Efficient greenish blue electrochemiluminescence from fluorene and spirobifluorene derivatives

Article abstract of DOI:10.1021/ja3054018

The spectroscopic and electrochemical behavior as well as electrogenerated chemiluminescence (ECL) of a series of donor-π-donor derivatives bearing triphenylamine groups as donor connected to a fluorene, 2,7-bis-(4-(N,N- diphenylamino)phen-1-yl)-9,9′-dimethylfluorene (1), or spirobifluorene core, 2,7-bis-(4-(N,N-diphenylamino)phen-1-yl)-9,9′-spirobifluorene (2) and 2,2′,7,7′-tetrakis(4-(N,N-diphenylamino)phen-1-yl)-9,9′- spirobifluorene (3), were investigated. Besides a high photoluminescence (PL) quantum yield in solution (between 81 and 87%), an efficient radical ions annihilation process induces intense greenish blue ECL emission that could be seen with the naked eye. Only the tetrasubstituted spirobifluorene derivative (compound 3) shows weak ECL obtained by a direct annihilation mechanism. Because the energy of the annihilation reaction is higher than the energy required to form the singlet excited state, the S-route could be considered the pathway followed by the ECL process in these molecules. The ECL emissions recorded by direct ion-ion annihilation show two bands compared to the single structureless PL band. The ECL spectra obtained by a coreactant approach using benzoylperoxide as a coreagent show no differences relative to that produced by annihilation, except for an increasing of ECL intensity for all compounds.

Full text of DOI:10.1021/ja3054018

Article
pubs.acs.org/JACS
Efficient Greenish Blue Electrochemiluminescence from Fluorene and
Spirobifluorene Derivatives
Federico Polo,*,†,∥ Fabio Rizzo,*,‡,§ Manoel Veiga-Gutierrez,† Luisa De Cola,† and Silvio Quici‡,§
†Physikalisches Institut, Westfalische Wilhelms-Universitat, Heisenbergstrasse 11, D-48149 Munster, Germany
̈
̈
̈
‡Istituto di Scienze e Tecnologie Molecolari (ISTM), CNR, Via Golgi 19, I-20133 Milano, Italy
§Polo Scientifico Tecnologico (PST), CNR, Via Fantoli 16/15, I-20138 Milano, Italy
S
* Supporting Information
ABSTRACT: The spectroscopic and electrochemical behav-
ior as well as electrogenerated chemiluminescence (ECL) of a
series of donor−π−donor derivatives bearing triphenylamine
groups as donor connected to a fluorene, 2,7-bis-(4-(N,N-
diphenylamino)phen-1-yl)-9,9′-dimethylfluorene (1), or spi-
robifluorene core, 2,7-bis-(4-(N,N-diphenylamino)phen-1-yl)-
9,9′-spirobifluorene (2) and 2,2′,7,7′-tetrakis(4-(N,N-
diphenylamino)phen-1-yl)-9,9′-spirobifluorene (3), were in-
vestigated. Besides a high photoluminescence (PL) quantum
yield in solution (between 81 and 87%), an efficient radical
ions annihilation process induces intense greenish blue ECL emission that could be seen with the naked eye. Only the
tetrasubstituted spirobifluorene derivative (compound 3) shows weak ECL obtained by a direct annihilation mechanism. Because
the energy of the annihilation reaction is higher than the energy required to form the singlet excited state, the S-route could be
considered the pathway followed by the ECL process in these molecules. The ECL emissions recorded by direct ion−ion
annihilation show two bands compared to the single structureless PL band. The ECL spectra obtained by a coreactant approach
using benzoylperoxide as a coreagent show no differences relative to that produced by annihilation, except for an increasing of
ECL intensity for all compounds.
