Synthetic, Optical and Theoretical Study of Alternating Ethylenedioxythiophene–Pyridine Oligomers: Evolution from Planar Conjugated to Helicoidal Structure towards a Chiral Configuration

Article abstract of DOI:10.1002/cphc.201601057

A series of alternating 3,4-ethylenedioxythiophene–alkynylpyridine oligomers (DA)_n with increased solubility are synthesized and their photophysical properties and nonlinear optical properties are investigated. Their quadratic polarizabilities are determined from hyper-Rayleigh scattering experiments to obtain information on their conformations in solution. These chromophores, based on the alternation of electron-rich (D) and electron-deficient (A) moieties, exhibit optical properties that arise from the combination of dipolar and helicoidal features in the (DA)_n homologue series where n=1–4. The transition from dipolar conjugated planar structures (n=1, 2) to helicoidal structures (n=3, 4) is clearly evidenced by results from symmetry-sensitive second-order nonlinear optical experiments. This suggests an approach towards highly efficient chiral chromophores for second-order nonlinear optics. Interestingly, this structural evolution also has significant impact on the photophysical properties: both absorption and fluorescence emission show bathochromic and hyperchromic shifts with increasing number of repeating units in the dipolar planar derivatives (n=1–2) but show saturation effects in the helicoidal structures (n=2–4). In addition, the helicoidal structures show sizeable two-photon absorption at 700–750 nm (40–100 GM) for compounds lacking either electron-donating or electron-withdrawing substituents.

Full text of DOI:10.1002/cphc.201601057

Accepted Article
Title: Synthesis,
Optical
and
Theoretical
Study
of
Alternated
Ethylenedioxythiophene-pyridine Oligomers: Evolution from Planar Con-
jugated over Helicoidal towards Chiral Configuration
Authors: Floris Chevalier; Marina Charlot; Florence Mongin; Benoˆıt Champagne;
Edith Franz; Koen Clays; Mireille Blanchard-Desce
This manuscript has been accepted after peer review and the authors have elected
to post their Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by using the Digital
Object Identifier (DOI) given below. The VoR will be published online in Early View as
soon as possible and may be different to this Accepted Article as a result of editing.
Readers should obtain the VoR from the journal website shown below when it is pu-
blished to ensure accuracy of information. The authors are responsible for the content
of this Accepted Article.
To be cited as: ChemPhysChem 10.1002/cphc.201601057
Link to VoR: http://dx.doi.org/10.1002/cphc.201601057
A Journal of
www.chemphyschem.org

ChemPhysChem
10.1002/cphc.201601057
Synthesis, Optical and Theoretical Study of Alternated Ethylenedioxythiophene-
pyridine Oligomers: Evolution from Planar Conjugated over Helicoidal towards
Chiral Configuration
Floris Chevallier,*[a,b] Marina Charlot,[a] Florence Mongin,[a,b] Benoît Champagne,*[c]
Edith Franz,[d] Koen Clays,*[d] and Mireille Blanchard-Desce*[a,e]
[a] Chimie et Photonique Moléculaires, UMR 6510 CNRS - Université de Rennes 1, 35042 Rennes, France.
[b] Institut des Sciences Chimiques de Rennes, UMR 6226 CNRS - Université de Rennes 1, 35042 Rennes, France.
[c] Laboratoire de Chimie Théorique, Université de Namur, B-5000 Namur, Belgium.
[d] Department of Chemistry, University of Leuven, B-3001 Leuven, Belgium
[e] Univ. Bordeaux, Institut des Sciences Moléculaires (UMR 5255 CNRS), 33405 Talence, France.
e-mail addresses: floris.chevallier@univ-rennes1.fr, benoit.champagne@unamur.be,
koen.clays@fys.kuleuven.be, mireille.blanchard-desce@u-bordeaux.fr
Abstract
A series of alternated 3,4-ethylenedioxythiophene-alkynylpyridine oligomers (DA)n with increased
solubility has been synthesized and their photophysical properties and nonlinear optical properties
have been investigated. Their quadratic polarizabilities  have been determined by performing HRS
experiments to get information on their conformation in solution. These chromophores based on the
alternation of electron-rich (D) and electron-poor (A) moieties exhibit optical properties that find their
genesis in the combination of dipolar and helicoidal features in the (DA)n homologue series for n = 1
to 4. The transition from dipolar conjugated planar structures (for n = 1 and 2) to helicoidal structures
(for n = 3 and 4) is clearly evidenced by the symmetry-sensitive second-order nonlinear optical
experiments. This suggests an approach towards highly efficient chiral chromophores for second-order
nonlinear optics. Interestingly, this structural evolution also has significant impact on the
photophysical properties: both absorption and fluorescence emission show a bathochromic and
hyperchromic shift with increasing number of repeating units in the dipolar planar derivatives (n = 1-
2) but saturation effect in the helicoidal structures (n = 2-4). In addition, the helicoidal structures show
sizeable two-photon absorption at 700-750 nm (40-100 GM) for compounds totally lacking either
electron-donating or electron-withdrawing substituents.
1
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Keywords
Nonlinear Optics, hyper-Rayleigh scattering, Helicoidal structures, Fluorescence, Two-photon
absorption
Introduction
The importance of electron-rich oligothiophenes in general,[1] and 3,4-ethylenedioxythiophene
(EDOT) in particular,[2] for the design of conjugated materials (including oligomers) with attractive
electronic and optical properties, is well recognized.[3] Among various structures, combining electron-
rich aromatic moieties with electron-poor ones, such as pyridine, is a viable strategy to enhance
properties that are derived from charge-transfer interaction, relevant for such applications as
photovoltaics[4] and second-order nonlinear optics.[5] We reported earlier the synthesis of oligomers
based on the alternation of electron-poor (pyridyl, A) and electron-rich (3,4-ethylenedioxythienyl, D)
heterocycles which show interesting fluorescence properties that strongly depend on their length
(Scheme 1).[6] The optical properties of these chromophores were discussed as a function of their
length and their specific electronic and geometrical (coiled structure) features.[7] One limit to the
earlier study being the poor solubility of the oligomers, we here consider more lipophilic derivatives,
bearing alkynyl chains at the 4 position of the pyridine rings which allows to expand the series with a
longer donor-acceptor DA homologue 4 (n = 4, see Scheme 1). To help elucidate the specific
relevance of the structural features (in particular the coiled configuration) for the spectroscopic
properties, the second-order nonlinear optical (NLO) properties were investigated. The coiled structure
in the longer alternated oligomers of these conjugated moieties is an intriguing feature. While for the
short homologues, planarity, and hence perfect conjugation over the complete oligomer, is achievable,
for the longer homologues, steric hindrance will prevent planarity and will cause breaking of the
conjugation towards a helicoidal configuration.[7] When steric hindrance becomes effective enough,
configurational locking allows chiral resolution into the two enantiomers. In this respect, these
oligomers are very reminiscent of helicenes.[8]
Because the lack of mirror symmetry relaxes the stringent symmetry requirements for second-
order NLO, chirality has been suggested as a viable approach towards (bulk) noncentrosymmetry.[9]
Chirality has also been used as a crystal engineering tool towards polar crystals of efficient dipolar DA
conjugated charge-transfer chromophores, to overcome the strong tendency towards the low-energy
centrosymmetric arrangement of such dipoles in aggregates in solution, and in the solid state.[10] Also,
unique contrast patterns in cellular imaging have been obtained from chiral membrane probes derived
from the binaphthyl group.[11]
2
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
The critical advantage of our material (series 1-4) with respect to helicenes is that the EDOT-
pyridine building block imparts inherently noncentrosymmetrical DA charge transfer properties in
addition to the noncentrosymmetry already derived from the helicoidal configuration. Unsubstituted
helicenes exhibit relatively modest second-order NLO properties only derived from the helicoidal
configuration,[8] so that computational and experimental efforts directed towards functionalization of
helicenes are needed to increase the nonlinear response.[12] They also differ from helical polymers on
which DA units are grafted to confer second-order NLO activity.[13] In that case the helicity drives the
spatial organization of the NLO-phores[14] and different strategies are needed to block the helical
conformation.[15]
We report here on a combined synthetic, spectroscopic and computational effort to rationalize
the optical properties of soluble alternating EDOT-pyridine oligomers. Apart from the standard
spectroscopy tools (linear optics), hyper-Rayleigh scattering (HRS) as a structural symmetry-sensitive
probe (second-order NLO) and two-photon excited fluorescence (TPEF, third-order NLO) were
determined. Advanced computational techniques were used to underpin the results from symmetry-
sensitive second-order NLO experiments. The effect of increasing number of DA repeating units on
the photophysical properties (including fluorescence and two-photon absorption properties) is
discussed, further evidencing the effect of structural transition from dipolar planar to helicoidal.
