Synthesis, spectroscopy, nonlinear optics, and theoretical investigations of thienylethynyl octopoles with a tunable core

Article abstract of DOI:10.1002/chem.200900702

We have synthesized three new molecules that have three thienylethynyl arms substituting a central benzene core and different electron donor/acceptor groups in the three remaining phenyl positions. The absorption, fluorescence, phosphorescence, and transi

Full text of DOI:10.1002/chem.200900702

FULL PAPER
DOI: 10.1002/chem.200900702
Synthesis, Spectroscopy, Nonlinear Optics, and Theoretical Investigations of
Thienylethynyl Octopoles with a Tunable Core
Marꢀa Moreno Oliva,[a] Juan Casado,*[a] Juan T. Lꢁpez Navarrete,*[a]
Gunther Hennrich,*[b] Stijn van Cleuvenbergen,[c] Inge Asselberghs,[c] Koen Clays,[c]
M. Carmen Ruiz Delgado,[d] Jean-Luc Brꢂdas,[d] J. Sꢂrgio Seixas de Melo,[e] and
Luisa De Cola[f]
Abstract: We have synthesized three
new molecules that have three thienyl-
ethynyl arms substituting a central ben-
yields, lifetimes, and rate constants)
could be derived. It was found that the
major deactivation pathway is internal
conversion, which competes with the
fluorescence and intersystem crossing
processes. For the three investigated
compounds, we provide convincing the-
oretical support corroborating these
findings and further conclusions based
on the theoretical information ob-
tained. These molecules are one of the
very few cases in which the depolariza-
tion ratios, obtained from the NLO op-
tical measurements, clearly reflect the
octopolar configuration. Molecular hy-
perpolarizabilities have been measured
and display a typical dependence on
the donor–acceptor substitution pat-
tern.
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
zene core and different electron donor/
acceptor groups in the three remaining
phenyl positions. The absorption, fluo-
rescence, phosphorescence, and transi-
ent triplet–triplet spectra are analyzed
in the light of the electronic structure
of the ground and excited states ob-
tained from quantum-chemical calcula-
tions. From the above, the relevant
photophysical data (including quantum
Keywords: density functional calcu-
lations · donor–acceptor systems ·
nonlinear optics · phosphorescence ·
sulfur heterocycles
Introduction
[a] M. M. Oliva, Dr. J. Casado, Prof. J. T. Lꢀpez Navarrete
Department of Physical Chemistry, University of Mꢁlaga
Campus de Teatinos s/n, Mꢁlaga 29071 (Spain)
E-mail: casado@uma.es
The synthesis of extended p-conjugated systems has been
the key to providing organic materials with electro-optical
properties needed for particular applications.[1] Over the
past few years, numerous synthetic routes have been investi-
gated to optimize the molecular hyperpolarizabilities in oc-
topolar molecules for nonlinear optical (NLO) response[2]
and, more recently, two-photon absorption (TPA) proper-
ties.[3] Many of these organic systems present C3-symmetric
architectures in which a central core is trigonally substituted
with conjugated branches in between donor–acceptor
groups and conjugative interactions with the center have
been established. These types of dyes are also examples of
molecules with p-electron delocalization extending in two
dimensions (2D), which were recently successfully exploited
as semiconductor elements in organic field-effect transistors
and in light-emitting diodes.[4] It is clear that the design of
these molecules, or improved derivatives, must be founded
on comprehensive knowledge of their electronic characteris-
tics, in particular, their ground- and excited-state properties.
For instance, it has been shown that donor–acceptor
[b] Dr. G. Hennrich
Departamento de Quꢂmica Orgꢁnica
Universidad Autꢀnoma de Madrid
Cantoblanco, Madrid 28049 (Spain)
E-mail: teodomiro@uma.es
[c] Dr. S. van Cleuvenbergen, Dr. I. Asselberghs, Dr. K. Clays
Department of Chemistry, Katholieke Universiteit Leuven
Celestijnenlaan 200D and F, B-3001 Leuven (Belgium)
[d] Dr. M. C. Ruiz Delgado, Prof. J.-L. Brꢃdas
School of Chemistry and Biochemistry and Center for Organic Pho-
tonics and Electronics, Georgia Institute of Technology
Atlanta, Georgia 30332-0400 (USA).
[e] Prof. J. S. Seixas de Melo
Department of Chemistry
University of Coimbra, 3004-535 Coimbra (Portugal)
[f] Prof. L. De Cola
Physikalisches Institut, Westfꢄlische Wilhelms-Universitꢄt Mꢅnster
Mendelstrasse 7, 48149 Mꢅnster (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900702.
Chem. Eur. J. 2009, 15, 8223 – 8234
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8223

