Arylamino-fluorene derivatives: Optically induced electron transfer investigation, redox-controlled modulation of absorption and fluorescence

Article abstract of DOI:10.1016/j.dyepig.2020.108325

A series of biarylaminofluorene-based systems with donor-π-donor (D-π-D) structure have been designed and synthesized in order to study the dependence on the π-conjugated bridge length of the intervalence charge-transfer transitions (IV-CT) and of the electronic coupling between the redox centers. To this purpose cyclic voltammetry, UV/Vis-NIR, fluorescence spectroscopy and computational investigations have been carried out to characterize the electronic structure of the compounds in the neutral as well as in the mono- and dication states. Additionally, a study of related D-π compounds has been performed to elucidate the effect of the interaction between two redox centers. Interestingly it was observed that the mono- and dication species exhibit intense transition bands in the NIR region, in the 10000-15000 cm^?1 range, whose intensity depends on the oxidation state and thus it can be reversibly tuned by an applied potential. In a similar way, all compounds show an oxidation state dependent fluorescence which leads to electrofluorochromism. Particularly significant is the mixed valence behavior that provides these systems singular optoelectronic properties, making them excellent active components for electrochromic and electrofluorochromic applications.

Full text of DOI:10.1016/j.dyepig.2020.108325

Journal Pre-proof
Arylamino-fluorene derivatives: Optically induced electron transfer investigation,
redox-controlled modulation of absorption and fluorescence
Agostina-Lina Capodilupo, Francesca Manni, Giuseppina Anna Corrente, Gianluca
Accorsi, Eduardo Fabiano, Antonio Cardone, Roberto Giannuzzi, Amerigo Beneduci,
Giuseppe Gigli
PII:
S0143-7208(20)30340-5
DOI:
https://doi.org/10.1016/j.dyepig.2020.108325
DYPI 108325
Reference:
To appear in: Dyes and Pigments
Received Date: 6 February 2020
Revised Date: 28 February 2020
Accepted Date: 29 February 2020
Please cite this article as: Capodilupo A-L, Manni F, Corrente GA, Accorsi G, Fabiano E, Cardone A,
Giannuzzi R, Beneduci A, Gigli G, Arylamino-fluorene derivatives: Optically induced electron transfer
investigation, redox-controlled modulation of absorption and fluorescence, Dyes and Pigments (2020),
doi: https://doi.org/10.1016/j.dyepig.2020.108325.
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.

GRAPHICAL ABSTRACT
Biarylaminofluorene-based systems with donor-π-donor (D-π-D) structure, with mixed-valence
behavior, exhibit redox-controlled modulation of absorption and fluorescence

1
2
3
Arylamino-fluorene derivatives: optically induced electron transfer
investigation, redox-controlled modulation of absorption and
fluorescence
4
5
Agostina-Lina Capodilupo,a* Francesca Manni,a,b Giuseppina Anna Corrente,c Gianluca Accorsi,a
Eduardo Fabiano,d,e Antonio Cardone,f Roberto Giannuzzi,a Amerigo Beneduci,c* Giuseppe Gigli.a,g
6
a
7
8
Institute of Nanotechnology (CNR-NANOTEC), c/o Campus Ecotekne, via Monteroni, Lecce
73100, Italy.
9
b Dipartimento di Ingegneria dell’Innovazione, Università del Salento, via Monteroni, Lecce 73100,
Italy.
10
11
12
c Department of Chemistry and Chemical Technologies, University of Calabria, Via P. Bucci, Cubo
15D, 87036 Arcavacata di Rende (CS), Italy
d
13
14
Institute for Microelectronics and Microsystems (CNR-IMM), c/o Campus Ecotekne, Via
Monteroni, Lecce 73100, Italy.
e
15
16
Centre for Biomolecular Nanotechnologies @UNILE, Istituto Italiano di Tecnologia (IIT), Via
Barsanti, Arnesano, Lecce 73010, Italy
17
18
f Institute of Chemistry of OrganoMetallic Compounds (CNR-ICCOM), Via Orabona, 4, 70125,
Bary, Italy
g
19
20
Dipartimento di Matematica e Fisica E. de Giorgi, Università Del Salento, Campus Ecotekne, via
Monteroni, Lecce, 73100, Italy
21

1
2
Francesca Manni,a,b E-mail: francesca.manni.fm@gmail.com
Giuseppina Anna Corrente,c E-mail: giuseppina.corrente@unical.it
Gianluca Accorsi,a E-mail: gianluca.accorsi@nanotec.cnr.it
Eduardo Fabiano,d,e E-mail: eduardo.fabiano@cnr.it
3
4
5
Antonio Cardone,f E-mail: cardone@ba.iccom.cnr.it
6
Roberto Giannuzzi,a E-mail: roberto.giannuzzi@nanotec.cnr.it
Giuseppe Gigli,a E-mail: giuseppe.gigli@unisalento.it
7
8
9
Corresponding authors:
10
11
12
13
Agostina-Lina Capodilupo, a* E-mail: agostina.capodilupo@nanotec.cnr.it
Amerigo Beneduci,c* E-mail: amerigo.beneduci@unical.it