Results and discussion
Synthesis
The synthesis of the oligomers 1-4 depicted in Scheme 1 was performed using deproto-metalation and
cross-coupling reactions. The bis-heterocycle 1 was prepared in three steps from 2-chloropyridine
(Scheme 2, path a). Treatment with an in situ prepared mixture of 0.5 equiv of ZnCl2·TMEDA
(TMEDA = N,N,N',N'-tetramethylethylenediamine) and 1.5 equiv of [Li(tmp)] (tmp = 2,2,6,6-
tetramethylpiperidino) in tetrahydrofuran (THF) at room temperature led to metalation at the C-3 and
C-4 position in an 83:17 ratio, a result evidenced by subsequent iodine trapping.[16] The resulting
dihalogenopyridine mixture was then reacted with [Li(da)] in THF at -75 °C to allow iodine migration
from the 3 to the 4 position.[17] Under these conditions, the 4-iodo isomer 5a was isolated in 51%
overall yield. As previously documented for similar compounds,[18] Sonogashira cross-coupling of the
heteroaryl iodide 5a with 1-heptyne selectively and quantitatively afforded the corresponding
monocoupled product 6a. The latter was converted to the bis-heterocycle 1 according to a Negishi-like
cross-coupling reaction.[19] To this purpose, EDOT was converted to the corresponding thienylzinc
chloride by deproto-metalation with butyllithium at 0 °C followed by transmetalation with
ZnCl2·TMEDA. Subsequent reaction with 6a using a combination of PdCl2 and 1,1'-
3
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
bis(diphenylphosphino)ferrocene (dppf) ligand as the catalyst in refluxing THF gave the bis-
heteroaromatic derivative 1 in 88% yield.
In order to progress toward the longer oligomers 2-4, we considered the synthesis of an
analogue of 1, bearing a chloro group to allow further functionalization. Compound 7 was synthesized
in three steps from 2,6-dichloropyridine (Scheme 2, path b). The 4-iodo derivative 5b was first
n
obtained in 56% yield by metalation/iodination using BuLi and molecular iodine.[20] It was then
converted to 7 through the alkynyl derivative 6b as described above from 5a.
The key building block 7 was then used as starting material for the synthesis of 2 using the
modified Negishi cross-coupling reaction after deprotonation/transmetalation of 1 (78% overall yield).
This strategy was extended with success to the preparation of the elongated alternating oligomers 3-4.
Indeed, using the same cross-coupling method in an iterative synthetic approach and making use of
selective deprotonation of the EDOT moiety, heterohexamer 3 and heterooctamer 4 were synthesized
in 81-87% yields by coupling 7 with thiophene-deprotonated 2 and 3 respectively (Scheme 3).
Photophysical properties
All photophysical measurements were conducted in chloroform and the corresponding data are
collected in Table 1. The data on oligomers (compound 1’-3’) lacking the alkynyl side chains reported
earlier have also been included for comparison purposes.
Linear absorption
The motivation for adding the heptynyl chain was to improve the solubility of the oligomers in
organic solvents, in order to include the higher homologue n = 4 in the study. Yet this modification
was found to induce a consistent red-shift of the low-energy absorption bands as well as of the
fluorescence emission. Such bathochromic shift can be related to (hyper)conjugation effect[21] as well
as to slight extension of conjugation due to the presence of the ethynyl linkage. We also note a (non-
systematically) varying red-shift in higher-energy absorption bands, and changes in contributions from
vibrational progressions. Hence the variation of the absorption maximum values (max) is not enough
to derive quantitative conclusion. Instead, comparison of the absorption spectra of the different
oligomers shows that the low-energy cut-off slope in the absorption spectra or the high-energy cut-on
slope in the emission spectra are shifted by about 20 nm, a value surprisingly valid for all the
homologues (see Supporting Information, Figure SI1).
In addition to this steady spectral shift observed for all heptynylated homologues, the comparison of
the absorption spectra (Figure 1 - left panel) shows that increasing the number of repeating unit (DA)
induces bathochromic and hyperchromic effects, indicative of extended electronic conjugation. These
4
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
two effects are strongly interlaced in all but the highest-energy band (at 260 nm), resulting in
superimpositions and/or shoulders. This is mostly obvious for the shoulders of the low-energy
absorption bands of compounds 3 and 4. Due to these intricacies, it is again informative to consider
the low-energy cut-off slope in the absorption spectra. A saturation of the cut-off values is clearly
observed, pointing to a levelling-off in the effective conjugation length for the higher homologues, or
alternatively, a distortion of the planarity.
Fluorescence properties
The fluorescence properties are found to depend markedly on the number of DA repeating
units, showing a non-monotonous variation: a marked bathochromic shift is observed on going from 1
to 2 while the bathochromic shift is vanishing on going from compound 2 to compound 4, evidencing
a levelling-off of the conjugation in the emissive excited state (Figure 1 - right panel). As observed
from the number of DA repeating units also has a major effect on the fluorescence quantum yield. An
increase of about one order of magnitude of the fluorescence quantum yield (f) is observed upon
going from 1 (which is only weakly fluorescent) to higher homologues (f = 0.3-0.4). This effect can
be related to both an increase of the radiative decay rate and a decrease of non-radiative rate, in
comparison to compound 1. The much lower radiative decay rate of compound 1, while its emission is
blue-shifted compared to its higher homologues, indicates a much lower transition dipole from the
lowest (slightly emissive) excited state to the ground state as compared to higher analogues. Its
comparatively much larger non-radiative decay rate also indicates that an efficient non-radiative
process competing with florescence emission is operating in this compound, suggesting effective
intersystem crossing from a singlet to a triplet excited state. This intersystem crossing is consistent
with the substantial singlet oxygen generation quantum yield ( ) value determined for compound 1’
( = 0.47). Interestingly, the singlet oxygen quantum yield values are observed to further decrease on
going from compound 2 ( = 0.35) to compound 3 and 4 ( = 0.28 and 0.25). Consistently, a
decrease of the non-radiative decay rate is observed on going from compound 2 to 3 or 4, suggesting a
less effective intersystem crossing. In parallel, a continuous slow decrease of the radiative decay rate is
observed on going from compound 2 to compound 4, in relation with the slight red-shift of the
emission.[22] As a result of these combined variations, compound 3 is found to show the largest
fluorescence quantum yield (f = 0.4).
Quite interestingly, while the heptynyl chain was mainly meant to improve solubility of the
prepared oligomers, the present study reveals that this side chain significantly affects the
photochemical behaviour of the oligomers compounds. In particular, a very interesting feature is the
effect of the presence of the heptynyl solubilizing chain on the fluorescence and singlet oxgen
quantum yield of the shorter analogues 1 and 2. In this case, the fluorescence quantum yield is
5
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
enhanced by a factor 2 upon ethynylating with respect to the values without alkynyl solubilizing chain
(compounds 1’ and 2’)[7] while no effect is noted for compound 3 compared to compound 3’. The
larger fluorescence quantum yield of compound 1 compared to its analogue 1’ can be ascribed to a
larger radiative rate (whereas the non-radiative rate are similar) suggesting that conjugation extends
over the ethynyl moiety in the emitting excited-state of compound 1, inducing a larger emission
transition dipole. In contrast, the larger fluorescence quantum yield of compound 2 compared to
compound 2’ can be ascribed to a smaller non-radiative decay rate as in this case the radiative rate are
similar, suggesting that the conjugation does extend on the ethynyl moiety in the relaxed excited state
in that case) . This suggests that the presence of the ethynyl alkyl chain hinders efficient non-radiative
decay processes in the case if compound 2. Such processes could be related either to inter-system
crossing (compatible with the much larger singlet oxygen generation quantum yield of compound 2’
compared to compound 2, resp. 0.85 and 0.35) or to aggregation of compound 2’ favouring competing
non-radiative process (such as electron-transfer). Interestingly, compound 3 also show lower singlet
oxygen generation quantum yield than his counterpart 3’ lacking the heptynyl chain (resp. 0.56 and
0.28) but the contrast is less marked than in the case of the shorter oligomer 2 and 2’. Along the same
line, we observed that compounds 3 and 3’ show similar quantum yield and lifetimes (as opposed to
the case of oligomers 2 and 2’) and consequently similar radiative and non-radiative rates. Hence the
effect of the ethynyl moiety on either aggregation or conjugation in the emissive excited state seems to
level off over two repeating units, most probably in relation with the transition from linear to coil
structure (vide infra).
Third-order nonlinear optical properties: Two-photon absorption
Thanks to their fluorescence properties, the two-photon absorption properties of compounds 1-
4 could be investigated using the well-known protocol based on two-photon excited fluorescence
experiments.[23] Two-photon absorption is a third-order optical process which accordingly does not
require absence of centrosymmetry. On the other hand two-photon absorption spectroscopy is of much
interest to investigate the various excited state as symmetry rules define one- and two-photon allowed
state.[24] For instance dipolar (or push-pull chromophores) show a lowest excited state which is both
one and two-photon allowed while perfect quadrupoles (having a center of symmetry) show one-
photon excited state which are two-photon forbidden and conversely two-photon excited state which
are not seen under one-photon excitation. In that framework, the investigation of two-photon
absorption spectra is of much interest. As seen in Figure 2, compounds 2-4 (i.e. oligomers) show
sizeable 2PA response in the spectral range of interest, although they do not bear electron-releasing or
electron-withdrawing substituents. Hence the alternation of electron-rich and electron-poor aromatic
6
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
units is enough to generate moderate 2PA responses. As observed in Figure 3 showing compared two-
photon and (rescaled) one-photon absorption spectra, the lowest excited-state (corresponding to the
absorption band in between 800 and 850 nm) is both one- and two-photon allowed. This is consistent
with the lack of a centrosymmetry of the molecular structure of all oligomers. We note that the
corresponding 2PA responses are rather small (8 to 15 GM) and slightly increases with the number of
repeating unit while their position is similar (810-820 nm). On the other hand distinct additional bands
are observed at higher energy leading to larger 2PA responses. In particular both oligomers 3 and 4
show a noticeable second maximum located at 750 nm (with corresponding 2PA values of 50 GM)
while compound 2 only second maximum located at 720 nm. This second band leading to higher 2PA
responses (and similar values for the three compounds) might be related to the specific coiled
conformation adopted by oligomers 2-4 (vide infra). Finally all oligomers show even higher 2PA
responses at higher energy in connection with a congestion of higher-energy excited states. Compound
4 shows the highest 2PA response at 700 nm (up to 100 GM). Finally oligomer 3 shows the largest
two-photon brightness (20 GM) at 750 nm while oligomers 3 and 4 show similar two-photon
brightness (30 GM) at 700 nm.
