Highly fluorescent, π-extended indenopyrido[2,1-a]isoindolone derivatives prepared by a palladium-catalysed cascade reaction

Article abstract of DOI:10.1002/ejoc.201201575

A new family of heterocyclic pentacyclic compounds have been prepared by a cascade reaction involving 2,5-dihalopyridines and (2-formylphenyl)boronic acids. Most of the compounds exhibit high quantum yields of fluorescence in dichloromethane. In some cases, small changes in the substitution pattern caused fluorescence quenching. To rationalize this effect, a detailed photophysical study combined with electrochemical and computational studies was performed on four representative derivatives. It appears that the fluorescence quenching is caused by a thermally activated non-radiative deactivation process that can be prevented in a rigid matrix such as ethanol at 77 K or a PMMA polymer.

Full text of DOI:10.1002/ejoc.201201575

FULL PAPER
DOI: 10.1002/ejoc.201201575
Highly Fluorescent, π-Extended Indenopyrido[2,1-a]isoindolone Derivatives
Prepared by a Palladium-Catalysed Cascade Reaction
Zein el abidine Chamas,[a] Enrico Marchi,[b] Alberto Modelli,[b] Yves Fort,[a]
Paola Ceroni,*[b] and Victor Mamane*[a]
Keywords: Cascade reactions / Polycycles / Nitrogen heterocycles / Chromophores / Fluorescence
A new family of heterocyclic pentacyclic compounds have
been prepared by a cascade reaction involving 2,5-dihalo-
pyridines and (2-formylphenyl)boronic acids. Most of the
compounds exhibit high quantum yields of fluorescence in
dichloromethane. In some cases, small changes in the substi-
tution pattern caused fluorescence quenching. To rationalize
this effect, a detailed photophysical study combined with
electrochemical and computational studies was performed on
four representative derivatives. It appears that the fluores-
cence quenching is caused by a thermally activated non-ra-
diative deactivation process that can be prevented in a rigid
matrix such as ethanol at 77 K or a PMMA polymer.
Introduction
During the last few years, considerable attention has
been devoted to the development of new organic fluoro-
phores, as can be seen from the rising number of applica-
tions in materials science[1] and biology.[2] Of these fluoro-
phores, nitrogen-containing, π-extended polycyclic systems
based on the isoindole[3] and indolizine[4] frameworks are
expected to exhibit excellent emitting properties. In particu-
lar, indolizino[3,4,5-ab]isoindoles prepared from pyrido[2,1-
Figure 1. Fluorescent molecules with a pyridoisoindole subunit.
We report herein a large series of fluorescent pentacyclic
b]isoindoles have been found to be excellent fluorophores compounds obtained by varying the substituents on the
with high quantum yields and tuneable colours (Figure 1).[5] starting pyridine (4- and/or 6-positions) and/or the boronic
Recently, other similar indolizine-based polyheterocyclic acid (meta or para positions with respect to the boronic acid
fluorescent dyes have been described that exhibit excellent function, Figure 2). Because two boronic acids are involved
photophysical properties.[6] We recently reported the syn- in the reaction, sequential coupling of two different boronic
thesis of indenopyrido[2,1-a]isoindolone derivatives, which acids allowed us to prepare the so-called “unsymmetrical”
constitute a new class of π-extended isoindolones pos- derivatives (different substitution of the A and C rings) in
sessing the same pyridoisoindole subunit[7] as in indoliz- contrast to the “symmetrical” ones (same substitution of
ino[3,4,5-ab]isoindoles (Figure 1).[8] These pentacyclic com- the A and C rings) obtained with two identical boronic ac-
pounds were prepared by a palladium-catalysed cascade re- ids. These pentacyclic compounds exhibit high fluorescence
action starting from simple monocyclic 2,5-dihalopyridines quantum yields in several cases, with the substitution
and (2-formylphenyl)boronic acids.
pattern having a significant effect. The photophysical inves-
[a] Laboratoire SRSMC (UMR CNRS 7565), Université de
Lorraine,
BP 70239, Bd. des Aiguillettes, 54506 Vandoeuvre les Nancy,
France
Fax: +33-3-83684785
E-mail: victor.mamane@univ-lorraine.fr
Homepage: http://www.srsmc.uhp-nancy.fr/spip/
[b] Department of Chemistry “G. Ciamician”, University of
Bologna,
Via Selmi 2, 40126 Bologna, Italy
E-mail: paola.ceroni@unibo.it
Homepage: http://www.photonanolab.com/
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201201575.
Figure 2. Functional group modification of the pentacyclic deriva-
tives.
2316
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 2316–2324

Highly Fluorescent π-Extended Heterocycles
tigation was accompanied by electrochemical as well as of the boronic acid. First, the effect of substituents on the
computational studies.
pyridine was evaluated by treating pyridine 3 as well as all
the 4- and/or 6-substituted 5-bromo-2-chloropyridines 1
and 4 with 2.5 equiv. of boronic acid 6a. In all cases, the
pentacyclic framework was formed and compounds 10 were
isolated in moderate-to-good yields (Table 1).
Results and Discussion
Synthesis of the Substituted Pyridines and Boronic Acids
5-Bromo-2-chloropyridines 1 bearing one or two methyl
groups and trihalopyridine 2 were prepared according to
literature procedures. The 4-substituted derivatives 1a and
2[9] were obtained by an LDA-mediated lithiation of 5-
bromo-2-chloropyridine (3) at –78 °C[10] followed by elec-
trophilic trapping whereas the 6-substituted compounds 1b
and 1c[8] were prepared from the corresponding 2-amino-
pyridines by a diazotization/substitution sequence. The tri-
halopyridine 2 was then used to prepare several 4-arylpyrid-
ines 4 by selective cross-coupling at C-4 bearing the iodine
atom (Scheme 1).
Table 1. Synthesis of “symmetrical” pentacyclic compounds by
modification of the pyridine substrates.
Scheme 1. Pyridine substrates for the cascade reaction.
