New push-pull dyes based on 2-(3-oxo-2,3-dihydro-1H-cyclopenta[b]naphthalen-1-ylidene)malononitrile: An amine-directed synthesis

Article abstract of DOI:10.1016/j.dyepig.2020.108182

A series of twelve dyes have been designed using 2-(3-oxo-2,3-dihydro-1H-cyclopenta[b]naphthalen-1-ylidene)malononitrile as the electron acceptor. While using piperidine as a classical base for the Knoevenagel reaction, a nucleophilic attack of the amine on the formed push-pull chromophore occurred, producing an azafluorenone derivative. By varying the amine, a series of azafluorenones could be obtained. By using an aldehyde of extended conjugation, a spontaneous aromatization following the cyclization reaction could also be demonstrated. The optical and electrochemical properties of the different dyes were examined. Theoretical calculations were also carried out to support the experimental results.

Full text of DOI:10.1016/j.dyepig.2020.108182

Journal Pre-proof
New push-pull dyes based on 2-(3-oxo-2,3-dihydro-1H-cyclopenta[b]naphthalen-1-
ylidene)malononitrile: An amine-directed synthesis
Corentin Pigot, Guillaume Noirbent, Sébastien Peralta, Sylvain Duval, Thanh-Tuan
Bui, Pierre-Henri Aubert, Malek Nechab, Didier Gigmes, Frédéric Dumur
PII:
S0143-7208(19)32307-1
DOI:
https://doi.org/10.1016/j.dyepig.2020.108182
DYPI 108182
Reference:
To appear in: Dyes and Pigments
Received Date: 27 September 2019
Revised Date: 29 November 2019
Accepted Date: 1 January 2020
Please cite this article as: Pigot C, Noirbent G, Peralta Sé, Duval S, Bui T-T, Aubert P-H,
Nechab M, Gigmes D, Dumur Fréé, New push-pull dyes based on 2-(3-oxo-2,3-dihydro-1H-
cyclopenta[b]naphthalen-1-ylidene)malononitrile: An amine-directed synthesis, Dyes and Pigments
(2020), doi: https://doi.org/10.1016/j.dyepig.2020.108182.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

Author Contribution Statement
C.P., G.N. and F.D. conceived this work. C.P. and F.D. designed and executed the
experimental work, analyzed the data and discussed the results. C.P. and G.N. realized the
chemical characterizations and the photophysical analysis of the compounds. S.D. solved the
crystal structures. S.P. carried out the theoretical calculations. T.-T. B. and P.-H. supervised
the electrochemistry experiments. M.N. and D.G. did a critical review of the manuscript. C.P.,
G.N. and F.D. contributed to writing and editing the manuscript.

N
O

New push-pull dyes based on 2-(3-oxo-2,3-dihydro-1H-
cyclopenta[b]naphthalen-1-ylidene)malononitrile: An amine-directed
synthesis
Corentin Pigot,*,a Guillaume Noirbent,a Sébastien Peraltab, Sylvain Duval,c Thanh-Tuan Buib,
Pierre-Henri Aubertb, Malek Nechab,a Didier Gigmesa and Frédéric Dumur*,a
a Aix Marseille Univ, CNRS, ICR UMR 7273, F-13397 Marseille, France, corentin.pigot@univ-amu.fr,
frederic.dumur@univ-amu.fr
b Laboratoire de Physicochimie des Polymères et des Interfaces (LPPI), Université de Cergy Pontoise, 5 mail
Gay Lussac, F-95000 Neuville-sur-Oise, France
c Université de Lille, CNRS, Centrale Lille, ENSCL, Univ. Artois, UMR 8181 - UCCS - Unité de Catalyse et
Chimie du Solide, F-59000 Lille, France
Abstract
A series of twelve dyes have been designed using 2-(3-oxo-2,3-dihydro-1H-
cyclopenta[b]naphthalen-1-ylidene)malononitrile as the electron acceptor. While using
piperidine as a classical base for the Knoevenagel reaction, a nucleophilic attack of the amine
on the formed push-pull chromophore occurred, producing an azafluorenone derivative. By
varying the amine, a series of azafluorenones could be obtained. By using an aldehyde of
extended conjugation, a spontaneous aromatization following the cyclization reaction could
also be demonstrated. The optical, electrochemical properties of the different dyes were
examined. Theoretical calculations were also carried out to support the experimental results.
Keywords
Knoevenagel reaction; Push-pull dyes; Indane-1,3-dione derivative; Cyclization reaction;
Azafluorenone
Introduction
During the past decades, a great deal of efforts has been devoted to design push-pull
chromophores as these structures can find applications in numerous research fields ranging
from Non-Linear Applications (NLO) [1-6] to Organic Photovoltaics (OPVs) [7-12], Organic
Fields Effects Transistors (OFETs) [13], Organic Light-Emitting Diodes (OLEDs) [14] and
photoinitiators of polymerization [15-19]. Typically, a push-pull compound is composed of an
electron donor connected to an electron acceptor by mean of a conjugated or a non-conjugated