Second-order nonlinear optical properties
The linear optical properties (UV-vis absorption and fluorescence) have a strong influence on the
nonlinear optical properties, especially with respect to the laser wavelength dependence and the
precision of the obtained results. The first hyperpolarizability HRS has been determined by hyper-
Rayleigh scattering (HRS) at 880 nm (see below). The strong two-photon excited fluorescence (TPEF)
contribution to the HRS signal at 440 nm is clear from Figure - left panel, showing the significant low
amplitude-modulated (AM) frequency fluorescence contribution to the total HRS signal, in
combination with the strong demodulation of this contribution for higher AM frequencies. The
substantial TPEF contribution is also evident from the values for the phase lag (Figure - middle
panel), approximating values as high as 90°, the maximum value for the phase delay in the case of
pure fluorescence, as observed in phase fluorometry.[25] Also the depolarization ratio ()[26]
measurement is strongly affected by TPEF, as shown in Figure - right panel. This accounts for the
rather low precision for the deduced values in presenting the results of the HRS experiments at 880 nm
in CHCl3. The dynamic first hyperpolarizability values HRS,880 (10-30 esu) show a very strong increase
upon going from the monomer 1 to the tetramer 4, that does not seem to suffer from saturation. There
is, however, a significant effect of resonance enhancement, which becomes more pronounced for the
longer structures, as the wavelength of maximal absorption gets closer to the second-harmonic
wavelength of 440 nm.
7
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
If we apply the simple two-level model[27] to crudely estimate the effect of the resonance
enhancement, the static hyperpolarizability values HRS,o (10-30 esu) are obtained. These values show a
rather nice linear dependence with number of DA units (within experimental error). The two-level
model might imply an oversimplification and involve an overcorrection (due to the closeness of the
max to the second-harmonic wavelength) for the magnitude of the resonance enhancement factor EF,
derived from the max for the lowest-energy transition), but this would affect the values in a very
gradual way, leaving the finding of the linear dependence of the hyperpolarizability on number of DA
units.
It is interesting to observe that both the dynamic and static hyperpolarizability values seem to
show a systematic increase with number of DA units, but the ever stronger enhancement of HRS,880 for
the longer compounds not only results in a very steep slope for HRS,880 as a function of n, but
particularly in a negative intercept for the dynamic HRS,880 values (at the mathematically relevant n =
0), while the linear dependence for the static HRS,0 values after correction for resonance enhancement
results in a physically much more acceptable zero intercept (-7 +/- 14 x 10-30 esu for n = 0) and, much
more physically relevant, a constant slope of (30 +/- 5) x 10-30 esu per DA unit. Especially the zero
intercept lends additional credibility to the resonance correction.
It might be correctly argued that this two-level model is oversimplified. However, from the
complexity of the absorption spectrum, it was deemed overambitious to attempt a more complete
analysis of the contributions from the different bands to the second-order NLO response. Irrespective
of the interpretation of the hyperpolarizability values in terms of resonance or not, the general
conclusion from the hyperpolarizability measurements should be that there is a linear increase going
from 1 to 4 without any sign of saturation. It is highly instructive to also review the other parameters
in Table 2 that could be deduced from the analysis. The retrieved fluorescence lifetime  found here is
confirming the trend of longer lifetimes reported in Table 1. There is no quantitative agreement, as the
experimental conditions are quite different (880 nm TPEF at 440 nm as a side effect in HRS, versus
dedicated experimental setup with one-photon excitation and detection at emission maximum). The
robustness of the hyperpolarizability results against detailed fluorescence decay profile has been
demonstrated earlier.[28]
More importantly, the TPEF contribution, revealed as its amplitude at 1 kHz AM frequency,
ATPEF, is confirming the earlier observations about the third-order nonlinear optical properties: Also in
this TPEF contribution, we see a negligible contribution for 1, and an almost linear increase for 2, 3
and 4.
To gain more insight in the symmetry of the chromophores, and the nature of the second-order
nonlinear response, the HRS depolarization  values were estimated.[26] Because of the significant
8
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
TPEF contribution,  values were seriously underestimated, inherently at 1 kHz because of the strong
fluorescence depolarization at low AM frequency, but still at the highest AM frequency of 880 MHz.
We had to resort to simultaneous fitting of all observables to arrive at still rather crude estimates for
the fluorescence-free  values[8] is providing the 3 sets of  values.
Theoretical results
Geometry optimizations were performed on the successive oligomers where the C5H11 groups are
replaced by CH3 groups and where the starting conformations match the X-ray diffraction data of
reference [7]. They show that for n = 1-3 the systems are mostly planar whereas from n = 4 a helix
starts to form in order to avoid steric hindrance (Figure 5). Additional calculations demonstrate that
linear oligomers obtained by a π-rotation of the dihedral angles between the pyridine and EDOT rings
are less stable. The calculated first hyperpolarizability values are listed in Table 3 and presented in
Figure . The HRS value HRS increases almost linearly for n ranging between 1 and 4 and then it levels
off. The corresponding depolarization ratio  is close to the value of 1.5, typical of octupolar systems,
when n = 3-5 whereas for smaller system a more dipolar character is observed. The EFISHG data are
different and show oscillations in the // values as a function of n. // is maximum for n = 2, then
decreases and increases again.
Analysis and effect of conformational transition
The linear optical properties, and also the third-order nonlinear optical properties can to a large extent
be analyzed in terms of number of EDOT and pyridine units, or, conversely, in terms of conjugation
length. There is a clear influence of the number of units on the oscillator strength of the highest-energy
transition at about 260 nm. This influence becomes less obvious in the lower-energy bands, where
there is a considerable red-shift of the bands observed. This red-shift of multiple bands is causing the
lowest-energy band to end up as a mere shoulder (at 422 nm) for compound 4, while it is still
observable as a local maximum at 415 nm for compound 3. The red-shift is indicative of increased
conjugation length, so the saturation in red-shift points to an effective conjugation length not growing
with the number of conjugated units when n > 3, or to an increasing distortion away from planarity.
From linear optics (and by the same token, from third-order nonlinear optics) nothing more can be
indisputably derived, since odd-order effects are not symmetry-sensitive, or more precisely do not
require lack of centrosymmetry.
This is where second-order nonlinear optical properties can provide more specific insight,
since even-order effects are critically sensitive to symmetry. We observe that the first
9
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
hyperpolarizability (second-order nonlinear optical molecular polarizability) is increasing linearly with
the number of units for n = 1-4. More instructively, the depolarization ratio shows that the shorter
compounds behave largely as a dipolar chromophore, but that the longer compounds do not continue
to extend along a single major molecular dipolar axis. This is most explicitly demonstrated in the
theoretically derived J=1 and J=3 values (Figure ). At n = 1, J=1 > J=3 so that NLA is smaller than 1
but from n = 1 to 4, the octupolar J=3 contribution increases linearly while the dipolar J=1 contribution
decreases. So, the octupolar contribution becomes dominant and dictates the evolution of HRS with n.
Going from n = 4 to 5, the octupolar contribution saturates and the dipolar one increases so that NLA
decreases. The transition from dominant dipolar (n = 1) to octupolar (n = 2-5) response is also visible
when analyzing the // EFISH response but the  angle between the dipole moment and the vector part
of  plays also a role. Indeed, for n = 1 to 5,  = 60°, 22°, 33°, 31°, and 33°, respectively. Both
analyses nicely corroborate the lower depolarization ratios for 3 and 4 indeed, pointing to a non-
dipolar symmetry.
Finally, reverting to the observed changes in fluorescence quantum yield, it is conjectured that
the effects of higher homologues and that of alkynylating are orthonormal in preventing self-
aggregation and quenching, either through the increased solubility by the chain, or by the deviation
from planarity. The lowest fluorescence quantum yield is observed for non-alkynylated DA n = 1,[7]
where the lower solubility and the planarity both are ideally conducive towards aggregation and
quenching (leading both to the lowest radiative rate and higher non-radiative rate). Either alkynylation
(increase by factor of 2) or oligomerization (1 to 3) results in a quantum yield of over 30%. This value
is obtained already for compound 2, which combines both structural features of coiling and
solubilizing. In the DA series without solubilizing tail, the higher fluorescence quantum yield is only
obtained for (DA)3,[7] which then has the stronger deviation from planarity, in agreement with the
rationale derived from depolarization ratios. An alternative explanation can be in terms of reduced
torsional degrees of freedom as a consequence of the intramolecular H-bridges and repulsion between
the N and O atoms of successive rings, also effects that become more pronounced for higher
homologues (longer oligomers), but this could not explain the effect of alkynylation.