Boronic acids 6 with a substituent para to the B(OH)2
group were prepared from the corresponding bromo deriva-
tives 5 by known procedures.[11] Boronic acids 8 containing
a substituent meta to the B(OH)2 group were prepared from
the corresponding benzaldehydes 7 by using the Comins
ortho-lithiation procedure[12] and trapping with B(OiPr)3.
Despite the low yields obtained, this method allowed direct
entry to rare and expensive (2-formylphenyl)boronic acids
from cheap benzaldehydes (Scheme 2).
Differently substituted boronic acids 6 and 8 were also
tested in the cascade process (Table 2). In these cases, the
use of 2-bromo-5-iodopyridine (9) instead of 3 led to higher
yields. The substituent on the boronic acid has a large in-
fluence on the yield of the reaction independently of its
electronic nature. Thus, moderate-to-good yields of com-
pounds 11a, 11b and 11d were obtained with boronic acids
Scheme 2. Boronic acid substrates used in the cascade reaction.
Cascade Reaction
The “symmetrical” compounds were obtained by the re- 6b, 6c and 8a, respectively, whereas compound 11c was iso-
action of 1 equiv. of the pyridine substrate with 2.5–3 equiv. lated in a poor yield of 12%.
Eur. J. Org. Chem. 2013, 2316–2324
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2317

P. Ceroni, V. Mamane et al.
FULL PAPER
Table 2. Synthesis of “symmetrical” pentacyclic compounds by coupling of pyridylboronic acid 14 and bromobenzaldehyde
modification of the boronic acid substrates.
15 in an excellent yield of 96% by using the combination
[Pd2(dba)3]/PCy3 as catalyst[13] (Scheme 3).
Table 3. Synthesis of the “unsymmetrical” pentacyclic derivatives
16.
The “unsymmetrical” compounds were prepared by
using a two-step procedure involving the isolation of the
products after the first cross-coupling followed by the cas-
cade reaction initiated by the second cross-coupling reac-
tion. Several pyridylbenzaldehyde derivatives 12 and 13
bearing either a chlorine or bromine at the 2-position of the
pyridine ring were prepared by coupling pyridines 3 and 9
with boronic acids 6a,b and 8a,b in the presence of
[Pd(PPh3)4] as catalyst. For 13c, to increase the yield (rela-
tive to boronic acid 8b), pyridine 9 was used in two-fold
excess. These conditions were unsuccessful with boronic
acid 6d and therefore compound 12c was synthesized by the
Scheme 3. Preparation of the pyridylbenzaldehyde substrates.
2318
[a] 1 equiv. of NaBr was added.
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 2316–2324

Highly Fluorescent π-Extended Heterocycles
Derivatives 12 and 13 were then used in the cascade reac-
tion with different boronic acids to obtain “unsymmetrical”
compounds 16 (Table 3). First, 12a and 13a were treated
with boronic acid 6a to form 10a in yields of 32 and 54%,
respectively (entries 1 and 2). The yield of 10a starting from
12a could be increased to 60% by the addition of 1 equiv.
of NaBr to the reaction mixture (entry 3). NaBr addition
was also beneficial for the synthesis of 16a from 12b (com-
pare entries 4 and 5). Therefore it was used in the cascade
reactions involving substrate 12c and boronic acids 6a or
6b to furnish 16b and 16c, respectively, in good yields (en-
tries 6 and 7). Compounds 13a,b reacted smoothly with
boronic acids 6b and 6d under the standard conditions to
give 16d–f in good yields (entries 8–10), however, substrate
16g was not isolated from the cascade reaction of 13c with
6a (entry 11).
We also explored the possibility of performing in a one-
pot procedure the successive coupling of two differently
substituted boronic acids without the isolation of the pyr-
idylbenzaldehyde intermediates. Good yields of 16a,b were
obtained when boronic acid 14 was cross-coupled success-
ively with bromobenzaldehydes 15 or 17 and then with 6a
by using the [Pd2(dba)3]/PCy3 catalytic system (Scheme 4).
The use of boronic acid 14 has two advantages during the
first step: It allows the formation of the intermediates in
high yields and avoids the generation of over-coupling
products enabling an easy isolation of the final cascade
products 16a,b.
Figure 3. a) Absorption and b) emission spectra of compound 10a
in air-equilibrated solutions of acetonitrile (solid lines) or dichloro-
methane (dashed lines) at 298 K (λex = 420 nm). The emission spec-
tra were recorded in optically matched solutions at the excitation
wavelength. For comparison purposes, the emission spectrum re-
corded in a rigid ethanol matrix at 77 K is also presented (dotted
line).
vent polarity, a solution of 10a in air-equilibrated dichloro-
methane was investigated (dashed lines in Figure 3). A
small redshift of both the absorption and emission maxima
is observed as well as substantial increases in the emission
quantum yield (Φem = 0.62) and lifetime (τ = 4.3 ns). Based
on these results, we decided to measure the absorption and
Scheme 4. “One-pot” coupling of three different partners.
Photophysical Properties
The absorption and emission spectra of the prototype emission spectra as well as the emission quantum yields of
compound 10a in acetonitrile are presented in Figure 3 (so- the other pentacyclic compounds in air-equilibrated di-
lid lines). The absorption spectrum presents a vibrationally chloromethane solution, in which compound 10a shows the
structured band with a maximum at 417 nm and a second best performances (Table 4).
peak in the UV region (250–270 nm).