1
ABSTRACT
2
3
A series of biarylaminofluorene-based systems with donor-π-donor (D-π-D) structure have been
designed and synthesized in order to study the dependence on the π-conjugated bridge length of the
intervalence charge-transfer transitions (IV-CT) and of the electronic coupling between the redox
centers. To this purpose cyclic voltammetry, UV/Vis-NIR, fluorescence spectroscopy and
computational investigations have been carried out to characterize the electronic structure of the
compounds in the neutral as well as in the mono- and dication states. Additionally, a study of
related D-π compounds has been performed to elucidate the effect of the interaction between two
redox centers. Interestingly it was observed that the mono- and dication species exhibit intense
transition bands in the NIR region, in the 10000-15000 cm-1 range, whose intensity depends on the
oxidation state and thus it can be reversibly tuned by an applied potential. In a similar way, all
compounds show an oxidation state dependent fluorescence which leads to electrofluorochromism.
Particularly significant is the mixed valence behavior that provides these systems singular
optoelectronic properties, making them excellent active components for electrochromic and
electrofluorochromic applications.
4
5
6
7
8
9
10
11
12
13
14
15
16

1
2
GRAPHICAL ABSTRACT
3
4
5
Biarylaminofluorene-based systems with donor-π-donor (D-π-D) structure, with mixed-valence
behavior, exhibit redox-controlled modulation of absorption and fluorescence
6

1
KEYWORDS:
2
3
Mixed-Valence, Fluorene, Triarylamines, Optically induced electron transfer, NIR-
Electrochromism, Electrofluorochromism.
4

1
1. INTRODUCTION
2
3
Organic conjugate compounds have been widely used as active materials in many optoelectronic
applications, including organic dyes solar cells (DSSCs),[1-4] organic light-emitting diodes
(OLEDs),[5-7] electrochromic devices (ECDs),[8-10] and electrofluorochromic devices
(ECFDs),[11-14] organic field-effect transistors (OFETs)[15-19] and as fluorescent materials for
bioimaging applications.[20-25] Since intramolecular electron transfer processes underlie
optoelectronic principles, their investigation and understanding are of crucial importance to afford a
rational design of materials with well-defined optoelectronic characteristics. Since decades, organic
mixed-valence systems (MVs) have been used as model compounds for electron transfer or hole
transfer investigations.[26-35] Typically, MV systems consist of two or more chemically equivalent
redox centers with different formal oxidation states and connected to each other by a saturated or
unsaturated molecular bridge.[36-42] Robin and Day in 1967[10] classified MV systems into three
classes, based on the charge distribution on the molecule and on the strength of electronic coupling
between the redox centers: class-I identifies a system in which the charge is localized on only one
redox center and the redox centers behave as separate entities (electron coupling is zero); class-II
includes the charge-localized systems with weak electron coupling, such that just a small
delocalization of charge between the redox centers occurs and a fraction of charge can be
transferred from one redox center to the other; class-III refers to systems with strong electron
coupling, where the charge is completely delocalized between the redox centers. A number of MV
species were prepared with various π-conjugated bridges having different electronic nature and
various lengths, with the aim to investigate the strength of the electron coupling and the charge
delocalization in these systems.[43, 44] MV systems show unique optoelectronic transitions
associated with intramolecular charge transfer; in particular, the electron coupling optically induced
between the redox centers gives rise to the intervalence charge transfer transition band, known as
IV-CT, that appears in the Vis or NIR region, depending on the nature of the molecular structure
and on the behavior of the redox centers. Arylamine moieties are the most widely used as redox
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

1
2
centers in a number of MVs in order to investigate hole transfer processes, since they are able to
form stable radical cations at low potential, showing excellent reversible redox processes.[33] For
this reason, the arylamino-derivatives are very interesting as hole-transporting organic materials in
organic light-emitting diodes[45-47] and in hybrid and all-organic solar cells.[48-51]
3
4
5
In our previous work,[36, 52] we have designed and synthesized interesting arylamine-based MV
compounds, belonging to class II/III, with an effective electron coupling between the redox centers,
which exhibit a very intense IV-CT band, well separated from other radical transition bands. Such
feature, together with the excellent reversibility of the redox processes, made them appealing for
NIR-electrochromic applications. The aforementioned materials were used as anodic components in
electrochromic devices showing a great ability to switch from almost colorless to colored states,
including the NIR range.[53] A further interesting application field for emissive materials showing
a MV behavior is the electrofluorochromism; as we have demonstrated in our previous work, the
arylamine-fluorene based MV system showed excellent electrofluorochromic properties due to the
stabilization of the radical ions by electron coupling between the redox centers.[13]
6
7
8
9
10
11
12
13
14
15
16
17
On the basis of these last results, in the present work we focused our interest on deepen the study on
biarylaminofluorene-based systems (Fig. 1), going to design and synthesize a series of MV
compounds (I-III) with D-π-D structure, based on two arylamine moieties (D) as redox centers
18
connected by π-conjugated spacers (π) of different length, based on the fluorene system. In order to
19
20
21
22
23
24
25
26
investigate the strength of the electron coupling in the resulting MV radical cations (D-π-D)+,
spacers with a different number of bonds between the redox centers have been used: fluorene for
compound I, phenyl-fluorene-phenyl for compound II and phenylethynyl-fluorene-phenylethynyl
for compound III. Our focus has been to study the dependence of electronic transitions associated
with different intra-molecular charge transfer processes on the nature of the π-conjugated bridge. To
better understand the role of the redox centers coupling in the different molecules, the mono-
arylamine derivatives IV, V and VI have been also prepared as model compounds for the
corresponding bis-arylamino derivatives, I, II and III, respectively. So, they have been synthesized