spacer but dyes comprising a conjugated spacer are the most popular ones. Among all
possible reactions to create a π-conjugated linker, the Knoevenagel reaction consisting in the
dehydrative condensation of an aldehyde with an activated methylene group is a reaction of
choice for the preparation of push-pull structures and this reaction is most generally carried
out under basic conditions. However, when strong electron acceptors are used, their anionic
forms are stable under basic conditions, rendering these conditions ineffective to prepare
push-pull dyes. In this case, use of acidic conditions are thus required, as exemplified with
Knoevenagel reactions carried out in acetic anhydride. When basic conditions are employed,
piperidine is without contest one of the most popular bases due to its relatively high pKa
(11.1, 25°C) which is sufficient to deprotonate activated methylene groups, its low cost, its
low toxicity and easiness to evaporate [20]. However, to prepare dyes, the selection of the
reaction conditions is of crucial importance as the amine can further react with the newly
prepared dye, giving rise to the formation of side-products. This drawback is notably observed
for all dyes prepared with 2-(3-oxo-2,3-dihydro-1H-inden-1-ylidene) malononitrile as the
electron acceptors or its derivatives. A nucleophilic attack of the amine onto the
dicyanomethylidene groups of the electron acceptor is occasionally observed. As a result of
this,
3-(dialkylamino)-1,2-dihydro-9-oxo-9H-indeno
[2,1-c]
pyridine-4-carbonitrile
derivatives can be obtained (See Figure 1) [21]. Reaction yields ranging from 17 to 51% were
obtained upon mixing one equivalent of aldehyde with one equivalent of base and one
equivalent of electron acceptor at room temperature for 14 hours.
In 2014, a better reaction yield was reported for the synthesis of C5, reaching 96%, by
opposing one equivalent of diethylamine to the preformed push-pull dye 2-(2-(furan-2-
ylmethylene)-3-oxo-2,3-dihydro-1H-inden-1-ylidene)malononitrile for 30 min. at room
temperature (see Figure 2) [22]. This unexpected nucleophilic attack on the
dicyanomethylidene group was notably observed during the synthesis of a series of
photochromes based on Donor−Acceptor Stenhouse Adducts (DASA).[22] More recently, a
fluorescent probe for live cell imaging was developed with a triphenylamine using a similar
approach and C6 could be obtained in 47% yield after heating for 12 hours at 50°C [23].
From these different works, it can be concluded that the formation of the push-pull compound
arises prior the nucleophilic attack of the amine onto the dicyanomethylidene group and
therefore the cyclization reaction.

H
N
O
O
NH
+
NC
EtOH, reflux
N
CN
OHC
NC
C1
17% yield
O
H
N
O
O
O
CHO
NH
+
NC
N
EtOH, reflux
CN
CN
CN
NC
NC
NC
C2
49% yield
H
N
O
S
+
NH
EtOH, reflux
NC
CHO
N
S
C3
37% yield
H
N
O
CHO
+
NH
N
EtOH, reflux
NC
C4
51% yield
Figure 1. Chemical structures of 3-(dialkylamino)-1,2-dihydro-9-oxo-9H-indeno[2,1-
c]pyridine-4-carbonitrile derivatives C1-C4 formed by nucleophilic attack of the amine
catalysts.

Figure 2. Chemical structure of C5 prepared from 2-(2-(furan-2-ylmethylene)-3-oxo-2,3-
dihydro-1H-inden-1-ylidene)malononitrile and structure of C6.
Also, the following mechanism was proposed to support the formation of 3-
(dialkylamino)-1,2-dihydro-9-oxo-9H-indeno [2,1-c]pyridine-4-carbonitrile derivatives (See
Figure 3).
Figure 3. Mechanism resulting in the synthesis of 3-(dialkylamino)-1,2-dihydro-9-oxo-9H-
indeno [2,1-c]pyridine-4-carbonitrile derivatives.
In this work, we report on the synthesis of two new push-pull dyes PP1 and PP2 based
on 2-(3-oxo-2,3-dihydro-1H-cyclopenta[b]naphthalen-1-ylidene)malononitrile EA1 as the
electron acceptor. To date, EA1 has only been scarcely used to prepare dyes [24-26] despite
its promising electron-withdrawing ability. We also drastically optimized the synthesis of 1H-

cyclopenta[b]naphthalene-1,3(2H)-dione EA2 so that this precursor of EA1 could be obtained
with reaction yields higher than 90%. To evaluate the benefit of EA1 in the optical properties
of PP1 and PP2, their analogues PP3 [27] and PP4 [28] using 2-(3-oxo-2,3-dihydro-1H-
inden-1-ylidene)malononitrile as the acceptor were prepared and the optical properties
compared to those of PP1 and PP2 (see Figure 4). It has to be noticed that the synthesis of
these two dyes (PP3 and PP4) has been previously reported in the literature. Parallel to this,
during the course of our investigations, choice of the amine used for the Knoevenagel reaction
was determined as being crucial to avoid the nucleophilic attack of this latter on the push-pull
structures so that a series of amines were tested as catalysts. The different products obtained
during these different reactions and their properties are also reported and discussed.
Figure 4. Chemical structures of push-pull dyes PP1 and PP2, their analogues PP3 and PP4
and the electron acceptors EA1 and EA2.
Results and Discussion
1- Synthesis of the different compounds
As mentioned in the introduction section, the first step consisted in preparing EA1
starting from EA2. Compared to previous reports [29,30], the reaction yield to get EA2 was
greatly improved, based on several modifications. Notably, a typical procedure to access to
EA2 consists in a Claisen reaction of a dialkyl naphthalene-2,3-dicarboxylate with ethyl
acetate in the presence of a base (See Figure 5).