Interestingly the transition from linear to coiled structure is also apparent in the onset of the
2PA band at higher energy (720-750 nm) to sizeable 2PA response (40-50 GM) for chromophores
which totally lack either electron-donating or electron-withdrawing substituents.
Conclusions and Perspectives
10
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
From an integrated synthetic, spectroscopic and computational approach, we have been able to
rationalize the linear and nonlinear optical properties of a homologue series of alternated EDOT-
alkynylpyridine oligomers in terms of an increasingly non-planar non-dipolar, coiled structure with a
limiting effective conjugation length. For the shorter (DA)n n = 1 or 2 homologues, a largely dipolar
response from unsaturated and planar effective conjugation is observed, while for the n = 2, 3 and 4
homologues, increasingly non-dipolar/octupolar response is attributed to the coiled structures with
deviation from planarity. The dipolar part of the second-order nonlinear optical response is shown to
follow this argumentation, while the total second-order nonlinear optical response is not levelling off
for the higher homologues. This confirms the unique features of coiled molecular structures for
second-order nonlinear optical properties.
In ensembles of molecules, or in thin film solid state format for practical applications in
second-order nonlinear optics, the non-centrosymmetry should be conserved at this supramolecular
level. To this end, resolution of the helicoidal structures in the two chiral enantiomers should be
attempted. It is expected that 3 is still configurationally unstable,[7] but steric hindrance in 4 should be
sufficient to lock the helical structure for chiral resolution. Alternatively, introducing more steric
hindrance in shorter oligomers has been shown to be an effective resolution strategy in
pentahelquats.[29]
Finally, synthetic efforts towards even higher, yet still soluble, homologues could lead towards
configurationally stable and even more efficient optical materials, since experimentally, no saturation
of the second- and third-order nonlinear optical properties has yet been observed, although the
computational result seems to suggest the onset of clamping of the HRS hyperpolarizability.
Acknowledgements
We are grateful to the Université de Rennes 1 and the CNRS for support (F.C. and F.M.).
MBD acknowledges financial support from Rennes Metropole and stimulating discussion with Olivier
Mongin. This work is supported by funds from the Francqui Foundation. The calculations were
performed on the computers of the Consortium des Équipements de Calcul Intensif and mostly those
of the Technological Platform of High-Performance Computing, for which we gratefully acknowledge
the financial support of the FNRS-FRFC (Conventions No. 2.4.617.07.F and 2.5020.11) and of the
University of Namur.
11
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Experimental Section
Synthesis
The reactions were performed in Schlenk tubes under an argon atmosphere. THF was freshly distilled
over sodium/benzophenone. Commercially available starting materials were used without further
purification. Liquid chromatography separations were achieved on silica gel Merck-Geduran Si 60
(63-200 μm). Nuclear magnetic resonance (NMR) spectra were acquired on Bruker AC-300 (300 and
75 MHz for 1H and 13C, respectively) spectrometer. 1H chemical shifts (δ) are given in ppm relative to
13
the solvent residual peak and C chemical shifts relative to the central peak of the solvent signal.[30]
Coupling constants (J) are given in Hz. High-resolution mass spectra measurements were performed at
the Centre Régional de Mesures Physiques de l'Ouest (CRMPO) in Rennes.
2-Chloro-4-iodopyridine (5a). To a stirred, cooled (0 °C) solution of 2,2,6,6-
n
tetramethylpiperidine (1.0 mL, 6.0 mmol) in THF (10 mL) were successively added BuLi (1.6 M
hexanes solution, 6.0 mmol) and ZnCl2∙TMEDA (0.51 g, 2.0 mmol). The mixture was stirred for 15
min at 0 °C before introduction of 2-chloropyridine (0.45 g, 4.0 mmol). After 2 h at room temperature,
a solution of I2 (1.5 g, 6.0 mmol) in THF (10 mL) was added. The mixture was stirred overnight before
addition of an aqueous saturated solution of Na2S2O3 (40 mL) and extraction with CH2Cl2 (3 × 40
mL). The combined organic layers were dried over Na2SO4, filtered and concentrated under reduced
pressure. The crude product was directly used in the next step. To a stirred, cooled (-75 °C) solution
nBuLi (1.6 M hexanes solution, 4.0 mmol) in THF (30 mL) were successively added diisopropylamine
(0.41 g, 4.0 mmol) and the crude mixture. The mixture was stirred for 4 h at -75 °C before addition of
water (50 mL) and extraction with diethyl ether (3 × 50 mL). The combined organic layers were dried
over MgSO4, filtered and concentrated under reduced pressure. Purification by flash chromatography
1
on silica gel (heptane/CH2Cl2: 4/1) gave 5a (488 mg, 51%) as a white solid (mp 40 °C). H NMR
(CDCl3, 300 MHz) 7.58 (dd, 1H, J = 1.4 and 5.2), 7.74 (dd, 1H, J = 0.5 and 1.4), 8.06 (dd, 1H, J = 0.4
13
and 5.2). C NMR (CDCl3, 75 MHz) 106.7, 131.7, 133.2, 149.8, 151.9. These values are consistent
with the literature.[17]
2,6-Dichloro-4-iodopyridine (5b). To a stirred, cooled (-75 °C) solution of 2,6-
n
dichloropyridine (4.4 g, 30 mmol) in THF (50 mL) was added BuLi (1.6 M hexanes solution, 30
mmol). After 45 min at -75 °C, a solution of I2 (7.6 g, 30 mmol) in THF (20 mL) was added. The
mixture was stirred 15 min before addition of an aqueous saturated solution of Na2S2O3 (25 mL) and
extraction with Et2O (3 x 20 mL). The combined organic layers were dried over MgSO4, filtered and
concentrated under reduced pressure. Precipitation in AcOEt gave 5b (4.6 g, 56%) a white powder
12
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
1
13
(mp 160 °C). H NMR (CDCl3, 300 MHz) 7.65 (s, 2H). C NMR (CDCl3, 75 MHz) 107.8, 131.6
(2C), 150.7 (2C). These values are consistent with the literature.[20]
2-Chloro-4-(1-heptynyl)pyridine (6a). To a stirred solution of 5a (3.6 g, 15 mmol) in
degassed THF (200 mL) were successively added 1-heptyne (1.6 g, 16 mmol), diethylamine (5.5 g, 76
mmol), CuI (0.29 g, 10 mol%) and PdCl2(PPh3)2 (0.53 g, 5 mol%) at room temperature. The reaction
mixture was stirred for 12 h at room temperature, quenched with water (1 mL) and filtered over
celite®. Purification by flash chromatography on silica gel (heptane/CH2Cl2: 4/1) gave 6a (3.0 g, 98%)
1
as a yellow oil. H NMR (CDCl3, 300 MHz) 0.92 (t, 3H, J = 7.0), 1.33-1.63 (m, 6H), 2.41 (t, 2H, J =
7.1), 7.15 (dd, 1H, J = 1.3 and 5.1), 7.29 (br s, 1H), 8.28 (dd, 1H, J = 0.7 and 5.1). 13C NMR (CDCl3,
75 MHz) 14.0, 19.5, 22.2, 28.0, 31.1, 77.5, 97.9, 124.5, 126.3, 135.3, 149.4, 151.6. HRMS (ESI):
calcd for C12H15ClN [M+H]+ 208.0893, found 208.0892.
2,6-Dichloro-4-(1-heptynyl)pyridine (6b). The above described procedure used for 6a was
employed starting from 5b (4.1 g, 15.0 mmol), THF (200 mL), 1-heptyne (1.6 g, 16 mmol),
diethylamine (5.5 g, 76 mmol), CuI (0.29 g, 10 mol%) and PdCl2(PPh3)2 (0.53 g, 5 mol%). Purification
1
by flash chromatography on silica gel (heptane/CH2Cl2: 1/1) gave 6b (3.6 g, 99%) as a yellow oil. H
NMR (CDCl3, 300 MHz) 0.92 (t, 3H, J = 7.0), 1.33-1.62 (m, 6H), 2.42 (t, 2H, J = 7.1), 7.20 (s, 2H).
13C NMR (CDCl3, 75 MHz) 14.1, 19.6, 22.3, 28.0, 31.2, 76.7, 99.6, 124.9 (2C), 137.5, 150.6 (2C).
HRMS (ESI): calcd for C12H14Cl2N [M+H]+ 242.05033, found 242.0502.
