Proaromaticity: Organic charge-transfer chromophores with small HOMO-LUMO gaps

Article abstract of DOI:10.1002/chem.201001051

Novel donor- and/or acceptor- substituted cross-conjugated carbocycles based on quinoids or expanded quinoids, with radiaannulene perimeters, were prepared and investigated to validate proaromaticity as a concept for reducing HOMO-LUMO gaps in push-pull chromophores. Analyses of IR, 1H NMR, and UV/Vis/NIR spectra in conjunction with molecular structures determined by X-ray diffraction show that these push-pull quinoids have significant charge-separated ground states. This feature results in small optical gaps (near IR region) and diatropic magnetic environments inside the carbocycles, as suggested by nucleus- independent chemical shift (NICS) calculations. The NICS results, together with the bond-length analysis of the quinoid spacers, provide strong support that proaromaticity, that is, aromatized zwitterionic mesomeric contributions in the ground state, is effective. A push-pull tetrakis(ethynediyl)- expanded quinoid chromophore represents the first proaromatic radiaannulene.

Full text of DOI:10.1002/chem.201001051

DOI: 10.1002/chem.201001051
Proaromaticity: Organic Charge-Transfer Chromophores with Small
HOMO–LUMO Gaps
Yi-Lin Wu,[a] Filip Buresˇ,[b] Peter D. Jarowski,[a] W. Bernd Schweizer,[a]
Corinne Boudon,[c] Jean-Paul Gisselbrecht,[c] and FranÅois Diederich*[a]
Abstract: Novel donor- and/or accept-
or-substituted cross-conjugated carbo-
cycles based on quinoids or expanded
quinoids, with radiaannulene perime-
ters, were prepared and investigated to
validate proaromaticity as a concept
for reducing HOMO–LUMO gaps in
push–pull chromophores. Analyses of
show that these push–pull quinoids
have significant charge-separated
ground states. This feature results in
small optical gaps (near IR region) and
diatropic magnetic environments inside
the carbocycles, as suggested by nu-
cleus-independent
chemical
shift
(NICS) calculations. The NICS results,
together with the bond-length analysis
of the quinoid spacers, provide strong
support that proaromaticity, that is, ar-
omatized zwitterionic mesomeric con-
tributions in the ground state, is effec-
tive. A push–pull tetrakis(ethynediyl)-
Keywords: annulenes · aromaticity ·
computational chemistry · conjuga-
tion · donor–acceptor systems
1
IR, H NMR, and UV/Vis/NIR spectra
in conjunction with molecular struc-
tures determined by X-ray diffraction
expanded
quinoid
chromophore
represents the first proaromatic
radiaannulene.
Introduction
are also popularly investigated by many research groups to
tailor the optoelectronic properties of monomeric chromo-
phores, denoted D-p-A systems. We reported in the past
years on families of nonplanar p-conjugated intramolecular
charge-transfer (ICT) chromophores featuring potent poly-
cyano acceptors and a variety of donors, which find applica-
Small band-gap organic chromophores have found applica-
tions in near infrared (NIR) fluorophores,[1] NIR photo-
ACHTUNGTRENNUNG
detectors/photovoltaics,[2] and nonlinear optical (NLO) ma-
terials.[3] These NIR organic materials are conventionally
produced and examined in the realm of p-conjugated poly-
mers; studies have revealed that the small band-gap could
be achieved by 1) increasing the conjugation length, 2) at-
taching powerful electron donors (D) and acceptors (A),
and 3) reducing the bond-length alternation (BLA) of the
conjugated backbones.[4] The first two approaches 1) and 2)
tronic devices.[5]
tion in optoACTHUNRGTENNUGelecCAHTUNGTRENNUGN
The third approach of varying BLA, on the other hand, is
based on the fact that the extent of p-electron delocalization
in polyene-type polymers is limited by Peierls distortion of
the polymer backbones that results in finite energy separa-
tions between the valence molecular orbitals. In polyarenes,
p-electron delocalization is confined within the aromatic
units, as polarization through the aromatic ring meets resist-
ance due to the partial loss of aromaticity. Increasing the
double-bond character of the s-bond linkage between
[a] Y.-L. Wu, Dr. P. D. Jarowski, Dr. W. B. Schweizer,
Prof. Dr. F. Diederich
Laboratorium fꢀr Organische Chemie
ETH-Zꢀrich, Hçnggerberg, HCI, 8093 Zꢀrich (Switzerland)
Fax : (+41)44.632.1109
arenes (or the quinACTHNUTRGNEUNGoid character of the arene units) would,
in turn, reduce the BLA of the whole polymer system, facili-
tate electron delocalization, and therefore reduce the
HOMO–LUMO gap. This is the case in polyisothianaph-
thene (PITN) and polyindenofluorene (PIF), for which NIR
electronic absorptions of about 1.2–1.5 eV have been report-
ed.[6]
The HOMO–LUMO gaps in monomeric chromophores
are also influenced by the BLA of the p spacers, as the elec-
tronic structures are immediate functions of molecular geo-
metries. In addition to the effect on electronic gap, Marder
E-mail: diederich@org.chem.ethz.ch
ˇ
[b] Prof. Dr. F. Bures
Institute of Organic Chemistry and Technology
Faculty of Chemical Technology, University of Pardubice
Studentskꢁ 573, Pardubice, 532 10 (Czech Republic)
[c] Prof. Dr. C. Boudon, Dr. J.-P. Gisselbrecht
Laboratoire d’Electrochimie et de Chimie Physique du Corps Solide
UMR 7177 C.N.R.S, Universitꢂ de Strasbourg
4, rue Blaise Pascal, 67000 Strasbourg (France)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001051.
9592
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9592 – 9605

FULL PAPER
et al. reported the interrelationship between molecular hy-
perpolarizabilities (b and g) and BLAs of oligomethines and
oligoenes in the 1990s.[7] Despite the fact that the fundamen-
tals of oligoene-based systems are better understood, organ-
ic materials that have been explored for applications are
mostly based on aromatic structural motifs for their intrinsic
superior stability; benzenoids in particular. Less aromatic
arenes, such as thiophenes, which pay smaller energetic pen-
alties of de-aromatization upon polarization, are frequently
used for organic optoelectronics. Conceptually, replacing
arenes by quinoid spacers would make polarization or CT
excitation even more favorable.
In this article, we report the preparation of the push–pull
quinoid systems 4–6 and related model systems and analyze
their spectra and molecular structures. Among them, push–
The highly colored compound 1 was published by Gomp-
per et al. in the late 1960s as the first example of D-A chro-
mophores with a quinoid spacer (abbreviated as D-Q-A af-
pull diphenoquinodimethane 5 is expected to be doubly
proaromatic and the expanded quinoid 6 would be the first
proaromatic radiaannulene macrocycle.[11] N,N-Dimethylani-
lino (DMA) or N,N-dibutylanilino (DBA) residues were
used as electron donors, while cyano or para-nitrophenyl
groups were employed as electron acceptors. Indeed, these
molecules show low-energy absorption maxima at 600–
850 nm, with the absorption edge down to 1200 nm (1.0 eV).
1
Their electronic absorption, H NMR, and IR spectra, and
their bond-length alternation are compared to those of ref-
erence compounds lacking quinoid spacers to highlight the
facile ground-state charge transfer in D-Q-A systems. Nu-
cleus-independent chemical shifts (NICS) were calculated to
analyze the aromatic characters in such quinoid systems,
and a preliminary evaluation of aromatic stabilization ener-
gies (ASE) of proaromatic systems was performed within a
set of hyperhomodesmotic reactions. Combining our results
from bond-length analysis and NICS calculations, the aro-
maticity of push–pull quinoid molecules was demonstrated.
terwards).[8] Over the past few decades, a similar design
principle has only been sparsely applied, incorporating
naphthoquinoid, anthraquinoid, or heteroquinoid p spacers
in some cases.[9] The intense, low-energy absorptions of
these quinoid chromophores were proposed to result from
the aromatized zwitterionic mesomeric contribution in the
ground state—the term “proaromatization” was thus coined.
The comparable degrees of charge separation in the elec-
tronic ground state and in the ICT excited state reduce the
energy that is needed for excitation. Large, negative solvato-
chromism and negative first molecular hyperpolarizabilities
have been observed for systems such as 2 and 3, implying
significant zwitterionic character in the ground states.[9a,b]
CT chromophores with proaromatic donors or acceptors in
recent years have found successful application as NLO-
phores and for molecular recognition purposes.[10]
Based on the neutral and zwitterionic limits of canonical
structures, the concept of proaromaticity seems to be ex-
tremely appealing; however, the fundamentals of proaroma-
ticity have not been systematically studied. Since the effi-
cient ground-state charge transfer might also originate from
a mechanism that does not involve aromaticity, a validation
of proaromaticity is clearly desirable.
Results and Discussion
Synthesis: The reference cyclohexyl and cyclohexenyl com-
pounds were published by our group,[12] and the synthesis is
briefly summarized in Scheme 1. Reaction of the cyclohexyl
compounds 7 and 8 with 2,3-dichloro-5,6-dicyano-1,4-benzo-
quinone (DDQ) resulted in partial oxidation to give the cy-
clohexenes 9 (50%) and 10 (43%), respectively. The quin-
oid linkages between the push/pull functionalities were syn-
thesized form their aromatic parents through a dehydration
reaction published for compound 4[9d] to avoid decomposi-
tion or polymerization under oxidative conditions. Electron-
Chem. Eur. J. 2010, 16, 9592 – 9605
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9593

F. Diederich et al.
ed the desired quinoids 4 and 5
in reasonable yields (88 and
51%, respectively, Scheme 1).
Preparation of the radia-
AHCTUNGTERGaNNUN nnulene-type expanded qui-
noid molecule 6 was achieved
in a modular approach. The
electron
acceptor, electron
donor, and the cis-hexa-3-en-
1,5-diyn-1,6-diyl moieties were
synthesized separately, and as-
sembled at the later stages to
construct
the
macrocycle
(Scheme 2). The pre-formed cis
geometry in the cyclopent-1-en-
1,2-diyl moiety[14] was exploited
not only to simplify the synthe-
sis, but also to avoid the p-elec-
tron localization in the syntheti-
cally more accessible ortho-di-
ethynylbenzene derivatives.
Scheme 1. Synthesis of push–pull quinoid chromophores 4 and 5, and their reference compounds with cyclo-
hex-1,4-diyl or cyclohex-2-en-1,4-diyl bridging units.
withdrawing, malononitrile-substituted phenyl 11 and bi-
phenyl 12 were obtained by a Pd-catalyzed Takahashi reac-
tion[13] with readily available para-dibromophenyl/biphenyl
precursors. Addition of the corresponding dilithiated aryl
anion to Michlerꢄs ketone, followed by dehydration, afford-
The acceptor unit 13 was obtained by dibromoolefination
of bis(4-nitrophenyl)methanone[15] in 72% yield. For the
donor part, Suzuki cross-coupling of gem-dibromo com-
pound 14[16] with 4-(dimethylamino)phenylboronic acid pro-
duced the triisopropylsilyl (TIPS) protected donor precursor
Scheme 2. Synthesis of push–pull chromophores 6, 19, and 20. m-CPBA=meta-chloroperbenzoic acid. For the specific Pd catalyst used for 14!15 and
15!16, please see the Experimental Section.
9594
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9592 – 9605