From the results reported in Table 4, the following con-
Compound 10a is strongly luminescent (Φem = 0.33) with clusions can be drawn. 1) Compounds 10a–c show very sim-
an emission band maximum at 494 nm in air-equilibrated ilar photophysical properties and so it can be deduced that
acetonitrile. This can be attributed to fluorescence deactiva- methyl substituents do not affect the optical features, as ex-
tion from the lowest singlet excited state (S1) to the ground pected. 2) Appending a phenyl group to the B ring (10e)
state (S0) based on the short-lived emission intensity decay lowers the emission quantum yield. 3) Electron-donating
(τ = 2.2 ns), which is not sensitive to the presence of di- substituents on both the lateral A and C rings in the para
oxygen. In an ethanol rigid matrix at 77 K, the emission position (compounds 11a, 11c and 16c) cause a strong de-
spectrum shows a better defined vibrational structure (dot- crease in the emission quantum yield, but this effect is not
ted line in Figure 3) with a maximum at 463 nm and a present by appending an electron-withdrawing substituent
longer lifetime (τ = 5.5 ns). To investigate the effect of sol- such as fluorine (11b). 4) If the electron-donating substitu-
Eur. J. Org. Chem. 2013, 2316–2324
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2319

P. Ceroni, V. Mamane et al.
FULL PAPER
Table 4. Photophysical data for the pentacyclic compounds in air- excited-state lifetimes. The emission quenching of 11a and
equilibrated dichloromethane at 298 K.
16a demonstrates that for these compounds there is an ad-
ditional non-radiative decay competing with the radiative
decay, which is prevented at 77 K. These results can be ra-
tionalized on the basis of the presence of a non-emitting
charge-transfer (CT) state. Population of this state is pos-
sible at 298 K, whereas it is prevented at 77 K because the
lack of solvent reorganization shifts the corresponding CT
state to a higher energy. However, this hypothesis is not
consistent with the electrochemical data reported below.
For example, the energies of the CT state of both 10a and
16a, calculated on the basis of the first reduction and oxi-
dation potentials (see Table 6), are very close to that of the
S1 excited state. Therefore no substantial difference between
10a and 16a is expected.
Based on the strong difference in the emission properties
at 298 and 77 K, we decided to carefully investigate the
emission as a function of temperature to determine whether
the quenching mechanism is a thermally activated deactiva-
tion pathway. The lifetime of the luminescent excited state
of 11a in ethanol was recorded as a function of temperature
in the range from 300 to 85 K (Figure 4). The S1 lifetime
Absorption
λmax [nm]
Emission
λmax [nm]
Φem
10a
10b
10c
10d
10e
10f
10g
10h
10i
11a
11b
11c
11d
16a
16b
16c
16d
16e
16f
423
424
421
424
425
425
426
429
427
430
420
468
424
429
470
465
422
438
440
497
505
500
511
510
510
515
520
510
518
498
597
504
516
620
609
498
518
520
0.62
0.71
0.67
0.71
0.20
0.01
0.09
0.02
0.05
0.11
0.63
0.08
0.64
0.04
0.03
0.03
0.78
0.54
0.56
ent is in the para position but only on the C ring, the com- and the fluorescence quantum yield are proportional, Φ =
pounds are poorly luminescent (16a and 16b), whereas they krτ, with kr being the radiative rate constant, which is usu-
show strong luminescence if the same substituent is on the ally slightly sensitive to temperature. The lifetime is ex-
A ring (16d and 16e). 5) Electron-donating substituents at tremely short (ca. 0.5 ns) in the range 200–300 K and in-
the meta position of the lateral A and C rings do not affect creases significantly close to the fluid-to-solid transition,
the photophysical properties (11d and 16f).
reaching around 4.8 ns at 85 K. The emission maximum un-
To rationalize the effects of the presence of electron-do- dergoes a blueshift from 513 to 480 nm on going from 300
nating substituents, we carried out a detailed photophysical to 85 K. The abrupt increase in lifetime and concomitant
investigation on compounds 10a, 11a, 16a and 16d in air- shift of the emission maximum suggest that the rigidity of
equilibrated solutions of acetonitrile and dichloromethane the matrix influences the photophysical properties. These
at 298 K as well as in an ethanol rigid matrix at 77 K results are consistent with a rigidochromic effect: in a rigid
(Table 5). The electrochemical properties of these com- matrix the solvent cannot reorganize around the excited
pounds were also studied in acetonitrile (see below).
state (which exhibits a different electronic distribution com-
Compound 16d shows photophysical properties very sim- pared with the ground state) and creates a cage-like struc-
ilar to those reported for 10a under all the investigated ex- ture in which some non-radiative deactivations are
perimental conditions, which confirms that the presence of slowed.[14]
a methoxy substituent on the A ring does not affect the
Because the rigidity of the matrix plays a crucial role in
photophysical properties. Compound 11a and 16a exhibit a determining the luminescence of compound 11a, we de-
redshifted maximum of both the lowest-energy absorption cided to investigate the luminescence properties of 11a in a
band and fluorescence band. At 298 K in solution, their polymeric film at 298 K. The film was obtained by spin-
emission quantum yields and the S1 lifetimes are consider- coating a dichloromethane solution containing poly(methyl
ably reduced, particularly in acetonitrile solution, whereas methacrylate) (PMMA) and 11a (50:1, w/w). The film is
in the ethanol rigid matrix at 77 K, they have much longer highly emitting and exhibits an emission lifetime of more
Table 5. Emission data for compounds 10a, 11a, 16a and 16d in air-equilibrated dichloromethane and acetonitrile (in parentheses) at
298 K and in an ethanol rigid matrix at 77 K.
Absorption at 298 K
λmax [nm]
ε [104 m–1 cm–1]
Emission at 298 K
Emission at 77 K
λmax [nm]
Φem
τ [ns]
λmax [nm]
τ [ns]
10a
11a
16a
16d
423
(417)
430
(426)
429
(425)
422
(419)
1.8
(1.8)
1.8
(1.8)
2.0
(2.0)
1.7
(1.7)
497
(494)
518
(513)
516
(519)
498
(495)
0.62
(0.33)
0.11
(0.025)
0.04
(0.009)
0.78
(0.66)
4.3
(2.2)
0.7
(0.4)
Ͻ 0.4
(Ͻ0.4)
5.1
463
480
484
464
5.5
4.8
4.5
5.8
(4.7)
2320
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 2316–2324

Highly Fluorescent π-Extended Heterocycles
Figure 4. Temperature dependence of the emission lifetime of 11a
in air-equilibrated ethanol. Tm represents the melting point of the
solvent.