instead follow the “T-route” in which the triplet state is first
produced (eq 5), and then the subsequent triplet−triplet
INTRODUCTION
■
Electrogenerated chemiluminescence (or electrochemilumines-
cence, ECL) is a very sensitive technique that can be used to
investigate radical ion formation and the luminescence
produced by annihilation reactions between radical cations
and anions, generated during electrochemical oxidation and
reduction processes.1 Although this technique is similar to the
photoexcitation method, it has the advantage that a light source
is not employed, thus avoiding the scattering phenomenon.2
Since the light intensity is usually proportional to the
concentration of the emitting species, ECL has found a very
wide application in diagnostics as a powerful analytical tool.3
The ECL mechanism can involve excited singlet state formation
by direct annihilation as follows:
1
annihilation generates the singlet excited state R* (eq 6) that
emits light (eq 4).4
3
R
•+ + R•− → R* + R (excited triplet state formation)
(5)
3R* + R* → 1R* + R (triplet−triplet annihilation)
3
(6)
Alternatively to S- and T-routes, ion annihilation reactions
can lead to direct formation of excimers or exciplexes:
R
•+ + R•− → R*2 (excimer formation)
•+ + R′•− → (RR′)* (exciplex formation)
(7)
(8)
R
R − e → R•+ (radical cation formation)
(1)
Another ECL pathway involves adding a reagent (or
coreactant) that can produce a strong reducing or strong
oxidizing agent, depending on its nature. For example, benzoyl
peroxide (BPO) is a commonly used coreactant. Upon
reduction, BPO decomposes to form a strong oxidizing agent
(Eox = +1.5 V vs saturated calomel electrode (SCE)),4,5 which
can react with the radical anion to produce the excited state:
R + e → R•− (radical anion formation)
(2)
R
•+ + R•− → 1R* + R (excited singlet state formation)
(3)
1R* → R + hν (ECL emission)
(4)
where R is the electroactive species. This pathway is defined as
the “S-route” because the singlet is directly generated (eq 3) in
systems with sufficient energy. Systems with insufficient energy
Received: June 4, 2012
Published: August 9, 2012
© 2012 American Chemical Society
15402
dx.doi.org/10.1021/ja3054018 | J. Am. Chem. Soc. 2012, 134, 15402−15409

Journal of the American Chemical Society
Article
Scheme 1. Synthetic Routes To Obtain 1−3
spirobifluorene derivatives showing ECL emission in the blue
region have a low efficiency compared to DPA.15
Herein we report the synthesis, photophysical properties,
electrochemical behavior, and greenish blue ECL of three
organic compounds based on fluorene and spirobifluorene
units, as shown in Scheme 1. In particular, the efficient ECL
pursued by both ion annihilation and coreactant methodologies
is described.
•
−
C6H5CO2 + R•− → R *+ C6H5CO2
(10)
(11)
R* → R + hν
The ECL quantum efficiency (ϕECL) of many polyaromatic
hydrocarbons (PAHs) and organometallic compounds has been
determined by relative methods using 9,10-diphenylanthracene
(DPA) or tris(2,2′-bipyridine)ruthenium(II) (Ru(bpy)32+) as
the standards.6 ECL has been widely investigated in the past 40
years,1−7 but to the best of our knowledge, only a few examples
of quantum efficiency higher than that of DPA have been
published.8,9
Among PAHs, fluorene- and spirobifluorene-based molecules
have been recently studied as suitable candidates for ECL.10
Spirobifluorene is characterized by two fluorene units
connected through the spiro-carbon atom. This spatial
configuration induces strong rigidity and increases the stability
of the system, hampering the oxidation process occurring at
position 9 of the fluorene ring. Moreover, spirobifluorene can
be chemically modified, like fluorene, with donor and/or
acceptor groups in order to tune the emission of the system.