1
2
using the same π- conjugated moiety used as spacers in the correlated bis-arylamine I, II and III.
Photophysical, electrochemical and time-dependent density functional theory (TD-DFT)
investigations have been carried out to understand the fundamental properties of the sinthesized
materials. Absorption and emission spectroelectrochemistry were performed in order to disclose
their electrochromic and electrofluorochromic potentialities. In detail, here we report the synthesis
and characterization of bis(arylamino)- (I-III) and monoarylamino-fluorene (I-VI) derivatives: 9,9-
dihexyl-N2,N2,N7,N7-tetrakis(4-methoxyphenyl)-9H-fluorene-2,7-diamine (I), 4,4'-(9,9-dihexyl-
9H-fluorene-2,7-diyl)bis(N,N-bis(4-methoxyphenyl)aniline) (II), 4,4'-((9,9-dihexyl-9H-fluorene-
2,7-diyl)bis(ethyne-2,1-diyl))bis(N,N-bis(4-methoxyphenyl)aniline) (III), 9,9-dihexyl-N,N-bis(4-
3
4
5
6
7
8
9
10
11
12
13
14
methoxyphenyl)-9H-fluoren-2-amine
methoxyphenyl)aniline (V),
methoxyphenyl)aniline (VI) (Fig.1).
(IV),
4-(9,9-dihexyl-9H-fluoren-2-yl)-N,N-bis(4-
4-((9,9-dihexyl-9H-fluoren-2-yl)ethynyl)-N,N-bis(4-
15
16
Fig. 1. Molecular structures of MV compounds I, II, III, IV, V and VI.
17
18

1
2
3
2. RESULTS AND DISCUSSION
4
2.1 Synthesis.
5
A straightforward synthetic strategy based on palladium-catalyzed C-N (Buchwald-Hartwig) and C-
C (Suzuki, Sonogashira) cross-coupling reactions was adopted to prepare the final compounds as
shown in scheme 1. 2,7-Dibromo-9,9-dihexyl-9H-fluorene (1) and 2-bromo-9,9-dihexyl-9H-
fluorene (2) were obtained by a C9 alkylation reaction of the corresponding starting reagents with
NatBuO, KI in THF. 9,9-Dihexyl-N2,N2,N7,N7-tetrakis(4-methoxyphenyl)-9H-fluorene-2,7-
diamine (I) and 4-(9,9-dihexyl-9H-fluoren-2-yl)-N,N-bis(4-methoxyphenyl)aniline (IV) were
prepared from compound 1 and compound 2, respectively, which were reacted with bis(4-
methoxyphenyl)amine, in the Buchwald-Hartwig reaction conditions, in the presence of
bis(dibenzylideneacetone)palladium (Pd(dba)2) and tri-tert-butylphosphine (P(t-Bu)3) as catalytic
system, NatBuO as the base, in a microwave reactor. By the Suzuki-Miyamura cross-coupling was
syntesized 4,4'-(9,9-dihexyl-9H-fluorene-2,7-diyl)bis(N,N-bis(4-methoxyphenyl)aniline) (II), by
reacting 2,2'-(9,9-dihexyl-9H-fluorene-2,7-diyl)bis(1,3,2-dioxaborinane) with 4-bromo-N,N-bis(4-
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
methoxyphenyl)aniline (3), in the presence of
tetrakis(triphenylphosphine)palladium(0)
(Pd(PPh3)4) as the catalyst, Na2CO3 2M as the base, in a mixture of toluene:ethanol:H2O (1:05:05,
v:v:v). The same procedure was also used to prepare 4-(9,9-dihexyl-9H-fluoren-2-yl)-N,N-bis(4-
methoxyphenyl)aniline (V), by reacting compound 2 with 4-methoxy-N-(4-methoxyphenyl)-N-(4-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)aniline (6). 4,4'-((9,9-Dihexyl-9H-fluorene-
2,7-diyl)bis(ethyne-2,1-diyl))bis(N,N-bis(4-methoxyphenyl)aniline) (III) and 4-((9,9-dihexyl-9H-
fluoren-2-yl)ethynyl)-N,N-bis(4-methoxyphenyl)aniline (VI) were prepared in the Sonogashira
reaction conditions, by reacting 1 and 2 with compound 5, respectively, using Pd(PPh3)2Cl2/ Ph3P
and CuI as the catalytic system, in a mixture of anhydrous dimethylformamide (DMF) and
triethylamine
(Et3N)
(5:1),
under
microwave
irradiation.
4-Bromo-N,N-bis(4-

1
2
methoxyphenyl)aniline (3) was prepared by reacting bis(4-methoxyphenyl)amine with 4-
bromoiodobenzene, in presence of tris(dibenzylideneacetone)palladium chloroform
3
(Pd2(dba)3·CHCl3) and 1,1'-bis(diphenylphosphino)ferrocene (dppf) as catalytic system, NatBuO as
the base, in toluene. To achieve compound 4 we reacted compound 3 with 2-methyl-3-butyn-2-ol in
presence of bis(triphenylphosphine)palladium(II) dichloride (Pd(PPh3)2Cl2) copper iodide (CuI) and
triphenylphosphine (Ph3P), as catalytic system, in piperidine. Afterwards, compound 4 was
deprotected with potassium hydroxide (KOH) in isopropanol to yield 4-ethynyl-N,N-bis(4-
methoxyphenyl)aniline (5). 4-Methoxy-N-(4-methoxyphenyl)-N-(4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)phenyl)aniline (6) was synthesized from compound 3, treated with n-butyllitium
(n-BuLi) in anhydrous tetrahydrofurane (THF) at -78°C and following was added 2-isopropoxy-
4,4,5,5-tetramethyl-1,3,2-dioxaborolane. Detailed synthetic protocols and product characterization
data of all new compounds are reported in the experimental section and Supporting Information.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

1
2
3
4
5
Scheme 1. Synthetic pathways of compounds I-VI.