strength, conjugation length, geometry, and nature of the
relevant low-lying electronic states significantly influence
the octopolar properties.[2,5]
Experimental and theoretical details
Materials and synthesis: The target compounds were obtained in good
yields by three-fold Sonogashira coupling of the 1,3,5-triiodobenzene de-
rivatives with 3-ethynylthiophene (Scheme 1).[12]
In the context of second-order nonlinear optics, one of
the main arguments advanced in favor of such 2D octopolar
molecules is the absence of a permanent dipole moment, in
contrast to conventional linear systems that tend to self-as-
semble in an antiparallel manner.[2] However, reaching non-
centrosymmetric superstructures remains a crucial prerequi-
site that is not trivial to achieve.[6] Recent accomplishments
in the preparation of second-order NLO devices from crys-
talline[7] and liquid-crystalline materials[8] based on octopolar
molecules have justified the expectations put into this kind
of chromophores as active building units in electro-optical
bulk materials.[9]
In our quest to understand the structure–property rela-
tionships in these molecules, we combined a common ben-
zene core trigonally substituted in the 1, 3, and 5 positions
with terminal thienyl groups connected through an acetylene
spacer. This spacer is recurrent because it offers efficient
electronic communication within the molecular scaffold
while minimizing the detrimental steric congestion at the
same time.[10] The central benzene core is substituted in its
2, 4, and 6 positions with methyl (donor) groups or fluorine
(acceptor) atoms.
Scheme 1. Synthesis of TEB, TEM, and TEBF.
General procedure for the synthesis of TEBF, TEB, and TEM: The re-
spective 1,3,5-triiodobenzene (1.0 mmol) was stirred with [PdCl2ACHTUNGTRENNUNG(PPh3)2]/
CuI (0.15 mmol) in degassed diisopropylamine (10 mL) under argon at
RT for 30 min before 3-ethynylthiophene (4.0 mmol) was added. After
heating the mixture at 708C for 18 h, the solvent was evaporated, water
(30 mL) added, and the mixture extracted with EtOAc (3ꢇ15 mL). The
combined organic layers were dried over MgSO4. The solvent was evapo-
rated to leave the crude solid product, which was purified by recrystalli-
zation from the solvents specified below.
In continuation of our recent interest in C3 octopoles,[5]
we report here on the photophysical properties of these
thiophene-based systems. Special emphasis has been placed
on characterizing the features of the main low-lying excited
states intervening in the properties of interest. This work ex-
plores the absorption, steady-state and time-resolved emis-
sion (fluorescence and phosphorescence), and transient trip-
let–triplet absorption to account for the energies of the
lowest-lying excited states together with the main excitation
and relaxation routes in photoluminescence. Furthermore,
we make use of vibrational Raman spectroscopy to scan the
molecular structure of the electronic ground state. An addi-
tional section is devoted to the second-order nonlinear opti-
cal (NLO) response, as measured by hyper-Rayleigh scatter-
ing.[11] The whole experimental study is guided by a theoreti-
cal analysis combining a variety of computational ap-
proaches, including Density Functional Theory and correlat-
ed Hartree–Fock-based methods. In the overall discussion,
we put strong emphasis on the role of the excited electronic
states of these conjugated octopolar molecules. Owing to
the importance of these states for a variety of applications
in organic electronic, this paper provides a perspective fo-
cused on the analysis of their features. The establishment of
structure–property relationships is strongly pursued. To the
best of our knowledge, there are few studies[5] that utilize
such a variety of spectroscopic tools, combined with theoret-
ical predictions, all directed to elucidating the excited-state
molecular properties.
1,3,5-Tris(thienylethynyl)benzene (TEB): Recrystallization from toluene
gave pure TEB as off-white crystals in 61% yield. Rf =0.57 (20:1 hex-
anes/EtOAc); m.p. 1968C; 1H NMR (CDCl3): d=7.60 (s, 3H), 7.55 (m,
3H), 7.33–7.19 ppm (m, 6H); 13C NMR: d=133.7, 129.8, 129.1, 125.5,
124.0, 121.8, 87.3, 85.6 ppm; (EI+-HRMS): m/z: calcd for C24H12S3:
396.0101; found: 396.0103.
1,3,5-Tris(thienylethynyl)mesitylene (TEM): Recrystallization from tolu-
ene/EtOH (5:1) gave pure TEM as colorless crystals in 70% yield. Rf =
0.52 (20:1 hexane/EtOAc); m.p. 2168C; 1H NMR (CDCl3): d=7.55–7.53
(m, 3H), 7.35–7.31 (m, 3H), 7.24–7.21 (m, 3H), 2.71 ppm (s, 9H);
13C NMR: d=141.8, 129.7, 128.2, 125.4, 122.5, 121.2, 92.3, 86.3, 20.3 ppm;
(EI+-HRMS): m/z: calcd for C27H18S3: 438.0571; found: 438.0553.
1,3,5-Tris(thienylethynyl)-2,4,6-trifluorobenzene (TEBF): Recrystalliza-
tion from CH3CN gave pure TEBF as pale purple crystals in 72% yield.
Rf =0.52 (20:1 hexanes/EtOAc); m.p. 194-1968C; 1H NMR (CDCl3): d=
7.56–7.55 (m, 3H), 7.27–7.24 (m, 3H), 7.18–7.15 ppm; 13C NMR: d=
161.8 (d), 131.2, 130.1, 129.7, 125.6, 120.8, 95.0, 73.5 ppm; (EI+-HRMS):
m/z: calcd for C24H9F3S3: 449.9819; found: 449.9818.
Spectroscopic measurements: Absorption and fluorescence spectra were
recorded on Shimadzu UV-2100 and Horiba–Jobin–Ivon SPEX Fluorog
3-22 spectrometers, respectively. The fluorescence spectra were corrected
for the wavelength response of the system. The fluorescence quantum
yields were measured using bithiophene (FF =0.014 in methylcyclohex-
ane)[13] as the standard. Fluorescence decays were measured using a
home-built TCSPC apparatus described elsewhere[14] and were analyzed
using the modulating functions method of Striker.[15] The experimental
setup used to obtain triplet spectra and triplet yields has been described
elsewhere.[14,16] First-order kinetics were observed in all cases for the
decay of the lowest triplet state, and the lifetime values were in the ms
range. When determining the triplet yields, special care was taken to
have optically matched dilute solutions (abs ꢀ0.2 in a 10 mm square cell)
and low laser energy (ꢁ2 mJ) to avoid multiphoton and T–T annihilation
effects. The triplet molar absorption coefficients were obtained by the
singlet-depletion and energy-transfer methods. Details of the experimen-
tal procedures and data analysis used can be found in Refs. [14,16].
Room-temperature singlet-oxygen phosphorescence was detected at
8224
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8223 – 8234

Thienylethynyl Octopoles with a Tunable Core
FULL PAPER
1270 nm with the equipment and procedures reported elsewhere.[14,16]
The singlet-oxygen quantum yields were obtained from these signals.
conjugated organic systems in conjunction with the INDO Hamiltoni-
an.[30] The FF approach relies on the fact that, by definition, bxxx is the
second derivative of the x component of the dipole moment mx with re-
spect to the x components of the applied electric fields (F) at zero field.
The electronic part of b can be cast rigorously as the integral over the
moments of the second derivative of the charge density 1. In an approxi-
mate way, that integral can be partitioned into a sum over derivatives of
point charges qi concentrated on the individual atoms i
Thin films were obtained with
a Desk-Top Precision Spin Coating
System, Model P6700 Series from Speedline Technologies with proce-
dures reported elsewhere.[14,16] The fluorescence emission spectra of the
thin films were obtained with a Horiba–Jobin–Yvon integrating sphere.
The solid-state photoluminescence quantum yields in thin films were ob-
tained as previously described.[14,16]
FT-Raman spectra (1064 nm) were measured using an FT-Raman acces-
sory kit (FRA/106-S) and a Bruker Equinox 55 FT-IR interferometer. A
continuous-wave Nd-YAG laser working at 1064 nm was employed for
excitation. A germanium detector operating at liquid nitrogen tempera-
ture was used. Raman scattering radiation was collected in a back-scat-
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Z
2
2
2
2
X
X
@ mx
@ 1ðrÞ
@ qi
bxxx
¼
¼
x
d3r ¼
xi
¼
xiqð2Þ
ð1Þ
i
@Fx2 ꢀ
@Fx2
@Fx2 ꢀ
Fx ¼0
Fx¼0
Fx ¼0
i
i
tering configuration with a standard spectral resolutions of 4 and 1 cmꢂ1
;
1000–3000 scans were averaged for each spectrum. A variable-tempera-
ture cell Specac P/N2100, with interchangeable pairs of quartz windows,
was used to record the FT-Raman spectra at different temperatures.
Quantum-chemical calculations: Depending on the specific spectroscopic
property addressed, a variety of theoretical methodologies have been
used. Density functional theory is very well suited for the study of p-con-
jugated molecules, in which electron correlation plays
a significant
role.[17] DFT has been used for the ground-state properties (i.e., opti-
mized geometries, vibrational spectra, etc.), and its time-dependent ex-
tension TD-DFT has been used to estimate the vertical and adiabatic ex-
cited-state transitions (i.e., energies and oscillator strengths).[18] DFT also
provides reliable geometry optimizations of the lowest triplet state (i.e.,
T1), whereas TD-DFT can be used to address vertical transitions within
the triplet manifold. In order to evaluate the optimized geometries of
higher-lying excited states, we have relied on the restricted configuration
interaction with singles approach (CIS) within the Hartree–Fock (HF)
approximation (RCIS/HF) in which the single determinant RHF wave-
function is used as the reference determinant in a CIS calculation of ex-
cited states.[19]
Here, bxxx is partitioned into local (atomic) contributions or so-called b
moments xiqð2Þ derived from the b charges qði2Þ. Note that the superscript
i
(2) represents the second derivative with respect to the applied electric
field in the x direction (with the x subscript dropped to simplify the nota-
tion). In this work, the charge derivatives were approximated by finite
differences obtained from INDO Mulliken charges, with fields of zero
and ꢃ5.14ꢇ1011 Vmꢂ1 (10ꢂ3 atomic units) applied to the octopolar mole-
cules.
Most of the computations were performed with the Gaussian03 pack-
age.[20] The DFT calculations were based on Beckeꢈs three-parameter gra-
dient-corrected exchange functional combined with the Lee–Yang–Parr
correlation functional (B3LYP).[21] The 6-31G** basis set was used in all
DFT and HF calculations.[22] The T1 state was optimized at the DFT level
with the corresponding unrestricted methodology (i.e., UB3LYP/6-
31G**). The molecular geometries of the ground state (excited states)
were optimized assuming C3h (CS) symmetry. In the TD-DFT calcula-
tions, at least the ten lowest-energy vertical electronic excited states were
evaluated. TD-DFT calculations were performed with the same function-
al (B3LYP) and basis set (6-31G**). Using the ground-state optimized
geometries, the harmonic vibrational frequencies and Raman intensities
were calculated analytically at the DFT//B3LYP/6-31G** level.
Results and Discussion
Absorption spectra: Figure 1 displays the absorption and
emission spectra of the investigated compounds. TD-DFT
excited-state calculations helped us assign the experimental
bands (Figure 2a).
Figure 1 and 2 and Table 1 show that, for example for
TEBF, the lowest energy, and strongest, band at 302 nm,
corresponds to the degenerate theoretical S0!S2 and S0!S3
transitions predicted at l=315 nm. This band is followed at
higher energies by a medium intensity band at l=291 nm
due to the degenerate S0!S5/S0!S6 transitions calculated at
l=295 nm and by the strong band measured at l=287 nm
arising from the S0!S7+S0!S8 excitations estimated theo-
retically at l=281 nm.
In Figure 2a, a comparison of the experimental and theo-
retical spectra highlights the relative experimental intensity
and oscillator strength of the three relevant bands for
TEBF. Even though these electronic excitations are best de-
scribed by multi-electron promotions involving their frontier
orbitals, some parallel tendencies can be extracted: i) The
theoretical HOMO–LUMO energy gaps in Figure 3 and the
experimental optical ones evolve similarly: the former as
4.21 eV in TEBF, 4.26 eV in TEB, and 4.17 eV in TEM; the
second as 4.09 eV in TEBF (i.e., 303 nm)!4.08 eV in TEB
(i.e., l=304 nm)!4.03 eV in TEM (i.e., l=308 nm). Hence,
theory predicts the narrowest gap for TEM to coincide with
Nonlinear optics calculations: Based on DFT//B3LYP/6-31G** ground-
state optimized geometries, the second-order polarizabilities (b) were
evaluated with the semiempirical Intermediate Neglect of Differential
Overlap (INDO) Hamiltonian.[23] The spectroscopic parameterization,
along with the Mataga–Nishimoto[24] electron-repulsion potential, was
used, as implemented in the ZINDO code.[25] For sum-over-states (SOS)
hyperpolarizabilities (see, e.g., Ref. [26] for the standard expressions),
the electronic properties of the ground and electronically excited states
(transition energies, state and transition dipole moments) were evaluated
using two configuration interaction schemes. The first considered singly
excited configurations (SCI) with an active space that included all occu-
pied and unoccupied p-molecular orbitals (MOꢈs). We note that INDO
calculations coupled to a SCI Scheme have been shown to provide relia-
ble descriptions of the second-order polarizabilities.[26] The second was
based on a SDCI (single and double configuration interaction) approach
that included single excitations between all occupied and unoccupied p-
MOꢈs as well as double excitations among the five highest occupied p-
MOꢈs and five lowest unoccupied p*-MOꢈs.
The components of the static b tensor along the x axis (bxxx) were also
calculated using the Finite Field (FF) approach, in its local contribution
version,[27] as proposed by Chopra et al.[28] and developed by Nakano
et al.[29] Geskin and co-workers have successfully applied this method to
Chem. Eur. J. 2009, 15, 8223 – 8234
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8225