The high versatility of this synthetic approach was recently
exploited, for example, to synthesize molecules bearing a
fluorene and spirobifluorene core that show nonlinear optical11
or high emissive properties.12 Because of their attractive
properties, spirobifluorenes have been proposed as suitable
compounds in the field of optoelectronic devices, including
solar cells13 and organic light-emitting diodes.14 The reported
EXPERIMENTAL SECTION
■
Electrochemistry. Cyclic voltammetry (CV) for the spirobifluor-
ene derivatives reported herein was performed in N,N-dimethylforma-
mide (DMF)/0.1 M tetrabutylammonium hexafluorophosphate
(TBAH). The concentration of the samples was 1.5 mM. Glassy
carbon was employed as the working electrode, platinum wire as the
counter electrode, and silver wire as the quasi-reference electrode
(QRE). DMF (Acros, extra dry, 99.8%) was used as received without
any further purification. TBAH (electrochemical grade, ≥99%, Fluka)
was used as the supporting electrolyte, which was recrystallized from a
1:1 ethanol−water solution and dried at 60 °C under vacuum.
For the electrochemical experiments, a CHI750C electrochemical
workstation (CH Instruments, Inc., Austin, TX) was used. The
electrochemical experiments were performed in a glass cell under an
Ar atmosphere. To minimize the ohmic drop between the working and
the reference electrodes, feedback correction was employed. The
electrochemical experiments were performed by using a homemade
glassy carbon (Tokai Inc.) disk electrode (3 mm diameter). The
working electrode was stored in ethanol and, before experiments, was
polished with a 0.05 μm diamond suspension (Metadi Supreme
diamond suspension, Buehler) and ultrasonically rinsed with ethanol
for 5 min. The electrode was electrochemically activated in the
background solution by means of several voltammetric cycles at 0.5 V/
s between the anodic and the cathodic solvent/electrolyte discharges,
until the same quality features were obtained. The reference electrode
15403
dx.doi.org/10.1021/ja3054018 | J. Am. Chem. Soc. 2012, 134, 15402−15409

Journal of the American Chemical Society
Article
using 2,7-dibromo-9,9′-spirobifluorene20 as the starting materi-
al. Compound 3 was synthesized by a cross-coupling reaction
with 62% yield, starting from the 2,2′,7,7′-tetrabromo-9,9′-
spirobifluorene.21 The purified product was obtained by
washing the crude with pentane.
was a silver QRE, which was separated from the catholyte by a glass frit
(Vycor). The reference electrode was calibrated at the end of each
experiment against the ferrocene/ferricenium couple, whose formal
potential is 0.464 V against the KCl SCE; in the following, all potential
values will be reported against the SCE. A platinum wire served as the
counter electrode.
Electrochemistry. Figure 1 shows CVs of 1−3 in DMF
containing 0.1 M TBAH as the supporting electrolyte. The
ECL. The ECL transients were recorded using an Autolab
PGSTAT101 electrochemical workstation (Metrohm, The Nether-
lands) coupled with a photosensor module with a photomultiplier tube
(PMT, Hamamatsu, H10723-01, Japan). The pulsing potential was set
to 50 mV past the peak potentials of the first oxidation and the first
(and/or the second) reduction waves of the samples investigated. The
pulse width was 0.3 s. The photocurrent produced by the PMT was
directly converted to a voltage signal through the photosensor module
and acquired by the external input channel of the analog-to-digital
converter (ADC) of the Autolab. The transients and the faradic
currents were managed by using the software NOVA provided with
the Autolab. The ECL spectra were acquired using a calibrated
electron-multiplying charge-coupled device (EM-CCD) camera
(Andor Technology, Newton EM-CCD) coupled with a spectrograph
(Andor Technology, Shamrock 163). A homemade glassy carbon
(Tokai Inc.) disk electrode (3 mm diameter) was employed as the
working electrode, which was close to (a few millimeters) and facing
the PMT. The working electrode was mechanically and electrochemi-
cally cleaned as described above for the electrochemical measure-
ments. The reference and counter electrodes employed were the same
as reported above.