2-(3,4-Ethylenedioxy-2-thienyl)-4-(1-heptynyl)pyridine (1). To a stirred, cooled (-78 °C)
n
solution of EDOT (0.43 g, 3.0 mmol) in dry THF (5 mL) was slowly added BuLi (1.6 M hexanes
solution, 3.0 mmol). After introduction of ZnCl2·TMEDA (0.76 g, 3.0 mmol), the temperature was
raised to 0 °C. The mixture was stirred for 1 h at 0 °C and PdCl2 (11 mg, 2 mol%), dppf (33 mg, 2
mol%) and 6a (0.62 g, 3.0 mmol) were successively added. The reaction mixture was stirred for 48 h
at 60 °C, quenched with water (1 mL) and filtered over celite®. After addition of an aqueous saturated
solution of NaCl (10 mL), the mixture was extracted with CH2Cl2 (3 x 10 mL). The combined organic
layers were dried over MgSO4, filtered and concentrated under reduced pressure. Purification by flash
chromatography on silica gel (heptane/CH2Cl2: 3/2) gave 1 (827 mg, 88%) as a yellow solid (mp 65
°C). 1H NMR (CDCl3, 300 MHz) 0.91 (t, 3H, J = 7.0), 1.35-1.64 (m, 6H), 2.41 (t, 2H, J = 7.1), 4.21-
4.36 (m, 4H), 6.42 (s, 1H), 7.01 (dd, 1H, J = 1.5 and 5.1), 7.89 (br s, 1H), 8.40 (dd, 1H, J = 0.9 and
13
5.1). C NMR (CDCl3, 75 MHz) 14.1, 19.6, 22.3, 28.2, 31.2, 64.4, 65.1, 79.0, 95.5, 101.6, 118.4,
122.3, 123.1, 132.7, 140.2, 142.1, 149.0, 151.6. HRMS (ESI): calcd for C18H19NO2S [M+H]+
336.1034, found 336.1047.
2-Chloro-6-(3,4-ethylenedioxy-2-thienyl)-4-(1-heptynyl)pyridine (7). The above described
n
procedure used for 1 was employed starting from EDOT (0.43 g, 3.0 mmol), THF (5 mL), BuLi (1.6
13
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
M hexanes solution, 3.0 mmol), ZnCl2·TMEDA (0.76 g, 3.0 mmol), PdCl2 (11 mg, 2 mol%), dppf (33
mg, 2 mol%) and 6b (0.73 g, 3.0 mmol). Purification by flash chromatography on silica gel
(heptane/CH2Cl2: 6/4) gave 7 (0.76 g, 73%) as a yellow solid (mp 90 °C). 1H NMR (CDCl3, 300 MHz)
0.92 (t, 3H, J = 7.0), 1.31-1.64 (m, 6H), 2.42 (t, 2H, J = 7.1), 4.22-4.38 (m, 4H), 6.45 (s, 1H), 7.04 (d,
1H, J = 1.1), 7.78 (d, 1J, J = 1.1). 13C NMR (CDCl3, 75 MHz) 14.1, 19.6, 22.3, 28.1, 31.2, 64.4, 65.2,
78.1, 96.9, 102.8, 117.1, 120.7, 123.0, 132.5, 140.9, 142.0, 150.6, 152.0. HRMS (ESI): calcd for
C18H19ClNO2S [M+H]+ 348.0825, found 348.0834.
2-(6-(3,4-Ethylenedioxy-2-thienyl)-4-(1-heptynyl)-2-pyridyl)-5-(4-(1-heptynyl)-2-pyridyl)-
3,4-ethylenedioxythiophene (2). The above described procedure used for 1 was employed starting
n
from 1 (0.94 g, 3.0 mmol), BuLi (1.6 M hexanes solution, 3.0 mmol), ZnCl2·TMEDA (0.76 g, 3.0
mmol), PdCl2 (11 mg, 2 mol%), dppf (33 mg, 2 mol%), 7 (1.0 g, 3.0 mmol) and THF (5 mL).
Purification by flash chromatography on silica gel (heptane/CH2Cl2: 4/1) gave 2 (1.46 g, 78%) as a
yellow solid (mp 128 °C). 1H NMR (CDCl3, 300 MHz) 0.93 (br t, 6H, J = 7.0), 1.42 (m, 8H), 1.65 (m,
4H), 2.43 (t, 2H, J = 7.1), 2.44 (t, 2H, J = 7.1), 4.24 (m, 2H), 4.36 (m, 2H), 4.43 (s, 4H), 6.43 (s, 1H),
7.04 (dd, 1H, J = 1.5 and 5.1), 7.70 and 7.72 (AB, d, 2H, J = 1.2), 7.93 (br s, 1H), 8.49 (dd, 1H, J =
0.8 and 5.1). 13C NMR (CDCl3, 75 MHz) 14.1, 14.2, 19.6, 19.7, 22.3, 22.4, 28.3, 28.4, 31.3, 31.4, 64.5,
64.8, 64.9, 65.1, 79.1, 79.5, 94.7, 95.5, 102.2, 119.0, 119.9, 120.0, 120.3, 120.4, 123.1, 123.4, 132.6,
132.9, 140.0, 140.1, 140.2, 141.9, 149.2, 150.5, 150.9, 151.7. HRMS (ESI): calcd for C36H37N2O4S2
[M+H]+ 625.2195, found 625.2198.
2-[6-[5-(4-(1-Heptynyl)-2-pyridyl)-3,4-ethylenedioxy-2-thienyl]-4-(1-heptynyl)-2-pyridyl]-
5-[6-(3,4-ethylenedioxy-2-thienyl)-4-(1-heptynyl)-2-pyridyl]-3,4-ethylenedioxythiophene (3). The
above described procedure used for 1 was employed starting from 2 (0.63 g, 1.0 mmol), nBuLi (1.6 M
hexanes solution, 1.0 mmol), ZnCl2·TMEDA (0.25 g, 1.0 mmol), PdCl2 (4 mg, 2 mol%), dppf (11 mg,
2 mol%), 7 (0.35 g, 1.0 mmol) and THF (5 mL). Purification by flash chromatography on silica gel
1
(heptane/CH2Cl2: 7/3) gave 3 (758 mg, 81%) as a yellow solid (mp 146 °C). H NMR (CDCl3, 300
MHz) 0.94 (m, 9H), 1.34-1.71 (m, 18H), 2.45 (m, 6H), 4.25 (m, 2H), 4.43 (m, 10H), 6.45 (s, 1H), 7.05
(dd, 1H, J = 1.3 and 5.1), 7.66 and 7.69 (AB, d, 2H, J = 1.2), 7.72 and 7.73 (AB, d, 2H, J = 1.3), 7.91
(dd, 1H, J = 0.9 and 1.3), 8.47 (dd, 1H, J = 0.9 and 5.1). 13C NMR (CDCl3, 75 MHz) 14.1, 14.2 (2C),
19.6, 19.7 (2C), 22.3, 22.4 (2C), 28.3, 28.4 (2C), 31.3, 31.4 (2C), 64.6, 64.8 (2C), 64.9 (2C), 65.2,
79.1, 79.6 (2C), 94.6, 94.7, 95.4, 102.0, 118.9, 119.8, 120.0, 120.2, 120.3, 120.4, 120.6, 120.7, 120.8,
123.0, 123.3, 132.5, 132.8, 132.9, 139.9, 140.0, 140.1, 140.2, 140.3, 141.9, 149.1, 150.7, 150.8, 150.9,
151.0, 151.8. HRMS (ESI): calcd for C54H54N3O6S3 [M+H]+ 936.3175, found 936.3165.
2-[6-[5-[6-(3,4-Ethylenedioxy-2-thienyl)-4-(1-heptynyl)-2-pyridyl]-3,4-ethylenedioxy-2-
thienyl]-4-(1-heptynyl)-2-pyridyl]-5-[6-[5-(4-(1-heptynyl)-2-pyridyl)-3,4-ethylenedioxy-2-
14
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
thienyl]-4-(1-heptynyl)-2-pyridyl]-3,4-ethylenedioxythiophene (4). The above described procedure
n
used for 1 was employed starting from 3 (0.94 g, 1.0 mmol), BuLi (1.6 M hexanes solution, 1.0
mmol), ZnCl2·TMEDA (0.25 g, 1.0 mmol), PdCl2 (4 mg, 2 mol%), dppf (11 mg, 2 mol%), 7 (0.35 g,
1.0 mmol) and THF (5 mL). Purification by flash chromatography on silica gel (heptane/CH2Cl2: 7/3)
gave 4 (1.09 g, 87%) as a yellow solid (mp 182 °C). 1H NMR (CDCl3, 300 MHz) 0.95 (m, 12H), 1.35-
1.69 (m, 24H), 2.45 (m, 8H), 4.16-4.60 (m, 16H), 6.11 (s, 1H), 6.91 (dd, 1H, J = 1.2 and 5.0), 7.58-
7.74 (m, 6H), 7.80 (br s, 1H), 8.21 (d, 1H, J = 5.0). 13C NMR (CDCl3, 75 MHz) 14.1 (4C), 19.6, 19.7
(3C), 22.3, 22.4 (3C), 28.4 (4C), 31.3, 31.4 (3C), 64.4, 64.9 (4C), 65.0 (3C), 79.3, 79.5, 79.6, 79.7,
94.3, 94.5, 94.6, 94.9, 102.3, 118.9, 119.6-120.7 (12C), 122.8, 123.0, 131.9, 132.7, 132.8 (2C), 139.7,
140.0, 140.1, 140.2, 140.3, 140.4, 140.6, 141.3, 148.8, 150.6, 150.7, 150.8, 150.9, 151.0, 151.3, 151.5.
HRMS (ESI): calcd for C72H71N4O8S4 [M+H]+ 1247.41548, found 1247.4173.