J. Casado, J. T. Lꢀpez Navarrete, G. Hennrich et al.
tals
(HOMOꢂ1,
HOMO,
LUMO and LUMO+1) of
TEBF relative to its counter-
parts is the larger participation
of the fluorine atoms, seemingly
consisting of a reorganization
of the charge in these atoms
along with the excitation
(i.e., HOMO!LUMO+1 or
HOMOꢂ1!LUMO
promo-
tions) that can be related to a
certain charge-transfer property
in TEBF. This argument will be
invoked in the NLO section as
well as in the Raman character-
ization.
Figure 1. Left: absorption and emission spectra in CH2Cl2 for ꢀ10ꢂ4 m solutions of a) TEM, b) TEB, and
c) TEBF. Right: absorption spectrum of TEM at different concentrations in cyclohexane obtained with cuv-
ettes with a pathlength of 100 mm.
Theory predicts the first ex-
cited state to appear at
lꢀ330 nm and to be radiatively
decoupled from the ground
electronic state. The S0!S1
transition is forbidden by the
dipole–dipole optical selection
rules for the C3h geometry. In
order to experimentally observe
this forbidden transition, spec-
tra were recorded in solution
for different TEM concentra-
tions (in cuvettes with
100 mm path length), see Fig-
ure 1b, and reproducible
a
a
band at l=332 nm was ob-
tained with an extinction coeffi-
cient of 1560mꢂ1 cmꢂ1 . Distor-
tion from the C3h planar geome-
try in solution can be expected
to contribute to the activation
of this band.
Figure 2. a) TD-DFT//B3LYP/6-31G** theoretical (bars) and experimental spectra of TEBF (in CH2Cl2).
b) Spectrum of TEM on cooling from room temperature to 77 K (in butyronitrile).
Table 1. Photophysical data in methylcyclohexane and nonlinear optical data in CH2Cl2 at 300 K.[a]
[a]
[b]
lmax
(abs)
log(e)
lmax(em)
[nm]
Ffl
tfl
bxxx,800
[10ꢂ30 esu]
bxxx,0
[10ꢂ30 esu]
1[c]
[nm]
[ns]
Inspection of the evolution of
TEB
TEM
TEBF
290 (313)
294 (317)
290 (315)
4.813
4.822
4.785
358
366
372
0.15
0.10
0.07
8.0ꢃ0.1
9.2ꢃ0.1
2.2ꢃ0.1
16ꢃ1
26ꢃ3
59ꢃ3
7ꢃ1
1.14ꢃ0.03
1.17ꢃ0.05
1.55ꢃ0.03
the absorption spectra as
a
11ꢃ1
21ꢃ1
function of temperature for
TEM reveals an enhancement
of the l=308 nm band relative
to the other two (Figure 2b).
The lowest temperature spec-
trum is more similar to the the-
[a] bxxx,800 are the dynamic hyperpolarizability values at 800 nm. [b] bxxx,0 are the static hyperpolarizability
values. [c] 1 denotes the depolarization ratio in the NLO measurements. See the NLO section for additional
details.
the most red-shifted bands, experimentally measured at l=
308 nm (i.e., 4.03 eV) in TEM as well. ii) Also seen in
Figure 3, the HOMO and HOMOꢂ1 levels are particularly
destabilized in TEM by 0.22 and 0.28 eV, respectively, rela-
tive to those in TEB and TEBF, thus providing information
on the positive inductive effect (i.e., +I) of the methyl
groups and on the negative or electron-withdrawing effect
of the three central fluorine atoms. iii) The same effects pro-
duce a stabilization/destabilization of the LUMO (and
LUMO+1) orbitals in TEBF/TEM relative to TEB. iv) A
distinctive feature of the wavefunctions of the frontier orbi-
oretical one in terms of relative intensities (Figure 2a). The
FT-Raman spectra of these samples in the solid state at low
temperatures display a continuous blue shift of the whole
spectrum on cooling to ꢂ1708C. This is interpreted as an in-
termolecular effect and rules out a possible intramolecular
structural or ground electronic state effect (see the section
on vibration Raman spectra below). In the next discussion,
we will see that the S2/S3 excited states feature electron de-
localization along the acetylene bridge and subsequent rigid-
ification (cumulenic character). This is favored by intermo-
lecular solute–solvent interactions at low temperature that
8226
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8223 – 8234