Figure 5. Cyclic voltammogram of 16a in acetonitrile/TEAPF6
solution at 298 K (scan rate: 1 V/s).
than 2 ns, which confirms that rigidity is an important pa-
rameter. This result is very important in view of future ap-
plications in solid-state devices.
Electrochemical Properties
Scheme 5.
Compounds 10a, 11a, 16a and 16d were characterized
electrochemically in acetonitrile with tetraethylammonium
hexafluorophosphate (TEAPF6) as supporting electrolyte
(Table 6). The compounds were chosen to study the effect
of the methoxy substituents on the pentacyclic moiety.
observed trend can again be rationalized on the basis of the
effect of the methoxy substituent. The electrochemical data
show slight changes in the reduction potentials of the inves-
tigated compounds, in agreement with the photophysical
properties (Table 4). The energy of the luminescent excited
state is slightly affected by the substituents.
Table 6. Half-wave potentials (E1/2 vs. SCE), unless otherwise
noted, in acetonitrile/TEAPF6 solution at 298 K.
Eox [V]
E(1)red [V]
E(2)red [V]
10a
11a
16a
16d
+1.06[a]
+0.93[a]
+0.97[a]
+1.03[a]
–1.52
–1.61
–1.56
–1.58
–1.85[a]
–1.94[a]
–1.91[a]
–1.90[a]
Computational Studies
The data reported above show that in 16a and 16d, which
differ in the position of the methoxy substituent, the ener-
gies of the luminescent excited state and the reduction po-
[a] Chemically irreversible process. Peak potential at 1 V/s.
All the investigated compounds show a chemically tentials are only slightly different, whereas both the emis-
irreversible oxidation process (see, for example, peak A for sion quantum yield and lifetime of the fluorescent excited
16a in Figure 5). The product generated after oxidation (P) state decrease strongly on going from 16d to 16a. These
is reduced at around –0.09 V versus SCE (peak B, Figure 5), changes are likely due not to different electronic distribu-
and this reduction is followed by a chemical reaction that tions of the π molecular orbitals (MOs) involved in the
is not reversible (Scheme 5).
Complete chemical reversibility was not attained by in- vation process, more favoured in 16a than in 16d.
creasing the scan rate up to 10 V/s at 298 and 233 K. The To support this interpretation, density functional theory
experiment at 233 K was performed to reduce the rate of calculations were carried out on the neutral states of 10a,
reaction (2) in Scheme 5.
16a, 16d and 11a by using the B3LYP functional[15] and
transition, but to a thermally activated non-radiative deacti-
In the cathodic region, a reversible process (peak I, Fig- the standard 6-31G(d) basis set to compare the (gas-phase)
ure 5) is observed followed by one chemically irreversible energies and localization properties of the highest- and low-
process. The E1/2 value of the first reduction process de- est-unoccupied MOs (HOMO and LUMO, respectively).
creases in the order 10aϾ16aϾ16dϾ11a, as expected by Moreover, the energies of their singlet excited states were
the presence of electron-donating methoxy substituents. An calculated by using the TD-B3LYP[16] method. The adia-
opposite trend is observed for the anodic peak potential of batic electron affinities (EAas) were evaluated as the total
the oxidation process: if we assume that the rate of reac- energy difference between the neutral and anion ground
tion (2) is the same for all the investigated compounds, the states, each in its optimized geometry. Owing to the spa-
Eur. J. Org. Chem. 2013, 2316–2324
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2321

P. Ceroni, V. Mamane et al.
FULL PAPER
tially diffuse electron distributions of the anions,[17] the Table 8. Representation of the HOMOs and LUMOs of com-
pounds 16a and 16d, as deduced by B3LYP/6-31G(d) calculations.
EAas were obtained by using the 6-31+G(d) basis set
(which also includes s- and p-type diffuse functions at the
non-hydrogen atoms). This computational method repro-
duced very accurately the (positive) EAas measured in poly-
aromatic hydrocarbons.[18] The results are reported in
Table 7, together with the experimental absorption energies
and reduction potentials for the sake of comparison.
Table 7. Calculated orbital energies, energies of the first singlet π–
π* transition and the adiabatic electron affinities. The experimental
excitation energies and reduction potentials are also reported.
B3LYP/6-31G(d)
Exp.
Abs.
[nm]
[a]
HOMO LUMO
Abs.
[nm]
EAa
E1/2
[eV]
[eV]
[eV]
[eV]
Conclusions
10a
11a
16a
16d
–5.290
–2.165
434.1
446.6
446.4
433.7
1.333
(1.436)
1.161
(1.269)
1.226
(1.330)
1.265
(1.366)
423
–1.52
–1.61
–1.56
–1.58
New highly fluorescent pentacyclic derivatives containing
the pyridoisoindole subunit have been obtained in one or
two steps from simple starting materials. The cross-coupling
of one dihalopyridine and two (identical) boronic acids de-
livers the pentacyclic framework in generally good chemical
yields. As a result of the flexibility of the reaction, different
substituents were introduced at various positions of the
pentacycles, thus allowing the preparation of a large
number of fluorophores. Their photophysical properties
demonstrate that the presence of electron-donating or
-withdrawing substituents does not cause substantial
changes in the absorption and emission maxima, but
strongly influences the fluorescence quantum yield and the
lifetime of the luminescent excited state. This result has
been rationalized on the basis of a thermally activated non-
radiative deactivation process. Indeed, a detailed investiga-
tion of a selected family of compounds has shown that the
quenching mechanism active at room temperature in fluid
solution is prevented at 77 K. Computational studies con-
firmed that the electronic distributions of the π molecular
orbitals involved in the transition are very similar for the
investigated compounds. Indeed, cyclic voltammetry shows
a reversible reduction process and a chemically irreversible
oxidation process at very similar potentials for the different
pentacyclic compounds.
–4.956
–5.067
–5.164
–1.936
–2.039
–2.058
430
429
422
[a] EAas calculated with the 6-31+G(d) basis set. Values in paren-
theses include zero-point vibrational energy corrections.