Proaromaticity
FULL PAPER
15a, which was then treated with tetrabutylammonium fluo-
ride trihydrate (TBAF·3H2O) in tetrahydrofuran (THF) to
remove the protecting group. Subsequently, Sonogashira re-
action between desilylated 15a and 2-bromocyclopent-1-
enecarbaldehyde (16)[17] afforded biscarbaldehyde 17a. One-
step alkynylation of 17a was achieved by treatment with
lithiated trimethylsilyldiazomethane [TMSC(Li)N2] to give
diethynyl derivative 18a.[18] Closure of the macrocyclic ring
was realized by a double Sonogashira cross-coupling reac-
tion between 13 and 18a. Though low-yielding, the D-Q-A
macrocycle 6a was synthesized under oxygen-free conditions
(41%), whereas the molecule of homocoupling, 19a, was
found as the major product if traces of dioxygen were pres-
ent; the 2:1 adduct 20a between 13 and 18a was occasional-
ly also isolated in a small amount. The DBA-substituted
macrocycle 6b was prepared by a similar route (14!15b!
17b!18b!6b) starting from the DBA–pinacol ester (19%
yield for the ring-closing step). The n-butyl substituents pro-
vide better solubility for both analysis and characterization.
gained from ICT; in contrast, for 4, the gain of onefold aro-
matization stabilization is concomitantly compensated by
the loss of aromaticity shared by the two aniline donors,
making CT in 4 less energetically favored (Scheme 3).
1
1
IR and H NMR spectra: Selected IR and H NMR data of
4–6a, 7, 9, 20a, and other model compounds are listed in
Table 1 to illustrate the extent of ground-state charge sepa-
Scheme 3. Contribution of charge-transfer aromatization (proaromatiza-
tion) to the ground state structures of 4 and 5.
1
The extent of downfield shifts of the H NMR resonance
Table 1. Selected IR and 1H NMR data for 4, 5, 6a, 7, 9, 19a, 20a, Li+
[C(CN)2Ph]ꢁ, and TCNQ.
of the N,N-dimethylamino moieties corroborates the find-
ings from CN vibration analysis. Additionally, 4 exhibits a
narrow range of the AA’BB’ signals from Ha and Hb at 7.14
and 7.17 ppm. The appearance of these proton signals in the
aromatic region, downfield from the olefinic region (com-
pared to 9), further supports the reduced quinoid character
in the ground state, due to ICT aromatization. In the case of
5, proton signals also appear in the aromatic region (6.87,
7.14, 7.23, and 7.45 ppm for Ha–d).
In the expanded systems 6 and 20, there is no formally
olefinic proton; however, the larger downfield shift of the
N,N-dimethylamino protons in 6a (3.02 ppm) suggests a
greater extent of charge separation in the ground state, as
compared to 20a (2.99 ppm), which in fact has more para-
nitrophenyl acceptor groups.
d [ppm][a]
n˜CN [cmꢁ1
]
Li+[C(CN)2Ph]ꢁ
TCNQ
7
9
4
5
19a
20a
6a
–
–
2178[b]
2226
2230
2212
2183
2168
–
2.95
3.02, 6.55, 7.01
3.16, 7.14, 7.17
3.31, 6.87, 7.14, 7.23, 7.45
3.04
2.99
3.02
–
–
[a] Only the 1H NMR chemical shifts (CDCl3) of N,N-dimethylamino
groups (dꢂ3 ppm) and/or the olefinic protons (dꢂ6.5–7.5 ppm) are
shown. [b] n˜CN of the corresponding Na+ salt is similar.
ration. Since the stretching frequency of the cyano groups is
sensitive to the vicinal electron density, it is a good probe
for the ground-state charge transfer. The bond strength and
therefore the vibrational frequency of CN groups are re-
duced by the presence of increased vicinal electron density,
Electronic absorption spectra: UV/Vis/NIR spectra of 4, 5,
7, and 9 are overlaid in Figure 1 to compare the effects of
the p spacers bearing similar carbon skeletons. Compound 7
exhibits aniline-centered bands at lmax =280 and 358 nm;[19]
the non-conjugated and distant dicyanovinyl group has little
influence on the absorption of the DMA ring. Conjugating
the two electronically active push–pull moieties by one
double-bond results in an ICT absorption at lmax =528 nm
(compound 9). Installation of another double bond into the
ꢀ
which interacts with the p* orbital of the C N group. For
reference, the stretching frequency, n˜CN =2178 cmꢁ1, for lithi-
um dicyanoACHTUNGTRENNUNG
(phenyl)methanide (Li+[C(CN)2Ph]ꢁ) models a
full formal vicinal negative charge, while the stretching fre-
quency, n˜CN =2226 cmꢁ1, for TCNQ serves as a model with-
out formal vicinal charge.[9a,d] The CN stretches at n˜CN =2230
and 2212 cmꢁ1 for compounds 7 and 9, respectively, point to
a small formal charge at the corresponding dicyanovinyl
cyclohexyl ring creates an even lower-energy band at lmax
=
698 nm with an energy gap of 1.55 eV estimated from the
absorption on-set in CH2Cl2 (compound 4). This striking dif-
ference in absorption spectra of 4 and 9 cannot be simply
explained by the difference of the electron donor/acceptor
nor the conjugation length, since identical donor/acceptor
groups are used, and identical linear conjugation paths are
sites. On the other hand, the lower values for 4 (n˜CN
=
2183 cmꢁ1) and 5 (n˜CN =2168 cmꢁ1) imply considerable CT
character in the ground state. The smaller n˜CN frequency for
compound 5 might be due to the twofold aromatization
Chem. Eur. J. 2010, 16, 9592 – 9605
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9595

F. Diederich et al.
cause of four para-nitrophenyl acceptors in compound 20a.
In view of the position of ICT bands, the accepting power of
the four para-nitrophenyl groups through the enediyne
spacers in 20a seems to be similar to that of the weakly
strained radiaannulenic cycle in 19a. As in the cases of quin-
AHCTUNGTREGoNNUN ids 4 and 5, when the macrocycle is closed in compound
6a, a lower energy absorption was found at lmax =586 nm
(estimated energy gap=1.62 eV). This decrease in CT tran-
sition energy cannot be attributed to the superior electron-
accepting power of the two para-nitrophenyl groups present,
since the number of accepting units here is in fact less then
that of 20a.
The absorption maxima of the lowest energy bands for 4,
5, and 6b exhibit various degrees of negative solvatochrom-
ism. This implies strong charge-separation in their ground
states which become more stabilized by increasing solvent
polarity compared to the excited states. For example, lmax of
4 shifts from 705 nm in THF, to 695 nm in acetone, and to
688 nm in EtOH (DE=372 cmꢁ1 or 0.05 eV). For 6b, lmax
appears at 622 nm in CHCl3, at 596 nm in acetone, and at
589 nm in EtOH (DE=901 cmꢁ1 or 0.11 eV). The largest
negative solvatochromism was found for diphenoquinodime-
thane 5: lmax shifts from 1005 nm in THF, to 802 nm in ace-
tone, and to 745 nm in MeCN (DE=3473 cmꢁ1 or 0.43 eV),
slightly more significant then reported for compound 2.[9b]
One should notice that the solvatochromism observed for
6b with its large p system is also partly explainable by the
changes in solvent polarizability. Plots of absorption maxima
of the lowest energy bands against the Dimroth–Reichardt
solvent polarity parameter ENT[21a] for 4, 5, and 6b, and a plot
against the Catalꢁn–Hopf SP parameter[21b] for 6b can be
found in the Supporting Information.
Figure 1. UV/Vis/NIR absorption spectra of compounds 4, 5, 7, and 9 in
CH2Cl2 at 298 K, cꢂ10ꢁ5 m.
present in both molecules (the directions of the transition
moments[20] for 4 and 9 are also similar; see illustration in
the Supporting Information). Extending the conjugation
pathway by an additional quinodimethane ring gives two
major absorptions (compound 5): one is located in the visi-
ble region with lmax =610 nm (estimated energy gap=
1.53 eV), and the other broad absorption is located in the
NIR region with lmax =845 nm (absorption edge ꢂ1200 nm;
1.03 eV). These lowest energy absorptions (lmax =698 nm for
4, lmax =845 nm for 5, and lmax =528 nm for 9) could be
quenched by the addition of trifluoroacetic acid.
The interesting behavior of the cross-conjugated cyclic p
spacers is further disclosed by comparing the series of ethy-
nediyl-expanded molecules (Figure 2). The weakly strained
Electrochemistry: Cyclic voltammetry (CV) and rotating-
disk voltammetry (RDV) of 4, 5, 6a, 6b, 19a, and 20a were
carried out in CH2Cl2 containing 0.1m nBu4NPF6 as support-
ing electrolyte. Potentials reported in Table 2 are referenced
to the ferricinium/ferrocene couple (Fc+/Fc). For most of
the species under investigation, due to electrode inhibition
during redox processes, reproducible data could only be ob-
served on freshly polished working electrodes (glassy carbon
disk, 3 mm in diameter).
Species 4 gave a well-resolved voltammogram in contrast
to 5, for which the CVs are not well resolved due to adsorp-
tions and inhibitions. For 4, two re-oxidation potentials cor-
responding to the electrogenerated species from the second
reduction steps were found at about ꢁ0.3 and +0.4 V by
CV. If the scan is reversed after the first reduction, these sig-
nals are not observed. Comparison of the first oxidation and
reduction potentials of 4 with those of 9 reflects that the de-
crease in optical HOMO–LUMO gap was mainly due to the
substantial effect of the proaromatic spacer on both HOMO
and LUMO levels. The first oxidation potential (E1/2)
changes from +0.33 V (9) to +0.17 V (4), whereas the first
reduction potential shifts from ꢁ1.45 V (9) to ꢁ1.30 V
(4).[12] The first oxidation was even shifted to ꢁ0.18 V for
diphenoquinodimethane 5.
Figure 2. UV/Vis/NIR absorption spectra of compounds 6a, 19a, and 20a
in CH2Cl2 at 298 K, cꢂ10ꢁ5 m.
(see below) macrocyclic Csp/Csp2 system acts as an electron
acceptor in molecule 19a, resulting in an ICT absorption at
lmax =540 nm. Compound 20a exhibits an absorption band
around 500 nm; however, comparing this molecule to 19a,
the two compounds exhibit different extinction profiles, be-
9596
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9592 – 9605