Thienylethynyl Octopoles with a Tunable Core
FULL PAPER
This suggests that the geome-
tries of the S0 and S1 states are
either significantly different or
that the absorbing and emitting
states are not the same. We
note that upon cooling to 77 K,
the maxima and shape of the
emission spectra do not change
significantly (see Figure S1),
thus indicating that the emitting
states are the same (identical
potential energy curves) at
293 K and 77 K. Because the
weak dipole–dipole forbidden
S0!S1 transition of TEM is lo-
cated at l=332 nm and the
S1!S0 emission is found at
366 nm, the SS is much smaller
and confirms the different
nature of the main transitions
in absorption (S0!S2) and
Figure 3. DFT//B3LYP/6-31G** wavefunction topologies of the orbitals around the gap.
probably improve the vertical alignment of the ground and
excited states, thus intensifying the S0!S2/S3 transition.
emission (S1!S0; Figure 1b).
The vibrational structure of the emission spectra remains
practically unchanged from 293 K to 77 K, which is compati-
ble with a planar quinoidal-like structure for the active ex-
cited state.[27] The following arguments are also in accord-
ance with the discussion above: i) For TEB, the S1!S0 theo-
retical energy is calculated at 347 nm close to the experi-
mental value at 358 nm in CH2Cl2. ii) Fluorescent quantum
yields are rather small and are therefore also in agreement
with the small electronic coupling between the S0 and S1
electronic states. iii) The evolution of the fluorescence quan-
tum yields in the three octopoles can be accounted for by a
competition between internal conversion and the intersys-
tem (ISC) crossing between S1 and the closest triplet state.
In fact, ISC to the triplet manifold is one of the main routes
proposed to quench the fluorescence in thiophene deriva-
tives, thus we have calculated the vertical and adiabatic
energy positions of the first triplet and singlet excited states
by TD-DFT (Figure 4). ISC is likely to take place between
S1 and the closest triplet state
Fluorescence spectra—the singlet states: Figure 1a shows
the absorption and fluorescence spectra of the three octo-
poles, and Table 1 summarizes the relevant spectral and
photophysical data. Table 2 presents the spectral characteris-
tics in the nonpolar solvent methylcyclohexane, including
singlet and triplet extinction coefficients, and Table 3 gives
the remaining photophysical parameters and rate constants,
including singlet-oxygen sensitization yields.
The bathochromic displacement l=358!366!372 nm
observed for the sequence TEB!TEM!TEBF together
with the gradual loss of vibrational structure of the emission
bands follows the expected trend in terms of enhanced
charge-transfer character (TEBF>TEM>TEB). From the
absorption and fluorescence spectra of TEM, TEB, and
TEBF, a significant displacement of the absorption relative
to the emission maxima [Stokes shift (SS)], can be observed.
of lower energy; in this context,
the S1!T7 ISC process is calcu-
Table 2. Spectral characteristics in methylcyclohexane at 293 K and 77 K.
lmax
N
lmax
A
lmax(em)293K
lmax(em)77K
eS
[mꢂ1 cmꢂ1
eT
[mꢂ1 cmꢂ1
]
lated to be the most probable.
When measuring the quantum
yield FT for the ISC process,
the values are found to be
rather similar for all three mol-
ecules (Table 3). These results
are well supported by the simi-
[nm]
[nm]
[nm]
[nm]
G
]
E
TEB
TEM
TEBF
290
294
290
290
291
290
358
366
372
354
357
365
65900
63160
65080
47009
44482
50435
lar exchange energy, or DACHTUNGTRENNUNG(S1-
Table 3. Photophysical parameters and rate constants in methylcyclohexane.
77
K
[a]
[b]
[c]
[d]
[e]
T7), predicted theoretically by
TD-DFT to be 0.07, 0.07/
0.06 eV in TEB, TEM/TEBF. In
this regard, internal conversion
seems to control the fluores-
cence efficiency because in 1,2-
Ffl
FD
FT
FIC
kT
[sꢂ1
4.05ꢇ104
3.20ꢇ105
4.37ꢇ104
tT
kF
[nsꢂ1
kNR
A
kISC
A
kIC
[nsꢂ1
G
]
[s]
G
]
[nsꢂ1
]
[nsꢂ1
]
G
]
TEB
TEM
TEBF
0.17
0.15
0.20
0.30
0.31
0.29
0.28
0.30
0.27
0.57
0.60
0.66
2.47ꢇ10ꢂ5
3.13ꢇ10ꢂ6
2.29ꢇ10ꢂ5
ꢀIC
0.012
0.013
0.06
0.11
0.10
0.40
0.04
0.03
0.13
0.08
0.06
0.27
ꢀF
1ꢂꢀF
ꢀT
[a] FIC =1ꢂFFꢂFT. [b] kF
¼
. [c] kNR
¼
. [d] kISC
¼
. [e] kIC
¼
.
tF
tF
tF
tF
Chem. Eur. J. 2009, 15, 8223 – 8234
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8227

J. Casado, J. T. Lꢀpez Navarrete, G. Hennrich et al.
this indicates that in the T1
state the electron delocalization
is less pronounced or restricted
to a smaller molecular domain
compared to the S1 state. This is
illustrated in Figure 7, which
shows the theoretical structures
of these excited states, and is in
accordance
with
previous
data.[31]
The electronic structure of
the S1 state displays symmetri-
Figure 4. TD-DFT//B3LYP/6-31G** excited-states energies in the singlet and triplet excited-state manifolds.
dichloromethane and methylcyclohexane at room tempera-
ture the less fluorescent molecule, TEBF, displays the larg-
est internal conversion quantum yield. At 77 K, however,
TEBF turns out to have the greatest fluorescence quantum
yield, indicating a more prominent blocking of the IC radia-
tionless channel on cooling in the molecule with the largest
fraction of it (FIC =0.66) at room temperature (Table 3).
The fluorescence decays obtained with nanosecond time
resolution were found to be single exponential. Moreover,
they remain mono-exponential on the ps timescale, indicat-
ing that no energy transfer or conformational relaxation
processes occur on this timescale, which supports the “rigid”
nature of the S1 emitting state.
Table 4. Phosphorescence data in methylcyclohexane at 77 K.
lmax(em)[nm]
Fphos
tphos [ms]
TEM
TEB
TEBF
477, 507, 525
482, 510, 526
475, 503, 522
0.10
0.11
0.15
284
363
387
Phosphorescence spectra—the triplet state: As discussed
above, the ISC process competes with fluorescence and in-
ternal conversion. This prompted us to explore the optical
properties of the triplet manifold by measuring the phos-
phorescence spectra and performing flash photolysis in non-
polar solvents (see below). The phosphorescence spectra at
77 K are shown in Figure 5, and Table 4 summarizes the cor-
responding data.
Figure 6 highlights the good correlation between the ex-
perimental and theoretical energies for the three relevant
excited electronic states, which further supports the use of
computational modeling in the following discussion. The
greatest difference between the phosphorescence maxima is
7/0.04 nmeVꢂ1 [i.e., 475/2.61 nmeVꢂ1 in TEBF, 477/
2.60 nmeVꢂ1 in TEM, and 482/2.57 nmeVꢂ1 in TEB] where-
as for fluorescence the corresponding difference is 14 nm;
Figure 6. Energy diagram for the electronic singlet and triplet states of
TEB comparing experimental (bold italics) and theoretical data.
cal distortion around the central benzene core and has an
equal effect on the acetylene arms and more slightly on the
peripheral thiophene rings; however, the T1 state is far more
similar to the S2/S3 states and is mainly located in one of the
three arms. These states are characterized by a cumulenic
character within the acetylene spacer and an asymmetrical
quinoidization of the central benzene ring. As a significant
difference between the T1 and S2/S3 states, the former is cen-
tered on the acetylene spacer with a stronger cumulenic
character. Hence the thiophene rings are less affected in T1
Figure 5. Absorption (293 K), fluorescence (293 K), and phosphorescence (77 K, grey) spectra in methylcyclohexane of a) TEBF, b) TEB, and c) TEM.
8228
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8223 – 8234