According to the calculations, the O···H hydrogen-bond
distances [10a: 1.958 Å; 16a: 1.942 Å; 16d: 1.954 Å;
11a:1.943 Å; data obtained with the B3LYP/6-31+G(d)
method] undergo small changes upon methoxy substitution
and are only slightly shorter than that (2.00 Å) measured in
10b in the solid state.[8] The calculated MO energies show
that each methoxy substitution causes a small destabiliza-
tion of the HOMO and an almost equal destabilization of
the LUMO. In agreement, the TD-B3LYP calculations pre-
dict quite similar HOMO–LUMO transition energies in the
four compounds, the largest difference being Ͻ0.1 eV. The
calculated absorption energies closely parallel the experi-
mental counterparts measured in acetonitrile solution (see
Table 6). The energies (not reported) of the second absorp-
tion band observed experimentally are also satisfactorily re-
produced by the calculations. The HOMO and LUMO lo-
calization properties are predicted to be similar in the four
compounds considered, in line with their relatively small
localization on the methoxy groups. In particular, the repre-
sentations determined by the B3LYP/6-31G(d) calculations
for the HOMOs and LUMOs of 16a and 16d are given in
Table 8.
Experimental Section
General: All reactions were performed under argon in oven-dried
glassware. THF and toluene were distilled from sodium/benzo-
phenone and stored over sodium. All other solvents and reagents
The calculated EAas are largely positive (Table 7) and were used as received. TLC was performed on silica gel plates and
close to that of pentacene,[19] thus confirming the expecta-
visualized with a UV lamp (254 nm). Chromatography was per-
formed on silica gel (70–230 mesh). Melting points were measured
tion that these polycyclic conjugated π systems are strong
electron acceptors. The unsubstituted compound (10a) is
1
with a Totoli apparatus. H and 13C NMR spectra were recorded
with Bruker AC-200 or AC-250 Fourier-transform spectrometers
predicted to possess the largest EAa, but that of the disub-
by using an internal deuterium lock. Chemical shifts are quoted in
stituted 11a is only 0.2 eV smaller, the EAas of 16a and 16d
parts per million (ppm) downfield of tetramethylsilane. Coupling
being intermediate and very close to each other. Note that
the trend in EA values closely parallels that of the calcu-
constants (J) are quoted in Hz. Electronic impact mass spectra (EI-
MS) were recorded with a Shimadzu QP 2010 apparatus. High-
lated LUMO energies of the neutral molecules and that of
the reduction potentials measured in solution (see Table 7),
although their variation along the series is even smaller.
resolution mass spectra were recorded with a Bruker micrOTOFQ
spectrometer. Photophysical experiments were carried out in air-
equilibrated acetonitrile or dichloromethane at 298 K or in an eth-
2322
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 2316–2324

Highly Fluorescent π-Extended Heterocycles
anol rigid matrix at 77 K. UV/Vis spectra were recorded with a
Perkin–Elmer λ650 spectrophotometer. Fluorescence spectra were
recorded with a Varian Cary Eclipse spectrofluorimeter equipped
with a Hamamatsu R928 phototube. Fluorescence quantum yields
were measured following the method of Demas and Crosby[20]
(standard used: perylene in ethanol solution, Φ = 0.92).[21] Fluores-
cence lifetime measurements were performed with an Edinburgh
FLS920 spectrofluorimeter equipped with a TCC900 card for data
acquisition in time-correlated single-photon counting experiments
(0.4 ns time resolution) with a LDH-P-C-405 pulsed diode laser.
The estimated experimental errors are 2 nm for the band maxi-
mum, 5% for the molar absorption coefficient and the fluorescence
lifetime and 15% for the fluorescence quantum yield. The electro-
chemical experiments were carried out in argon-purged acetonitrile
at 298 and 233 K. Cyclic voltammetry (CV) was performed with a
glassy carbon the working electrode (0.08 cm2), a Pt spiral counter
electrode, and a silver wire was employed as a quasi-reference elec-
trode (AgQRE). The potentials reported were referenced to SCE
1 H), 3.04 (s, 6 H, NMe2) ppm. 13C NMR (CDCl3, 50 MHz): δ =
169.5, 150.2, 143.7, 135.9, 135.0, 133.5, 132.0, 130.5, 129.3, 129.2,
129.1, 129.0, 128.3, 128.1, 126.1, 124.7, 123.6, 123.4, 120.4, 112.4,
109.1, 66.7, 40.5 ppm. HRMS: calcd. for C27H22N2O2 406.1691
[MH]+; found 407.1770.
7,8-Dihydro-8-hydroxy-13-(4-methoxyphenyl)-7H-indeno[1Ј,2Ј:4,5]-
pyrido[2,1-a]isoindol-5-one (10i): M.p. 184 °C. 1H NMR (CDCl3,
200 MHz): δ = 7.92 (d, J = 6.0 Hz, 1 H), 7.75–7.48 (m, 4 H), 7.27–
7.37 (m, 3 H), 7.09 (t, J = 7.6 Hz, 1 H), 7.00 (d, J = 8.8 Hz, 2 H),
6.91 (d, J = 7.8 Hz, 1 H), 6.27 (d, J = 1.4 Hz, 1 H), 5.94 (d, J =
1.4 Hz, 1 H), 5.55 (d, J = 6.0 Hz, 1 H), 4.76 (d, J = 6.0 Hz, 1 H),
3.89 (s, 3 H, OMe) ppm. 13C NMR (CDCl3, 62.5 MHz): δ = 169.5,
159.5, 143.9, 135.6, 135.0, 133.7, 132.1, 131.2, 130.8, 129.5, 129.4,
128.8, 128.3, 128.2, 124.9, 123.6, 123.5, 120.4, 114.4, 108.5, 77.9,
66.7, 55.3 ppm. HRMS: calcd. for C26H19NO3 393.1373 [M +
Na]+; found 416.1270.