Photophysical studies
All photophysical properties have been performed with freshly-prepared solutions of the
chromophores in air-equilibrated toluene at room temperature (298 K). UV/Vis absorption spectra
were recorded on a Jasco V-570 spectrophotometer. Steady-state and time-resolved fluorescence
measurements were performed on dilute solutions (ca. 10-6 M, optical density < 0.1) contained in
standard 1 cm quartz cuvettes using an Edinburgh Instruments (FLS920) spectrometer in photon-
counting mode. Emission spectra were obtained, for each compound, under excitation at the
wavelength of the absorption maximum. Fluorescence quantum yields were measured according to
literature procedures using quinine in 0.5 M H2SO4 as a standard (quantum yield f= 0.55).[31] The
reported fluorescence quantum yields are within ±10%. The fluorescence lifetime values were
obtained from the reconvolution fit analysis (Edinburgh F900 analysis software) of decay profiles
obtained using the FLS920 instrument under excitation with a nitrogen-filled nanosecond flashlamp.
The quality of the fits was evidenced by the reduced χ2 value (χ2 < 1.1). Measurements of singlet
oxygen quantum yield () were performed on a Fluorolog-3 (Horiba Jobin Yvon), using a 450 W
Xenon lamp. The emission at 1272 nm was detected using a liquid nitrogen-cooled Ge-detector model
(EO-817L, North Coast Scientific Co). Singlet oxygen quantum yields  were determined in
chloroform solutions, using tetraphenylporphyrin (TPP) in chloroform as reference solution (ΦΔ=
0.55) and were estimated from 1O2 luminescence at 1272 nm.
Third-order nonlinear optical method: two-photon excited fluorescence measurements
Two-photon excitation cross-sections (2) of chromophores 1-4 were determined by
investigating their two-photon-excited fluorescence (TPEF) in solution. These measurements provide
15
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
the TPEF action cross-section 2f The corresponding 2 values were derived by determining the
fluorescence quantum yield f from standard fluorescence measurements (vide supra). TPEF
measurements were conducted using a mode-locked Ti:sapphire laser operating between 700 and 1000
nm and delivering 80 fs pulses at 80 MHz, using the well-established method described by Xu and
Webb,[23a] and the appropriate solvent-related refractive index corrections.[23a] The quadratic
dependence of the fluorescence intensity on the excitation intensity was verified for each data point,
indicating that the measurements were carried out in intensity regimes, in which saturation or
photodegradation do not occur. TPEF measurements were calibrated relative to the absolute TPEF
action cross-section determined by Xu and Webb for fluorescein in 0.01M aqueous NaOH for 700-980
nm.[23a] The experimental uncertainty amounts to  10%.
Second-order nonlinear optical method: hyper-Rayleigh scattering
As a more widely applicable characterization method for second-order nonlinear optical properties at
the molecular level, hyper-Rayleigh scattering (HRS) has been used.[32] This method provides two
experimental observables, the total HRS scattering intensity, which can be related to the tensor
component averaged HRS(-2;,), and the HRS depolarization ratio (), which can provide
information about the molecular symmetry.
Based on the linear optical properties (UV-vis absorption spectra), it was decided to perform
HRS experiments at 880 nm, as a compromise between resonance enhancement and absorption of the
HRS signal at the second harmonic wavelength of 440 nm. The other rationale, for not shifting to even
longer fundamental wavelengths, is to make use of the high power of a Titanium-sapphire
femtosecond laser (a longer wavelength implies a serious drop of power from the Titanium-sapphire
laser, or the use of an optical parametric oscillator pumped by this Titanium-sapphire laser, in either
case a serious reduction of laser power).
Because of the fluorescent properties of the chromophores, it was imperative to use amplitude-
modulated (AM) frequency-resolved femtosecond hyper Rayleigh scattering.[33] A two-photon excited
fluorescence (TPEF) contribution to the HRS signal at low AM frequency (chopper frequency of 1
kHz) at 440 nm was observed for all samples. At the higher AM frequencies (up to 880 MHz) the
TPEF contribution was almost completely demodulated and the phase shift between immediate
nonlinear scattering for a non-fluorescent reference and the fluorescent samples had passed its
maximum.[34] From the simultaneous data analysis of modulus and phase of the HRS signal as a
function of AM frequency towards the TPEF contribution at 1 kHz, the fluorescence lifetime and the
first hyperpolarizability, an accurate and fluorescence-free hyperpolarizability can be extrapolated for
the high AM frequency limit. This approach in the frequency domain is our Fourier-transform
implementation of the discrimination between scattering and fluorescence in the time domain. The
16
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
AM frequency components are obtained as the inherent harmonic content of the femtosecond laser
pulse with a repetition frequency of 80 MHz for the Titanium sapphire laser (model Tsunami, Spectra-
Physics).[33]
To gain more insight in the molecular symmetry, or the nature of the second-order NLO
response, depolarization ratios () were also experimentally determined.[26] However, due to the TPEF
contribution, this  had also to be determined as a function of AM frequency, since the TPEF
contribution is strongly depolarized, resulting in a  of 1.0 at 1kHz, and only going up to physically
relevant values for HRS  for the higher AM frequencies.[8] To obtain reliable experimental values for
the , a simultaneous fitting of modulus, phase and  was implemented. For the  as a function of AM
frequency, this analysis results in the low-frequency  value (depolarized because of the fluorescence
depolarization) and the high-frequency  value, without fluorescence effects, hence the true HRS .
Because of the substantial TPEF contribution to both HRS total signal and  at low AM
frequencies, the extrapolation at high AM frequencies towards TPEF-free HRS and  yields accurate
values, yet with rather low precision.
Theoretical and computational methods
The ground state geometries were optimized at the density functional theory (DFT) level of
approximation by employing the B97-D exchange-correlation (XC) functional and the 6-311G(d)
basis set. The B97-D XC functional includes semi-empirical dispersion corrections and is therefore
well-suited to account for the London dispersion forces between the successive helix turns.[35]
Neglecting these interactions would indeed lead to an overestimated helical pitch as that was obtained
for pyridine-pyrimidine helical oligomers when using the standard B3LYP XC functional as well as
the AM1, PM3, or PM5 semi-empirical schemes.[13]
The time-dependent DFT (TDDFT) method was adopted to calculate the first
hyperpolarizabilities, in combination with the long-range corrected LC-BLYP XC functional[36] (with
a range-separating parameter  = 0.47) and the 6-311+G(d) basis set. In the LC-BLYP treatment, the
electron repulsion operator and thereof the exchange (X) is partitioned using the error function into
short- and long-range parts. The short-range part modifies a conventional DFT exchange potential
(here the Becke functional), while the long-range part is inserted into the Hartree-Fock exchange
expression.[37] This functional enables to avoid the chain-length-dependent overpolarization that is
characteristic of conventional X functionals[38] and that has been traced back to their short-
sightedness.[39] Owing to recent contrasted results for calculating  of extended systems,[40] the
standard range-separated parameter ( = 0.47) was used rather than a physically-motivated value
17
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
determined to reproduce Koopmans’ theorem.[39] Time-dependent Hartree-Fock (TDHF) calculations
were also performed and are consistent with the TDDFT/LC-BLYP results (SI).
The effects of the solvent were accounted for using the polarizable continuum model within
the integral equation formalism (IEF-PCM),[41] where the solvent is approximated as a structureless
polarizable continuum characterized by its macroscopic dielectric permittivity. Dichloromethane was
used as solvent (n = ∞ = 2.03). The calculations were done at 1064 nm, which using the LC-BLYP
XC functional can still be considered essentially "off-resonance", or for any practical comparative
purposes within this series, with very small resonance enhancement. The choice of a dynamic
dielectric constant enables to avoid the large static dielectric constant o of the solvent that would
overestimate the solvent effects on  and therefore to get close to the experiment situation where
measurements are performed at optical wavelengths and then extrapolated to infinite wavelength.
Starting from the 18 independent (-2;,) tensor components, two measured quantities
have been calculated, i.e. i) the HRS(-2;,) hyper-Rayleigh scattering (HRS) response that is
related to the HRS intensity for plane-polarized incident light and observation made perpendicularly to
the propagation direction,
bHRS - 2w;w,w =
b2 + bZ2XX
(1)
(
)
{
}
ZZZ
and ii) its depolarization ratio () that corresponds to the intensity ratio for measurements performed in
the VV (vertically polarized incident and scattered lights) and HV (horizontally polarized incident
light and vertically polarized scattered light) configurations, respectively
 Z2ZZ
 Z2XX
2
VV
I
(2)
 

2
HV
I
bZ2ZZ
bZ2XX
, quadratic products of  tensor components, which
where the full expressions for
and
correspond to isotropic averaging over the molecular motions, can be found in several textbooks and
publications.[5a, 42] Calculated  was also decomposed into its dipolar (J = 1) and octupolar (J = 3)
tensorial components, which are related to the quadratic averages. Assuming Kleinman’s conditions,
2
2
9
45
6
105
bZ2ZZ
=
=
bJ=1
+
+
bJ=3
(3)
(4)
2
2
1
45
4
105
bZ2XX
bJ=1
bJ=3
18
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
The nonlinear anisotropy parameter
compares the relative contributions of the
NLA= bJ=3 bJ=1
octupolar and dipolar components to the first hyperpolarizability tensor  while it is linked to the
depolarization ratio. The extreme cases correspond to fully octupolar and dipolar responses and their
NLA values amount to ∞ and 0 ( = 3/2 and 9) whereas for a one-dimensional push-pull system with a
unique dominant diagonal  tensor component, NLA = 0.82 ( = 5), respectively. In addition the //(-
2;,) quantity has been calculated, which can be deduced from electric field-induced second-
harmonic generation (EFISHG) measurements, and that corresponds to the projection of the vector
part of  on the dipole moment vector:
1
m
m bi
53å
i
i
(5)
b - 2w;w,w = b =
(b + b + b ) =
ijj jij jji
(
)
5 å å
//
//
||m ||
||m ||
i
j
i
where
is the norm of the dipole moment,  and  are the components of the  and  vectors. All
||m ||
i
i
 values are given within the B convention. All calculations were performed using Gaussian09.[43]
19
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
References
[1]
a) T.-G. Zhang, Y. Zhao, I. Asselberghs, A. Persoons, K. Clays, M. J. Therien, J. Am. Chem.