Thienylethynyl Octopoles with a Tunable Core
FULL PAPER
Figure 7. Representation of the geometric/chemical structures of the excited states of TEB as deduced from RCIS/6-31G** calculations. The optimized
geometry of the ground state (S0) was calculated with DFT//B3LYP/6-31G**. Bond lengths are given in ꢉ.
than in S2/S3 with these high-energy singlets readily involv-
ing the sulfur atoms of the outermost thiophenes. This theo-
retical consideration is in very good agreement with the
structural properties deduced from the above experimental
optical features relative to the different shifts of the fluores-
cent/phosphorescent emissions in the three molecules. The
participation of the sulfur atoms in the optically relevant
frontier orbitals results in the heavy atom effect coming into
play,[32] which can lead to spin flipping in the ISC process. In
any case, an additional mechanism for spin-orbit coupling
will be discussed in the following paragraph.
The three compounds display high phosphorescence quan-
tum yields. The case of TEBF is remarkable because the
largest phosphorescence quantum yield (i.e., 0.15) is ob-
served in the molecule where ISC is most favored. For acet-
ylene derivatives, rapid intersystem crossing is expected be-
cause of the presence of two p-orbitals, p1 and p2, that are
perpendicular to each other; one conjugates strongly with
the phenyl and thiophene rings, whereas the other is mainly
localized on the acetylene bonds. Due to the p1–p2 configu-
ration, the excited state will contribute to enhancing the
spin-orbit coupling responsible for intersystem crossing.[32,33]
There is vibronic structure in the three phosphorescence
spectra. The peak spacing in this vibrational structure is
around 1100–1300 cmꢂ1, which corresponds to vibrations of
C bonds, nACTHNGUTRENNU(G C-C). These bonds are especially affected on ex-
citation at the origin of the vibronic activity.
TD-DFT calculations provide the theoretical energy for
the S0!T1 transitions, which would relate with the phos-
phorescence wavelengths. For instance, in TEB, this transi-
tion is predicted at 518 nm, in agreement with the set of sub-
peaks at 482/510/526 nm. This correlation provides further
support for the assignment of the low-temperature emission
to phosphorescence.
Transient triplet–triplet absorption spectra and yields: The
transient triplet–triplet absorption spectra in Figure 8 dis-
play a depletion at ꢀ300 nm followed by a maximum at
ꢀ500 nm.
In order to assess the transient spectra in terms of triplet–
triplet transitions, we have calculated the TD-DFT//
UB3LYP/6-31G** transitions starting from the DFT//
UB3LYP/6-31G** T1 geometry. These theoretical results are
shown in Figure 9 for TEB compared with the correspond-
ing flash-photolysis spectrum taken 0.39 mseconds after exci-
tation.
In our experimental interval, the most active and optically
relevant transitions are recorded for these samples. It must
be pointed out that these bands involve relatively high
energy excitations where, very usually, theoretical calcula-
tions such as TD-DFT fail in the estimation. Thus, while
ꢂ
the acetylene spacer, in particular, stretches of the single C
Chem. Eur. J. 2009, 15, 8223 – 8234
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8229

J. Casado, J. T. Lꢀpez Navarrete, G. Hennrich et al.
tively, to T1!T19 and T1!T14 triplet–triplet transitions; in
addition, these two bands are the strongest predicted transi-
tions.
The ISC yield values indicate that ISC is an efficient
decay channel for S1. The quantum yields for singlet-oxygen
sensitization are practically identical to the fT values
(Table 3), thus supporting these considerations and showing
that the compounds efficiently sensitize molecular singlet
oxygen. In addition, the nature of the T1 state could be in-
ferred from the phosphorescence lifetimes, whose values are
indicative of allowed p,p* triplet states.[32–34] This is corrobo-
rated theoretically because the S0!T1 excitation corre-
sponds to the HOMO!LUMO dipole-forbidden transition
with two orbitals of p-character, as discussed previously.
The fphos values show that these compounds are moderately
phosphorescent (10% of the deactivation), which is uncom-
mon for thiophene compounds, although it has been ob-
served for terthiophene derivatives using gated detection in
combination with nanosecond excitation in frozen solutions
at 80 K.[31] The photophysical parameters in Table 3 show
that the radiationless internal conversion dominates the de-
activation of the S1 state over the radiative and ISC process-
es because the sum FF + FT only accounts for 0.3–0.4
(Table 3). However, from the rate constant values (kF, kISC
,
kIC), the three processes are kinetically competitive with
values of the same order of magnitude (Table 3). From the
molar extinction coefficients of the S0!S1 transitions and
the kF values it is clear that S1 and S2 are of p,p* origin, al-
though in the first case S0!S1 is symmetry forbidden
(Tables 2 and 3).
Solid-state electronic spectra: Studies carried out in dilute
solution essentially probe intramolecular excitations. How-
ever, it has become evident that the optical properties of
conjugated molecules are greatly affected by intermolecular
interactions that are predominant in the solid state.[35] Con-
sequently, optical data from solid thin films are more likely
to reflect the excitation dynamics present in electrolumines-
cent devices. Figure 10 displays the absorption and emission
spectra (fluorescence and phosphorescence) of TEM in the
Figure 8. Transient triplet–triplet spectra of the investigated octopolar
compounds in methylcyclohexane at room temperature (293 K).
a) TEBF, b) TEB, and c) TEM
Figure 9. Transient triplet–triplet spectrum of TEB in methylcyclohexane
recorded 0.39 ms after excitation (black line) and its theoretical TD-
DFT//UB3LYP/6-31G** triplet–triplet spectrum (grey).
Figure 10. Absorption and emission (fluorescence and phosphorescence)
spectra in the solid state (c) and in methylcyclohexane solution (a)
of TEM. The phosphorescence spectra both for solution and solid state
were obtained at 77 K.
considering this correlation with much caution, we suggest
that the pair of bands at 460/491 nm could be related to
those calculated at 416/465 nm, which correspond, respec-
8230
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8223 – 8234

Thienylethynyl Octopoles with a Tunable Core
FULL PAPER
solid state. We see that the absorption and fluorescence
spectra of TEM in thin films are slightly shifted to lower en-
ergies relative to the spectra in solution. Phosphorescent
profiles are rather different in solution compared to the
solid phase. The first vibronic peak is displaced by 0.06 eV
(i.e., 477 nm in methylcyclohexane and 489 nm in the solid)
in the phosphorescent spectrum of TEM and by 0.03 eV (by
4 nm in Figure 10) in the fluorescent spectrum. In addition,
the component at 546 nm becomes stronger in the solid, in
contrast to that at 477 nm in solution.
This indicates that the absorbing and singlet emitting
states both in solution and in the solid state have very simi-
lar structures. Although larger distortions in solution rela-
tive to the solid would be expected, the small differences in
the spectra indicate that the molecules keep almost the
same molecular conformations (close-to-planarity conform-
ers) in both phases, thus supporting the choice of planar C3h
models for the theoretical studies. In view of the alteration
of the vibronic profile, this does not appear to be valid for
the triplet state. Solid-state packing would tend to planarize
the molecules, an effect that would likely have greater inci-
dence than in solution phase on the triplet state owing to its
significant cumulenization (i.e., cumulenization would be fa-
cilitated in a pro-planar solid-state environment).
Figure 11. Left: 1064 nm FT-Raman spectra of the three compounds in
the solid state. Right: DFT//B3LYP/6-31G** Raman spectra of the three
compounds.
As expected, the solid-state fluorescence quantum yields
decrease by about one order of magnitude relative to the
liquid phase; this decrease is more marked in TEB and
more modest in TEBF (from 0.07 to 0.04), indicating that
internal conversion for TEBF is already efficient in solution.
In solid TEB, the effect of intermolecular interactions (i.e.,
solid-state aggregation) reduces the emission activity, hence
leading to the largest changes of Ffl. In TEM, intermolecu-
lar packing is expected to be less favorable with respect to
TEB because the three methyl groups in the molecular core
introduce steric hindrance; this is consistent with the moder-
ate change of Ffl. Moreover, the radiative and radiationless
rate constant values did not change significantly relative to
the solution behavior (Tables 3 and 5), which once more in-
dicates that the deactivation pathways are identical in both
liquid and solid media.
Figure 12. 1064 nm FT-Raman spectra of TEB as a function of the tem-
perature.
The prominent bands in the three spectra are recorded at
2210 cmꢂ1 in TEB and TEM and at 2219 cmꢂ1 in TEBF.
ꢄ
They are related to the C C stretching vibration of the acet-
ylene spacer. This assignment allows us to establish a struc-
ture–property relationship because the corresponding triple
bond is significantly shorter in TEBF, in agreement with its
larger vibrational frequency: 2210 cmꢂ1 (TEB)!2210 cmꢂ1
(TEM)!2219 cmꢂ1 (TEBF) and 1.216 ꢉ (TEB)!1.217 ꢉ
(TEM)!1.215 ꢉ (TEBF).
Table 5. Fluorescence and phosphorescence data as well as the rate con-
stants in the solid state.
lmax
G
N
tfl [ns] tph [ms]
The stretching modes of the CC bonds of the central ben-
zene ring appear in the 1600-1550 cmꢂ1 spectral region.
There is also a direct relationship between this vibrational
frequency evolution and the CC bond lengths: 1584 cmꢂ1
(TEB)!1560 cmꢂ1 (TEM)!1604 cmꢂ1 (TEBF) and 1.406/
1.405 ꢉ (TEB)!1.410/1.417 ꢉ (TEM)!1.404 ꢉ (TEBF).
As for the medium intensity lines at 1322 cmꢂ1 in TEM,
1329 cmꢂ1 in TEB, and 1379 cmꢂ1 in TEBF, they can be de-
TEM
TEB
TEBF 365
360, 375
355, 369
489
482
480
115
218
248
0.06 5.56
0.04 2.54
Vibrational Raman spectra: Figure 11 shows the FT-Raman
spectra of the three molecules in the solid state with an exci-
tation wavelength of 1064 nm. It is possible to correlate the
Raman frequencies with the molecular structure of the S0
ground electronic state. As a representative case, Figure 12
displays the FT-Raman spectra of TEB between ꢂ1708C
and 1708C in the solid state.
ꢂ
scribed as C C stretching modes of the single bonds of the
ꢂ
spacers, or n
A
with the bond lengths calculated for the CC bonds connect-
ed to the core: 1.428 ꢉ (TEM), 1.426 ꢉ (TEB), and 1.418 ꢉ
Chem. Eur. J. 2009, 15, 8223 – 8234
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8231