Figure 5. Synthetic routes to EA1 and EA2.
By increasing the quantity of sodium hydride (2.5 eq. instead of 1.45 eq.) for the first
step and by elongating the decarboxylation time (1h30 instead of 20 min.) for the second step,
the reaction yield could be increased up to 91%, higher than the 65% yield reported in the
literature. Noticeably, safer conditions were used for the recrystallization step of EA2, in
which benzene was replaced by toluene. Finally, the functionalization of EA2 by
malononitrile was carried out under standard conditions, using sodium acetate as the base.
EA1 could be obtained in 67% yield, as previously described in the literature [31]. After the
preparation of EA1, the synthesis of the different push-pull dyes described in the Figure 4
were realized, starting with PP1. This synthesis consisted in a Knoevenagel reaction between
EA1 and 4-N,N-dimethylaminobenzaldehyde ED1 in the presence of a base, currently
piperidine, in ethanol for 30 min.
Unexpectedly, the reaction media turned red and the structure of a new compound,
PP6, was confirmed by NMR spectroscopy (See Figure 6). Nevertheless, this reactivity of
secondary amines, like piperidine, has already been highlighted in the literature, in 2008, by
Landmesser et al. [21]. As no such reaction has been previously described concerning α,β-
unsaturated aldehydes, the same procedure was used for PP2, using the same conditions.
After 30 min. of reaction, if the Nuclear Magnetic Resonance (NMR) analyses confirmed the
absence of the targeted chromophore PP2 in the reaction media, a red colour was also
observed for this second reaction opposing EA1 to ED2 (See Figure 6).

Figure 6. Synthetic route to the targeted PP1 and PP2 dyes, and chemical structures of the
unexpected PP5 and PP6 adducts formed using piperidine as the base.
1
Surprisingly, analyses by H NMR spectroscopy in CDCl3 revealed the electron-
donating moiety ED2 not to be connected to EA1. On the opposite and unexpectedly, the
1
covalent linkage of piperidine to EA1 could be easily evidenced by H NMR analyses, with
the presence of two sets of signals at 1.85 and 3.86 ppm. Presence of a singlet at 5.87 ppm
integrating for one proton could also be detected. Finally, the exact structure of PP5 could be
revealed by X-ray diffraction (see Figure 7).
8.6
8.4
8.2
8.0
7.8
7.6
7.4
8
7
6
5
4
3
2
(ppm)

Figure 7. 1H NMR spectrum of PP5 done in CDCl3 (7.26 ppm) (left) and crystal structure of
PP5 (right).
The single crystals of PP5 suitable for X-ray crystallographic analyses were grown by
slow evaporation of CDCl3 solvent in the NMR tube. The detailed crystallographic data are
given in the Electronic Supporting Information.
As evidenced in the Figure 8. Mechanism inducing the formation of PP5 results from
the nucleophilic attack of piperidine on the polarized ketone group. After prototropy, a
dehydration reaction resulting from the deprotonation of the activated methylene group by
piperidine can occur, providing PP5 as a red powder. Interestingly, attempts to prepare PP5
by simply refluxing EA1 in ethanol, in the presence of piperidine failed to produce PP5 and
only traces of product were detected by thin layer chromatography (TLC) even after two
hours of reflux (See Figure 6). Upon addition of ED2, immediately, the reaction colour turned
deep red and a complete consumption of EA1 was observed within 20 minutes. If the exact
role of ED2 is not fully understood, ED2 clearly favours the synthesis of PP5. It can be
tentatively assigned to a role of base, this electron donor comprising a dimethylaniline moiety
probably facilitating the deprotonation occurring in last step.
Considering that PP6 was also obtained as a red solid, that its absorption properties
are clearly centered in the visible range with a small Highest Occupied Molecular Orbital
(HOMO)-Lowest Unoccupied Molecular Orbital (LUMO) gap, [21] the design and synthesis
of a series of dyes still based on EA1 as the electron acceptor but differing by the electron-
donating groups and the amine pendant groups were realized. The first parameter investigated
was the influence of the amine on the reaction yields and the optical properties of the resulting
products while keeping ED1 as the electron donor. In this aim, five different amines were
investigated. Secondly, the influence of the conjugation length between the electron-donating
part and the electron-withdrawing group on both the reaction yields and the optical properties
was examined. For this purpose, another electron donor, namely ED2 was used (See Figure
10). A summary of the different product obtained during these syntheses and the
corresponding reaction yields are provided in the Table 1. Even if no clear conclusion can be
made concerning the impact of the basicity of the amine on this reaction, we clearly
demonstrate that higher reaction yields were obtained when cyclic aliphatic amines were used
during the synthesis of PP6-PP10 (See Figure 9). It might be explained by the higher
availability of the nitrogen lone pair electrons in this case. In fact, the cycle drastically inhibits
the pyramidal inversion of the nitrogen atom.