Soc. 2005, 127, 9710-9720; b) S. M. Bouzzine, M. Hamidi, M. Bouachrine, F. Serien-
Spirau, J. P. Lere Porte, J. M. Sotiropoulos, A. Iraqi, J. Phys. Chem. A 2012, 116, 9730-9738.
a) J. Roncali, P. Blanchard, P. Frère, J. Mater. Chem. 2005, 15, 1589-1610; b) P. Camurlu,
T. Durak, A. Balan, L. Toppare, Synth. Met. 2011, 161, 1898-1905; c) S. Narayanan, S. P.
Raghunathan, A. C. Poulose, S. Mathew, K. Sreekumar, C. Sudha Kartha, R. Joseph, New J.
Chem. 2015, 39, 2795-2806.
[2]
[3]
a) G. Barbarella, L. Favaretto, G. Sotgiu, M. Zambianchi, V. Fattori, M. Cocchi, F. Cacialli,
G. Gigli, R. Cingolani, Adv. Mater. 1999, 11, 1375-1379; b) A. Cazzato, M. L. Capobianco,
M. Zambianchi, L. Favaretto, C. Bettini, G. Barbarella, Bioconjugate Chem. 2007, 18, 318-
322; c) A. B. Nepomnyashchii, R. J. Ono, D. M. Lyons, C. W. Bielawski, J. L. Sessler, A. J.
Bard, Chem. Sci. 2012, 3, 2628-2638; d) M. L. Capobianco, G. Barbarella, A. Manetto,
Molecules 2012, 17, 910-933.
[4]
[5]
H. Spanggaard, F. C. Krebs, Sol. Energy Mater. Sol. Cells 2004, 83, 125-146.
a) T. Verbiest, K. Clays, V. Rodriguez, Second-order Nonlinear Optical Characterization
Techniques: An Introduction, CRC. Press, New York, 2009; b) Y. Shen, C.-F. Chen, Chem.
Rev. 2012, 112, 1463-1535; c) M. Gingras, Chem. Soc. Rev. 2013, 42, 1051-1095.
D. Tilly, F. Chevallier, F. Mongin, Synthesis 2016, 48, 184-199.
[6]
[7]
F. Chevallier, M. Charlot, C. Katan, F. Mongin, M. Blanchard-Desce, Chem. Commun. 2009,
692-694.
[8]
K. Clays, K. Wostyn, A. Persoons, S. Maiorana, A. Papagni, C. A. Daul, V. Weber, Chem.
Phys. Lett. 2003, 372, 438-442.
[9]
T. Verbiest, A. Persoons, in Topics in Stereochemistry, Vol. 24 (Eds.: M. M. Green, R. J. M.
Nolte, E. W. Meijer), Wiley, New York, 2003, pp. 519-570.
[10]
a) J. Zyss, J. F. Nicoud, M. Coquillay, J. Chem. Phys. 1984, 81, 4160-4167; b) R. Andreu, I.
Malfant, P. G. Lacroix, H. Gornitzka, K. Nakatani, Chem. Mater. 1999, 11, 840-848.
P. Yan, A. C. Millard, M. Wei, L. M. Loew, J. Am. Chem. Soc. 2006, 128, 11030-11031.
a) B. Champagne, J.-M. André, E. Botek, E. Licandro, S. Maiorana, A. Bossi, K. Clays, A.
Persoons, ChemPhysChem 2004, 5, 1438-1442; b) A. Bossi, E. Licandro, S. Maiorana, C.
Rigamonti, S. Righetto, G. R. Stephenson, M. Spassova, E. Botek, B. Champagne, J. Phys.
Chem. C 2008, 112, 7900-7907; c) B. Champagne, S. N. Labidi, Chem. Phys. Lett. 2016,
644, 195-200.
[11]
[12]
[13]
[14]
[15]
E. Botek, F. Castet, B. Champagne, Chem. Eur. J. 2006, 12, 8687-8695.
M. Ohkita, J.-M. Lehn, G. Baum, D. Fenske, Chem. Eur. J. 1999, 5, 3471-3481.
a) I. Huc, Eur. J. Org. Chem. 2004, 17-29; b) A. Petitjean, L. A. Cuccia, M. Schmutz, J.-M.
Lehn, J. Org. Chem. 2008, 73, 2481-2495.
[16]
a) J.-M. L'Helgoual'ch, A. Seggio, F. Chevallier, M. Yonehara, E. Jeanneau, M. Uchiyama, F.
Mongin, J. Org. Chem. 2008, 73, 177-183; b) K. Snégaroff, S. Komagawa, F. Chevallier, P.
C. Gros, S. Golhen, T. Roisnel, M. Uchiyama, F. Mongin, Chem. Eur. J. 2010, 16, 8191-
8201; c) K. Snégaroff, T. T. Nguyen, N. Marquise, Y. S. Halauko, P. J. Harford, T. Roisnel,
V. E. Matulis, O. A. Ivashkevich, F. Chevallier, A. E. H. Wheatley, P. C. Gros, F. Mongin,
Chem. Eur. J. 2011, 17, 13284-13297; d) M. Hedidi, G. Bentabed-Ababsa, A. Derdour, Y. S.
Halauko, O. A. Ivashkevich, V. E. Matulis, F. Chevallier, T. Roisnel, V. Dorcet, F. Mongin,
Tetrahedron 2016, 72, 2196-2205.
[17]
[18]
[19]
E. Marzi, A. Bigi, M. Schlosser, Eur. J. Org. Chem. 2001, 1371-1376.
J. V. Mello, N. S. Finney, Org. Lett. 2001, 3, 4263-4265.
a) E.-i. Negishi, Handbook of Organopalladium Chemistry for Organic Synthesis, Vol. 1,
Wiley-Interscience, New York, 2002; b) A. O. King, N. Okukado, E.-i. Negishi, J. Chem.
20
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Soc., Chem. Commun. 1977, 683-684; c) A. Seggio, G. Priem, F. Chevallier, F. Mongin,
Synthesis 2009, 3617-3632.
[20]
[21]
M. Schlosser, C. Bobbio, T. Rausis, J. Org. Chem. 2005, 70, 2494-2502.
a) K. Nassau, The Physics and Chemistry of Color: The Fifteen Causes of Color, John Wiley
& Sons, New York, 1983; b) J. B. Birks, Photophysics of Aromatic Molecules, Wiley-
Interscience, London, 1970.
[22]
[23]
S. J. Strickler, R. A. Berg, J. Chem. Phys. 1962, 37, 814-822.
a) C. Xu, W. W. Webb, J. Opt. Soc. Am. B 1996, 13, 481-491; b) M. A. Albota, C. Xu, W.
W. Webb, Appl. Opt. 1998, 37, 7352-7356.
[24]
a) M. Pawlicki, H. A. Collins, R. G. Denning, H. L. Anderson, Angew. Chem., Int. Ed. 2009,
48, 3244-3266; b) F. Terenziani, C. Katan, E. Badaeva, S. Tretiak, M. Blanchard-Desce,
Adv. Mater. 2008, 20, 4641-4678.
[25]
[26]
K. Clays, J. Jannes, Y. Engelborghs, A. Persoons, J. Phys. E Sci. Instrum. 1989, 22, 297-305.
G. J. T. Heesink, A. G. T. Ruiter, N. F. van Hulst, B. Bölger, Phys. Rev. Lett. 1993, 71, 999-
1002.
[27]
[28]
J. L. Oudar, D. S. Chemla, J. Chem. Phys. 1977, 66, 2664-2668.
K. Clays, K. Wostyn, G. Olbrechts, A. Persoons, A. Watanabe, K. Nogi, X.-M. Duan, S.
Okada, H. Oikawa, H. Nakanishi, H. Vogel, D. Beljonne, J.-L. Brédas, J. Opt. Soc. Am. B
2000, 17, 256-265.
[29]
L. Severa, D. Koval, P. Novotná, M. Ončák, P. Sázelová, D. Šaman, P. Slavíček, M.
Urbanová, V. Kašička, F. Teplý, New J. Chem. 2010, 34, 1063-1067.
H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 62, 7512-7515.
a) G. A. Crosby, J. N. Demas, J. Phys. Chem. 1971, 75, 991-1024; b) D. F. Eaton, Pure
Appl. Chem. 1988, 60, 1107-1114.