General Procedure for the Preparation of Pentacycles 16: Degassed
solutions of boronic acid 6 (0.75 mmol) in methanol (1 mL) and
Na2CO3 (110 mg, 1 mmol) in water (1 mL) were successively added
to a degassed solution of toluene (2 mL) containing [Pd(PPh3)4]
(29 mg, 0.025 mmol), NaBr (only with chlorinated substrates 12;
50 mg, 0.5 mmol) and pyridylbenzaldehyde 12 or 13 (0.5 mmol).
After heating for 12 h at 100 °C, the reaction mixture was cooled
to room temperature, extracted with ethyl acetate and dried with
MgSO4. After concentration, the residue was purified by
chromatography on silica gel (cyclohexane/ethyl acetate) to give the
desired compounds.
by measuring the AgQRE potential relative to ferrocene (E1/2
=
0.39 V vs. SCE for Fc+/Fc). The concentrations of the compounds
were approximately 1ϫ 10–3 m; 0.1 m tetraethylammonium hexaflu-
orophosphate (TEAPF6) was added as supporting electrolyte. Cy-
clic voltammograms were obtained at scan rates in the range 0.05–
10 Vs–1. The estimated experimental error in the E1/2 value is
Ϯ10 mV.
General Procedure for the Preparation of Pentacycles 10: To a de-
gassed solution of toluene (15 mL) containing [Pd(PPh3)4] (116 mg,
0.01 mmol) and 2,5-dihalopyridine 1, 3 or 4 (1 mmol), degassed
solutions of (2-formylphenyl)boronic acid (6a; 2.5 mmol) in meth-
anol (2.5 mL) and Na2CO3 (5 mmol) in water (5 mL) were success-
ively added. After heating for 12 h at 100 °C, the reaction mixture
was cooled to room temperature, extracted with ethyl acetate
(3ϫ20 mL) and dried with anhydrous magnesium sulfate (MgSO4).
After filtration through Celite and concentration, the residue was
purified by chromatography on silica gel (cyclohexane/ethyl acet-
ate) to give the desired compound.
7,8-Dihydro-8-hydroxy-10-methoxy-7H-indeno[1Ј,2Ј:4,5]pyrido[2,1-a]-
1
isoindol-5-one (16a): M.p. 152 °C. H NMR (CDCl3, 200 MHz): δ
= 7.83 (d, J = 7.0 Hz, 1 H), 7.65–7.30 (m, 4 H), 7.12 (s, 1 H), 6.88
(dd, J = 2.0, 8.0 Hz, 1 H), 6.32 (dd, J = 2.0, 6.0 Hz, 1 H), 6.23 (d,
J = 6.0 Hz, 1 H), 5.83 (s, 1 H, OH), 5.35 (d, J = 6.0 Hz, 1 H), 4.60
(dd, J = 2.0, 6.0 Hz, 1 H), 3.84 (s, 3 H, OMe) ppm. 13C NMR
(CDCl3, 50 MHz): δ = 169.5, 161.6, 145.2, 136.7, 135.0, 132.9,
131.8, 128.8, 128.6, 127.7, 123.1, 122.2, 120.1, 116.8, 110.5, 108.4,
104.2, 78.1, 66.0, 55.5 ppm. HRMS: calcd. for C20H15NO3
317.1047 [M + Na]+; found 340.0944.
13-(4-Formylphenyl)-7,8-dihydro-8-hydroxy-7H-indeno[1Ј,2Ј:4,5]pyr-
ido[2,1-a]isoindol-5-one (10f): M.p. 127 °C. 1H NMR (CDCl3,
250 MHz): δ = 10.10 (s, 1 H, CHO), 7.99 (d, J = 5.0 Hz, 2 H), 7.94
(d, J = 5.0 Hz, 1 H), 7.72–7.53 (m, 6 H), 7.36 (t, J = 5.0 Hz, 1 H),
7.08 (t, J = 5.0 Hz, 1 H), 6.78 (d, J = 5.0 Hz, 1 H), 6.25 (s, 1 H),
5.89 (s, 1 H, OH), 5.58 (d, J = 4.0 Hz, 1 H), 4.77 (d, J = 4.0 Hz, 1
H) ppm. 13C NMR (CDCl3, 62.5 MHz): δ = 191.6, 169.5, 145.0,
144.3, 136.0, 134.9, 134.3, 132.3, 132.0, 130.4, 130.0, 129.8, 129.1,
128.3, 127.8, 125.1, 123.6, 123.5, 120.5, 106.8, 77.7, 66.7 ppm.
HRMS: calcd. for C26H17NO3 391.1215 [M + Na]+; found
414.1112.
10-(Dimethylamino)-7,8-dihydro-8-hydroxy-7H-indeno[1Ј,2Ј:4,5]pyr-
ido[2,1-a]isoindol-5-one (16b): M.p. 193 °C. 1H NMR (CDCl3,
200 MHz): δ = 7.86 (d, J = 7.0 Hz, 1 H), 7.70–7.35 (m, 4 H), 6.90
(s, 1 H), 6.70 (d, J = 8.0 Hz, 1 H), 6.28 (s, 2 H), 5.94 (s, 1 H, OH),
5.37 (d, J = 6.0 Hz, 1 H), 4.63 (d, J = 6.0 Hz, 1 H), 3.03 (s, 6 H,
NMe2) ppm. 13C NMR (CDCl3, 50 MHz): δ = 169.5, 152.1, 145.2,
137.8, 135.1, 131.9, 131.7, 128.4, 127.6, 124.0, 123.1, 122.2, 119.9,
113.1, 108.3, 107.1, 105.1, 78.2, 66.1, 40.5 ppm. HRMS: calcd. for
C21H18N2O2 330.1368 [M + Na]+; found 353.1260.