Spectroscopy Measurements. Suprasil quartz cells and spec-
trophotometric-grade solvents were employed. Absorption spectra
were measured on a Varian Cary 5000 double-beam UV−vis−NIR
spectrometer and baseline-corrected. Steady-state emission spectra
were recorded on an Edinburgh FS920 spectrometer equipped with a
450-W xenon arc lamp, excitation and emission monochromators (1.8
nm/mm dispersion, 1800 grooves/mm blazed at 500 nm), and a
Hamamatsu R928 PMT. Emission and excitation spectra were
corrected for source intensity (lamp and grating) by standard
correction curves.
The phosphorescence measurements at low temperature were
performed in quartz tubes inserted in a liquid-nitrogen-filled quartz
Dewar. The solvent used was methylcyclohexane, which forms a clear
glass at 77 K. Phosphorescence spectra were measured using a gated
camera and a pulsed laser excitation source. Briefly, a Nd:YAG (355
nm, fwhm 10 ns, 10 Hz, Spectra Physics) pumped an optical
parametric oscillator (MOPO-FDO Spectra Physics) was used to tune
the excitation wavelength (300 nm, 0.5 mJ/pulse). The generated
emission was passed through a cutoff filter (375 nm) placed at 90°
with respect to the excitation, directed through an optical fiber to a
spectrograph (Acton), and detected by an intensified charge-coupled
device camera (iCCD, Princeton Instruments). The delay between
pump laser and high-voltage pulse amplifier (Princeton Instruments)
used to gate iCCD was controlled by a programmable delay pulse
generator (Berkeley Nucleonics). For all samples, accumulation of
phosphorescence was performed in the time window from 1 to 51 ms.
RESULTS AND DISCUSSION
■
Synthesis. Details of the synthetic procedures and
characterizations are given in the Experimental Section. Scheme
1 shows the synthesis of 2,7-bis(4-(N,N-diphenylamino)phen-
1-yl)-9,9′-dimethylfluorene (1),16 2,7-bis(4-(N,N-
diphenylamino)phen-1-yl)-9,9′-spirobifluorene (2),17 and
2,2′,7,7′-tetrakis(4-(N,N-diphenylamino)phen-1-yl)-9,9′-spiro-
bifluorene (3)18 obtained by Suzuki reactions. The fluorene-
based molecule was obtained by reacting 2,7-dibromo-9,9′-
dimethylfluorene19 with 4-(N,N-diphenylamino)phenylboronic
acid in the presence of Pd(PPh3)4 as the catalyst and aqueous
Na2CO3 as the base in THF, leading to 1 with a yield of 60%
after purification by column chromatography. Compound 2 was
prepared with a yield of 60% following the same procedure,
Figure 1. CVs of 1 (a), 2 (b), and 3 (c) in DMF/0.1 M TBAH. The
concentration for each sample is 1.5 mM. Scan rate, 0.1 V/s; working
electrode, glassy carbon; counter electrode, platinum wire; and
reference electrode, silver wire.
standard potential of the redox processes was referenced against
SCE and was calculated as the average value between cathodic
and anodic peak potentials as the scan rate is varied in the range
0.1−5 V/s.
