Kityk, J. Phys. Chem. A 2010, 114, 9440–9446.
[12] S. H-I. Kato, F. Diederich, Chem. Commun. 2010, 46, 1994–2006.
[13] a) K. D. Belfield, D. J. Hagan, E. W. Van Stryland, K. J. Schafer,
R. A. Negres, Org. Lett. 1999, 1, 1575–1578; b) E. Ortꢇ, P. M. Virue-
la, R. Viruela, F. Effenberger, V. Hernꢅndez, J. T. Lꢆpez Navarrete,
J. Phys. Chem. A 2005, 109, 8724–8731.
[14] a) M. Yeung, A. C. H. Ng, M. G. B. Drew, E. Vorpagel, E. M. Brei-
tung, R. J. McMahan, D. K. P. Ng, J. Org. Chem. 1998, 63, 7143–
7150; b) S. M. LeCours, H.-W. Guan, S. G. DiMagno, C. H. Wang,
M. J. Therien, J. Am. Chem. Soc. 1996, 118, 1497–1503; c) N. Jiang,
G. Zuber, S. Keinan, A. Nayak, W. Yang, M. J. Therien, D. N. Bera-
tan, J. Phys. Chem. C 2012, 116, 9724–9733.
794, 984, 1015, 1075, 1157, 1193, 1260, 1298, 1372, 1464, 1499, 1538, 1705
( CHO), 2852, 2923, 2958 cmÀ1; EIMS (neat matter): m/z calcd for
À
C63H68BF2N5O: 959.5; found: 959.3 (100) [M]+, 940.3 (35) [MÀF]+; ele-
mental analysis calcd (%) for C63H68BF2N5O: C 78.82, H 7.14, N 7.29;
found: C 78.54, H 6.72, N 6.82.
Compound TXCV (8): Malonitrile (7 mg, 0.112 mmol) and several drops
of piperidine/acetic acid were added to a Schlenk tube equipped with a
stirrer bar and containing a solution of aldehyde precursor 7 (68 mg,
0.056 mmol) in THF (3 mL). The mixture was stirred at 958C for 20–
30 min. After purification by column chromatography (silica gel, toluene/
CH2Cl2 8:2, v/v), the product was obtained as a dark blue solid (22 mg,
50%). Rf =0.64 (toluene/CH2Cl2 8:2); 1H NMR (CDCl3, 400 MHz): d=
0.78–0.85 (m, 9H), 1.21–1.33 (m, 18H), 1.41 (s, 3H), 1.47 (s, 3H), 1.94–
2.03 (m, 6H), 2.65 (s, 3H), 4.91–4.98 (m, 6H), 6.05 (s, 1H), 6.76 (s, 1H),
7.36 (t, 3J=7.4 Hz, 2H), 7.46 (t, 3J=8.1 Hz, 2H), 7.54 (d, 3J=7.6 Hz,
1H), 7.58 (d, 3J=8.3 Hz, 2H), 7.36–7.66 (m, 2H), 7.69 (d, 3J=8.5 Hz,
1H), 7.75 (s, 1H), 7.85 (d, 3J=16.2 Hz, 1H), 7.86 (s, 1H), 8.08 (d, 3J=
[15] a) N. S. Lewis, D. G. Nocera, Proc. Natl. Acad. Sci. USA 2006, 103,
15729–15735; b) N. Armaroli, V. Balzani, Angew. Chem. 2007, 119,
52–67; Angew. Chem. Int. Ed. 2007, 46, 52–66; c) J. Liu, G. Cao, Z.
Yang, D. Wang, D. Dubois, X. Zhou, G. L. Graff, L. R. Pederson, J.-
G. Zhang, ChemSusChem 2008, 1, 676–697.
13536
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13528 – 13537

Charge-Recombination Fluorescence
FULL PAPER
[16] a) J. Jortner, J. Chem. Phys. 1976, 64, 4860–4867; b) J. Jortner, J.
Am. Chem. Soc. 1980, 102, 6676–6686; c) S. Franzen, R. F. Gold-
stein, S. G. Boxer, J. Phys. Chem. 1993, 97, 3040–3053; d) C. C.
Page, P. Moser, X. Chen, P. L. Dutton, Nature 1999, 402, 47–52.
[17] a) M. R. Wasielewski, Chem. Rev. 1992, 92, 435–461; b) D. Gust,
T. A. Moore, A. L. Moore, Acc. Chem. Res. 1993, 26, 198–205;
c) H. B. Gray, J. R. Winkler, Ann. Rev. Biochem. 1996, 65, 537–561.
[18] R. A. Marcus, J. Chem. Phys. 1956, 24, 966–978.
[35] A. Harriman, M. Hissler, R. Ziessel, Phys. Chem. Chem. Phys. 1999,
1, 4203–4211.
[36] Z. Shuai, J. L. Brꢀdas, W. P. Su, Chem. Phys. Lett. 1994, 228, 301–
306.
[37] R. Englman, J. Jortner, Mol. Phys. 1970, 18, 145–170.