Proaromaticity
FULL PAPER
Table 2. Cyclic voltammetry (CV; scan rate v=0.1 Vsꢁ1) and rotating
disk voltammetry (RDV) data for compounds 4–6, 9, 19a, and 20a in
CH2Cl2 (+0.1m nBu4NPF6).[a]
can be found for 5. Thus, the oxidative potential might no
longer be simply attributable to the anilino “donor” units.
Macrocyclic compounds 6a and 6b showed similar redox
behavior by CV if one accepts the fact that for 6a, none of
the oxidation process was reversible in contrast to 6b. Only
at scan rates higher than 1 Vsꢁ1, do these processes became
reversible. The irreversible behavior may result from elec-
trode inhibition or low solubility of the oxidized species.
Studies by RDV gave, for both species, well-resolved waves
of equal amplitude with no electrode inhibition. Macrocycle
19a gave by CV, on reduction, a reversible one-electron
transfer followed by an irreversible reduction. On oxidation,
the first oxidation step was irreversible at all scan rates (up
to 10 Vsꢁ1), whereas the second oxidation was reversible
but of small amplitude due to possible electrode inhibition.
It had been previously shown that the ring system of radi-
aannulenes comprised of buta-1,3-diynediyl units is reduci-
ble at rather low potentials.[11b] In the present study, com-
pound 19a is reduced at ꢁ1.87 V, which is more negative
and may be due to a lower number of conjugated double
and triple bonds. Open-chain molecule 20a gave several re-
duction and oxidation steps, which all seem to be irreversi-
ble. However, scanning over a potential range including
only the first two reductions and the first oxidation shows
that the electron transfers were reversible at all scan rates.
From the peak amplitude, it turns out that the oxidation in-
volves one electron, whereas the two reductions each in-
volve two electrons, indicative of the reduction of the four
nitrophenyl moieties.[22]
CV
RDV
E8
DEp
Ep
E1/2
Slope
[V][b]
[mV][c]
[V][d]
[V][e]
[mV][f]
4
+0.76
70
+0.75 (1eꢁ)
+0.17 (1eꢁ)
ꢁ1.30 (1eꢁ)
ꢁ2.16 (1eꢁ)
60
60
75
90
+0.18
ꢁ2.14
ꢁ1.32
70
5
+0.75
ꢁ1.08
60
70
+0.65
ꢁ0.18
ꢁ1.09
ꢁ1.97
60
60
60
ꢁ0.16
ꢁ1.99
6a
+0.85
+0.60
+0.20
+0.90 (1eꢁ)
+0.60 (1eꢁ)
+0.16 (1eꢁ)
ꢁ1.38 (1eꢁ)
ꢁ1.54 (1eꢁ)
ꢁ1.70 (1eꢁ)
100
100
60
80
70
ꢁ1.38
ꢁ1.51
ꢁ1.67
80
80
80
90
ꢁ2.19
6b
+0.79
+0.54
+0.10
ꢁ1.38
ꢁ1.54
ꢁ1.71
70
70
60
70
70
60
+0.78 (1eꢁ)
+0.53 (1eꢁ)
+0.10 (1eꢁ)
ꢁ1.39 (1eꢁ)
ꢁ1.54 (1eꢁ)
ꢁ1.71 (1eꢁ)
60
60
70
70
70
70
ꢁ2.23
9
+0.70
80
+0.32
+0.33
ꢁ1.45
50
60
ꢁ1.71[g]
The effect of the expanded quinoid ring structure in 6a to
reduce the HOMO–LUMO gap was elucidated by compar-
ing the electrochemical potentials to those of 20a. The first
oxidation potential (E1/2) at +0.16 V for 6a is smaller than
that for 20a (+0.25 V), and the first reduction potential
(E8) at ꢁ1.38 V is also lower than that for 20a (ꢁ1.41 V).
All together, both the decrease in LUMO energies and the
increase in HOMO energies account for the small optical/
electrochemical gaps of push–pull proaromatic molecules, as
clearly suggested by the electrochemical data for 4–6.
19a
+0.79
+0.24
+0.63 (1eꢁ)
+0.26[h]
ꢁ1.87
90
ꢁ1.92 (1eꢁ)
60
70
ꢁ2.34
ꢁ2.39 (1eꢁ)
20a
+0.77
+0.57
+0.77 (1eꢁ)
+0.57 (1eꢁ)
+0.25 (1eꢁ)
ꢁ1.62[i]
120
125
100
+0.23
ꢁ1.41
ꢁ1.55
75
100
100
ꢁ2.05
ꢁ2.40
[a] All potentials are given versus the Fc+/Fc couple used as internal
standard. [b] E8=(Epc +Epa)/2, in which Epc and Epa correspond to the
cathodic and anodic peak potentials, respectively. [c] DEp =EpaꢁEpc.
[d] Ep =Irreversible peak potential. [e] E1/2 =Half-wave potential.
Structural properties: The molecular structures 6b, 19a, and
20a were analyzed by single-crystal X-ray diffraction
(Figure 3). The crystal structures of 9 and 10 have been pre-
viously reported.[12] For comparisons, we use the X-ray data
of 10, for which a much better structure refinement value
had been obtained. Also, the two compounds 9 and 10
behave similarly in absorption spectra and electrochemistry.
Crystals of 4 and 5 with sufficiently good quality were un-
available. Thus, their computational gas-phase structures
were optimized with Gaussian 09 at the level of B3LYP/6-
31G(d) and used in place of experimental data. Solvated
structures in MeCN were also calculated using the polariza-
ble continuum model (PCM).[23]
[f] Logarithmic analysis of the wave obtained by plotting
Log[I/
signal. [i] Large amplitude unresolved wave.
E versus
ACHTUNGTRENNUNG
Analyzing the frontier molecular orbitals of 4, 5, and 9
gave some qualitative insight into the unusual redox behav-
ior and hence the small optical gaps of 4 and 5 (Supporting
Information). While the HOMO of 9 has significant coeffi-
cients on the DMA moieties, as one would expect, the
HOMO of 4 is much more concentrated on the quinoid
spacer and shifted closer toward the dicyanovinyl moiety;
this observation is in line with what was seen in the IR
study. A similar yet more noticeable HOMO distribution
Bond-length alternation (BLA, dr) for a 1,4-disubstituted
hexagon can be numerically defined as depicted in Table 3,
and this value is frequently used as a measure of how much
the structure differs from perfectly delocalized benzene by
Chem. Eur. J. 2010, 16, 9592 – 9605
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9597

F. Diederich et al.
Table 3. Bond-length alternation (BLA) for push–pull quinoids 4 and 5,
tetrakis(ethynediyl)-expanded quinoid 6b, and their reference com-
pounds 10, 19a, and 20a; standard uncertainties are given in parenthesis.
BLA (dr) [ꢅ]
Aniline
Formal quinoid
0.031 (vacuum)
0.043 (MeCN)
0.033 (vacuum)
0.049 (MeCN)
0.075 (vacuum)
0.053 (MeCN)
0.060, 0.062 (vacuum)
0.032, 0.033 (MeCN)
4[a]
5[a]
10
6b
20a
19a
0.013 (3), 0.015 (4)
0.032 (6), 0.037 (6)
0.027 (12), ꢁ0.001 (12)
0.021 (3), 0.016 (3)
–
–
–
–
[a] Computational structures in either the gas-phase or solvated in
MeCN with PCM model at the level of B3LYP/6-31G(d).
the effect of substituents.[24] The dr value is zero for benzene
and 0.08–0.12 ꢅ for fully quinoid systems, for instance, dr=
0.08–0.10 ꢅ for TCNQ. Since N,N-dialkylanilino groups
were used as the electron donors in all the molecules under
study, the BLA of the aniline rings can serve as a general
probe for the degree of charge transfer in the ground state.
For the cyclohex-2-en-1,4-diyl-bridged compound 10, dr=
0.014 (4) ꢅ (average value for the two DMA rings) was
found in its solid-state structure, showing only a small
degree of charge transfer.[12] Much higher dr values of 0.031
(0.043 in MeCN with PCM model) and 0.033 ꢅ (0.049 in
MeCN) were found for the DMAs in push–pull quinoids 4
and 5, respectively. The ethynediyl-expanded molecules
present small dr values of 0.019 (3) ꢅ for the DMA donors
of 19a, and 0.013 (12) ꢅ for those of the open-chain com-
pound 20a (or 0.027 (12) for the ring with N67, since dr for
the other aniline ring is chemically unreasonable); again, a
much larger dr value of 0.035 (6) ꢅ was observed for the
DMAs of quinoid radiaannulene 6b. Taken together, these
data demonstrate the significance of the degree of charge
transfer in the case of push–pull quinoid system, in addition
1
to the data from the IR and H NMR analysis.
Analyzing the bond-length pattern of the quinoid p
spacers in 4 and 5 and the twisting between the two quinoid
rings in 5 gave structural hints of proaromaticity. As shown
in Table 3, dr=0.075 and 0.061 ꢅ were found for the formal-
ly quinoid rings of 4 and 5, respectively, based on the com-
puted structures in vacuum. These values are slightly smaller
than that of TCNQ, indicating some degree of bond-length
equalization. Calculations of these values in MeCN gave
greatly reduced dr numbers of 0.053 (4) and 0.033 (5) ꢅ,
which are comparable or even smaller than the correspond-
ing values of their formally aromatic aniline donors (0.043
and 0.049 ꢅ, respectively)! The smaller dr for the quinoid
ring and larger dr for the aniline donor of 5, compared to 4,
are in accordance with the greater degree of ground-state
charge transfer in 5. Additionally, a large dihedral angle of
248 between the two hexagonal spacers of 5 was found in
Figure 3. ORTEP plots of a) the expanded quinoid macrocycle 6b (at
173 K, the bond to disordered C70 is drawn with a dashed line), b)
DMA-substituted radiaannulene 19a (123 K), and c) open-chain mole-
cule 20a (223 K). The thermal ellipsoids were shown at the 50% proba-
bility level. Heteroatoms and some carbon atoms are labeled with arbi-
trary numbering. Hydrogens and crystal solvents (CDCl3 for 6b and
THF for 19a) were removed for clarity.
9598
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9592 – 9605

Proaromaticity
FULL PAPER
MeCN (138 in vacuum); this number is close to the twisting
of aromatic biphenyl derivatives.
macrocycle 6b is the first experimental, structural implica-
tion of aromatization in a quinoid radiaannulene system. It
should be emphasized here that all the previously proposed
geometric indices of aromaticity (Julgꢄs value, for in-
stance)[26] are implicitly based on the assumption of bond-
length equalization as a measure of aromaticity; as a result,
they are applicable to systems bearing simply alternating
single and double bonds. Since the concept of “bond-length
equalization” is inappropriate for systems such as dehy-
droannulenes or radiaannulenes, which contain CC triple
bonds as structural units; more structures of aromatic dehy-
droannulenes and radiaanulenes should be analyzed to give
another way to estimate their degree of electron delocaliza-
tion.
For donor substituted radiaannulene 19a, as expected, the
buta-1,3-diyne-diyl segment deviates from linearity; an aver-
age bond-angle of 167.1 (2)8 was observed at each of the
four Csp centers. As a result, the Csp-rich strained ring
system exhibits electron-accepting power and polarizes its
DMA donors, which have similar BLA number compared to
those of 6b. Though weakly strained, the molecule is stable
not only under ambient condition, but also towards Berg-
man cyclization. Bergman cyclization is known to occur
under physiological conditions when the terminal acetylenic
atoms of endiynes are separated by 3.15–3.31 ꢅ;[27] numbers
larger than this range suggest that heating is necessary for
cyclization. In the case of bis-endiyne moieties in 19a, dis-
The interplay of the quinoid radiaannulene ring and its
push–pull substituents could be seen by comparing the X-
ray crystal structures of 6b with those of symmetrically sub-
stituted 22,[25a] 23,[25a] and 24,[25b] and other related molecules
(Table 4; structures 22–24 are shown here). The average
Table 4. Selected bond lengths of ethynediyl-expanded molecules 6b,
19a, 20a, and 22–24; standard uncertainties are given in parenthesis.
Average bond length [ꢅ]
ꢁ
ꢀ
Csp Csp2
C
sp2=C
N
C
sp2=C
(endo)
Csp Csp
6b
22
23
24
20a
19a
1.422 (6)
1.432 (2)
1.430 (3)
1.432 (7)
1.455 (13)
1.424 (3)
1.384 (6)
1.367 (2)
1.369 (3)
1.376 (7)
1.354 (12)
1.391 (3)
1.349 (6)[a]
1.411 (2)[b]
1.409 (3)[b]
1.409 (8)[b]
1.362 (13)[a]
1.357 (3)[a]
1.203 (6)
1.199 (2)
1.199 (3)
1.195 (7)
1.127 (12)
1.208 (3)
[a] Csp2=Csp2 bond in the cyclopentenyl rings, a formal double bond.
ꢁ
[b] Benzo-annulated Csp2 Csp2 bond, a formal 1.5 bond.
ꢁ
ꢁ
tances of C2 C10 (3.826 ꢅ) and C11 C19 (3.959 ꢅ) are
longer than the optimal number. Indeed, only traces of mol-
ecules corresponding to [19a+H2] or [19a+2H2] were de-
tected by mass spectroscopy after heating 19a in cyclohexa-
1,4-diene at reflux for weeks.[28]
Evidence of electron delocalization based on nucleus-inde-
pendent chemical shifts (NICS) and aromatic stabilization
energies (ASE): In addition to experimental characteriza-
tions presented in previous sections, symmetric and asym-
metric quinoids, diphenoquinoids, and expanded quinoids,
which include compounds 4, 5, and truncated 6, respectively,
were subjected to NICS and ASE calculations to examine
the proposed concept of proaromatization in these systems.
Nucleus-independent chemical shifts (NICS) were intro-
duced by Schleyer and co-workers as a measure of electron
delocalization and induced ring current.[29] This index is de-
fined as the negative value of the calculated magnetic
shielding of the dummy atoms assigned at the specific posi-
tions of interest; frequently, positions at the ring center
(NICS(0)) or at 1 ꢅ above the center relative to the ring
plane (NICS(1)) are chosen. A negative NICS value is indi-
cative of the presence of diatropic ring current; as referen-
ces, the NICS(0) of highly aromatic benzene is ꢁ8.03 ppm,
about +20 ppm for antiaromatic D2h-cyclobutadiene, and
+1.7 ppm for non-aromatic para-quinodimethane. The
NICS calculations have been done for many annulenic com-
pounds, but few analogous calculations were conducted for
dehydroannulenes and radiaannulenes that bear acetylenic
units.
length of the endocyclic CC single bonds in 6b is slightly
shorter than of those of the symmetric compounds 22–24,
and also shorter than the corresponding CC single bonds of
push–pull compound 20a; the lengths of the endocyclic CC
triple bonds and the exocyclic CC double bonds in 6b are
uniformly longer than those of 22–24 and the corresponding
bonds of 20a as well. The comparison between the endocylic
C
sp2=Csp2 bonds cannot be easily made, since those bonds of
22–24 are part of the highly aromatic benzene rings; howev-
er, the endocyclic Csp2=Csp2 bond of 6b is apparently shorter
than the corresponding bonds in 20a.
The shortening of the formal CC single bonds and the
elongation of the formal CC double and triple bonds in
Chem. Eur. J. 2010, 16, 9592 – 9605
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9599