The CVs show only one reversible oxidation peak, attributed
to the oxidation of the triphenylamines at E°ox = 0.95, 0.96, and
15404
dx.doi.org/10.1021/ja3054018 | J. Am. Chem. Soc. 2012, 134, 15402−15409

Journal of the American Chemical Society
Article
0.97 V versus SCE for 1−3, respectively. The ratio between the
anodic and the cathodic currents is approximately 1 down to a
scan rate of 100 mV/s. By comparison with the peaks in the
negative bias, integration of the peak in the positive bias reveals
that the charge flowing in the oxidation process is about double
the value of the reduction process. This would indicate a
bielectronic process involving the simultaneous oxidation of the
two amine units and can be explained by taking into account
the chemical structure of the compounds. In fact, the two
triphenylamine substituents are not well conjugated because of
the tilted phenyl ring between the spirobifluorene (fluorene in
1) core and the electron-donating groups. Thus, a positive
charge in one triphenylamine substituent does not affect the
oxidation potential of the second amine, so that they have the
same value, and a single two-electron wave is observed for the
oxidation.22 On the other hand, with respect to the negative
bias, by comparison of the cathodic scans for the three
compounds, one can notice that 2 shows a shoulder around −2
V, followed by the formation of the radical anion as a reversible
process at −2.15 V, which could be associated with the
presence of the spirobifluorene core. In fact, this feature in the
electrochemical pattern is missing in 1, which is a substituted
fluorene. Two reversible reduction peaks at −2.22 and −2.44 V
(1) and at −2.15 and −2.39 V (2) versus SCE were observed
and associated with the formation of the mono- and dianions.23
In compound 3, only an irreversible reduction peak was
observed at −2.07 V vs SCE. The electrochemical data of all
molecules are summarized in Table 1.
Figure 2. Absorption and emission spectra of 1 (blue dashed line), 2
(black solid line), and 3 (red dash−dot line) in DMF solutions.
Excitation at 350 nm.
Table 2. Optical Data Recorded in DMF Solution
λmax (nm)
a
b
c
c
compd abs
PL
τPL (ns) ϕPL (%) kr (×108/s) knr (×108/s)
1
2
3
371
374
375
451
456
462
1.2
1.3
1.3
81
87
86
6.9
6.8
6.6
1.6
1.0
1.1
a
b
Recorded with λexc = 350 nm. Measured by integrated sphere in
c
dichloromethane. Calculated using the following equations: kr = τPL/
ϕPL, knr = (1 − ϕPL)/τPL.
compounds, the nanosecond deactivation regime of the excited
state is in agreement with singlet emitters, and high
fluorescence quantum yields were recorded in dichloromethane
by means of the integrated sphere, leading to 0.81, 0.87, and
0.86 for 1−3, respectively.
Table 1. Electrochemical Data in DMF/0.1 M TBAH vs SCE
compd
Eox (eV)
E1,red (eV)
E2,red (eV)
ΔE (eV)
1
2
3
0.95
0.96
0.97
−2.22
−2.15
−2.07
−2.44
−2.39
−
3.17
3.11
3.04
ECL. Figure 3 shows the ECL spectra of 1−3 obtained by
pulsing between the oxidation in the positive bias and the
The peak-to-peak separation was 90 mV at 0.2 V/s, which is
larger than expected for an ideal Nernstian behavior (59 mV).24
However, the behavior of the redox couple ferrocene/
ferricenium (Fc+|Fc0), used as internal standard, showed the
same trend. This effect can be attributed to the ohmic drop of
the system, as was previously observed for aprotic media.7a The
peak currents instead were found to be linearly dependent on
the square root of the scan rate, as expected for a diffusion-
controlled redox process.24
Spectroscopic Characterization. Absorption and emis-
sion spectra of 1−3 shown in Figure 2 have been recorded in
the same solvent used for the electrochemical measurements
for the sake of consistency, and the photophysical parameters
are gathered in Table 2.
Figure 3. ECL emission spectra in DMF solution of 1 (red), 2 (black),
and 3 (blue). The ECL emission scale is arbitrary, and the spectra have
been offset for comparison.
All molecules show very similar photophysical behavior. The
absorption bands in the UV region can be assigned to the π−π*
transitions of the entire conjugated backbone25 with the
absorption maxima occurring at 371, 374, and 375 nm for 1−3,
respectively. Another absorption band occurs at 300 nm for all
compounds and could be attributed to the n−π* transition of
triphenylamine groups.26 The lack of shift in the electronic
absorption profile of 2 and 3 indicates that electronic
communication between the two fluorene parts is missing.