[38] a) R. Richert, Chem. Phys. Lett. 1992, 199, 355–359; b) C. Reich-
ardt, Chem. Rev. 1994, 94, 2319–2358.
[39] W. W. Qin, T. F. Rohand, W. Dehaen, J. N. Clifford, K. Driessen, D.
Beljonne, B. Van Averbeke, M. Van der Auweraer, N. Boens, J.
Phys. Chem. A 2007, 111, 8588–8597.
[19] D. Holten, D. F. Bocian, J. S. Lindsey, Acc. Chem. Res. 2002, 35, 57–
69.
[20] a) N. Armaroli, Photochem. Photobiol. Sci. 2003, 2, 73–87; b) T. S.
Balaban, Acc. Chem. Res. 2005, 38, 612–623.
[40] R. Ziessel, S. Rihn, A. Harriman, Chem. Eur. J. 2010, 16, 11942–
11953.
[21] L. Sꢅnchez, N. Martꢇn, D. Guldi, Angew. Chem. 2005, 117, 5508–
5516; Angew. Chem. Int. Ed. 2005, 44, 5374–5382.
[22] F. J. M. Hoeben, P. Jonkheijm, E. W. Meijer, A. P. H. J. Schenning,
Chem. Rev. 2005, 105, 1491–1556.
[41] G. U. Bublitz, S. G. Boxer, J. Am. Chem. Soc. 1998, 120, 3988–3992.
[42] a) M. Bixon, J. Jortner, J. W. Verhoeven, J. Am. Chem. Soc. 1994,
116, 7349–7355; b) J. W. Verhoeven, T. Scherer, B. Wegewijs, R. M.
Hermant, J. Jortner, M. Bixon, S. Depaemelaere, F. C. De Schryver,
Rec. Trav. Chim. Pays-Bas 1995, 114, 443–449.
[43] F. A. Bovey, S. S. Yanari, Nature 1960, 186, 1042–1044.
[44] a) I. Renge, J. Phys. Chem. 1995, 99, 15955–15962; b) T. Renger, B.
Grundkçtler, M. E. A. Madjet, F. Mꢉh, Proc. Natl. Acad. Sci. USA
2008, 105, 13235–13240.
[45] A. Filarowski, M. Kluba, K. Cieslik-Boczula, A. Koll, A. Kochel, L.
Pandey, W. M. de Borggraeve, M. Van der Auweraer, J. Catalan, N.
Boens, Photochem. Photobiol. Sci. 2010, 9, 996–1008.
[46] a) E. Lippert, Z. Naturforsch. A: Phys. Sci. 1955, 10, 541–545; b) N.
Mataga, Y. Kaifu, M. Koizumi, Bull. Chem. Soc. Jpn. 1955, 28, 690–
691; c) N. Mataga, Y. Kaifu, M. Koizumi, Bull. Chem. Soc. Jpn.
1956, 29, 465–470.
[47] a) K. Rurack, M. Kollmannsberger, J. Daub, New J. Chem. 2001, 25,
289–292; b) Y. H. Yu, A. B. Descalzo, Z. Shen, H. Rohr, Q. Liu,
Y. W. Wang, M. Spieles, L. Z. Li, K. Rurack, X. Z. You, Chem.
Asian J. 2006, 1, 176–187.
[48] M. Baruah, W. W. Qin, C. Flors, J. Hofkens, R. A. L. Vallee, D. Bel-
jonne, M. Van der Auweraer, W. M. De Borggraeve, N. Boens, J.
Phys. Chem. A 2006, 110, 5998–6009.
[23] N. Saki, T. Dinc, E. U. Akkaya, Tetrahedron 2006, 62, 2721–2725.
[24] a) R. Ziessel, G. Ulrich, A. Harriman, New J. Chem. 2007, 31, 496–
501; b) R. Ziessel, Compt. Rendus Acad. Sciences Chimie 2007, 10,
622–629; c) A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891–
4932; d) G. Ulrich, R. Ziessel, A. Harriman, Angew. Chem. 2008,
120, 1202–1219; Angew. Chem. Int. Ed. 2008, 47, 1184–1201.
[25] a) S. Kolemen, Y. Cakmak, S. Erten-Ela, Y. Altay, J. Brendel, M.
Thelakkat, E. U. Akkaya, Org. Lett. 2010, 12, 3812–3815; b) T.
Rousseau, A. Cravino, J. Roncali, T. Bura, G. Ulrich, R. Ziessel,
Chem. Commun. 2009, 1673–1675; c) T. Rousseau, A. Cravino, J.
Roncali, T. Bura, G. Ulrich, R. Ziessel, J. Mater. Chem. 2009, 19,
2298–2300; d) T. Rousseau, A. Cravino, E. Ripaud, P. Leriche, S.
Rhin, A. De Nicola, R. Ziessel, J. Roncali, Chem. Commun. 2010,
46, 5082–5084; e) B. Kim, B. Ma, V. R. Donuru, H. Liu, J. M. J. Fre-
chet, Chem. Commun. 2010, 46, 4148–4150.