J. Casado, J. T. Lꢀpez Navarrete, G. Hennrich et al.
ꢂ
ꢄ
(TEBF). The Raman data related to n
(C C) and n
E
At the highest modulation frequencies, the fluorescence
contribution is almost completely demodulated. Thus, an ac-
curate value for the first hyperpolarizability can be ob-
tained. The reported hyperpolarizability values are the dy-
namic, fluorescence-free values at 800 nm (bxxx,800) and the
static, dispersion-free values (bxxx,0), as calculated using the
three-level model.[11] The depolarization ratios 1 obtained
for the target compounds, referenced to the value for crystal
violet, prove their octopolar symmetry, which in theory
amounts to a value of 1.5. There exists a subtle modulation
of b (either b000 or bxxx) as a function of the substitution pat-
tern. It shows the largest activity for TEBF in accordance
with the charge-transfer interaction between the active ex-
cited states.
The components of the b tensor along the x axis (bxxx),
which is taken along one of the arms of the octopolar com-
pounds, were calculated using two complementary ap-
proaches, the sum-over-states (SOS) method (see, e.g.,
ref. [26]) and the finite-field (FF) method.[30a] The results are
given in Table 6. The agreement among the values obtained
flect the cumulenic character of the spacer in TEBF as a
result of ground-state polarization by donor–acceptor inter-
actions; this has a significant impact on the spectroscopic
and NLO properties (vide infra).
These correlations highlight the possibility of tuning the
spectroscopic properties as a function of the core substitu-
tion. As for the ground electronic state, TEBF can be
viewed as a central acceptor core competing with three sur-
rounding ethynyl groups that are also electron-withdrawing
moieties. As a consequence, the S0 electronic polarization is
limited owing to partial cancellation of opposite effects. The
contrary is expected in TEM because +I inductive electron
donation of the methyl groups facilitates polarization of the
benzene groups and of the acetylene spacers; overall TEM
displays the largest/lowest bond lengths/Raman frequencies.
It must be pointed out that although Raman data show a
collapse of the electronic polarization in TEBF, the UV/Vis
results indicate that this compound displays some charge-
transfer character. Obviously, Raman frequencies reflect the
S0 molecular structure, whereas the electronic spectroscopic
data are related to the S0!S2/S3 or S1!S0 electronic excita-
tions or relaxations.
Table 6. Theoretical static bxxx components of the first hyperpolarizability
calculated with the SOS and FF methods for TEBF, TEB, and TEM.[a]
The temperature evolution of the Raman frequencies in
the solid state primarily indicates that the molecule has sig-
nificant thermal stability; the spectra fully recover after
heating to 2008C (i.e., no appreciable changes are observed
after thermal cycling). A continuous downshift is recorded
with temperature for all the CC stretching bands that are in-
Com- bxxx0 [10ꢂ30 esu] bxxx [10ꢂ30 esu] bxxx [10ꢂ30 esu] bxxx 10ꢂ30 esu]
pound exptl[b]
FF
SOS
SOS
INDO/SCI[c]
INDO/SDCI[d]
TEBF 21ꢃ1
15
2
3
18
8
1
11
4
1
TEB
7ꢃ1
TEM
11ꢃ1
[a] Absolute values. [b] HRS data obtained in CH2Cl2, units in 10ꢂ30 esu.
[c]All the p-molecular orbitals (MOs) were active in the SCI procedure.
[d] Configuration interaction here includes single (S) excitations between
all occupied and unoccupied p-MOs, and double (D) excitations among
the five highest occupied p-MOs and five lowest unoccupied p*-MOs.
volved in the p system; the most pronounced variation is
ꢂ1
ꢄ
observed for the n
G
by ꢂ11 cmꢂ1. This behavior has been found for other p-con-
jugated molecules and planar molecules, and ascribed to the
relatively small expansion/compression of the solid during
the heating/cooling cycle resulting in the decrease/increase
of the molecular interactions.[36] A decrease of these interac-
tions would soften the slope of the potential energy surface
of the ground electronic state causing the vibrational fre-
quencies to decrease. This interpretation agrees reasonably
well with the possible formation of aggregates for these mol-
ecules.[8]
with the two different approaches is good. Compared to ex-
periments, the FF results reproduce the experimental trends
better; the two methods provide the largest b values for
TEBF, which is in agreement with experiment and with our
assertion that the fluorine atoms exert an electron-with-
drawing effect, in combination with the electron-richness of
the terminal thienyl groups. This charge transfer can be
viewed as a donor-to-acceptor electron polarization. The in-
ductive effect of the three methyl groups, which causes the
hyperpolarizability to increase very slightly from TEB to
TEM, is well reproduced by the FF approach. The inclusion
of the acceptor groups (-F) into the central benzene ring ap-
pears to be a crucial structural modification responsible for
the large bxxx component in TEBF.
In order to gain deeper insight into structure–property re-
lationships, the calculated bond-length alternation (BLA)
values and the HOMO–LUMO energy gap (DE) are dis-
cussed and displayed for TEBF, TEB, and TEM in Table 7.
BLA was calculated by subtracting the triple bond length of
the conjugation bridge from the average length of the two
Nonlinear optical spectroscopy: Owing to the octopolar
nature of the investigated compounds, hyper Rayleigh scat-
tering (HRS) is the method of choice for determining the
first hyperpolarizability b.[11] As expected from the linear
one-photon fluorescence spectra, all three compounds ex-
hibit significant (multiphoton) fluorescence when excited
with strong IR pulses. Here, frequency-resolved femtosec-
ond HRS with fluorescence demodulation at 800 nm has
been applied.[11] As a reference value, 330ꢇ10ꢂ30 esu was
used for bxxx of the octopolar molecule crystal violet in
methanol, also at 800 nm. Upon simultaneously fitting the
demodulation and the phase data as a function of modula-
tion frequency, the fluorescence-free hyperpolarizability
values were obtained together with the multiphoton fluores-
cence contribution, and the fluorescence lifetime (Table 1).
ꢂ
C C bonds connected to it. The inclusion of the fluorine
8232
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8223 – 8234