F. Diederich et al.
Table 5. Isotropic NICS(0) values calculated in vacuum at GIAO-
B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) level of theory for quinoid-type
molecules with various ring sizes and containing electronically symmetric
or asymmetric substituents. Values calculated in MeCN are indicated.
molecules. The seemingly incomplete quenching of the elec-
tron-donating power was not too surprising: based on spec-
troscopic evidence, Bunz and co-workers recently found that
protonated N,N-dibutylanilino groups behave similarly to
the weakly electron-donating phenol.[30]
ACHTUNGTRENNUNG
Entry
Class
R1
R2
NICS(0) [ppm]
1
2
CN
NH2
CN
NH2
+0.66
+3.04
Too much interpretation about the degree of aromatiza-
tion should not be made based on the absolute NICS values.
ꢁ1.79 (vacuum)
ꢁ4.90 (MeCN)
ꢁ0.86[a]
3
4
5
6
CN
CN
CN
CN
NH2
NH3
ꢀ
The ring-size and the anisotropic effects from the C N, C=
+
ꢀ
C, and C C functionalities all contribute to the overall mag-
ꢁ0.83 (vacuum)
ꢁ3.11 (MeCN)
+1.07[a]
netic properties of the system. However, observations of 1)
the homogeneous sign reversal from the positive values to
the negative ones, which accompanies electronically asym-
metric substitution, and 2) the much more negative NICS
values for the asymmetric molecules in MeCN both clearly
support the arguments of proaromaticity.
PhNMe2
PhNMe2H+
A: ꢁ1.58
7
CN
NH2
B: ꢁ1.13
A: ꢁ1.80
B: ꢁ1.16 (vacuum)
A: ꢁ4.67
8
9
CN
CN
PhNMe2
Besides the magnetic criteria of aromaticity, it has long
been recognized that the resonance stabilization energy
present in the p systems is an essential indication for the ex-
istence of aromaticity. Though compared to NICS, an appre-
ciably good quality of aromatic stabilization energy (ASE)
is much more difficult to reach computationally; the extent
of resonance stabilization of proaromaticity was evaluated
here within the scheme of hyperhomodesmotic reactions.
A hyperhomodesmotic reaction was originally designed to
predict/examine the thermochemistry of hydrocarbons. The
reaction is balanced under two requirements to minimize
the systematic errors from the methods of calculation/ex-
B: ꢁ4.52 (MeCN)
A: ꢁ0.19
PhNMe2H+
B: ꢁ0.07[a]
10
11
12
13
14
15
CN
NH2
CN
CN
PhNO2
CN
CN
NH2
NH2
NH3
PhNH2
PhNH2
+0.56
+2.88
ꢁ3.76
ꢁ1.13[a]
ꢁ0.97
ꢁ2.64
+
[a] Geometries of the dicationic molecules were optimized at B3LYP/6-
311+G(d,p).
ACHTUNGTRENNUNG
ꢁ
periment. It has 1) the same number of each bond type (C
ꢁ
ꢁ
ꢁ
C, C CH, C CH2, H2C CH2, C=C, HC=CH, etc.), and 2)
Isotropic NICS(0) indices of quinoid-type molecules with
various ring sizes and substituents were calculated; the
values are shown in Table 5. NICS(1) was not used here,
since we found that the number is highly contaminated by
the contribution from the magnetic anisotropy of CN
groups. Calculations were performed with Gaussian 09 at
the same number of each carbon atom types (Csp3, Csp3(H),
C
sp3(H2), Csp3(H3), Csp2, Csp2(H), Csp2(H2), etc.) in products as
in reactants.[31] The concept of constructing an error-cancel-
ing reaction to predict molecular thermochemistry has also
been applied to molecules containing heteroatoms.
The aromatic stabilization energies were inferred from
the computed reaction energies of the hyperhomodesmotic
reactions shown in Table 6; positive numbers indicate stabi-
lization favoring the reactant side. Even though unlimited
the GIAO-B3LYP/6-311+GACTHNUTRGENU(GN d,p)//B3LYP/6-31G(d) level of
theory; symmetry constraints were not imposed. Positive
NICS(0) values were found for the electronically symmetric
compounds (entries 1, 2, 10, and 11). Moving to the asym-
metric molecules (entries 3, 5, 7, 8, 12, 14, and 15), negative
NICS(0) values ranging from ꢁ0.8 to ꢁ3.8 ppm were ob-
tained. Calculated in high-dielectric acetonitrile, NICS(0) in-
dices for the asymmetric molecules were even more nega-
tive (entries 3, 5, and 8), implying the greater degree of aro-
maticity in the more charge-separated quinoid molecules.
This finding is parallel to the solvent effect on the BLA (dr)
values, and highlights the interplay between ICT interaction
and electron delocalization.
Table 6. Hyperhomodesmotic reaction energies (zero-point energy
(ZPE) corrected) as the estimates for the aromatic stabilization energies
(ASEs) of quinoid molecules. Methods are indicated, and DFT calcula-
tion was performed at B3LYP/6-31G(d).
In silico protonation was attempted to diminish the elec-
tron-donating power of the amino groups. As depicted by
entries 4, 6, 9, and 13 in Table 5, the resulting NICS(0)
values became less negative compared to the corresponding
neutral species; in particular, protonation of 4 (entry 6)
even reverses its sign of NICS(0) to positive. These varia-
tions of NICS(0) again emphasize the necessity of having
both the electron-donating and -accepting substituents to
create a diatropic magnetic environment in the quinoid-type
D[E
G
DE0ACTHNGUTERNN(UG G3MP2)
Entry
1
R1
R2
[kcalmolꢁ1
[kcalmolꢁ1
]
1.29 (vacuum)
3.71 (MeCN)
–
–
–
CN
PhNMe2
–
2
3
4
H
CN
NH2
H
CN
NH2
2.87
2.55
0.21
0.97 (vacuum)
6.67 (MeCN)
5
CN
NH2
–
9600
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9592 – 9605

Proaromaticity
FULL PAPER
number of hyperhomodesmotic reactions could in principle
be written, the chosen reaction is expected to extract only
the electronic effect of the quinoid p spacer with push–pull
substituents by comparing itself with the non-quinoid ana-
logues. Here, effects from molecular sizes, the longest conju-
gation paths, hyperconjugation, ring-strain, allylic strain, and
any other geometrical factors are preserved at both ends of
the equations.
For the large systems (Table 6, entry 1), the zero-point
corrected reaction energies in vacuum and in acetonitrile
were calculated for the optimized structures at B3LYP/6-
31G(d) level of theory. For the small ones (entries 2–5), the
highly accurate, yet economic compound method of
G3MP2[32] was used to reproduce effectively QCISD(T)/G3
large energies. For all calculations, vibration analysis was
done to establish that a stationary point on the potential
energy surface was located.
As shown in Table 6, there is a small amount of stabiliza-
tion for the prototypical molecule 4 (1.29 kcalmolꢁ1); as ex-
pected, the number becomes larger in acetonitrile (3.71 kcal
molꢁ1). By comparing the valence structures of the mole-
cules under study, the stabilization here suggests an observa-
ble stabilization of the push–pull quinoid, even though this
number is much smaller in magnitude than the generally ac-
cepted approximate 36 kcalmolꢁ1 resonance energy for ben-
zene. Since such small numbers are on the order of the
mean absolute deviation (MAD) of B3LYP/6-31G(d)
(7.9 kcalmolꢁ1 assessed on G2 molecule set),[33] model
chemistry of G3MP2 at higher level of theory was sought.
Due to the rather expensive computational resources
Conclusion
Combining experimental and theoretical approaches, the
push–pull quinoid D-Q-A systems were found to possess
highly charge-separated ground-state characters, small
HOMO–LUMO gaps, and noticeable degrees of aromaticity.
Evidence of proaromaticity has been presented herein based
on structural, spectroscopic, electrochemical, and computed
data. The effect of proaromaticity is subtle, yet readily no-
ticeable, even in the expanded quinoid macrocycle; in fact,
compound 6 represents the first member of the radiaannu-
lenes for which a certain degree of aromaticity is demon-
strated. Although evidence from IR, NMR, and absorption
spectra, electrochemistry, X-ray structures, and NICS calcu-
lations provide support and insights for proaromaticity, as
shown above, a more well-designed theoretical treatment is
clearly needed to describe the energetics of these systems.
Experiments to test the first- and second-order molecular
hyperpolarizabilities of our systems are being sought. This
work provides the impetus for further application of quinoid
p spacers in molecular wires in the context of molecular
electronics.
Experimental Section
Materials and general methods: Reagents were purchased at reagent
grade from Acros, Sigma–Aldrich, and Fluka and used as received. THF
was freshly distilled from Na/benzophenone and CH2Cl2 from CaH2
under N2 atmosphere. Flash column chromatography (FC) was carried
out with SiO2 60 (particle size 0.040–0.063 mm, 230–400 mesh; Fluka)
and technical solvents. Thin-layer chromatography (TLC) was conducted
on aluminum sheets or glass plate coated with SiO2 60 F254 obtained from
Merck; visualization with a UV lamp (254 nm). Melting points (m.p.)
were measured on a Bꢀchi B-540 melting-point apparatus in open capil-
laries and were uncorrected; decomp indicates decomposition. 1H NMR
and 13C NMR spectra were measured on a Varian Gemini 300, Varian
Mercury 300, Bruker DRX 400, Bruker AV 400, or Bruker Avance III
600 instrument at 208C. Chemical shifts were reported in ppm relative to
the signal of tetramethylsilane. Residual solvent signals in the 1H and
13C NMR spectra were used as an internal reference. Coupling constants
(J) were given in Hz. The apparent resonance multiplicity was described
as s (singlet), d (doublet), t (triplet), tt (triplet of triplet), and m (multip-
let). Infrared spectra (IR) were recorded on a Perkin–Elmer Spectrum
One FTIR Spectrometer. UV/Vis spectra were recorded on a Varian
Cary-500 spectrophotometer in a quartz cuvette (1 cm). The absorption
wavelengths are reported in nm with the molar extinction coefficient e
(mꢁ1 cmꢁ1) in parenthesis; shoulders are indicated as sh. High-resolution
HR-EI-MS spectra were measured on a Hitachi–Perkin–Elmer VG-Tri-
brid spectrometer; HR-MALDI-MS spectra were measured on a Bruker
Daltonics UltraFlex II instrument with 3-hydroxypicolinic acid (3-HPA)
as matrix. The signal of the molecular ion (M+) is reported in m/z units.
ꢀ
needed for G3MP2, small molecules with simply C N as ac-
ceptors and NH2 as donors bridged by hexagonal spacers
were considered, and the electronically neutral (Table 6,
entry 2), symmetric (entries 3–4), and asymmetric (entry 5)
cases were treated. Surprisingly, in the gas-phase calculation,
small values of stabilization resulted (0.21–2.87 kcalmolꢁ1
;
for G3MP2, MAD ꢂ1.3 kcalmolꢁ1),[32a] irrespective to the
pattern of substituents, in contrast to our inferences based
on spectroscopic data and findings from NICS calculations.
Nonetheless, upon inclusion of solvent effects, a much great-
er value for the stabilization in the D-Q-A case was found,
6.67 kcalmolꢁ1 (reaction 5), with an approximate sixfold in-
crease. The ASEs we obtained here reflect the subtlety of
“proaromatic stabilization” for the quinoid molecules, while
the effects of electronically asymmetric substitution are
more perceptible in terms of the strength of ring currents
(NICS indices). Only when applying large basis sets (includ-
ing diffuse functions) and consideration of solvent effect
were noticeable energetic effects of proaromatization found.
The difficulty in elucidating aromaticity by energetic means
has been pointed out by Prof. Schleyer: “The difficulty in
writing about aromaticity is that it is encrusted by two cen-
turies of tradition … Energetic properties are most impor-
tant, but you need to keep in mind that aromaticity is only
5% of the total energy. But if you want to get as close to
the phenomenon as possible, then one has to go to the prop-
erty most closely related, which is magnetic properties.”[34]
X-ray analysis: X-ray data collection was carried out on a Bruker Kap-
paCCD diffractometer equipped with a graphite monochromator (MoKa
radiation, l=0.71073 ꢅ) and an Oxford Cryostream low-temperature
device. Cell dimensions were obtained by least-squares refinement of all
measured reflections (HKL, Scalepack[35a]), qmax =27.58. All structures
were solved by direct methods (SIR97[35b]). All non-hydrogen atoms were
refined anisotropically, H-atoms isotropically by full matrix least-squares
with SHELXL-97[35c] using experimental weights (1/[s2(Io)+(Io +Ic)2/
900]).
X-ray crystal structure of compound 6b: Single crystals were obtained by
slow evaporation of a solution of 6b in CDCl3 at room temperature:
Chem. Eur. J. 2010, 16, 9592 – 9605
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9601