Figure 8. Mechanism inducing the formation of PP5
Concerning the disparity of the reaction yields, two factors might also be considered.
Firstly, for morpholine and pyrrolidine, no column chromatography was necessary to get the
product in pure form, the dyes directly precipitating by addition of acetone to the reaction
media. In these conditions, the products could be obtained in high yields. As further detailed
in the NMR section, a facile protonation of the dihydro-2-azafluorenes in chloroform was
1
clearly evidenced, resulting in the duplication of the NMR signals (see H NMR spectra in
SI).

Figure 9. Synthetic route to PP6-PP10 and crystal structure of PP10.
Considering that CDCl3 is sufficiently acidic to protonate the dyes in solution, it also
means that during the purification process, a non-negligible part of the products was lost by
adsorption onto the stationary phase (silicagel), chloroform being used as the eluent.
Therefore, it may also explain why PP9 or PP10 were isolated in pure form only in low
yields. The formation of the aromatized product observed in the case of PP10 and producing
PP10’ also contributes to reduce the final reaction yield. To confirm the chemical structure of
these dihydro-2-azafluorenes, crystals could notably be obtained upon slow evaporation of
chloroform. The crystal structure of PP10 in presented in the Figure 9.
Regarding the condensation of ED2 onto EA1 and the subsequent nucleophilic attack
of the amine, adducts totally differing in structures from that obtained with ED1 were
obtained, evidencing a significant difference of reactivity with the elongated aldehyde ED2.
Notably, a spontaneous aromatization resulting in the formation of a pyridine group was
notably evidenced with PP11, PP12 and PP13 (See Figure 10). The crystal structure of PP12
is presented in the Figure 10.

Table 1 Isolated yields obtained with different amines during the synthesis of azafluorenones
using ED1 or ED2 as the donors.
Amine
ED1
ED2
PP6 (14%)
PP5 (86%)
PP11 (6%)
PP12 (18%)
PP13 (5%)
PP2 (60%)
PP7 (79%)
NH
PP8 (35%)
PP9 (10%)
PP10 (5%) + PP10’ (25%)
Figure 10. Synthetic route to PP11-PP13 and crystal structure of PP12.
Elongation of the π-conjugation therefore favoured the oxidation (and the
aromatization process), providing the pyridine adducts PP11, PP12 and PP13 as the main
products of a one-pot synthesis. Formation of such aromatized structures has been previously
reported by Landmesser et al. [21] While using aldehydes of short conjugation length, by

developing a two-step procedure where chemical oxidants (KMnO4, chloranil) were used, the
access to aromatized structures was possible. Besides, it has to be noticed that Landmesser et
al. also detected the coexistence of the aromatized and none-aromatized forms during the
course of their investigations, but the aromatized products were only detected as a minor side-
product of the reaction. Surprisingly, while using dioctylamine as the amine, the cyclization
product was not detected in the reaction media. Instead, PP2 could be isolated in pure form in
60% yield, demonstrating either a steric hindrance of the amine or a better kinetic favouring
the Knoevenagel reaction. After investigating the synthesis of the different azafluorenones
PP6-PP13, we focused our work on the synthesis of the initially desired push-pull dyes PP1
and PP2 and their analogues PP3 and PP4 (see Figure 11).
As previously evidenced, it appeared that the main drawback encountered with
piperidine was its nucleophilicity. To overcome this drawback, a none-nucleophilic base was
selected for the Knoevenagel reactions, namely N,N-diisopropylethylamine (DIPEA). Indeed,
the steric hindrance of this tertiary amine prevents its nucleophilic addition onto PP1-PP4. As
expected, PP1-PP4 were obtained with reaction yields ranging between 55 and 65% (See
Figure 11). Formation of these products were confirmed by 1H and 13C NMR analyses.

Figure 11. Synthetic route to PP1-PP4.
Finally, after the successful synthesis of PP1-PP4 by using DIPEA as the base, the possibility
to convert these push-pull structures into azafluorenones was examined. As a proof of
concept, the conversion of PP1 to PP6 and PP2 to PP12 was examined. In the two cases,
slightly higher reaction yields were obtained compared to the direct one-pot synthesis,
resulting from the fact that the push-pull dyes PP1 and PP2 are already formed and purified
prior to the cyclization reaction occurring in final step. (See Figure 12).