10-(Dimethylamino)-7,8-dihydro-8-hydroxy-3-methoxy-7H-indeno-
[1Ј,2Ј:4,5]pyrido[2,1-a]isoindol-5-one (16c): M.p. 144 °C. 1H NMR
(CDCl3, 200 MHz): δ = 7.56 (d, J = 8.5 Hz, 1 H), 7.39–7.31 (m, 2
H), 7.12 (d, J = 8.5, Hz, 1 H), 6.92 (s, 1 H), 6.72 (d, J = 8.0 Hz, 1
H), 6.27 (d, J = 6.0 Hz, 1 H), 6.17 (d, J = 6.5 Hz,1 H), 5.91 (s, 1
H, OH), 5.36 (d, J = 5.5 Hz, 1 H), 4.61 (d, J = 4.5 Hz, 1 H), 3.89
(s, 3 H, OMe), 3.04 (s, 6 H, NMe2) ppm. 13C NMR (CDCl3,
50 MHz): δ = 169.5, 160.5, 152.0, 145.1, 136.9, 131.9, 129.3, 128.2,
124.4, 122.1, 121.3, 120.6, 113.3, 108.5, 107.3, 105.4, 103.8, 78.3,
66.3, 55.7, 40.6 ppm. HRMS: calcd. for C22H20N2O3 360.1481 [M
+ H]+; found 361.1561.
7,8-Dihydro-8-hydroxy-13-(4-methylthiophenyl)-7H-indeno[1Ј,2Ј:4,5]-
pyrido[2,1-a]isoindol-5-one (10g): M.p. 109 °C. 1H NMR (CDCl3,
200 MHz): δ = 7.94 (d, J = 6.0 Hz, 1 H), 7.71–7.50 (m, 5 H), 7.34
(s, 4 H), 7.10 (t, J = 8.0 Hz, 1 H), 6.91 (d, J = 8.0 Hz, 1 H), 6.25
(s, 1 H, H), 5.93 (s, 1 H, OH), 5.55 (d, J = 6.0 Hz, 1 H), 4.75 (d, J
= 6.0 Hz, 1 H), 2.56 (s, 3 H, SMe) ppm. 13C NMR (CDCl3,
50 MHz): δ = 169.5, 143.9, 138.7, 135.4, 135.1, 134.9, 133.8, 132.1,
131.3, 129.6, 129.5, 128.7, 1284, 128.3, 128.2, 126.6, 124.9, 123.6,
123.5, 120.4, 108.1, 66.7, 15.6 ppm. HRMS: calcd. for C26H19NO2S
409.1142 [M + Na]+; found 432.1029.
13-(4-Dimethylaminophenyl)-7,8-dihydro-8-hydroxy-7H-indeno-
[1Ј,2Ј:4,5]pyrido[2,1-a]isoindol-5-one (10h): M.p. 201 °C. 1H NMR
(CDCl3, 200 MHz): δ = 7.92 (d, J = 6.0 Hz, 1 H), 7.67–7.48 (m, 4
H), 7.31 (m, 3 H), 7.08 (m, 2 H), 6.80 (d, J = 8.0 Hz, 2 H), 6.31
(s, 1 H), 5.97 (s, 1 H), 5.55 (d, J = 6.0 Hz, 1 H), 4.76 (d, J = 6.0 Hz,
7,8-Dihydro-8-hydroxy-3-methoxy-7H-indeno[1Ј,2Ј:4,5]pyrido[2,1-a]-
1
isoindol-5-one (16d): M.p. 209 °C. H NMR (CDCl3, 200 MHz): δ
= 7.58 (m, 2 H), 7.48 (dd, J = 1.5, 4.0 Hz, 1 H), 7.40–7.31 (m, 3
H), 7.17 (dd, J = 1.5, 4.0 Hz, 1 H), 6.52 (dd, J = 1.5, 4.0 Hz, 1 H),
Eur. J. Org. Chem. 2013, 2316–2324
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2323

P. Ceroni, V. Mamane et al.
FULL PAPER
[2] a) K. E. Sapsford, L. Berti, I. L. Medintz, Angew. Chem. 2006,
118, 4676–4704; Angew. Chem. Int. Ed. 2006, 45, 4562–4588;
b) M. S. T. Gonçalves, Chem. Rev. 2009, 109, 190–212; c) L. D.
Davis, R. T. Raines, ACS Chem. Biol. 2008, 3, 142–155.
[3] T. S. A. Heugebaert, B. I. Roman, C. V. Stevens, Chem. Soc.
Rev. 2012, 41, 5626–5640.
[4] Y. Cheng, B. Ma, F. Wudl, J. Mater. Chem. 1999, 9, 2183–2188.
[5] a) T. Mitsumori, M. Bendikov, O. Dautel, F. Wudl, T. Shioya,
H. Sato, Y. Sato, J. Am. Chem. Soc. 2004, 126, 16793–16803;
b) Y.-M. Chen, G. Grampp, N. Leesakul, H.-W. Hu, J.-H. Xu,
Eur. J. Org. Chem. 2007, 3718–3726.
[6] a) J. Wan, C.-J. Zheng, M.-K. Fung, X.-K. Liu, C.-S. Lee, X.-
H. Zhang, J. Mater. Chem. 2012, 22, 4502–4510; b) B. Liu, Z.
Wang, N. Wu, M. Li, J. You, J. Lan, Chem. Eur. J. 2012, 18,
1559–1603; c) E. Kim, M. Koh, B. J. Lim, S. B. Park, J. Am.
Chem. Soc. 2011, 133, 6642–6649; d) P. J. Aittala, O. Cramar-
iuc, T. I. Hukka, M. Vasilescu, R. Bandula, H. Lemmetyinen,
J. Phys. Chem. A 2010, 114, 7094–7101; e) M. Aginagalde, Y.
Vara, A. Arrieta, R. Zangi, V. L. Cebolla, A. Delgado-Camon,
F. P. Cossio, J. Org. Chem. 2010, 75, 2776–2784.
[7] V. Mamane, Y. Fort, Tetrahedron Lett. 2006, 47, 2337–2340.
[8] Z. Chamas, O. Dietz, E. Aubert, Y. Fort, V. Mamane, Org.
Biomol. Chem. 2010, 8, 4815–4818.
[9] F. Cottet, M. Schlosser, Tetrahedron 2004, 60, 11869–11874.