F. Diederich et al.
(C62H64N4O4) ·
G
CH2Cl2). M.p.> 2508C (decomp); 1H NMR (CDCl3, 300 MHz): d=3.31
(s, 12H), 6.78 (d, J=9 Hz, 4H), 6.87 (d, J=8.4 Hz, 2H), 7.14 (d, J=
8.4 Hz, 2H), 7.23 (d, J=8.4 Hz, 2H), 7.27 (d, J=9 Hz, 4H), 7.45 ppm (d,
J=8.4 Hz, 2H); 13C NMR (CDCl3, 75 MHz): d=40.86, 112.98, 119.65,
124.43, 125.57, 125.94, 126.51, 126.95, 135.36, 136.47, 140.26, 144.73,
146.85, 156.10, 176.72 ppm (1 signal was missing); IR (neat): n˜ =2165,
2123, 1614, 1520, 1481, 1359, 1287, 1169 cmꢁ1; UV/Vis (CH2Cl2): lmax
¯
0.015 mm, triclinic space group P1, Z=1, a=10.2879 (3), b=14.0169 (3),
c=22.1651 (8) ꢅ, a=81.7157 (8), b=78.2439 (13), g=81.6699 (14)8,V=
3074.5 (2) ꢅ3 at 173 K. Number of measured and unique reflections
17945 and 10408, respectively (Rint =0.068). Final R(F)=0.0995,
wR(F2)=0.2540 for 816 parameters and 7793 reflections with I>2s(I)
(corresponding
0.2682).
(e)=611 (105 000), 845 nm (30 500); HR-MALDI-MS: m/z (%):
+
470.2422 (31), 469.2394 (100) [M+H]+ (m/z calcd for C32H29N4
:
X-ray crystal structure of compound 19a: Single crystals were obtained
by slow diffusion of pentane into a solution of 19a in THF at room tem-
perature: C36H32N2·C4H8O, Mr =564.773, crystal dimensions 0.18ꢆ0.15ꢆ
0.03 mm, orthorhombic space group Pbcn, Z=8, a=32.7508 (6), b=
12.5342 (2), c=15.0546 (3) ꢅ, V=6180.0 (2) ꢅ3 at 123 K. Number of
469.2397).
1,1’-(2,2-Dibromoethene-1,1-diyl)bis(4-nitrobenzene) (13): Bis(4-nitro-
phenyl)methanone (300 mg, 1.1 mmol) was added to the solution of CBr4
(731 mg, 2.20 mmol) and PPh3 (1156 mg, 4.40 mmol) in anhydrous
CH2Cl2 (50 mL) in one portion. The mixture was heated at reflux for 4 h
measured and unique reflections 13032 and 7009, respectively (Rint
0.062). Final R(F)=0.0628, wR(F2)=0.1631 for 516 parameters and 4714
reflections with I>2s(I) (corresponding R values based on all 7009 re-
flections 0.1022 and 0.1922).
under Ar. The crude mixture was filtered through
a plug of SiO2
(CH2Cl2) and concentrated. Subsequent FC (SiO2; CH2Cl2/hexane 15:1)
afforded 13 as a white solid (340 mg, 72%). M.p. 195–1978C; 1H NMR
(CDCl3, 400 MHz): d=7.50 (d, J=8.5 Hz, 4H), 8.23 ppm (d, J=8.5 Hz,
4H); 13C NMR (CDCl3, 100 MHz): d=94.79, 123.97, 129.88, 143.83,
146.25, 147.50; IR (neat): n˜ =2924, 1603, 1523, 1344, 1106, 704 cmꢁ1; HR-
EI-MS: m/z (%): 429.8806 (51) [M]+, 427.8833 (100) [M]+, 425.8844 (49)
[M]+ (m/z calcd for C14H879Br2N2O4+: 425.8851), 268.0468 (36), 176.0618
(61).
X-ray crystal structure of compound 20a: Single crystals were obtained
by slow evaporation of a solution of 20a in a mixture of Et2O, CH2Cl2,
and hexane at room temperature; crystals probably lost solvent mole-
cules before measurement; cavities in the crystals are big enough for
Et2O, CH2Cl2 with some electron density: C64H48Br2N6O8, Mr =1188.930,
crystal dimensions 0.3ꢆ0.06ꢆ0.045 mm, monoclinic space group P21/c,
Z=4, a=14.3098 (5), b=19.1052 (8), c=22.9260 (10) ꢅ, b=94.449 (2)8,
V=6248.9 (4) ꢅ3 at 223 K. Number of measured and unique reflections
26658 and 8023, respectively (Rint =0.107). Final R(F)=0.1028, wR(F2)=
0.2556 for 721 parameters and 4239 reflections with I>2s(I) (corre-
sponding R values based on all 8023 reflections 0.1722 and 0.2942).
N,N-Dibutyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)aniline:
A
mixture of 4-bromo-N,N-dibutylaniline (2.416 g, 8.5 mmol) in anhydrous
THF (120 mL) was slowly treated with nBuLi (6.4 mL, 10.2 mmol, 1.6m
solution in hexane) at ꢁ788C under Ar, and the resulting solution was
stirred at ꢁ788C for 1.5 h. Triisopropyl borate (5.9 mL, 25.5 mmol) was
then added in one portion to the above solution, and the temperature
was slowly increased to room temperature. The mixture was stirred at
this temperature overnight, and subsequently sat. NH4Cl(aq) (80 mL) was
added. After 2 h of stirring, the mixture was extracted with CH2Cl2. The
organic phase was collected, dried over Na2SO4, and concentrated, and
the residue was re-dissolved in a solution of NH4Cl (0.7 g, 13 mmol) and
pinacol (1.310 g, 11.05 mmol) in anhydrous toluene (80 mL). The mixture
was heated at reflux for 9 h. After cooling, the mixture was extracted
with H2O and CH2Cl2. The organic phases were collected, dried over
Na2SO4, and the residue was purified by FC (SiO2; CH2Cl2/hexane 1:2!
1:1) to give the title product as white solid (1.872 g, 66%). M.p. ꢂ308C;
1H NMR (CDCl3, 300 MHz): d=0.94 (t, J=7.3 Hz, 6H), 1.26–1.41 (m,
4H), 1.31 (s, 12H), 1.51–1.61 (m, 4H), 3.28 (t, J=7.5 Hz, 4H), 6.60 (d,
J=8.9 Hz, 2H), 7.64 ppm (d, J=8.9 Hz, 2H); 13C NMR (CDCl3,
75 MHz): d=14.24, 20.52, 25.01, 29.54, 50.60, 53.50, 82.98, 110.43, 136.03,
149.98 ppm; IR (neat): n˜ =2959, 2932, 2870, 1601, 1399, 1351, 1142, 1094,
861, 814, 655 cmꢁ1; HR-MALDI-MS: m/z (%): 408.3509 (35), 322.2757
(69) [M+H]+ (m/z calcd for C20H35BNO2+: 332.2755), 235.0709 (100).
CCDC-773117 (6b), -773119 (19a), and -773118 (20a) contain the supple-
mentary crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Electrochemistry: Electrochemical measurements were carried out in
CH2Cl2 containing 0.1m nBu4NPF6 in a classical three-electrode cell by
cyclic voltammetry (CV) and rotating-disk voltammetry (RDV). The
working electrode was a glassy carbon disk (3 mm in diameter), the auxil-
iary electrode a Pt wire, and the pseudo reference electrode a Pt wire.
The cell was connected to an Autolab PGSTAT30 potentiostat (Eco
Chemie, Holland) driven by a GPSE software running on a personal
computer. All potentials are referenced to the ferricinium/ferrocene (Fc+
/Fc) couple used as internal standard and are uncorrected from ohmic
drop.
2-(4’-Bromobiphenyl-4-yl)malononitrile (12): NaH (95%, 0.189 g,
7.5 mmol) was added in small portions to an ice-water cooled solution of
malonitrile (0.363 g, 5.5 mmol) in anhydrous THF (30 mL); the suspen-
sion was then stirred under N2 atomosphere at room temperature. After
4,4’-{4-(Triisopropylsilyl)-2-[(triisopropylsilyl)ethynyl]but-1-en-3-yne-1,1-
30 min, 4,4’-dibromobiphenyl (1.56 g, 5 mmol) and [PdACHTUNGTRNEG(UN PPh3)2Cl2] (0.14 g,
diyl}bis(N,N-dimethylaniline) (15a):
0.606 mmol), 4-(N,N-dimethylamino)benzeneboronic acid (300 mg,
1.818 mmol), [Pd(PPh3)2Cl2] (43 mg, 0.06 mmol), and Na2CO3 (193 mg,
A
solution of 14 (331 mg,
0.2 mmol) were added and the resulting mixture was heated for 42 h at
reflux. The mixture was acidified with 1 n HCl at 08C and extracted with
CH2Cl2. The organic layer was dried over MgSO4, the solvent removed
under reduced pressure, and the residue purified by FC (SiO2, CH2Cl2/
hexane 1:1) to give the product as a white solid (0.83 g, 54%). M.p. 128–
1308C; 1H NMR (CDCl3, 300 MHz): d=5.11 (s, 1H), 7.45 (d, J=8.4 Hz,
2H), 7.58 (d, J=8.4 Hz, 2H), 7.61 (d, J=8.4 Hz, 2H), 7.68 ppm (d, J=
8.4 Hz, 2H); 13C NMR (CDCl3, 75 MHz): d=27.92, 111.52, 122.59,
125.21, 127.69, 128.36, 128.63, 132.07, 138.10, 142.18 ppm; IR (neat): n˜ =
2922, 2899, 2852, 2255, 1480, 1074, 1002, 821, 791 cmꢁ1; HR-EI-MS: m/z
AHCTUNGTRENNUNG
1.82 mmol) in THF/H2O (10 mL, 4:1) was stirred at 708C under Ar over-
night. The mixture was extracted with H2O and CH2Cl2, and the organic
layers were collected and the solvents removed under reduced pressure.
The residue was purified by collecting the yellow portions from FC
(SiO2, CH2Cl2/hexane 1:5!1:3) (360 mg, 95%). M.p. 112–1148C;
1H NMR (CDCl3, 400 MHz): d=1.10 (s, 42H), 3.30 (s, 12H), 6.65 (d, J=
8.9 Hz, 4H), 7.48 ppm (d, J=8.9 Hz, 4H); 13C NMR (CDCl3, 100 MHz):
d=11.45, 18.68, 40.35, 92.23, 97.05, 107.38, 111.01, 128.47, 131.98, 150.36,
156.94 ppm; IR (neat): n˜ =2939, 2886, 2860, 2141, 2119, 1604, 1461, 1162,
816, 672 cmꢁ1; HR-MALDI-MS: m/z (%): 628.4545 (46), 627.4524 (100)
[M+H]+ (m/z calcd for C40H63N2Si2+: 627.4524), 626.4472 (74) [M]+.
(%): 297.9923 (85) [M]+, 295.9944 (85) [M]+ (m/z calcd for C15H979BrN2
+
: 295.9944), 217.0756 (100), 190.0645 (48), 152.0616 (29).
7,7-Bis[4-(dimethylamino)phenyl]-7’,7’-dicyanodiphenoquinodimethane
(5): A mixture of 12 (0.298 g, 1 mmol) in anhydrous THF (20 mL) at
ꢁ788C under N2 was treated with nBuLi (1.38 mL, 2.2 mmol, 1.6m solu-
tion in hexane) and the resulting solution was stirred at ꢁ788C for 5 h to
ensure the complete generation of the dianionic species. Michlerꢄs ketone
(0.241 g, 0.90 mmol) was then added and the resulting mixture stirred
overnight at room temperature. The crude mixture was filtered through a
plug of SiO2 (MeCN) and concentrated. Subsequent FC (SiO2; CH2Cl2/
MeCN 3:2) afforded 5 (216 mg, 51%) as a purple metallic solid (green in
4,4’-{4-(Triisopropylsilyl)-2-[(triisopropylsilyl)ethynyl]but-1-en-3-yne-1,1-
diyl}bis(N,N-dibutylaniline) (15b):
1.917 mmol), N,N-dibutyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-
yl)aniline (1905 mg, 5.752 mmol), and Na2CO3 (620 mg, 5.752 mmol) in
THF/H2O (30 mL, 5:1) was purged with Ar for 25 min. [Pd(PPh3)4]
A
solution of 14 (1048 mg,
ACHTUNGTRENNUNG
(200 mg, 0.173 mmol) was added to the above mixture, and the solution
was purged with Ar for another 15 min. The mixture was stirred at 708C
9602
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9592 – 9605