[30]
[31]
[32]
[33]
[34]
[35]
[36]
K. Clays, A. Persoons, Phys. Rev. Lett. 1991, 66, 2980-2983.
G. Olbrechts, R. Strobbe, K. Clays, A. Persoons, Rev. Sci. Instrum. 1998, 69, 2233-2241.
K. Wostyn, K. Binnemans, K. Clays, A. Persoons, Rev. Sci. Instrum. 2001, 72, 3215-3220.
S. Grimme, J. Comput. Chem. 2006, 27, 1787-1799.
a) H. Iikura, T. Tsuneda, T. Yanai, K. Hirao, J. Chem. Phys. 2001, 115, 3540-3544; b) Y.
Tawada, T. Tsuneda, S. Yanagisawa, T. Yanai, K. Hirao, J. Chem. Phys. 2004, 120, 8425-
8433.
[37]
[38]
H. Stoll, A. Savin, in Density Functional Methods in Physics (Eds.: R. M. Dreizler, J. da
Providencia), Plenum, New York, 1985, pp. 177-208.
a) M. Kamiya, H. Sekino, T. Tsuneda, K. Hirao, J. Chem. Phys. 2005, 122, 23411101-
23411110; b) H. Sekino, Y. Maeda, M. Kamiya, K. Hirao, J. Chem. Phys. 2007, 126,
0141071-0141076; c) P. A. Limacher, K. V. Mikkelsen, H. P. Lüthi, J. Chem. Phys. 2009,
130, 1941141-1941147; d) M. de Wergifosse, B. Champagne, J. Chem. Phys. 2011, 134,
07411301-07411313; e) S.-I. Lu, C.-C. Chiu, Y.-F. Wang, J. Chem. Phys. 2011, 135,
1341041-1341047.
[39]
[40]
L. Kronik, T. Stein, S. Refaely-Abramson, R. Baer, J. Chem. Theory Comput. 2012, 8, 1515-
1531.
a) H. Sun, J. Autschbach, ChemPhysChem 2013, 14, 2450-2461; b) K. Garrett, X. A. Sosa
Vazquez, S. B. Egri, J. Wilmer, L. E. Johnson, B. H. Robinson, C. M. Isborn, J. Chem. Theory
Comput. 2014, 10, 3821-3831; c) M. Hidalgo Cardenuto, B. Champagne, Phys. Chem.
Chem. Phys. 2015, 17, 23634-23642.
[41]
a) J. Tomasi, M. Persico, Chem. Rev. 1994, 94, 2027-2094; b) J. Tomasi, B. Mennucci, R.
Cammi, Chem. Rev. 2005, 105, 2999-3093.
[42]
[43]
R. Bersohn, Y.-H. Pao, H. L. Frisch, J. Chem. Phys. 1966, 45, 3184-3198.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P.
Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K.
21
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T.
Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C.
Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, 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, R. L. Martin, K. Morokuma, V. G.
Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, Revision A.1, Gaussian,
Inc., Wallingford CT, 2009.
22
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Table 1. Photophysical and two-photon absorption data of oligomers 1-4 in chloroform and their analogues lacking an heptynyl chain (1’-3’).[7]
max3
max2
max1
max
a
b
c
d
d
max3
max3
max2
max2
max1
max1
 abs
(nm) (cm-1 M-1) (nm) (cm-1 M-1)
10-4max3 abs
10-4max2
abs
10-4max1 b
(cm-1 M-1)
1.38
em
f

f
10-9kr 10-9knr
TPA
2
TPA
2
TPA
(nm)
2
2max1
(s-1)
0.09
0.04
0.28
0.29
0.18
0.21
0.12
(s -1)
2.69
2.74
0.54
1.67
0.27
0.30
(nm)
334
(nm)
391
372
437
408
(%)
3.1
1.3
34
(%)
(ns)
0.36
0.36
1.2
(nm) (GM)
(nm) (GM)
(GM)
257
2.03
286
286
354
343
365
353
391
1.55
1.22
2.59
2.30
4.62
3.62
5.31
668
-
-
-
1
1’
2
315.5
1.71
47
35
85
28
56
25
309
299
322
310
329
2.55
2.49
3.98
3.47
5.76
388(sh),402
373,389(sh)
3.03
804 <700
804 <700
840 <700
40
70
720
750
750
40
50
50
810
810
820
8
3.12
15
0.5
2’
3
402, 415(sh) 3.23
439,464(sh) 40
426,450(sh) 41
444,472(sh) 29
2.2
13
15
390
3.11
3.57
1.95
2.45
3’
4
420 sh
0.29
105
a
b
Fluorescence quantum yield determined in chloroform relative to quinine in 0.5 M H2SO4.
Singlet oxygen quantum yield determined relative to
d
c
e
f
tetraphenylporphyrin. Experimental fluorescence lifetime. Radiative (kr) and non-radiative (knr) decay rates. Two-photon absorption maxima. Two-photon
absorption cross-section (1GM= 1050 cm4 s. photon-1)
.
23
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Table 2. Second-order nonlinear optical properties of chromophores 1-4: dynamic fluorescence-free first
hyperpolarizability, HRS,880; estimated static fluorescence-free first hyperpolarizability, HRS,0; parameters used
in this estimation: resonance wavelength, max; derived resonance enhancement factor, EF; parameters from
TPEF contribution: amplitude, ATPEF; and lifetime,  ; and depolarization ratios : at low frequency (1 kHz,
TPEF affected), at indicated high frequency (a: 320 MHz, b: 880 MHz, still TPEF affected), and as derived from
global fitting. As a reference for the values, the octopolar crystal violet chloride is used, resulting in values
of 1.47 at 1 kHz, 1.43 at 320 MHz and 1.55 at 880 MHz, or essentially 1.5 as expected for the octopole.
a,b
MHz
EF
ATPEF
(10-30 esu)
HRS,880
(10-30 esu)
63
max
(nm)
333
HRS,0
(10-30 esu)
23



(ns)
1 kHz
global fit
NA
2.73
70
1.06 0.2
1.12 0.2a
1
2
3
4
1.0 0.2
2.0 0.3
3.0 0.3
3.0 0.2
5
2
4
1.09 0.2
1.05 0.2
1.06 0.2
1.71 0.2b
1.78 0.2b
1.49 0.2b
42
32
21
350
402
415
420
7.65
46
2680
80
180
1130
130
1550
150
23
11.65
14.58
96
6000
200
8000
200
11
106
10
NA: not available, not enough signal and not even an onset of increase in  value for higher AM frequencies, since shortest
fluorescence lifetime , as also to be observed in Figure SI1 lower panel.
24
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Table 3. EFISHG first hyperpolarizability values, //; HRS first hyperpolarizability values, HRS; and
depolarization ratios, ; for increasingly large DAn (n = 1-5) helices calculated at the IEF-PCM/TDDFT/LC-
BLYP level of approximation with the 6-311+G(d) basis sets and at  = 1064 nm.
// (10-30 esu)

HRS (10-30 esu)
n

1
2
3
4
5
2.9
5.0
3.6
3.1
3.4
4.1
6.2
4.68
2.20
1.73
1.58
1.61
8.2
10.7
10.9
25
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Scheme 1.
Scheme 2. a) ZnCl2·TMEDA (0.5 equiv), [Li(tmp)] (1.5 equiv), THF, RT, 2 h, then I2, then iPr2NLi, THF, -75
°C, 4 h (5a, 51%). b) nBuLi, THF, -75 °C, 45 min, then I2 (5b, 56%). c) 1-heptyne, Et2NH, CuI (10 mol%),
PdCl2(PPh3)2 (5 mol%), THF, RT, 12 h (6a, 98%; 6b, 99%). d) EDOT, nBuLi, ZnCl2·TMEDA, PdCl2 (2 mol%),
dppf (2 mol%), THF, reflux, 48 h (1, 88%; 7, 73%).
Scheme 3. a) nBuLi, then ZnCl2·TMEDA, THF, 0 °C, 1 h. b) 7, PdCl2 (2 mol%), dppf (2 mol%), reflux, 48 h.
26
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Figure 1. Absorption spectra (left panel) and normalized emission spectra (right panel) of 4 EDOT-
ethynylpyridine oligomers 1-4 in CHCl3.
Figure 2. . Two-photon absorption spectra of EDOT-ethynylpyridine oligomers 1-4 in CHCl3.
Figure 3. Two-photon absorption spectra of compound 2 (left), 3 (middle) and 4 (right) compared with their
rescaled one-photon absorption spectra.
27
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Figure 4. Experimental results from hyper-Rayleigh scattering for compound 4 in CHCl3 at 880 nm: Data
(black) points: left panel: demodulation of the AM frequency-dependent apparent first hyperpolarizability;
middle panel: phase lag between pure scattering from non-fluorescent reference and total signal; right panel: AM
frequency dependence of HRS depolarization ratio. Solid (red) line: fitting towards single fluorescence lifetime,
, TPEF amplitude, ATPEF, and fluorescence-free first hyperpolarizability HRS. Results for compounds 1 to 3 are
provided in the Supporting Information section.
Figure 5. Lateral and top views of the B97-D/6-311G(d) optimized structure of compound 4.
28
This article is protected by copyright. All rights reserved

ChemPhysChem
10.1002/cphc.201601057
Figure 6. HRS (HRS, J=1, J=3, and NLA) and EFISHG (//) quantities as a function of the number of units
calculated at the IEF-PCM/TDDFT/LC-BLYP/6-311+G(d) level of approximation at  = 1064 nm.
29
This article is protected by copyright. All rights reserved