[10] For the importance of temperature in the ortho-lithiation of 3,
see: M. Abboud, V. Manane, E. Aubert, C. Lecomte, Y. Fort,
J. Org. Chem. 2010, 75, 3224–3231.
[11] a) For 6b, see: S. J. Kumar, J. Chem. Soc. Perkin Trans. 1 2001,
1018–1032; b) for 6c, see: S. Lozano Yeste, M. E. Powell, S. D.
Bull, T. D. James, J. Org. Chem. 2009, 74, 427–430; c) for 6d,
see ref.[8]
[12] D. L. Comins, J. D. Brown, J. Org. Chem. 1984, 49, 1078–1083.
[13] N. Kudo, M. Perseghini, G. C. Fu, Angew. Chem. 2006, 118,
1304–1306; Angew. Chem. Int. Ed. 2006, 45, 1282–1284.
[14] M. Wrighton, D. L. Morse, J. Am. Chem. Soc. 1974, 96, 998–
1003.
6.19 (d, J = 4.0 Hz,1 H), 5.73 (s, 1 H, OH), 5.45 (d, J = 4.0 Hz, 1
H), 4.68 (dd, J = 1.5, 3.0 Hz, 1 H), 3.84 (s, 3 H, OMe) ppm. 13C
NMR (CDCl3, 50 MHz): δ = 169.6, 161.0, 143.1, 136.1, 133.9,
129.7, 129.6, 128.8, 127.9, 125.1, 121.6, 120.9, 120.6, 112.9, 105.7,
102.4, 78.4, 66.1, 55.8 ppm. HRMS: calcd. for C20H15NO3
317.1053 [M + Na]+; found 340.0950.
3-(Dimethylamino)-7,8-dihydro-8-hydroxy-7H-indeno[1Ј,2Ј:4,5]pyr-
ido[2,1-a]isoindol-5-one (16e): M.p. 195 °C. 1H NMR (CDCl3,
200 MHz): δ = 7.62–7.33 (m, 5 H), 7.07 (s, 1 H), 6.88 (dd, J = 2.0,
8.0 Hz, 1 H), 6.47 (dd, J = 3.0, 6.0 Hz, 1 H), 6.04 (d, J = 6.0 Hz,1
H), 5.83 (s, 1 H, OH), 5.40 (d, J = 6.0 Hz, 1 H), 4.62 (dd, J = 2.5,
5.5 Hz, 1 H), 3.04 (s, 6 H, NMe2) ppm. 13C NMR (CDCl3,
50 MHz): δ = 170.4, 151.4, 142.9, 136.3, 134.7, 134.6, 129.6, 129.2,
128.6, 125.0, 122.8, 121.4, 120.7, 116.0, 113.3, 104.8, 100.4, 78.5,
66.1, 40.5 ppm. HRMS: calcd. for C21H18N2O2 330.1365 [M + H]
+; found 331.1445.
10-(Dimethylamino)-7,8-dihydro-8-hydroxy-3-methoxy-7H-indeno-
[1Ј,2Ј:4,5]pyrido[2,1-a]isoindol-5-one (16f): M.p. 196 °C. 1H NMR
(CDCl3, 200 MHz): δ = 7.51 (s, 2 H), 7.48 (d, J = 2.5 Hz, 1 H),
6.95–6.87 (m, 3 H), 6.51 (d, J = 6.0 Hz, 1 H), 6.04 (d, J = 6.0 Hz,
1 H), 5.74 (s, 1 H, OH), 5.36 (d, J = 5.5 Hz, 1 H), 4.61 (dd, J =
3.0, 5.0 Hz, 1 H), 3.84 (s, 3 H, OMe), 3.06 (s, 6 H, NMe2) ppm.
13C NMR (CDCl3, 50 MHz): δ = 170.4, 160.4, 151.4, 137.7, 135.5,
134.7, 134.6, 129.7, 126.0, 122.8, 121.4, 116.4, 116.0, 113.3, 104.9,
104.7, 100.3, 78.0, 66.4, 55.5, 40.6 ppm. HRMS: calcd. for
C22H20N2O3 360.1479 [M + Na]+; found 383.1376.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures and analytical data for pyridines 4,
1
boronic acids 8 and pyridylbenzaldehydes 12 and 13, H and 13C
NMR spectra for all new compounds.
[15] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[16] R. E. Stratmann, G. E. Scuseria, M. J. J. Frisch, Chem. Phys.
1998, 109, 8218–8224.
Acknowledgments
[17] T. H. Dunning, Jr., K. A. Peterson, D. E. Woon in Basis Sets:
Correlation Consistent Sets in the Encyclopedia of Computa-
tional Chemistry (Ed.: P. v. R. Schleyer), Wiley, Chichester,
UK, 1998.
This work has been supported by the Centre National de la Recher-
che Scientifique (CNRS), the Université de Lorraine, by the Min-
istère de l’Enseignement Supérieur et de la Recherche (grant to
Z. C.), the Ministero dell’Università e della Ricerca (MIUR)
(PRIN 20085ZXFEE, PRIN 2009SLKFEX and FIRB
RBAP11C58Y) and by MAE DGPCC (Dendrimeri come nuovi
materiali per celle solari ecocompatibili).
[18] A. Modelli, L. Mussoni, D. Fabbri, J. Phys. Chem. A 2006,
110, 6482–6486.
[19] L. Crocker, T. Wang, P. Kebarle, J. Am. Chem. Soc. 1993, 115,
7818–7822.
[20] J. N. Demas, G. A. Crosby, J. Phys. Chem. 1971, 75, 991–1024.
[21] M. Montalti, A. Credi, L. Prodi, M. T. Gandolfi, in: Handbook
of Photochemistry, 3rd ed., Taylor & Francis Group, New York,
1996.
[1] a) M. Pawlicki, H. A. Collins, R. G. Denning, H. L. Anderson,
Angew. Chem. 2009, 121, 3292–3316; Angew. Chem. Int. Ed.
2009, 48, 3244–3266; b) J. Kalinowski, Opt. Mater. 2008, 30,
792–799.
Received: November 22, 2012
Published Online: March 13, 2013
2324
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 2316–2324