1
2
2.2 Neutral state photophysical properties
3
4
The photophysical properties of all the samples (I-VI) have been studied in solvents with different
polarity (increasing dielectric constant: MCH < THF < DCM < ACN < DMSO), in order to
investigate the effect of the environment on their optical properties.
5
6
Figs. 2a-3a show the absorption profiles in DCM (those obtained in other solvents are reported in
the ESI section, Figs. S1-S8), while the photophysical data are reported in Table 1. They are
characterized by two bands in the UV and blue range (c.a. 250-450 nm) that can be mainly assigned
to π−π* and partial charge transfer (CT) transitions, respectively. [54] TD-DFT calculations confirm
this assignment (see Table2) showing, in particular, that the higher level bands always corresponds
to a CT excitation, while those at longer wavelength originate by both π−π* and CT excitations.
For MV compounds I-III, the increase of the absorption intensity (I < II < III, Table 1) can be
ascribed to the increasing conjugation (phenyl and phenylethynyl rings, see scheme 1). The same
trend is found for compounds IV-VI.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

1
2
Fig. 2 a) Absorption and b) Emission spectra of I, II and III in DCM, at room temperature.
3
4
5
6
Fig. 3 a) Absorption and b) Emission spectra of VI, V and VI in CH2Cl2, at room temperature.
For the sample, the absorption spectra are not or slightly affected by the solvent polarity
given the low polarization of the ground state (see Table 1).
7
8
9
10
By contrast, the emission profiles (Figs. 2a-3b) reveal a substantial red shift (up to c.a. 130
nm for III) by increasing the solvent polarity (see Table 1). This indicates that a significant charge
transfer contribution to the emission is present, in particular for samples with longer spacers
(phenylethynyl: III and VI or phenyl: II and V) that show larger effect than I and IV. This trend is
also confirmed by the Stokes-shift parameters (see Table 1), thus highlighting the ability of the
spacer to act as a CT tuner.
11
12
13
14
15
16

1
Table 1. Photophysical properties of compounds I-VI in solution at room temperature.
Sample Solvent
ε / M-1 cm-1
ΦPL
τ/ns
λabs/nm
385
λem/nm ∆λ/ nma
MCH
35.638
34.355
45.807
41.410
31.148
68.093
66.516
64.960
72.759
52.143
90.525
83.053
86.560
84.704
81.445
24.431
22.916
22.362
22.825
22.917
33.834
31.944
35.123
35.866
34.252
39.422
40.748
40.406
39.611
39.615
405
411
417
422
422
419
456
464
497
496
424
475
491
555
535
394
441
431
441
441
408
456
472
501
496
412
472
479
522
524
20
0.38
0.48
0.41
0.53
0.70
0.89
0.72
0.92
0.71
0.83
0.90
0.71
0.65
0.19
0.23
0.16
0.39
0.41
0.40
0.67
0.40
0.81
0.89
0.60
0.75
0.51
0.97
0.65
0.47
0.52
1.0
1.5
1.3
1.7
3.3
1.0
1.9
2.2
2.7
3.3
0.9
2.1
2.4
1.3
1.7
1.0
2.2
1.7
3.3
4.0
1.1
2.8
3.0
3.6
4.5
1.0
2.6
2.8
2.6
3.1
I
THF
383
28
DCM
ACN
385
34
370
52
DMSO 378
44
MCH
THF
373
378
376
373
46
II
III
IV
V
78
DCM
ACN
88
124
116
34
DMSO 380
MCH
THF
390
395
395
391
80
DCM
ACN
96
164
138
50
DMSO 397
MCH
THF
344
345
345
341
96
DCM
ACN
86
100
96
DMSO 345
MCH
THF
348
352
348
348
60
104
124
153
141
46
DCM
ACN
DMSO 355
MCH
THF
366
368
368
364
VI
104
111
158
156
DCM
ACN
DMSO 368
2
3
4
a ss = Stockes shift
Most of the samples shows remarkable photoluminescence efficiencies (up to > 90%) under
different polarity environment. The corresponding decays range from c.a. 1 to 4 nanosecond (Table
5

1
2
1), indicating a gradual increase of the CT character by passing from MCH up to DMSO and ACN,
as a consequence of the larger stabilization of the charge transfer levels in polar media.
3
4
5
Table 2. Absorption energies, oscillator strengths (o.s), main single-particle transitions, and
character
Energy (nm)
o.s.
Character
CT
362
293
292
291
383
286
0.73
0.17
0.36
0.10
0.85
0.21
I
π−π*
CT
CT
CT
II
π−π *
CT
421
315
305
334
287
278
220
354
283
0.94
0.25
0.41
0.66
0.23
0.09
0.21
0.72
0.32
III
π−π *
CT
CT
IV
π−π *
CT
CT
CT
V
π−π *
CT
382
301
280
0.80
0.36
0.29
VI
π−π *
CT
6
7
8
2.3 Electrochemical characterization
9
10
11
12
13
14
Cyclic voltammograms, show reversible oxidations in the potential window between +0 and 1.2 V
vs. Ag/AgCl (Figs. 4a-c, Fig. S10 and Table 3). Specifically, two reversible one-electron oxidations
can be observed for compound I, while the monoamines (IV-VI) and the compounds II and III
show a one-reversible oxidation wave. The monoamines (IV-VI), reference compounds, show a
very similar behavior, characterized by redox processes occurring at 0.69 V, 0.77 V and 0.74 V for
IV, V and VI, respectively, indicating that the length of the π-conjugation does not significantly