Figure 12. Synthetic route to PP6 and PP12 starting from PP1 and PP2.
2- 1H NMR spectroscopy study
As above-mentioned in the text, the different cyclization products proved to be highly
sensitive to the protic character of the solvent. This point was notably evidenced by the low
isolated reaction yields obtained for the different compounds after chromatographic
purification. This also explains that most of the product remaining immobilized on the acidic
silica gel during the complex separation of numerous products by column chromatography.
Face to this consideration, this point was also explored by NMR spectroscopy. Two different
solvents enabling to perfectly solubilize the different compounds (PP6, PP7, PP8, PP9 and
PP10) were tested, namely, a protic solvent i.e. chloroform and an aprotic one i.e. DMSO.
Figure 13 displays the NMR data of PP8 in the two selected solvents.
water
DMSO
8.0
7.5
7.0
6.5
6.0
5.5
(ppm)
8
7
6
5
4
3
2
(ppm)

CHCl3
*
* *
* *
*
*
* **
*
*
9.0
8.5
8.0
7.5
7.0
6.5
(ppm)
*
*
*
*
*
*
*
*
9
8
7
6
5
4
3
2
(ppm)
1
1
Figure 13. Top: H NMR spectra of PP8 in DMSO-d6. Bottom: H NMR spectra of PP8 in
CDCl3 with the protonated form representing 31% of the sample. (stars indicate the NMR
signals corresponding to the product (*) and its protonated form (*)
1
If a clear H NMR spectrum corresponding to the targeted molecule was obtained in
DMSO-d6, the coexistence of two different forms could be found in the 1H NMR spectrum of
PP8 in deuterated chloroform (see Figure 13). Indeed, the protonated and the none-protonated
form could be determined as being in a 31:69 ratio, the none-protonated state constituting the
main form in the sample. Examination of the behaviour of these five chromophores in the two
solvents revealed PP7, PP8, and PP9 to be the most sensitive compounds of the series, ratios
of 21:79 and 28:72 being respectively determined for PP7 and PP9. As specificity for PP7,
traces of the protonated form could be even detected in DMSO, evidencing its easy
1
protonation (See H NMR spectra in SI). Conversely, PP6 proved to be insensitive to the
acidic character of chloroform, no trace of protonation being detected by NMR. It has to be
noticed that all cyclization reactions have been carried out in the presence of an excess of base
1
so that the protonation detected by H NMR can only originate from acidic traces in the
deuterated solvents.
3-UV-visible absorption and photoluminescence spectroscopy
All compounds reported in this work are stable under air, even when diluted in
solution so that photophysical measurements could be carried out. As evidenced by NMR
analyses, if a protonation for some of the dyes in protic solvents could be demonstrated, no
photodegradation of the dyes were detected by NMR during the course of our investigations.
Especially, no modification of the NMR signals was detected even for NMR tubes exposed to
sunlight during several weeks. In this context, the dyes being stable in solution, these latter
could be characterized by UV-visible absorption spectroscopy and the different absorption

spectra were recorded in three different solvents, namely chloroform, dichloromethane and
DMSO. Interestingly, four different but characteristic UV-visible spectra were found for the
cyclized compounds, depending of the length of the conjugated spacer between the electron-
donating dimethylaminophenyl group and the electron-accepting azafluorenone moiety, but
also to the fact that the pyridine group of the azafluorenone moiety is aromatized or not. Thus,
the UV/Vis absorption spectra of the dihydro-2-azafluorenones PP6, PP7, PP8, PP9, and
PP10 are dominated by an intensive intramolecular charge-transfer (ICT) band extending
between 440 and 600 nm irrespective of the solvent (See Figure 14).
a)
b)
35000
30000
25000
20000
15000
10000
5000
0
30000
25000
20000
15000
10000
5000
0
PP6
PP7
PP8
PP9
PP10
PP6
PP7
PP8
PP9
PP10
300
400
500
600
300
400
500
600
Wavelength (nm)
Wavelength (nm)
c)
d)
50000
40000
30000
20000
10000
0
35000
30000
25000
20000
15000
10000
5000
0
PP6
PP7
PP8
PP9
PP10
PP6
PP7
PP8
PP9
PP10
300
400
500
600
7
300
400
500
600
7
Wavelength (nm)
Wavelength (nm)
Figure 14. UV-visible absorption spectra of dihydro-2-azafluorenones PP6-PP10 in DMSO
(a), dichloromethane (b), chloroform (c) and the theoretical UV-visible absorption spectra in
dichloromethane (d).

By changing the appended amine, no shift of the ICT band was found for all
compounds, evidencing the push-pull effect to be independent of this substituent. The lack of
contribution on the push-pull effect were confirmed by theoretical calculations. As illustrated
in the Figure 15 and the corresponding Figures in SI, only the nitrogen atom of the pendant
amine was determined as contributing to the highest Occupied Molecular Orbital (HOMO)
and the Lowest Unoccupied Molecular Orbital (LUMO). No extension over the alkyl chain of
the different amines (piperidine, morpholine, dioctylamine, diethylamine or pyrrolidine) was
found, justifying the dihydro-2-azafluorenones to be insensitive to this substitution pattern.
For all dyes, an absorption maximum peaking at 500 nm was determined for the five
chromophores PP6-PP10 and in the three different solvents. A HOMO-LUMO gap of 2.48
eV was calculated for all dyes. Comparison of the experimental absorption spectra with those
determined by theoretical calculations revealed the absorption maxima to be red-shifted by
about 13 nm relative to that determined theoretically, this latter peaking at 487 nm.
Compounds
HOMO
LUMO
PP6
PP7
PP8