Proaromaticity
FULL PAPER
overnight and then extracted with H2O and CH2Cl2. After removal of the
solvents, preliminary H NMR analysis indicated a mixture of starting di-
bromo compound and partially reacted monoanilino-substituted com-
pound present in the residue. Thus, the residue (ca. 500 mg) and 14
(484 mg, 0.886 mmol) were again reacted with N,N-dibutyl-4-(4,4,5,5-tet-
ramethyl-1,3,2-dioxaborolan-2-yl)aniline (1410 mg, 4.26 mmol), Na2CO3
pad of SiO2 (CH2Cl2/hexane 5:1) to give diethynylated 18a. A 1:1 mix-
ture of anhydrous Et3N/THF (15 mL) was freshly subjected to the freeze-
pump-thaw cycle 5 times before use. The inner atmosphere of the flask
containing 18a, from the previous transformation, 13 (51 mg, 0.12 mmol),
1
CuI (3 mg, 0.014 mmol), and [PdACHTNUGRTNEUNG(PPh3)4] (8 mg, 0.007 mmol) was
changed to Ar by three evacuation-refilling cycles. The oxygen-free Et3N/
THF mixture was then introduced to the above flask, and the mixture
was heated at 708C for 24 h under Ar. After cooling, the crude mixture
was passed through a pad of SiO2 (MeCN). After concentration under re-
duced pressure, the residue from the filtrate was purified by FC (SiO2,
CH2Cl2/hexane 1:1!CH2Cl2) to afford 6a as green metallic solid (35 mg,
38%). M.p.> 4008C; 1H NMR (CDCl3, 600 MHz): d=1.85–1.97 (m,
4H), 2.45 (t, J=6.2 Hz, 4H), 2.52 (t, J=7.1 Hz, 4H), 3.02 (s, 12H), 6.64
(brs, 4H), 7.46 (d, J=9.0 Hz, 4H), 7.67 (d, J=9.0 Hz, 4H), 8.20 ppm (d,
J=9.0 Hz, 4H); 13C NMR (CDCl3, 150 MHz): d=35.66, 36.49, 40.28,
87.75, 93.36, 95.26, 100.98, 108.33, 110.52, 123.15, 127.31, 127.93, 131.21,
132.81, 134.90, 145.29, 146.11, 147.33, 158.19 ppm (3 peaks were missing);
IR (neat): n˜ =2920, 2850, 2162, 1600, 1512, 1469, 1331, 1187 cmꢁ1; UV/
(450 mg, 4.26 mmol), and [PdACHTNUTRGNEUNG(PPh3)4] (200 mg, 0.173 mmol) according to
the procedure described above. The mixture was heated at 808C for 3 d,
and the product was purified according to the method described for 15a
(1827 mg, 82%). M.p. 96–988C; 1H NMR (CDCl3, 400 MHz): d=0.95 (t,
J=7.3 Hz, 12H), 1.03 (s, 42H) 1.27–1.41 (m, 8H), 1.50–1.62 (m, 8H),
3.26 (t, J=7.5 Hz, 8H), 6.49 (d, J=9.0 Hz, 4H), 7.39 ppm (d, J=9.0 Hz,
4H); 13C NMR (CDCl3, 100 MHz): d=11.48, 13.97, 18.67, 20.33, 29.47,
50.74, 91.73, 95.89, 107.81, 110.04, 127.11, 132.24, 148.06, 157.49 ppm; IR
(neat): n˜ =2938, 2861, 2131, 1600, 1519, 1189, 813, 674 cmꢁ1
; HR-
MALDI-MS: m/z (%): 796.6429 (55), 795.6398 (100) [M+H]+ (m/z calcd
for C52H87N2Si2+: 795.6402), 794.6334 (51) [M]+.
2,2’-(3-{Bis[4-(dimethylamino)phenyl]methylene}penta-1,4-diyne-1,5-
diyl)biscyclopent-1-ene-1-carbaldehyde (17a): A solution of TBAF·3H2O
(252 mg, 0.80 mmol) and the TIPS-protected ethynyl compound 15a
(100 mg, 0.16 mmol) in THF (15 mL) was stirred at room temperature
for 40 min. The mixture was extracted with H2O and CH2Cl2, and the ter-
minal acetylene compound was obtained from the organic layer. The di-
ethynyl intermediate obtained from the above operation was mixed with
Vis (CH2Cl2): lmax (e)=586 nm (38500); HR-MALDI-MS: m/z (%):
+
762.3153 (46), 761.3111 (100) [M+H]+ (m/z calcd for C50H41N4O4
7601.3122), 760.3056 (91) [M]+, 745.3193 (24).
:
4,4’-{[14-[Bis(4-nitrophenyl)methylene]-4,5,7,8,12,13,15,16-octadehydro-
2,3,9,10,11,14-hexahydrodicyclopentaACTHNUGRTNENUG[a,h]cyclotetradecen-6ACHTUNGTRENN(UGN 1H)-ylide-
ne]methylene}bis(N,N-dibutylaniline) (6b): The synthesis of 18b was per-
formed by using the method described for 18a. Lithium diisopropylamide
(0.56 mL, 1.0 mmol, 1.8m in THF/heptane/ethylbenzene), trimethylsilyl-
diazomethane (0.56 mL, 1.12 mmol, 2.0m in Et2O) THF (15+8 mL) were
used to prepare 16b (300 mg, 0.447 mmol). An oxygen-free 3:1 mixture
of anhydrous EtACHTUNRGTNEUNG(iPr)2N/THF (15 mL) was freshly prepared by freeze-
pump-thaw cycles. The inner atmosphere of the flask containing 18b
(93 mg, 0.14 mmol), 13 (60 mg, 0.14 mmol), CuI (3 mg, 0.014 mmol), and
16 (70 mg, 0.40 mmol), [PdACHTNUTRGNEUNG(PPh3)4] (20 mg, 0.016 mmol), CuI (5.7 mg,
0.030 mmol), THF (1.5 mL), and Et3N (15 mL). The solution was stirred
at room temperature overnight, while the color of the mixture quickly
turned deep red. The mixture was passed through a pad of SiO2 (MeCN).
After concentration under reduced pressure, the residue was purified by
FC (SiO2, CH2Cl2!CH2Cl2/MeCN 2:3) to afford 17a as red solid (80 mg,
99%). M.p. 178–1808C; 1H NMR (CDCl3, 300 MHz): d=1.87–2.02 (m,
4H), 2.58 (t, J=8.0 Hz, 4H), 2.70 (t, J=6.0 Hz, 4H), 3.02 (s, 12H), 6.63
(d, J=9.0 Hz, 4H), 7.39 (d, J=9.0 Hz, 4H), 9.61 ppm (s, 2H); 13C NMR
(CDCl3, 100 MHz): d=22.27, 29.48, 38.52, 40.22, 84.91, 93.40, 102.49,
110.70, 127.33, 132.60, 143.52, 147.02, 151.39, 162.80, 189.30 ppm; IR
[PdACHTNUGRTNEUNG(PPh3)4] (8 mg, 0.007 mmol) was changed to Ar by three evacuation-
refilling cycles. The solvent was then introduced to the above flask and
the mixture heated at 908C for 24 h under Ar. After cooling, the crude
mixture was passed through a pad of SiO2 (MeCN). After concentration
under reduced pressure, the residue was purified by FC (SiO2, CH2Cl2/
hexane 2:3!1:1) to afford 6b as green metallic solid (25 mg, 19%). The
formation of 19b and 20b was not observed. M.p. 261–2628C; 1H NMR
(CDCl3, 300 MHz): d=0.97 (t, J=7.3 Hz, 12H), 1.30–1.43 (m, 8H), 1.53–
1.66 (m, 8H), 1.84–1.98 (m, 4H), 2.43–2.56 (m, 8H), 3.32 (t, J=7.5 Hz,
8H), 6.55 (d, J=9.0 Hz, 4H), 7.44 (d, J=9.0 Hz, 4H), 7.67 (d, J=9.0 Hz,
4H), 8.18 ppm (d, J=9.0 Hz, 4H); 13C NMR (CDCl3, 75 MHz): d=14.25,
20.55, 23.59, 29.67, 35.77, 36.59, 50.84, 87.63, 93.46, 94.05, 95.27, 101.66,
108.25, 109.70, 122.93, 126.29, 126.59, 131.00, 132.91, 134.77, 144.64,
145.88, 146.93, 148.53, 158.59 ppm; IR (neat): n˜ =2954, 2926, 2869, 2159,
1594, 1515, 1456, 1337, 1190 cmꢁ1; HR-MALDI-MS: m/z (%): 930.5045,
929.5013 (100) [M+H]+ (m/z calcd for C62H65N4O4+: 929.5000), 928.4960
(75) [M]+, 913.5063 (14).
(neat): n˜ =2900, 2848, 2807, 2176, 2154, 1655, 1596, 1360, 822, 813 cmꢁ1
;
HR-MALDI-MS: m/z (%): 504.2726 (41), 503.2684 (100) [M+H]+ (m/z
calcd for C34H35N2O2+: 503.2693), 502.2599 (23) [M]+.
2,2’-(3-{Bis[4-(dibutylamino)phenyl]methylene}penta-1,4-diyne-1,5-diyl)-
biscyclopent-1-ene-1-carbaldehyde (17b): The synthesis is similar to that
described for 17a. The TIPS groups of 15b (1.77 g, 2.23 mmol) were re-
moved by reacting it with TBAF·3H2O (3.51 g, 11.13 mmol) in THF
(80 mL) for 1.5 h. The diethynyl intermediate obtained from extraction
was mixed with 16 (976 mg, 5.58 mmol), [PdACTHNUGTRNEG(UN PPh3)2Cl2] (160 mg,
0.23 mmol), CuI (87 mg, 0.46 mmol), THF (3 mL), and Et3N (30 mL).
After purification by FC (SiO2, CH2Cl2!CH2Cl2/MeCN 2:3), the butyl
product 17b was obtained as red solid (1 g, 67%). M.p. 125–1278C;
1H NMR (CDCl3, 400 MHz): d=0.96 (t, J=7.3 Hz, 12H), 1.30–1.44 (m,
8H), 1.54–1.66 (m, 8H), 1.93 (tt, J=7.7, 7.7 Hz, 4H), 2.58 (t, J=7.5 Hz,
4H), 2.69 (t, J=7.6 Hz, 4H), 3.31 (t, J=8.0 Hz, 8H), 6.56 (d, J=9.0 Hz,
4H), 7.38 (d, J=9.0 Hz, 4H), 9.70 ppm (s, 2H); 13C NMR (CDCl3,
100 MHz): d=13.92, 20.27, 22.22, 29.32, 29.41, 38.49, 50.72, 84.79, 91.96,
103.12, 109.93, 126.02, 132.92, 143.60, 146.50, 149.29, 163.01, 189.10 ppm;
IR (neat): n˜ =2954, 2928, 2869, 2176, 2160, 1655, 1597, 1522, 1356, 1189,
4,4’-(4,5,6,7,11,12,14,15-Octadehydro-1,2,3,8,9,10-hexahydro-13H-
dicyclopentaACTHNUTRGNE[UNG a,g]cyclotridecen-13-ylidenemethylene)bis(N,N-dimethyla-
niline) (19a): Radiaannulene 19a was occasionally isolated from the syn-
thesis of 6a; the yield is varying. M.p.>2488C (explosion); 1H NMR
(CDCl3, 400 MHz): d=2.00 (tt, J=8.0, 8.0 Hz, 4H), 2.53 (t, J=8.0H,
8H), 3.03 (s, 12H), 6.66 (d, J=8.7 Hz, 4H), 7.49 ppm (d, J=8.7 Hz, 4H);
13C NMR (CDCl3, 100 MHz): d=23.76, 33.68, 35.79, 40.24, 84.77, 88.49,
91.22, 94.64, 100.95, 110.50, 127.89, 128.16, 133.88, 140.17, 150.95,
161.01 ppm; IR (neat): n˜ =2919, 2846, 2182, 2160, 2124, 1596, 1521, 1467,
1357, 813 cmꢁ1; UV/Vis (CH2Cl2): lmax (e)=432 (9500), 531 nm (14900);
HR-MALDI-MS: m/z (%): 493.2605 (62), 492.2553 (100) [M+H]+ (m/z
calcd for C36H33N2+: 493.2638), 491.2473 (20) [M]+.
815 cmꢁ1
; HR-MALDI-MS: m/z (%): 672.4618 (54), 671.4581 (100)
[M+H]+ (m/z calcd for C46H59N2O2+: 671.4571), 670.4501 [M]+.
4,4’-{[14-[Bis(4-nitrophenyl)methylene]-4,5,7,8,12,13,15,16-octadehydro-
2,3,9,10,11,14-hexahydrodicyclopentaACTHNUGRTNENUG[a,h]cyclotetradecen-6ACHTUNGTRENN(UGN 1H)-ylide-
ne]methylene}bis(N,N-dimethylaniline) (6a): A solution of TMSC(Li)N2
was freshly prepared by reacting lithium diisopropylamide (150 mL,
0.27 mmol, 1.8m in THF/heptane/ethylbenzene) with trimethylsilyldiazo-
methane (150 mL, 0.30 mmol, 2.0m in Et2O) in anhydrous THF (5 mL) at
ꢁ788C for 30 min. A mixture of 17a (60 mg, 0.12 mmol) in anhydrous
THF (5 mL) was then introduced to the above solution. The whole mix-
ture was stirred at ꢁ788C for 1 h, followed by heating at reflux for 3 h.
After cooling, the mixture was extracted with water and CH2Cl2. The or-
ganic layers were collected and dried over Na2SO4, solvents were re-
moved under reduced pressure, and the residue was purified by a short
4,4’-[4-{2-[3-Bromo-4,4-bis(4-nitrophenyl)but-3-en-1-yn-1-yl]cyclopent-1-
en-1-yl}-2-({2-[3-bromo-4,4-bis(4-nitrophenyl)but-3-en-1-yn-1-yl]cyclo-
pent-1-en-1-yl}ethynyl)but-1-en-3-yne-1,1-diyl]bis(N,N-dimethylaniline)
(20a): Compound 20a was occasionally isolated from the synthesis of 6a;
the yield varied and about 1 mg was collected in total. 1H NMR (CDCl3,
400 MHz): d=1.91 (tt, J=7.5, 7.5 Hz, 4H), 2.51 (t, J=7.5 Hz, 4H), 2.57
(t, J=7.5 Hz, 4H), 2.99 (s, 12H), 6.64 (d, J=9.0 Hz, 4H), 7.39 (d, J=
9.0 Hz, 4H), 7.47 (d, J=9.0 Hz, 8H), 8.14 (d, J=9.0 Hz, 4H), 8.23 ppm
Chem. Eur. J. 2010, 16, 9592 – 9605
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9603