1
2
influence the oxidative process. The two redox processes for compound I, ascribed to the radical
cation/neutral (N●+/0) and dication/radical cation (N++/●+) couples, are separated by a potential
splitting of 342.6 mV, suggesting a strong coupling among the amine centers (Table 4). The
stability of the radical cation species (I●+) is clearly evidenced by the lower value of the first
oxidative potential when compared with the reference monoarylamine (IV●+) (0.57 V vs 0.69 V). A
very different situation is found in the CVs of compound II and III, where only one redox process
is detectable. In order to better resolve the two oxidation processes, a DPV analysis was performed.
For compound II the DPV clearly shows two distinctive processes separated by a potential splitting
of 72 mV (Fig. S9); while, for compound III the two oxidation processes were hardly
distinguishable. The oxidative potential of II and its reference monoarylamine V is almost the same
(0.76 V vs 0.77 V), suggesting a weak electron coupling between the two redox centers. This is a
typical example of “potential compression”.[55, 56]. The redox process of III occurs at 80 mV
more positive than the reference monoarylamine, VI. In this case, therefore, the oxidation process
can be considered as n one-step two-electron process, at least from a thermodynamic point of view.
As we will see shortly, this could be an example of the so-called “potential inversion”, [55, 56] for
which the structural rearrangements associated with the one-electron oxidation causes a high
instability of the radical cation with respect to the neutral and dicationic species. From the data
reported above, we note that the increase of the distance between the redox centers, within the series
I-III, causes a destabilization of the radical cation, resulting in a shift towards higher potentials of
the first oxidation.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

1
2
Fig. 4. Cyclic voltammetry of I (a) (50 mV/s); II (b) (50 mV/s); III (c) (50 mV/s) at c = 10-3 M in
CH2Cl2/TBAPF6 (0.1 M) vs AgCl/Ag.
3
Table 3. Electrochemical Properties and Corresponding Energy Levelsa
Egopt
E1/2(1) (mV) E1/2(2) (mV)
b∆E1
cEHOMO
(eV)
dELUMO
(eV)
(eV)
(mV)
-9.2
190.5
267.1
189.1
132
333.4
262.5
342.6
72
-5.08
-5.26
-5.35
-5.28
-5.35
-5.33
-2.25
-2.31
-2.32
-2.08
-2.2
2.83
I
II
III
IV
V
2.95
3.03
3.20
3.14
3.05
/
/
/
/
/
/
/
/
242.8
-2.28
VI
3
b
4
5
ac = 1 × 10− mol/L in CH2Cl2/TBAPF6 (0.1 M) vs. Fc+/Fc at 50 mV/s. ΔE = E1/2(2) − E1/2(1).
d ELUMO = Egopt − EHOMO
cEHOMO = −(Eonset(ox1) + 5.16) (eV).
6
7
8
2.4 Vis/NIR spectra of radical cation species
9
10
11
12
13
14
15
16
17
18
19
20
21
The radical cations of compounds I-VI were generated upon oxidation in stepwise additions of
antimony pentachloride (SbCl5) (Fig. 5 and Fig. S11) and copper perchlorate (CuClO4) (Figs. S12-
S14) in CH2Cl2 and MeCN solutions, respectively, and their UV/vis/NIR changes were
investigated. In addition, the electrochromic and electrofluorochromic properties of compounds I-
III were also investigated by spectroelctrochemical experiments in CH2Cl2 and
tetrabutylammonium hexafluorophosphate (TBAPF6) as supporting electrolyte. (See Fig. S15).
All MV compounds show the rise of a broad absorption band in the range 4000-9000 cm-1 which is
associated with an optically induced hole transfer, IV-CT, from the oxidized arylamine moiety to
the second neutral arylamine unit. For compound I+ this band is rather intense and broad, well
separated from the adjacent radical bands, and centered at 6590 cm-1; instead, for the compounds
II+ and III+ the IV-CT (at 7592 cm-1 and 8536 cm-1, respectively) displays a weaker intensity than
for I+ and an evident overlap with other radical bands, in particular for compound III+. An
analogous behavior is nicely observed during the electrochemical oxidation of the above

1
2
3
4
5
6
7
8
9
compounds (Fig. S15). Actually, the spectroelectrochemical experiment combined with the CVs
(Fig. 4) allowed us to clearly assign the bands of the radical cation and dication species generated at
the working electrode from the neutral species of I and II, while, it was not possible to do the same
for compound III, because the IV-CT band of its radical cation was hardly detectable. However, as
previously shown this band appears of very low intensity and overlapped with the band of the
dication species, due to its strong shift to higher energy in the absorption spectra. The weak
intensity of the IV-CT bands of II+ and III+, is probably due to the low oscillator strengths of the
associated excitations, indicating a weak electronic coupling (see Table 4).
10
11
12
13
14
15
16
17
18
Fig. 5 Absorption spectra of I (a) (1.3× 10−5 M), II (b) (1.7 × 10−5 M), III (c) (1.7 × 10−5 M) and VI
(d) (1.8 × 10−5 M) in CH2Cl2 while adding dropwise a solution of SbCl5 in CH2Cl2 (1 mM).

1
2
3
4
5
6
7
Table 4 Absorption energies, oscillator strengths (o.s), main single-particle transitions, and
character
Energy
6627
o.s.
Character
IVCT
BNCT
LRCT
IVCT
BNCT
π−π
I+
0.803
0.11
0.55
0.10
0.32
0.29
0.13
0.09
0.44
0.29
0.28
0.740
0.37
0.26
0.16
0.42
0.29
0.23
0.55
0.29
0.20
0.13
17825
23041
5627
II+
11919
15083
22727
5733
LRCT
IVCT
BNCT
π−π
III+
10977
14881
20449
2341
LRCT
LRCT
BNCT
π−π*
IV+
13624
15480
24630
11468
15083
27027
10616
14925
19417
24331
LRCT
BNCT
π−π
V+
LRCT
BNCT
π−π
VI+
LRCT
LRCT
8
9
10
The assignment of the IV-CT transition character is supported by the Vis/NIR measurements of
reference compounds IV+-VI+, which do not show low energy transitions at the same energy
frequency in oxidized forms (Fig. 6d and Fig. S11), as well as by the fact that the IV-CT band
disappears when the dication species are generated. A further confirmation for the nature of this
11
12
13