PP9
PP10
Figure 15. Optimized geometries and contour plots of the HOMO and LUMO energy levels
for the non-aromatized azafluorenones compounds (PP6-PP10)
Besides, a good agreement between the experimental and the theoretical values are
thus found for the absorption maxima. Apart from the ICT band, three absorption peaks could
also be detected between 350 and 430 nm and origins of these transitions were assigned to
HOMO-5/-4/-3 → LUMO transitions admixed with HOMO-1 → LUMO+1, HOMO →
LUMO+1 transitions (See Table 2).
Table 2. Summary of simulated absorption characteristics in dilute dichloromethane of a
selected set of synthetized compounds. Data were obtained in dichloromethane solution.
λmax (nm) Transitions
Compounds EHOMO (eV) ELUMO (eV)
-5.955
-2.914
477
HOMO->LUMO (92%)
PP5
HOMO-2->LUMO (7%)
HOMO->LUMO (95%)
HOMO->LUMO (96%)
HOMO->LUMO (89%)
HOMO-1->LUMO (9%)
HOMO->LUMO (95%)
HOMO->LUMO (92%)
HOMO->LUMO (95%)
HOMO->LUMO (96%)
HOMO->LUMO (95%)
-5.392
-5.504
-5.283
-2.217
-2.320
-2.103
496
496
496
PP6
PP7
PP8
-5.401
-5.367
-5.283
-5.389
-5.321
-2.211
-2.176
-2.379
-2.588
-2.454
496
487
496
506
496
PP9
PP10
PP11
PP12
PP13

Finally, the intense high-energy bands in the 300-350 nm region typically correspond
to localized aromatic π-π* transitions of the chromophores. Examination of their UV-visible
absorption properties in chloroform revealed the absorption spectra of all dyes to be similar to
those determined in DMSO or dichloromethane (See Figure 14), despites the partial
protonation of the dihydro-2-azafluorenones in chloroform. Unexpectedly, elongation of the
spacer between the electron donor and the electron acceptor in PP11, PP12 and PP13 resulted
in a hypsochromic shift of the absorption maxima irrespective of the amine substituents.
Indeed, absorption maxima peaking at 480 nm was found for the three dyes, blue-shifted by
about 20 nm relative to the previous series. Here again, absence of contribution of the amino
group to the push-pull effect was evidenced, the three absorption spectra superimposing each
other. Conversely and as anticipated, elongation of the spacer resulted in a significant increase
of the molar extinction coefficients of the ICT bands for PP11, PP12 and PP13 relative to
those of the previous dihydro-2-azafluorenones (PP6-PP10), resulting from an enhancement
of the oscillator strength. Here again, comparison of the theoretical and experimental
absorption spectra revealed a good agreement between both. Two main transitions were
detected in DMSO, in the 350-420 nm and 420-600 nm regions, the second one corresponding
to the ICT band of the chromophores. Positions of these transitions are close to those
determined theoretically, the simulated spectra displaying two main absorption bands at
comparable positions (See Figure 16).
60000
70000
PP11
PP12
PP13
50000
40000
30000
20000
10000
0
60000
50000
40000
30000
20000
10000
0
PP11
PP12
PP13
300
400
500
600
700
300
400
500
600
700
Wavelength (nm)
Wavelength (nm)
Figure 16. Comparison between the theoretical (right) an the experimental (left) UV-visible
absorption spectra recorded in DMSO for PP11-PP13.
A difference as small as 14 nm were found between the theoretical (492 nm) and the
experimental (478 nm) absorption maxima. Comparison of the theoretical calculations carried
out on the dihydro-2-azafluorenones and the 2-azafluorenones revealed the HOMO energy

level of dihydro-2-azafluorenones to benefit from both the electron-donating effects of the
dimethylaminophenyl group and the amino pendant group. Conversely, as evidenced in the
Figure 17, only the dimethylaminophenyl group of 2-azafluorenones contributes as the
electron donor, supporting the blue-shift detected for the ICT bands.
Compounds
HOMO
LUMO
PP11
PP12
PP13
PP10’
Figure 17. Optimized geometries and contour plots of the HOMO and LUMO energy levels
for the aromatized azafluorenones compounds (PP11-PP13 and PP10’)

Upon aromatization, a clear modification of the absorption spectra could be evidenced
between the dihydro-2-azafluorenone PP10 and the 2-azafluorenone PP10’ corresponding to
the aromatized version of PP10. As shown in the Figure 18, a blue-shift of the ICT band was
observed upon aromatization of the pyridine moiety, the absorption maximum shifting from
500 nm for PP10 to 430 nm for PP10’. These results are consistent with a reduced electronic
delocalization upon aromatization, what is notably observed when quinones are converted to
hydroquinone derivatives.[32] Parallel to this, the two intense high-energy bands in the 300-
350 nm region detected for PP10 were not visible anymore in the absorption spectrum of
PP10’. While using the same theoretical model, only a slight shift of the ICT band is observed
upon aromatization. Notably, absorption maxima peaking at 481 and 448 nm are theoretically
determined for PP10 and PP10’ respectively. If a good accordance is found between the
experimental and the theoretical absorption maximum for PP10 (496 nm and 482 nm
respectively) and PP10’ (434 and 447 nm respectively), a complete mismatch between the
theoretical and experimental molar extinction coefficients was found. This is attributable to
the long alkyl chains that rendered the optimization of structures and the subsequent
calculations difficult to be carried out.
25000
40000
20000
30000
PP10
PP10'
15000
PP10
PP10'
20000
10000
10000
5000
0
0
300
400
500
600
700
300
400
500
600
700
Wavelength (nm)
Wavelength (nm)
Figure 18. UV-visible absorption spectra of PP10 and PP10’ in chloroform (left) and the
theoretical ones (right).
Finally, the most intriguing absorption spectrum was obtained for PP5, resulting from
the nucleophilic addition of piperidine on the electron acceptor EA1. Presence of numerous
optical transitions could be found between 300 and 600 nm irrespective of the solvent, making
this dye a panchromatic chromophore (see Figure 18). Based on the theoretical calculations, a