F. Diederich et al.
(d, J=9.0 Hz, 4H); 13C NMR (CDCl3, 100 MHz): d=23.25, 36.33, 37.44,
40.27, 87.42, 92.83, 96.17, 100.17, 103.60, 110.52, 123.44, 123.64, 125.57,
127.64, 130.68, 130.83, 132.88, 135.46, 143.55, 144.20, 147.28, 147.49,
150.95 ppm (3 peaks were missing); UV/Vis (CH2Cl2): lmax (e)=504 nm
(sh, 23 100); HR-MALDI-MS: m/z (%): 1191.2016 (40), 1190.1919 (56),
1189.1954 (67) [M+H]+ (m/z calcd for C64H4979Br81BrN6O8+: 1189.1958),
1188.1903 (60), 1187.1892 (48), 1186.1883 (32), 1171.1726 (20), 1111.2756
(41), 1110.2740 (72), 1109.2710 (100), 1108.2693 (74), 1107.2666 (74),
1106.2833 (42), 1093.2748 (33), 1030.3631 (45), 1029.3585 (73), 1028.3493
(85), 1027.3487 (80), 1011.3530 (43).
T. Cheng, B. G. Tiemann, A. C. Friedli, M. Blanchard-Desce, J. W.
Perry, J. Skindhoj, Science 1994, 263, 511–514.
[8] R. Gompper, H.-U. Wagner, E. Kutter, Chem. Ber. 1968, 101, 4123–
4143.
[9] a) R. Andreu, M. J. Blesa, L. Carrasquer, J. Garꢇn, J. Orduna, B. Vil-
lacampa, R. Alcalꢁ, J. Casado, M. C. R. Delgado, J. T. L. Navarrete,
M. Allain, J. Am. Chem. Soc. 2005, 127, 8835–8845; b) S. Inoue, Y.
Aso, T. Otsubo, Chem. Commun. 1997, 1105–1106; c) S. Inoue, S.
Mikami, Y. Aso, T. Otsubo, T. Wada, H. Sasabe, Synth. Met. 1997,
84, 395–396; d) T. Kawase, M. Wakabayashi, C. Takahashi, M. Oda,
Chem. Lett. 1997, 1055–1056; e) A. S. Batsanov, M. R. Bryce, M. A.
Coffin, A. Green, R. E. Hester, J. A. K. Howard, I. K. Lednev, N.
Martꢇn, A. J. Moore, J. N. Moore, E. Ortꢇ, L. Sꢁnchez, M. Savirꢈn,
P. M. Viruela, R. Viruela, T.-Q. Ye, Chem. Eur. J. 1998, 4, 2580–
2592.
[10] a) A. Abbotto, L. Beverina, S. Bradamante, A. Facchetti, C. Klein,
G. A. Pagani, M. Redi-Abshiro, R. Wortmann, Chem. Eur. J. 2003,
9, 1991–2007; b) R. Andreu, J. Garꢇn, J. Orduna, R. Alcalꢁ, B. Villa-
campa, Org. Lett. 2003, 5, 3143–3146; c) A. J. Kay, A. D. Wool-
house, Y. Zhao, K. Clays, J. Mater. Chem. 2004, 14, 1321–1330; d) S.
Alꢇas, R. Andreu, M. J. Blesa, M. A. Cerdꢁn, S. Franco, J. Garꢇn, C.
Lꢈpez, J. Orduna, J. Sanz, R. Alicante, B. Villacampa, M. Allain, J.
Org. Chem. 2008, 73, 5890–5898; e) R. Andreu, M. A. Cerdꢁn, S.
Franco, J. Garꢇn, A. B. Marco, J. Orduna, D. Palomas, B. Villacampa,
R. Alicante, M. Allain, Org. Lett. 2008, 10, 4963–4966; f) J. Schmidt,
R. Schmidt, F. Wꢀrthner, J. Org. Chem. 2008, 73, 6355–6362; g) R.
Andreu, L. Carrasquer, S. Franco, J. Garꢇn, J. Orduna, N. M. de Bar-
oja, R. Alicante, B. Villacampa, M. Allain, J. Org. Chem. 2009, 74,
6647–6657; h) R. Andreu, E. Galꢁn, J. Garꢇn, V. Herrero, E. Lacar-
ra, J. Orduna, R. Alicante, B. Villacampa, J. Org. Chem. 2010, 75,
1684–1692.
[11] For chemistry of radiaannulenes and expanded radialenes, see:
a) M. B. Nielsen, M. Schreiber, Y. G. Baek, P. Seiler, S. Lecomte, C.
Boudon, R. R. Tykwinski, J.-P. Gisselbrecht, V. Gramlich, P. J. Skin-
ner, C. Bosshard, P. Gunter, M. Gross, F. Diederich, Chem. Eur. J.
2001, 7, 3263–3280; b) F. Mitzel, C. Boudon, J.-P. Gisselbrecht, P.
Seiler, M. Gross, F. Diederich, Chem. Commun. 2003, 1634–1635;
c) F. Mitzel, C. Boudon, J.-P. Gisselbrecht, P. Seiler, M. Gross, F.
Diederich, Helv. Chim. Acta 2004, 87, 1130–1157; d) M. Gholami,
R. R. Tykwinski, Chem. Rev. 2006, 106, 4997–5027; e) F. Diederich,
M. Kivala, Adv. Mater. 2008, 20, 803–812.
Acknowledgements
We are grateful for access to the Competence Center for Computational
Chemistry (C4) Obelix cluster (ETH). Y.-L.W. appreciates being the re-
cipient of a scholarship from the Schweizerischen Chemischen Industrie
(SSCI). The research was funded by a grant from the ETH research
council and, in part, by the ERC Advanced Grant No. 246637 (“OPTE-
LOMAC”). We thank R. R. Tykwinski (Nꢀrnberg–Erlangen) and K.
Mꢀllen (Mainz) for valuable discussions.
[1] a) G. Qian, Z. Zhong, M. Luo, D. Yu, Z. Zhang, Z. Y. Wang, D. Ma,
Adv. Mater. 2009, 21, 111–116; b) S. T. Meek, E. E. Nesterov, T. M.
Swager, Org. Lett. 2008, 10, 2991–2993; c) G. Qian, B. Dai, M. Luo,
D. Yu, J. Zhan, Z. Zhang, D. Ma, Z. Y. Wang, Chem. Mater. 2008,
20, 6208–6216.
[2] a) Y. Yao, Y. Liang, V. Shrotriya, S. Xiao, L. Yu, Y. Yang, Adv.
Mater. 2007, 19, 3979–3983; b) special issue on “Low-Band-Gap
Polymer Materials for Organic Solar Cells”: Sol. Energy Mater. Sol.
Cells 2007, 91, 953–1036, edited by F. C. Krebs.
[3] a) L. Beverina, J. Fu, A. Leclercq, E. Zojer, P. Pacher, S. Barlow,
E. W. V. Stryland, D. J. Hagan, J.-L. Brꢂdas, S. R. Marder, J. Am.
Chem. Soc. 2005, 127, 7282–7283; b) S. Zheng, A. Leclercq, J. Fu, L.
Beverina, L. A. Padilha, E. Zojer, K. Schmidt, S. Barlow, J. Luo, S.-
H. Jiang, A. K.-Y. Jen, Y. Yi, Z. Shuai, E. W. V. Stryland, D. J.
Hagan, J.-L. Brꢂdas, S. R. Marder, Chem. Mater. 2007, 19, 432–442;
c) J. M. Hales, S. Zheng, S. Barlow, S. R. Marder, J. W. Perry, J. Am.
Chem. Soc. 2006, 128, 11362–11363.
[12] F. Bure, W. B. Schweizer, C. Boudon, J.-P. Gisselbrecht, M. Gross, F.
Diederich, Eur. J. Org. Chem. 2008, 994–1004.
[4] a) D. F. Perepichka, M. R. Bryce, Angew. Chem. 2005, 117, 5504;
Angew. Chem. Int. Ed. 2005, 44, 5370—5373; b) H. A. M. van Mulle-
kom, J. A. J. M. Vekemans, E. E. Havinga, E. W. Meijer, Mater. Sci.
Eng. R 2001, 32, 1–40; c) J. Roncali, Chem. Rev. 1997, 97, 173–205.
[5] a) M. Kivala, C. Boudon, J.-P. Gisselbrecht, P. Seiler, M. Gross, F.
Diederich, Chem. Commun. 2007, 4731–4733; b) P. Reutenauer, M.
Kivala, P. D. Jarowski, C. Boudon, J.-P. Gisselbrecht, M. Gross, F.
Diederich, Chem. Commun. 2007, 4898–4900; c) M. Kivala, C.
Boudon, J.-P. Gisselbrecht, B. Enko, P. Seiler, I. B. Mꢀller, N.
Langer, P. D. Jarowski, G. Gescheidt, F. Diederich, Chem. Eur. J.
2009, 15, 4111–4123; d) S.-i. Kato, M. Kivala, W. B. Schweizer, C.
Boudon, J.-P. Gisselbrecht, F. Diederich, Chem. Eur. J. 2009, 15,
8687–8691; e) C. Koos, P. Vorreau, T. Vallaitis, P. Dumon, W. Bo-
gaerts, R. Baets, B. Esembeson, I. Biaggio, T. Michinobu, F. Dieder-
ich, W. Freude, J. Leuthold, Nat. Photonics 2009, 3, 216–219; f) J.
Leuthold, W. Freude, J.-M. Brosi, R. Baets, P. Dumon, I. Biaggio,
M. L. Scimeca, F. Diederich, B. Frank, C. Koos, Proc. IEEE 2009,
97, 1304–1316; g) T. Vallaitis, S. Bogatscher, L. Alloatti, P. Dumon,
R. Baets, M. L. Scimeca, I. Biaggio, F. Diederich, C. Koos, W.
Freude, J. Leuthold, Opt. Express 2009, 17, 17357–17368.
[13] M. Uno, K. Seto, M. Masuda, W. Ueda, S. Takahashi, Tetrahedron
Lett. 1985, 26, 1553–1556.
[14] For p-conjugated compounds incorporating 1,2-diethynyl-cyclopent-
1-ene moieties, see: a) A. J. Boydston, M. Laskoski, U. H. F. Bunz,
M. M. Haley, Synlett 2002, 981–983; b) C. Kosinski, A. Hirsch, F. W.
Heinemann, F. Hampel, Eur. J. Org. Chem. 2001, 3879–3890.
[15] B. Halton, Q. Lu, P. J. Stang, Aust. J. Chem. 1990, 43, 1277–1282.
[16] A. M. Boldi, J. Anthony, C. B. Knobler, F. Diederich, Angew. Chem.
1992, 104, 1270–1273; Angew. Chem. Int. Ed. Engl. 1992, 31, 1240–
1242.
[17] A. Balsamo, I. Coletta, P. Domiano, A. Guglielmotti, C. Landolfi, F.
Mancini, C. Milanese, E. Orlandini, S. Rapposelli, M. Pinza, B. Mac-
chia, Eur. J. Med. Chem. 2002, 37, 391–398.
[18] K. Miwa, T. Aoyama, T. Shioiri, Synlett 1994, 107–108.
[19] M. Ballester, J. Riera, Spectrochim. Acta Part A 1967, 23, 1533–
1539.
[20] Transition moments were calculated at the TD-PBE1PBE/6-311+
GACTHNUTRGNEUNG(d,p)//B3LYP/6–31G(d) level of theory. However, TD-DFT calcu-
lations using various functionals all erroneously predicted the optical
transition energies for most of the chromophores under study.
[21] a) C. Reichardt, Solvents and Solvent Effects in Organic Chemistry,
3rd ed., Wiley-VCH, Weinheim, 2003; b) J. Catalꢁn, H. Hopf, Eur. J.
Org. Chem. 2004, 4694–4702.
[6] a) F. Wudl, N. Kobayashi, A. J. Heeger, J. Org. Chem. 1984, 49,
3382–3384; b) J.-L. Brꢂdas, A. J. Heeger, F. Wudl, J. Chem. Phys.
1986, 85, 4673–4678; c) H. Reisch, U. Wiesler, U. Scherf, N. Tuytuyl-
kov, Macromolecules 1996, 29, 8204–8210.
[22] F. Ammar, J. M. Savꢂant, J. Electroanal. Chem. Interfacial Electro-
chem. 1973, 47, 115–125.
[7] a) S. R. Marder, J. W. Perry, G. Bourhill, C. B. Gorman, B. G. Tie-
mann, K. Mansour, Science 1993, 261, 186–189; b) S. R. Marder, L.-
9604
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9592 – 9605