1
2
3
4
5
low-energy band comes from the analysis of the TD-DFT density difference plots (Fig. 6), that
shows the variation of the electron density upon excitation. For all compounds, in fact, the
excitation involves a charge transfer between the two redox centers. Note anyway that in going
from I+ to III+, an increasing charge separation is observed upon excitation. This is quantitatively
described by the ratio between ∆R,[57] i.e. the distance of centers of the excited- and ground-state
6
7
charge densities, and ∆ΝΝ, the geometric distance between the N atoms in the redox centers; at the
TD-DFT level of theory the ratio is computed to be 0.11, 0.90 and 0.92 for I+, II+, and III+,
respectively. This trend suggests that the coupling between the redox centers is diminishing in the
series I+ to III+.
8
9
10
11
12
13
14
15
16
Fig. 6 TD-DFT density difference plots for the lowest energy excitation of radical compounds I+-
III+. The blue regions denote an increase of electron density upon excitation; the red regions denote
a depletion of electron density upon excitation.

1
2
In addition to the IV-CT band, in all radical species a band is observable at about 20000 cm-1, that is
due to long-range charge-transfer excitations (LR-CT). For all MV species an absorption band is
recorded between 10500 cm-1 and 13000 cm-1 and it is associated with the hole transfer from the
oxidized arylamine to the bridge (BN-CT), see Table 5. This transition band for radical I+ is not
present, it starts to rise upon the second oxidation with the generation of the I2+ (see isosbestic point
9860 cm-1). For compounds II+ and III+ the NB-CT band appears already during the first oxidation
process (before the addition of an equivalent of oxidant) and continues to grow during the second
oxidation step with the formation of II2+ and III2+ species, respectively. These absorption bands are
also present, at the same energy frequencies, into vis/NIR spectra of reference compounds V+-VI+.
In addition to the IV-CT and the bridge band, all MV species, as well as the monoarylamine radical
cations, exhibit an additional transition band attributes to π-π* transition typical of arylamine
radical cations which lies in the range 13000-16000 cm-1. In order to analyze in detail the IV-CT
band of the various compounds and identify the class to which the MV belong, a quantitative
analysis of the IV-CT band was performed by adapting the experimental spectra with Gaussian
functions and extrapolating the main parameters that characterize it (Figs. S16-S17). All the IV-CT
parameters are collected in Table 5.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Table 5. Band shape data of the IV-CT band of I+, II+ and III+.
aCH2Cl2
bMeCN
νmax
c
νmax
ϵ [M-1cm-1] ν1/2(HTL)
ν1/2(high)
͂
͂
͂
͂
[cm-1]
6590
[cm-1]
[cm-1]a
3524
[cm-1]b
2936
[cm-1]c
2348
͂
͂
[cm-1]
7336
I+
42919
3902
1.50
0.90
II+
7592
8536
10960
9309
4188
4440
4766d
4766d
4766d
1.00d
1.14d
9099
10126
4673d
4673d
4673d
1.00d
1.06d
III+
a
18
19
20
Acquired using chemical oxidation with SbCl5. b Acquired using chemical oxidation with
CuClO4. c Calculated according to eq 1.
21
22
For I+ only one Gaussian function was employed, while for II+ and III+, due to the overlap with the
adjacent radical band (Fig. S14), a deconvolution with three Gaussian functions was required to

1
2
3
4
5
6
obtain a satisfactory fit (Figs. S16-S17). The spectra used for the band fits were those obtained after
addition of ca 0.25 equiv. of oxidizing agent, in order to minimize overlap with dication bands and
avoid disproportionation of the mixed-valence state.[38, 58] Given the amount of added oxidant, the
molar absorptivities were calculated considering that all of the oxidant added resulted in the
formation of the monocation. The observed width of the IV-CT band (ꢀꢁ ꢂ ꢄꢅꢆꢇ) was compared with
́
⁄ ꢃ
that expected from Hush theory for class-II compounds, ꢀꢁ ꢂ ꢈꢉꢆꢊꢇ eq. 1,
́
⁄ ꢃ
7
8
ꢀꢁ ꢂ ꢈꢉꢆꢊ
́
=
ꢇ
2310 × ꢀ
́
(1)
ꢋ
⁄ ꢃ
ꢌꢍꢎ
where both the ꢀꢁ ꢂ ꢈꢉꢆꢊꢇand ꢀ́ꢌꢍꢎ
are given in cm-1.
́
⁄ ꢃ
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
The asymmetry of the IV-CT band was estimated by the ratio ꢀ́ꢁ ꢂ ꢈꢏꢐꢊꢇ ∕ ꢀ́ꢁ ꢂ ꢑꢄꢒꢇ, where ꢀ́ꢁ ꢂ ꢊꢏꢐꢊ
⁄ ꢃ ⁄ ꢃ ⁄ ꢃ ꢇ
and ꢀꢁ ꢂ ꢑꢄꢒ
́
ꢇ are twice the half-width on high- and low-energy side of the band, respectively. For I+
is narrower than ꢀꢁ ꢂ ꢈꢉꢆꢊ and the IV-CT band shows an evident asymmetry (Fig. S16),
⁄ ꢃ
ꢀꢁ ꢂ ꢄꢅꢆ
́
́
⁄ ꢃ
⁄ ꢃ
ꢇ
ꢇ
these parameters commonly belong to class III or mixed-valence compounds borderline of class
II/III. More specifically, for I+ the ratio ꢀꢁ ꢂ ꢈꢏꢐꢊ
́
ꢇ ∕ ꢀꢁ ꢂ ꢑꢄꢒ
́
ꢇ value is 1.50, comparable to other values of
⁄ ꢃ
⁄ ꢃ
class III systems, therefore we can assign I+ to class III. [33] For II+ and III+ ꢀꢁ ꢂ ꢄꢅꢆ
́
ꢇ are slightly broader
⁄ ꢃ
than the Hush limit and ꢀꢁ ꢂ ꢊꢏꢐꢊ are larger ꢀ́ꢁ ꢂ ꢈꢉꢆꢊꢇ; thus, these IV-CT parameters support the
́
⁄ ꢃ
ꢇ ⁄ ꢃ
assignment of II+ and III+ to class II. [33, 38, 59] [28] As a further test to determine the class of
MV compounds, the solvent-dependence of the IV-CT was estimated recording the oxidation
experiments in two different solvents, CH2Cl2 and MeCN. The IV-CT of I+ showed a modest blue
shift of 746 cm-1, while the IV-CT of II+ and III+ exhibited a more significant hypsochromic shift
of ca. 1507 and 1590 cm-1, respectively. This proves that, the latter systems show a more localized
charge with respect to I+, and therefore the IV-CT is shifted to higher energies due to the increase of
the solvent reorganization energy in MeCN vs CH2Cl2 (Pekar factor γMeCN = 0.526 and γDCM
=
0.381).[60] As expected for MV systems with a larger charge delocalization, I+, the
solvatochromism is less prominent.