Thienylethynyl Octopoles with a Tunable Core
FULL PAPER
Table 7. Experimental bxxx values and theoretical DFT//B3LYP/6-31G**
BLA values, HOMO and LUMO gaps (DE) for the thienyl-substituted
benzene derivatives.
Acknowledgements
The present work was supported in part by the Direcciꢀn General de En-
seÇanza Superior (DGES, MEC, Spain) through research projects
CTQ2006-14987-C02-01 and CT2007-65683. The authors are also indebt-
ed to Junta de Andalucꢂa for the project FQM1678/2006. J.C. and G.H.
are grateful to the MEC of Spain for I3 professorships. M.M.O is indebt-
ed to the MEC for a predoctoral FPU fellowship. J.S.D.M. acknowledges
financial support from the Portuguese Science Foundation (FCT) through
FEDER and POCI. MCRD acknowledges the MEC for a MEC/Ful-
bright Postdoctoral Fellowship at the Georgia Institute of Technology.
The work at Georgia Tech was partly supported by the STC Program of
the National Science Foundation (Award DMR-0120967).
Compound
bxxx0 [10ꢂ30 esu]
BLA[ꢉ]
DE[b]
exper.[a]
TEBF
TEB
TEM
21ꢃ1
7ꢃ1
0.203
0.207
0.207
4.212
4.253
4.181
11ꢃ1
[a] HRS data obtained in CH2Cl2, units in 10ꢂ30 esu. [b] DE=
ELUMOꢂEHOMO in eV.
groups in the central benzene core decreases the lengths of
the single CC bonds and increases the length of the triple
CC bond, thus resulting in a decrease in the calculated BLA
values. As expected from Table 7, a decrease in BLA and
DE within the (-F) substitution of the benzene core leads to
an increase in b.
[1] a) A. Tsumura, H. Koezuka, T. Ando, Appl. Phys. Lett. 1986, 49,
1210; b) C. D. Dimitrakopoulos, P. Malenfant, Adv. Mater. 2002, 14,
99; c) J. A. Rogers, Z. Bao, A. Dodabalapur, B. Crone, V. R. Raju,
H. E. Katz, V. Kuck, K. J. Ammundson, P. Drzaic, Proc. Natl. Acad.
Sci. USA 2001, 98, 4817; d) F. Garnier, Acc. Chem. Res. 1999, 32,
209; e) J. H. Burroughes, D. D. C. Bradley, R. N. Marks, R. H.
Friend, P. L. Burn, A. B. Holmes, Nature 1990, 347, 539; f) A. Kraft,
A. C. Grimsdale, A. B. Holmes, Angew. Chem. 1998, 110, 416;
Angew. Chem. Int. Ed. 1998, 37, 402; g) U. Mitschke, P. Bꢄuerle, J.
Mater. Chem. 2000, 10, 1471; h) H. Sirringhaus, Adv. Mater. 2005, 17,
2411; i) J. Roncali, P. Leriche, A. Cravino, Adv. Mater. 2007, 19,
2045; j) T. Noda, H. Ogawa, N. Noma, Y. Shirota, J. Mater. Chem.
1999, 9, 2177; k) A. Facchetti, M. Mushrush, H. E. Katz, T. J. Marks,
Adv. Mater. 2003, 15, 33; l) V. Coropceanu, J. Cornil, D. A. Da Silva
Filho, Y. Olivier, R. Silbey, J.-L. Bredas, Chem. Rev. 2007, 107, 926.
[2] a) C. Bosshard, K. Sutter, P. PrÞtre, J. Hulliger, M. Flçrsheimer, P.
Kaatz, P. Gꢅnther, Organic Nonlinear Optical Materials, Vol. 1,
Gordon & Breach, Amsterdam, 1995; b) F. Meyers, S. R. Marder,
J. W. Perry in Chemistry of Advanced Materials: An Overview (Eds.:
L. V. Interrante, M. J. Hamden-Smith),Wiley-VCH, Weinheim, 1998;
c) T. Verbiest, K. Clays, C. Samyn, J. Wolff, D. Reinhoudt, A. Per-
soons, J. Am. Chem. Soc. 1994, 116, 9320; d) J. Zyss, J. Chem. Phys.
1993, 98, 6583; e) J. Zyss, I. Ledoux, Chem. Rev. 1994, 94, 77.
[3] a) L. Porrꢊs, O. Mongin, C. Katan, M. Charlot, T. Pons, J. Mertz, M.
Blanchard-Desce, Org. Lett. 2004, 6, 47–50; b) W.-H. Lee, H. Lee,
J.-A. Kim, J.-H. Choi, M. Cho, S.-J. Jeon, B. R. Cho, J. Am. Chem.
Soc. 2001, 123, 10658; c) G. S. He, J. Swiatkiewicz, Y. Jiang, P. N.
Prasad, B. A. Reinhardt, L. S. Tan, R. Kannan, J. Phys. Chem. A
2000, 104, 4805—4810; d) C. Katan, F. Terenziani, O. Mongin,
M. H. V. Werts, L. Porres, T. Pons, J. Mertz, S. Tretiak, M. Blan-
chard-Desce, J. Phys. Chem. A 2005, 109, 3024–3037; e) G. Bordeau,
R. Lartia, G. Metge, C. Fiorini-Debuisschert, F. Charra, M. P. Teu-
lade-Fichou, J. Am. Chem. Soc. 2008, 130, 16836–16837; f) D. Bel-
jonne, W. Wenseleers, E. Zojer, Z. Shuai, H. Vogel, S. J. K. Pond,
J. W. Perry, S. R. Marder, J. L. Brꢃdas, Adv. Funct. Mater. 2002, 12,
631.
[4] a) A. Cravino, S. Roquet, O. Aleveque, P. Leriche, P. Frere, J. Ron-
cali, Chem. Mater. 2006, 18, 2584; b) R. de Bettignies, Y. Nicolas, P.
Blanchard, E. Levillain, J.-M. Nunzi, J. Roncali, Adv. Mater. 2003,
15, 1939; c) F. Chꢃrioux, L. Guyard, P. Audebert, Chem. Commun.
1998, 2225; d) J. Bras, S. Guillerez, Pꢃpin- B. Donat, Chem. Mater.
2000, 12, 2372; e) Y. Geng, A. Fechtenkçtter, K. Mꢅllen, J. Mater.
Chem. 2001, 11, 1634; f) T. M. Pappenfus, K. R. Mann, Org. Lett.
2002, 4, 3043–3046; g) S. A. Ponomarenko, S. Kirchmeyer, A. Elsch-
ner, B.-H. Huisman, A. Karbach, D. Drechsler, Adv. Funct. Mater.
2003, 13, 591; h) S. A. Ponomarenko, E. A. Tatarinova, A. M. Muza-
farov, S. Kirchmeyer, L. Brassat, A. Mourran, M. Moeller, S. Se-
tayesh, D. De Leeuw, Chem. Mater. 2006, 18, 4101; i) M. Y. Cho, S. J.
Kim, D. Y. Han, D. H. Park, K. H. Kim, D. H. Choi, J. Joo, Adv.
Funct. Mater. 2008, 18, 2905; j) K. Kreger, M. Baete, C. Neuber, H.-
W. Schmidt, P. Strohriegl, Adv. Funct. Mater. 2007, 17, 3456.
Conclusion
We have synthesized three new molecules comprising three
thieno-acetylene arms substituting a central benzene core
with different electron donor/acceptor groups in the free
phenyl positions. The absorption and fluorescence properties
are consistent with the fact that the absorbing and emitting
states are different, as a consequence of the C3 planar (i.e.,
C3h) molecular symmetry. Fine modulation of these proper-
ties is observed as a result of chemical functionalization of
the central core; however, such a modulation does not occur
for the ground electronic properties, as observed from
Raman spectroscopy data, which show the intrinsic effect of
donor–acceptor (TEB and TEM) and acceptor–acceptor
(TEBF) interactions.
The photophysical properties are satisfactorily explained
by the coupling of the singlet and triplet manifolds. We find
ISC to be an important de-excitation route, based on the
measurements of the phosphorescence spectra and triplet-
state formation quantum yields. The molecules can be con-
sidered as rather good phosphorescent dyes given the low
propensity of pure p, p* states to phosphoresce. As a result,
the internal conversion route coupled to the ground elec-
tronic state seems to be a competitive mechanism with fluo-
rescence and phosphorescence. In all the cases, we have pro-
vided theoretical support for our conclusions. As for the hy-
perpolarizabilities, these are moderate and display a typical
dependence on the donor–acceptor substitution pattern.
In summary, a complete study of the nature of the elec-
tronic states in new molecules has been carried out. It de-
scribes interesting molecular properties directed to the un-
derstanding of the structure of their low-lying excited states.
These molecular features are responsible for their perfor-
mance in organic electronics, such as in nonlinear optical
and light-emitting devices. Comprehensive analyses, such as
those carried out here, are needed in order to design ad-
vanced or improved molecules from a property–function
point of view.
[5] a) M. Moreno Oliva, J. Casado, J. T. Lopez Navarrete, G. Hennrich,
M. C. Ruiz Delgado, J. Orduna, J. Phys.Chem. C 2007, 111, 18778;
b) M. C. Ruiz Delgado, J. Orduna, M. Moreno Oliva, J. Casado, J. T.
Chem. Eur. J. 2009, 15, 8223 – 8234
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8233