Upon excitation at 350 nm, photoluminescence (PL) spectra
were generated with emission maxima at 451, 456, and 462 nm
for 1−3, respectively. The fluorescence spectra follow the
mirror-image rule with a broad, structureless band. In all three
second reduction peak in the negative bias (+1.2 and −2.7 V
for 1, +1.1 and −2.65 V for 2, +1.1 and −2.5 V for 3). Radical
cations and radical anions annihilate to form an excited state
whose subsequent emission signal was integrated for 5 s using a
1 nm slit width. Such a short exposition time employed to
collect the ECL emission spectra suggests that the annihilation
process in these molecules is highly efficient. The ECL
occurring by annihilation of radical anion and cation in 1
shows two maxima, the first at 453 nm and the second at longer
15405
dx.doi.org/10.1021/ja3054018 | J. Am. Chem. Soc. 2012, 134, 15402−15409

Journal of the American Chemical Society
Article
wavelength centered at 484 nm. In 2 the ECL emission
spectrum shows a shoulder at 450 nm and an intense peak at
lower energy with emission maximum at 493 nm. A similar
behavior was reported by Fungo et al.27 for a spirobifluorene
bearing a single diphenylamino group as substituent. The weak
ECL emission of 3 could be ascribed to the instability of the
radical ions and the formation of a polymer film on the
electrodic surface.
The ECL quantum yields were calculated by comparative
methods as the photons emitted per redox event relative to
DPA under similar conditions using DMF as solvent. For 1 and
2 we found higher quantum yields than for DPA (see Table 3).
Table 3. Photophysical and ECL Data Measured in DMF
Figure 4. Comparison of CVs of 2 before (solid black line) and after
(dashed red line) pulsing the potential for several cycles between the
oxidation and the second reduction peaks (+1.1 and −2.65 V vs SCE;
DMF/0.1 M TBAH; scan rate, 0.1 V/s).
λmax (nm)
b
c
f
Es
ΔHann
ϕPL
g
compd abs
PL
ECL
(eV)
(eV)
(%)
ϕECL
d
3.07,
1
371
451 453, 484
2.75
81
3.0
e
e
3.29
a
d
λ, where λ (in nm) is the wavelength at the maximum emission
peak.8 The estimated values for the energy of the first single
excited state are 2.75, 2.72, and 2.68 eV for 1−3, respectively. If
the annihilation energy ΔHann is higher than Es, the system can
be defined as “energy-sufficient”, and the luminescence comes
from a species in a singlet excited state; in this case the process
is known as the singlet or S-route.1−4 ΔHann (in eV)1 can be
calculated from the equation −ΔHann = Eox − Ered − 0.1, i.e.,
based on the difference between the potentials of the oxidation
and the first and second reductions in the CV with an estimated
entropy effect subtracted out (∼0.1 eV).
Upon comparison of the values of Es and −ΔHann, as shown
in Table 3, we can conclude that for all compounds the
annihilation process follows the S-route (−ΔHann > Es). The
light emission generated by the reaction of radical cations and
anions occurs via singlet−singlet annihilation, as the energy
difference between the oxidation and the reduction is larger
than the energy required to populate the singlet excited state
directly.1−4 The three molecules can be defined as energy-
sufficient systems. With a large Stokes shift, a more correct
calculation of the energy of the excited singlet state energy can
be estimated from the crossing point between the absorption
and the emission spectra; in our case, this would yield an
excited singlet state energy of 3.02 eV (at 411 nm for 2), for
which our estimated −ΔHann = 3.01 eV for oxidation and first
reduction would be insufficient to populate it.
Experimentally, we observe that, upon pulsing into the
second reduction peak, where −ΔHann ≈ 3.29 eV, the ECL
emission profile is maintained (vide infra Figure 8b), suggesting
that the same excited state is formed and is thus energy-
sufficient, even for the oxidation and first reduction case. Given
the apparent high ECL efficiency, even with the uncertainties
presented in the calculations, we assign it to an energy-sufficient
system by considering the maximum of the PL emission, as is
usually done for ECL systems.8,22
The ECL emission of 1 and 2 superimposed to the
corresponding PL is depicted in Figures 5 and 6, respectively.