1
2
2.5 Electron Coupling
3
4
The electron coupling, V, between diabatic states in MV systems was estimated, according to the
Mulliken-Hush theory, from the equation:
ꢓ
͂
ꢗ
ꢔꢕꢖ ꢘꢙ
5
6
V
=
(2)
͂νmax is the absorption maximum of the IV-
ꢚꢛ
where µge is the transition dipole moment (in Debye) and
7
8
CT, e is the electronic charge and R (in Å) is effective electron-transfer distance.
For class-III systems, the coupling can also be estimated directly by the formula
͂
ꢜ = ꢓ
(3)
ꢔꢕꢖ
9
ꢂ
10
11
12
13
14
About the R value in equation 2, several studies report that the effective electron coupling distance
does not correspond to the geometrical distance between redox centers (N-N), especially in MV
systems where the charge is more delocalized. The R value can be estimated by either Stark
spectroscopy or via semiempirical calculations. In this work, we will consider for the R value both
N-N geometrical distance (RNN) and Rab estimated by the following equation (Ref):
ꢠ
ꢠ
ꢡꢢꢗ
ꢚ
ꢞ∆ꢗ
ꢘꢙ
ꢕꢟ
15
16
ꢝꢍꢅ
=
(4)
where µge is the adiabatic transition moment (experimental or calculated) and ∆µab is the adiabatic
17
18
19
20
21
22
23
dipole moment difference. Since ∆µab is difficult to measure experimentally, we used DFT
computed adiabatic moment differences.[31] The estimate of the electron coupling VNN and Vab
(Table5) has been made by the eq 2, for the three MV systems using the N-N geometric distance
(RNN) and the Rab values, respectively, and considering experimental value of µge and νmax from
Table 6. The equation 3 has been applied only for I+ compound, because it is the only one that
belongs to class III. As expected from previous discussion we find that the electron coupling
increases in the series III+ < II+ < I+.

1
2
3
4
5
Table 6 Transition dipole moment (µge (D), Diabatic electron transfer, Nitrogen-Nitrogen, distance
(RNN), Estimated adiabatic electron-transfer distance (Rab), Electron coupling, V, calculated
according to equation 2 and according the equation 3 (for class-III species).
Compounds
µge (D)
RNN (Å)
Rab (Å)
VNN (cm-1)
Vab (cm-1)
V eq 3
(cm-1)
I+
15.75
8.15
6.81
9.60
6.67
2252
3228
3295
II+
18.13
23.14
10.21
10.64
764
523
1257
1134
-
-
III+
6
7
8
2.6 Chemical fluorescence switching and electrofluorochromism
All compounds are good emissive materials with high quantum yield. They show fluorescence in
the blue and green range, due to the molecular extended π-conjugation and to intramolecular
charge-transfer processes. The fluorescence quenching experiments by chemical oxidation were
carried out in order to investigate the contribution of the electron-coupling between the redox
centers to the fluorescence quenching (Fig. 7 and Figs. S18-S21). As known, the light emission of
arylamine derivatives is quenched by generation of their radical species.[13, 61-63] Through
chemical oxidation with SbCl5 in CH2Cl2, and analyzing the decrease of fluorescence intensity by
linear fitting as a function of the amount of oxidant added, we observed that the presence of two
interacting redox centers further affects the fluorescence quenching. This effect is particularly
evident for compound I, in which a stronger interaction between the redox centers is present.
Comparing the fluorescence quenching trend of I and IV (Figs. 7a-b), we observe that I shows a
larger linear-slope grade than IV (Fig. 7c).
9
10
11
12
13
14
15
16
17
18
19
20