few transitions could be assigned. Notably, in the 300-380 nm region, the different transitions
can be assigned to an admixture of HOMO-1→LUMO+1 and HOMO-2 → LUMO
transitions. Conversely, in the 380-440 nm region, the absorption band is dominated by the
HOMO-1→LUMO and HOMO→LUMO+1 transitions. At low energy, a broad peak
composed of multiple transitions extending from 450 nm to 600 nm and dominated by the
HOMO→LUMO transition was found. An absorption maximum peaking at 506 nm was
found for PP5, close to that obtained for the dihydro-2-azafluorenones PP6-PP10 (500 nm).
40000
16000
30000
12000
HOMO
LUMO
20000
10000
0
8000
4000
0
300
350
400
450
500
550
600
300
350
400
450
500
550
600
Wavelength (nm)
Wavelength (nm)
Figure 19. UV-visible absorption spectrum of PP5 in DMSO (red) and chloroform (green).
PP1
PP1
PP2
PP3
PP4
PP2
PP3
PP4
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
400
450
500
550
600
650
700
750
450
500
550
600
650
700
750
Wavelength (nm)
Wavelength (nm)
Figure 20. UV-visible spectra of PP1-PP4 in chloroform (left) and the theoretical absorption
spectra (right).
It must be noticed that the complexity of the transitions detected by absorption
spectroscopy can be tentatively assigned to the coexistence of two forms in solution, the
neutral form and its corresponding zwitterion, complexifying the absorption spectrum. By
examining the UV spectra of PP1-PP4 in chloroform (see Figure 20), several trends can be
deduced. As described at the beginning of the study, influence of two factors are studied in

this paper, namely the impact of the additional phenyl ring on the electron-withdrawing
ability of EA1 and the role of the conjugated spacer on the optical properties of PP1-PP4.
Concerning the elongation of the aromatic scaffold of EA1, a red-shift absorption by about 30
nm is observed while comparing PP1 and PP3, PP2 and PP4 i.e. at similar electron-donating
group and spacer (see Figure 20). It can be thus concluded that a stabilization of the LUMO
energy level is observed for EA1 compared to its analogue 2-(3-oxo-2,3-dihydro-1H-inden-1-
ylidene) malononitrile. Concerning the extension of the π-conjugation of the spacer in PP2
and PP4, an even bigger red-shift of 80 nm is detected. As previously reported in the
literature, the HOMO of push-pull structure comprises both the electron-donating group and
the π-conjugated spacer [33]. Upon elongation of the spacer, a destabilization of the HOMO
level occurs, resulting in a decrease of the HOMO-LUMO gap and a red-shift of the
absorption spectrum towards long wavelengths. Theoretical calculations carried out on PP1-
PP4 confirmed the order determined in solution (See Figure 20) and the distribution of the
electronic density on the HOMO and LUMO energy levels (See Figure 21).
Compounds
HOMO
LUMO
PP1
PP2
PP3
PP4
Figure 21. Optimized geometries and contour plots of the HOMO and LUMO energy levels
for PP1-PP4.

Finally, the emission properties of all dyes were examined in dichloromethane as the
solvent. Here again, each series displayed a typical fluorescence spectrum, with an emission
centered in the visible range (See Figure 22). Thus, for the dihydro-2-azafluorenones PP6-
PP10, an emission peaking between 568 and 570 nm could be determined. The emission
peaks for PP11-PP13 were found at 574 nm, close to that obtained for the former series
despites the presence of the elongation of the conjugated spacer introduced between the donor
and the acceptor. Upon aromatization of dihydro-2-azafluorenones, no significant change was
found in the PL spectrum of PP10’, displayed an emission maximum superimposing that of
the two previous series. Therefore, it can be concluded that the energetic transition involved
in these different dyes are similar. Interestingly, PP5 that drastically differs from the others by
the absence of the dimethylaminophenyl group could emit at 562 nm. Stokes shift values of
70 and 95 nm were determined for the dihydro-2-azafluorenones PP6-PP10 and the 2-
azafluorenones PP11-PP13 respectively. It can be therefore concluded that a higher electronic
redistribution occurs upon excitation of the 2-azafluorenones relative to the dihydro-2-
azafluorenones.
a)
b)
PP11
PP12
PP13
PP6
PP7
PP8
PP9
PP10
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
500
550
600
650
700
500
550
600
650
700
750
Wavelength (nm)
Wavelength (nm)
d)
c)
1.0
1.0
0.8
0.6
0.4
0.2
0.0
0.8
0.6
0.4
0.2
0.0
500
550
600
650
700
750
500
550
600
650
700
750
Wavelength (nm)
Wavelength (nm)