[26] a) G. Ulrich, C. Goze, L. Charbonniꢈre, R. Ziessel, Chem. Eur. J.
2003, 9, 3748–3755; b) S. Atilgan, T. Ozdemir, E. U. Akkaya, Org.
Lett. 2008, 10, 4065–4067; c) S. Atilgan, T. Ozdemir, E. U. Akkaya,
Org. Lett. 2010, 12, 4792–4795.
[27] S. L. Niu, C. Massif, G. Ullrich, R. Ziessel, P.-Y. Renard, A.
Romieu, Org. Biomol. Chem. 2010, 8, 66–69.
[49] R. Ziessel, G. Ulrich, A. Harriman, M. A. H. Alamiry, B. Stewart, P.
Retailleau, Chem. Eur. J. 2009, 15, 1359–1369.
[28] a) R. Ziessel, P. Retailleau, K. J. Elliott, A. Harriman, Chem. Eur. J.
2009, 15, 10369–10376; b) X. Yin, Y. Li, Y. Zhu, X. Tang, H. Zheng,
D. Zhu, Tetrahedron 2009, 65, 8373–8377; c) X. Yin, Y. Li, Y. Zhu,
X. Jing, Y. Li, D. Zhu, Dalton Trans. 2010, 39, 9929–9935; d) W.-J.
Shi, P.-C. Lo, A. Singh, I. K. P. Ledoux-Rak, D. Ng, Tetrahedron
2012, 68, 8712–8718.
[29] a) S. Niu, G. Ullrich, P. Retailleau, R. Ziessel, Org. Lett. 2011, 13,
4996–4999; b) S. L. Niu, G. Ullrich, P. Retailleau, R. Ziessel, Tetra-
hedron Lett. 2011, 52, 4848–4853.
[30] a) J. J. Holt, B. D. Caltree, J. Vincek, M. K. Gannon, M. R. Detty, J.
Org. Chem. 2007, 72, 2690–2693; b) A. R. Katritzky, B. Rachwal, S.
Rachwal, J. Org. Chem. 1996, 61, 3117–3125; c) K. H. Lee, M. H.
Park, S. M. Kim, Y. K. Kim, S. S. Yoon, Jpn. J. Appl. Phys. 2010, 49,
08JG02–08JG04.
[50] a) J. A. Levitt, P. H. Chung, M. K. Kuimova, G. Yahioglu, Y. Wang,
J. L. Qu, K. Suhling, ChemPhysChem 2011, 12, 662–672; b) E.
Ishow, R. Guillot, G. Buntinx, O. Poizat, J. Photochem. Photobiol. A
2012, 234, 27–36; c) F. J. Hoogesteger, C. A. van Walree, L. W. Jen-
neskens, M. R. Roest, J. W. Verhoeven, W. Schuddeboom, J. J. Piet,
J. M. Warman, Chem. Eur. J. 2000, 6, 2948–2959; d) D. L. Phillips,
I. R. Gould, J. W. Verhoeven, D. Tittelbach-Helmrich, A. B. Meyers,
Chem. Phys. Lett. 1996, 258, 87–93; e) J. M. Lawson, M. N. Paddon-
Row, W. Schuddeboom, J. M. Warman, A. H. A. Clayton, K. P.
Ghiggino, J. Phys. Chem. 1993, 97, 13099–13106; f) M. Koeberg, M.
de Groot, J. W. Verhoeven, N. R. Lokan, M. J. Shephard, M. N.
Paddon-Row, J. Phys. Chem. A 2001, 105, 3417–3424.
[51] F. B. Dias, S. Pollock, G. Hedley, L. O. Palsson, A. R. Monkman, I. I.
Perepichka, I. F. Perepichka, M. Tavasli, M. R. Bryce, J. Phys.
Chem. B 2006, 110, 19329–19339.
[52] C. Streck, R. Richert, Ber. Bunsen.-Gesell. Phys. Chem. Chem. Phys.
1994, 98, 619–625.
[31] T. Bura, N. Leclerc, S. Fall, P. LꢀvÞque, T. Heiser, R. Ziessel, Org.
Lett. 2011, 13, 6030–6033.
[32] A. Zhu, B. Wang, J. O. White, H. G. Drickamer, J. Phys. Chem. A
2003, 107, 6932–6935.
[33] C.-H. Lee, H.-J. Yoon, J.-S. Shim, W.-D. Jang, Chem. Eur. J. 2012,
18, 4513–4516.
[34] a) R. Ziessel, L. Bonardi, P. Retailleau, G. Ulrich, J. Org. Chem.
2006, 71, 3093; b) T. Bura, R. Ziessel, Tetrahedron Lett. 2010, 51,
2875–2879.
Received: March 18, 2013
Revised: June 14, 2013
Published online: August 14, 2013
Chem. Eur. J. 2013, 19, 13528 – 13537
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13537