1
2
3
Fig. 7 Fluorescence emission spectra (a) of I (ca 2×10−5 M in CH2Cl2) and (b) of IV (ca 2×10−5 M)
upon SbCl5 titration. (c) Correlation of fluorescent quenching trend of I (red line) and IV (blue line)
vs. equivalent SbCl5 added.
4
5
6
7
The above-mentioned studies have been carried out by exciting I and IV at 345 and 342 nm
(isosbestic point), respectively. Compounds II, III, V and VI also show a fluorescence quenching
upon oxidation process although, the fluorescent variation trend for II and III is comparable with V
and VI compounds, probably due to the relatively weak electron coupling between the redox
centers in these compounds. We have also monitored the fluorescence variation of the
bisdiarylamine and monodiarylamine systems in CH2Cl2 solutions as a function of the voltages (vs
AgCl/Ag). In the system I the fluorescence is almost constant up to about 0.6 V but, above this
value, a stepwise increase of the contrast ratio occurs (Fig. 8a) which increases with the voltage, as
already observed in the similar compound 9,9-bis(2-(2-methoxyethoxy)ethyl)-N2,N2,N7,N7-
8
9
10
11
12
13
14
15

1
2
3
4
5
6
7
8
9
tetrakis(4-methoxyphenyl)-9H-fluorene-2,7-diamine.[13]. In the homologous monoamine system
(IV), a significant fluorescence quenching (high contrast ratio) is observed only at voltages close to
the fluorene oxidation, that is, for potential values larger than that leading to the radical cation. The
last observation is also shared by the other two monoamine systems (Figs. 8b-c). In II and III,
where there is not a clear potential splitting, fluorescence quenching occurs in correspondence to
potential values at which both mono and dication species are quite simultaneously generated (Figs.
4b-c). It is therefore clear that, for practical applications in electrofluorochromism, it is convenient
to use a coupled system like I because a) a significant contrast ratio is achieved at low voltages and
b) the quenching process is fully reversible.
10
11
12
Fig.8 Fluorescence contrast ratio as a function of the applied voltage.
13
14
15

1
2
3
4
5
6
7
3. CONCLUSIONS
We have synthesized six diarylaminofluorene–based compunds, three with D-π-D (I-III) structure
and three with D-π (IV-VI) structure. The D-π systems have been used as model compounds for
their symmetric diarylamine homologous, in order to better characterize the electronic transitions in
the D-π-D systems. The correlation between electronic properties, i.e. the electronic coupling
between the redox centers, and the molecular structure has been investigated, focusing the study on
the mixed-valence (MV) response of the three diarylaminofluorene D-π-D radical cation derivatives
8
9
(I-III) with respect to the π-conjugated bridge and its length. The three MVs compounds I-III show
intense intervalence charge transfer (IV-CT) in the NIR region, which has been exploited to
quantify the electron coupling, V. All compounds showed excellent photophysical properties, such
as high quantum yield (up to > 90%) and high molar absorptivity of the IV-CT (up to > 40000),
with the length of the π-conjugated bridge exploitable as a tool to tune their optoelectronic
properties.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
All compounds exhibited chromatic switching up to NIR by varying their oxidized states. The
spectroelectrochemical measurements revealed also the ability of these compounds to work as
anodic components for electrochromic applications and as promising materials for energy saving.
The fluorescence switching experiments by applied potential highlighted that these class of
materials can modulate the light emission intensity by changing their oxidized states.
The above results show that these compounds are of potential use in a wide range of applications
such as in displays, molecular logic gates, sensors and memory devices.
4. EXPERIMENTAL SECTION
4.1 General Methods
1-Bromohexane, 2-bromo-9H-fluorene, bis(4-methoxyphenyl)amine, 2,7-dibromo-9H-fluorene and
9,9-dihexylfluorene-2,7-bis(trimethyleneborate) were purchased from commercial sources (Sigma

1
2
Aldrich) and used without further purification. All solvents and reagents were used as received,
unless otherwise claimed. The syntheses microwave-assisted were performed with CEM Discover
Labmate reactor. The reactions were monitored by TLC using 0.25 mm silica gel plates with UV
3
1
13
4
indicator (60F-254). The H and C NMR spectra were recorded on a Bruker AVANCE III 400
MHz instrument using chloroform-d (CDCl3) and acetone-d6 as solvents, and splitting patterns
were described as singlet (s), doublet (d), triplet (t), quartet (q) or multiplet (m). Elemental analyses
were done by Carlo Erba CHNS-O EA1108-Elemental Analyzer. LC-MS spectra were acquired
with an Agilent 6300 Series Ion Trap source (APCI).
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
4.2 UV-Vis-NIR absorption and fluorescence spectroscopy
UV-Vis absorption spectra were recorded on Varian-Cary 3000 spectrophotometer. For
luminescence experiments, the samples were placed in fluorimetric 1-cm path cuvettes. Uncorrected
emission spectra were obtained with an Edinburgh FLS980 spectrometer equipped with a peltier-
cooled Hamamatsu R928 photomultiplier tube (185-850 nm). An Edinburgh Xe900 450 W Xenon
arc lamp was used as exciting light source. Corrected spectra were obtained via a calibration curve
supplied with the instrument. Luminescence quantum yields (Φem) in solution obtained from
spectra on a wavelength scale (nm) were measured according to the approach described by Demas
and Crosbya using air-equilibrated [Ru(bpy)3Cl2 in water solution Φem = 0.028, or quinine sulphate
Φem = 0.546][64] as standard. Emission lifetimes in the ps-µs range were determined with the
single photon counting technique by means of the same Edinburgh FLS980 spectrometer using a
laser diode as excitation source (1 MHz, λexc = 407 nm) and a Hamamatsu MCP R3809U-50 (time
resolution 20 ps) as detector. Experimental uncertainties are estimated to be ± 8% for lifetime
determinations, ± 20% for emission quantum yields, ± 2 nm and ± 5 nm for absorption and
emission peaks, respectively.
Redox titrations monitored by UV/Vis/NIR spectroscopy were performed in CH2Cl2 and ACN
solvents using a Varian-Cary 5000 UV-Vis-NIR absorption spectrometer and 10×10 mm quartz-