J. Casado, J. T. Lꢀpez Navarrete, G. Hennrich et al.
Lopez Navarrete, J. Chem. Phys. 2007, 127, 164704; c) M. More-
no Oliva, J. Casado, J. T. Lopez Navarrete, R. Berridge, P. J. Ska-
bara, A. L. Kanibolotsky, Perepichka, F. Igor, J. Phys. Chem. B
2007, 111, 4026; d) M. Moreno Oliva, J. Casado, G. Hennrich, J. T.
Lopez Navarrete, J. Phys. Chem. B 2006, 110, 19198.
H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Pittsburgh,
PA, 2003.
[6] Y. Tu, Y. Luo, H. ꢉgren, J. Phys. Chem. B 2005, 109, 16730.
[7] V. Le Floch, S. Brasselet, J. Zyss, B. R. Cho, S. H. Lee, S.-J. Jeon, M.
Cho, K. S. Min, M. P. Suh, Adv. Mater. 2005, 17, 196.
[8] G. Hennrich, A. Omenat, I. Asselberghs, S. Foerier, K. Clays, T.
Verbiest, J. L. Serrano, Angew. Chem. 2006, 118, 4310; Angew.
Chem. Int. Ed. 2006, 45, 4203.
[9] a) S. R. Marder, Chem. Commun. 2006, 131; b) Nonlinear Optical
Properties of Matter: From Molecules to Condensed Phases (Eds.:
M. G. Papadopoulos, J. Leszczynski, A. J. Sadlej), Kluwer, Dor-
drecht, 2006.
[10] J. M. Seminario, A. G. Zarcarias, J. M. Tour, J. Am. Chem. Soc.
1998, 120, 3970.
[21] A. D. Becke, J. Chem. Phys. 1993, 98, 1372.
[22] M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon,
D. J. Defrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654.
[23] A. Pople, D. L. Beveridge, P. A. Dobosh, J. Chem. Phys. 1967, 47,
2026.
[11] a) K. Clays, K. Wostyn, A. Persoons, Adv. Funct. Mater. 2002, 12,
557; b) E. Hendrickx, K. Clays, A. Persoons, Acc. Chem. Res. 1998,
31, 675; c) G. Olbrechts, R. Strobbe, K. Clays, A. Persoons, Rev. Sci.
Instrum. 1998, 69, 2233; d) K. Wostyn, K. Binnemans, K. Clays, A.
Persoons, Rev. Sci. Instrum. 2001, 72, 3215.
[24] N. Mataga, K. Nishimoto, Z. Phys. Chem. 1957, 13, 140.
[25] M. C. Zerner, G. H. Loew, R. F. Kichner, U. Mueller-Westerhoff, J.
Am. Chem. Soc. 1980, 102, 589.
[26] F. Meyers, S. R. Marder, B. M. Pierce, J. L. Brꢃdas, J. Am. Chem.
Soc. 1994, 116, 10703.
[12] R. Chinchilla, C. Nꢁjera, Chem. Rev. 2007, 107, 874.
[13] R. S. Becker, J. Seixas de Melo, A. L. Macanita, F. Elisei, J. Phys.
Chem. 1996, 100, 18683.
[27] H. D. Cohen, C. C. J. Roothaan, J. Chem. Phys. 1965, 43, 34.
[28] P. Chopra, L. Carlacci, H. F. King, P. N. Prasad, J. Phys. Chem. 1989,
93, 7120.
[14] J. Pina, H. D. Burrows, R. S. Becker, F. B. Dias, A. L. Macanita, J.
Seixas de Melo, J. Phys. Chem. B 2006, 110, 6499.
[29] M. N. Nakano, I. Shigemoto, S. Yamada, K. Yamaguchi, J. Chem.
Phys. 1995, 103, 4175.
[15] G. Striker, V. Subramaniam, C. A. M. Seidel, A. Volkmer, J. Phys.
Chem. B 1999, 103, 8612.
[16] J. Pina, J. Seixas de Melo, H. D. Burrows, A. Bilge, T. Farrell, M.
Forster, U. Scherf, J. Phys. Chem. B 2006, 110, 15100.
[17] P. J. Stephens, F. J. Devlin, F. C. F. Chabalowski, M. J. Frisch, J. Phys.
Chem. 1994, 98, 11623; J. J. Novoa, C. Sosa, J. Phys. Chem. 1995, 99,
15837; E. Casida, C. Jamorski, K. C. Casida, D. R. Salahub, J. Chem.
Phys. 1998, 108, 4439; R. E. Stratmann, G. E. Scuseria, M. J. Frisch,
J. Chem. Phys. 1998, 109, 8218.
[18] E. Runge, E. K. U. Gross, Phys. Rev. Lett. 1984, 52, 997; E. K. U.
Gross, W. Kohn, Adv. Quantum Chem. 1990, 21, 255; H. Heinze, A.
Goerling, N. Roesch, J. Chem. Phys. 2000, 113, 2088.
[30] a) V. M. Geskin, J. L. Brꢃdas, J. Chem. Phys. 1998, 109, 6163;
b) V. M. Geskin, C. Lambert, J. L. Brꢃdas, J. Am. Chem. Soc. 2003,
125, 15651.
[31] a) D. Beljonne, J. Cornil, R. H. Friend, R. A. J. Janssen, J. L. Brꢃdas,
J. Am. Chem. Soc. 1996, 118, 6453; b) D. Beljonne, Z. Shuai, R. H.
Friend, J. L. Brꢃdas, J. Chem. Phys. 1995, 102, 2042.
[32] D. Wasserberg, P. Marsal, S. C. J. Meskers, R. A. J. Janssen, D. Bel-
jonne, J. Phys. Chem. B 2005, 109, 4410.
[33] R. S. Becker, Theory and Interpretation of Fluorescence and Phos-
phorescence, Wiley-Interscience, New York, 1969.
[34] N. J. Turro, Modern Molecular Photochemistry, University Science
Books, Sausalito, 1991.
[19] J. B. Foresman, M. Head-Gordon, J. A. Pople, M. J. Frish, J. Phys.
Chem. 1992, 96, 135.
[20] Gaussian03, RevisionB.05, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
[35] R. A. J. Janssen, Primary Photoexcitations in Conjugated Polymers
(Ed.: N. S. Sariciftci), World Scientific, Singapore, 1997, pp. 524–
558.
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom
ery, Jr.,
[36] J. Casado, T. M. Pappenfus, K. R. Mann, E. Ortꢂ, P. M. Viruela, B.
Miliꢁn, V. Hernꢁndez, Lꢀpez J. T. Navarrete, ChemPhysChem 2004,
5, 529.
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
Received: March 18, 2009
Published online: July 16, 2009
8234
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 8223 – 8234