Proaromaticity
FULL PAPER
[23] Gaussian 09, Revision A.1, 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. Son-
nenberg, M. Hada, M. Ehara, K. 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, ꢉ.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian,
Inc., Wallingford CT, 2009.
[24] C. Dehu, F. Meyers, J.-L. Brꢂdas, J. Am. Chem. Soc. 1993, 115,
6198–6206.
[25] a) M. Gholami, M. N. Chaur, M. Wilde, M. J. Ferguson, R. McDo-
nald, L. Echegoyen, R. R. Tykwinski, Chem. Commun. 2009, 3038–
3040; b) G. Chen, L. Wang, D. W. Thompson, Y. Zhao, Org. Lett.
2008, 10, 657–660.
[26] a) A. Julg, Ph. FranÅois, Theor. Chim. Acta 1967, 8, 249–259;
b) C. W. Bird, Tetrahedron 1985, 41, 1409–1414.
[27] H. S. Blanchette, S. C. Brand, H. Naruse, T. J. R. Weakley, M. M.
Haley, Tetrahedron 2000, 56, 9581–9588.
[29] a) H. Fallah-Bagher-Shaidaei, C. S. Wannere, C. Corminboeuf, R.
Puchta, P. von R. Schleyer, Org. Lett. 2006, 8, 863–866; b) P. von R.
Schleyer, M. Manoharan, Z.-X. Wang, B. Kiran, H. Jiao, R. Puchta,
N. J. R. v. E. Hommes, Org. Lett. 2001, 3, 2465–2468; c) P. von R.
Schleyer, C. Maerker, A. Dransfeld, H. Jiao, N. J. R. v. E. Hommes,
J. Am. Chem. Soc. 1996, 118, 6317–6318; d) Z. Chen, C. S. Wannere,
C. Corminboeuf, R. Puchta, P. von R. Schleyer, Chem. Rev. 2005,
105, 3842–3888.
[30] S. M. Brombosz, A. J. Zucchero, P. L. McGrier, U. H. F. Bunz, J.
Org. Chem. 2009, 74, 8909–8913.
[31] a) S. E. Wheeler, K. N. Houk, P. von R. Schleyer, W. D. Allen, J.
Am. Chem. Soc. 2009, 131, 2547–2560; b) B. A. Hess, Jr., L. J.
Schaad, J. Am. Chem. Soc. 1983, 105, 7500–7505.
[32] a) L. A. Curtiss, P. C. Redfern, K. Raghavachari, V. Rassolov, J. A.
Pople, J. Chem. Phys. 1999, 110, 4703–4709; b) D. Bond, J. Org.
Chem. 2007, 72, 7313–7328.
[33] a) J. B. Foresman, E. Frisch, Exploring Chemistry with Electronic
Structure Methods, 2nd ed., Gaussian, Pittsburgh, PA, 1996, pp. 147–
148 and J. B. Foresman, E. Frisch, Exploring Chemistry with Elec-
tronic Structure Methods, 2nd ed., Gaussian, Pittsburgh, PA, 1996,
pp. 157–158; b) D. C. Young, Computational Chemistry, Wiley-Inter-
science, New York, 2001, pp. 137–141.
[34] S. M. Bachrach, Computational Organic Chemistry, Wiley-Inter-
science, New York, 2007, pp. 103–106.
[35] a) Z. Otwinowski, W. Minor, Meth. Enzymol. 1997, 276, 307–326;
b) A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Gia-
covazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna,
J. Appl. Crystallogr. 1999, 32, 115; c) SHELXL97, Program for the
Refinement of Crystal Structures, G. M. Sheldrick, University of
Gçttingen (Germany), 1997.
[28] After heating 19a in cyclohexa-1,4-diene at reflux for 1 week, a tiny
mass peak at m/z=496.2849 (relative intensity 7.38%, [M+4]+) was
observed by analyzing the crude mixture with high resolution
MALDI-MS. This peak was not found by the same analysis of 19a
(see the Supporting Information).
Received: April 21, 2010
Published online: July 20, 2010
Chem. Eur. J. 2010, 16, 9592 – 9605
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9605