Figure 22. Photoluminescence spectra of the different dyes recorded in DCM upon excitation
at 490 nm. a) PP6-PP10, b) PP11-PP13, c) PP10’ and d) PP5.
4-Cyclic Voltammetry
The electrochemical properties of all compounds have been investigated by cyclic
voltammetry (CV) in a dilute solution of dichloromethane. All solutions have been deaerated
with argon for 10 minutes prior to CV measurements. Besides this, traces of oxygen could be
still detected on certain CV spectra, attributable to a strong interaction between dyes and
oxygen. In this context, determination of the reduction potentials for several dyes was affected
by the presence of oxygen, underestimating the real energy level of the LUMO orbitals. The
redox potentials of the compounds PP1-PP13 are summarized in the Table 3 in which the
redox potentials are given against the half wave oxidation potential of the ferrocene/
ferrocenium cation couple. The values of the HOMO and LUMO energy levels of the
different chromophores were also determined using as the standard the value of the oxidation
potential of ferrocene (Fc) based on the calculations of Pommerehne et al. (4.8 eV vs.
vacuum). [34] The obtained values of EHOMO and ELUMO are summarized in the Table 3.
Figure 23 allows a comparison of the redox potentials of PP1-PP4 molecules. Notably, PP1
and PP2 possess the same electron acceptor and the oxidation potentials vary with the
contribution of the electron-donating part that includes the electron donor for PP1, but also
the conjugated spacer in the case of PP2.

a)
b)
PP1
PP3
PP1
PP2
-1.5
-1.0
-0.5
0.0
0.5
1.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
E (V/ Fc)
E (V/ Fc)
c)
d)
PP8
PP9
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
E (V/ Fc)
f)
e)
PP13
PP10'
PP12
PP13
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
E (V/Fc)
E (V/ Fc)
g)
PP10'
PP10
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
E (V/ Fc)
Figure 23. Cyclic voltammograms of a) PP1 and PP2, b) PP1 and PP3, c) PP5, d) PP8 and
PP9, e) PP12 and PP13, f) PP10’ and PP10’, g) PP10’ and PP13 in dichloromethane
solutions (10-3 M) with tetrabutylammonium perchlorate (0.1 M) as the supporting electrolyte.
Scan rate: 100 mV/s

As a result of this, elongation of the conjugated spacer facilitates the oxidation process
in PP2 relative to PP1 and a decrease of the oxidation potential of 250 mV was observed (See
Figure 23a). Parallel to this, a modification of the reduction process centered onto EA1 was
also detected, a cathodic shift of 150 mV being evidenced for the reduction peak of PP2
relative to PP1. This cathodic shift of the reduction peak upon elongation of the π-conjugated
spacer is quite unusual, considering that the same electron acceptor is present in the two
molecules. To support this shift, repartition of the respective HOMO and LUMO energy
levels of PP1 and PP2 have to be considered. As shown in the Figure 21, the HOMO levels of
PP1 and PP2 is mainly centered onto the electron-donating group with a contribution of the
spacer. Conversely, the LUMO levels extend over the whole molecule. As a result of this, and
due to the contribution of the electron-rich electron donating group to the LUMO level, a
reduced electron-withdrawing ability is observed for PP2, shifting the reduction peak at lower
potential than that of PP1. While comparing the dyes at identical electron donor, EA1 proved
to be a better electron acceptor than 2-(3-oxo-2,3-dihydro-1H-inden-1-ylidene)malononitrile,
the irreversible reduction of the electron acceptor in PP1 occurring at less cathodic potentials
than for PP3 (-1.27 V vs. -1.35 V). By electrochemistry, an experimental determination of
both the HOMO and the LUMO energy levels can be done from the onset of the oxidation and
the reduction peaks. An electrochemical bandgap can be thus determined. As shown in the
Table 3, a good accordance between the theoretical and the optical bandgaps of PP1-PP4 can
be found. On the opposite, the electrochemical bandgaps underestimate the values of
approximately 0.5 eV for all dyes. A different behaviour was evidenced with PP5 for which
an irreversible oxidation and a reversible reduction process could be evidenced (See Figure
23c). Considering that PP5 is substituted by a piperidine group and that the oxidation process
is centered on the nitrogen atom, an oxidation peak is logically detected at a higher potential
than that observed for PP1-PP4, the oxidation of aliphatic amines being more difficult than
for the aromatic ones.
Compared to PP1-PP4, reduction of PP5 is detected at a more cathodic potential,
attributable to the lower electron-withdrawing ability resulting from a LUMO energy level
extending over the whole molecule and notably the electron-donating piperidine moiety.
Figure 23d presents the voltammograms of PP8 and PP9 which differ only by the nature of
the pendant amine. As evidenced by theoretical calculations, only the nitrogen atom of the
pendant amines contributes to the HOMO and LUMO energy levels (see Figure 17). The
oxidation processes for PP8 and PP9 are centered on the same fragments of the molecule so