The component of ECL emission at higher energy could be
associated with the radiative decay of the singlet excited state.
The lower energy band could originate from the formation of
side products and excimers or by triplet states.
450 (sh) ,
493
3.01,
3.25
2
3
374
375
456
2.72
87
86
1.7
h
462 ∼490
b
2.68 2.94
−
a
sh = shoulder. Es estimated from fluorescence emission maximum.
−ΔHann = Eox − Ered − 0.1.1 With E1,red. With E2,red. Measured by
c
d
e
f
g
integrated sphere in dichloromethane. Relative to DPA, taking ϕECL
h
of DPA = 1. Too weak to be measured accurately (see inset in Figure
3).
To the best of our knowledge, these are among the highest
values achieved with PAHs emitting in the greenish blue region.
Nevertheless, it is worth mentioning that the ECL efficiency of
DPA was demonstrated to be solvent dependent.28 However,
the stability of ECL of all compounds decreased with the time,
as was also observed in photocurrent decay reported in the
Supporting Information (Figure S1). We hypothesized that this
phenomenon is due to the formation of a thin polymer film on
the electrodic surface. In fact, once the electrode was polished
after this loss of ECL, the intensity of the emission was
restored, corroborating the formation of a thin film over the
electrode while pulsing the potential. To further investigate this
phenomenon, we recorded CVs of 2 before and after pulsing
the potential for several cycles between the oxidation and the
second reduction potential in the positive and negative bias,
respectively. The outcome, reported in Figure 4, clearly shows
that the compound underwent a chemical reaction after pulsing
the potential. Notably, the loss of reversibility involved both the
oxidized and reduced species.
In fact, scanning the potential for several cycles around the
oxidation peak (centered at E0 = 0.96 V) does not produce any
change. The same consideration is valid upon scanning the
potential for several cycles around the first and second
reduction peaks. Therefore, it can be concluded that the
electropolymerization of the triphenylamine, as described in the
literature,29 does not cause the formation of secondary
products. Furthermore, a similar outcome has been reported
in the literature for 3,6-dispirobifluorene-N-phenylcarbazole.15b
The product of the annihilation process formed an insoluble
film on the electrodic surface, which altered the electrochemical
behavior of the compound under investigation, thus supporting
our hypothesis.
To understand which pathway is followed by ECL reaction in
our systems, we calculated the energy of the excited state from
the fluorescence spectrum by the equation Es (eV) = 1239.81/
To rule out possible emission from the triplet states, we
performed phosphorescence measurements at 77 K in a
methylcyclohexane matrix. Figure 7 depicts the structured
15406
dx.doi.org/10.1021/ja3054018 | J. Am. Chem. Soc. 2012, 134, 15402−15409

Journal of the American Chemical Society
Article
investigated in many fluorene-based compounds.31 As reported
in the literature,4 the intensity of ECL emission increases
remarkably in the presence of BPO. This behavior is observed
for all the three compounds. In particular, ECL of 3 with BPO
has been well detected with a shoulder at 465 nm and a
maximum at 488 nm (Figure S2, Supporting Information),
showing a shape very similar to that of 2. Although the ECL
arising in the presence of the coreactant is not decisive, it
strongly suggests that the formation of an excimer is not the
reason for the band peaked at about 490 nm in the spectra of all
three compounds.22a Moreover, the formation of aggregates
due to the high concentrations used in the electrochemical
experiments is also excluded. In fact, the PL emission profile is
maintained in a concentrated solution (10−3 M) in the presence
of the electrolyte, even when exciting at a longer wavelength
(430 nm) where the absorbance is very low (see Figure S3,
Supporting Information). Besides, the PL spectra of thin films
of 1 and 2 (see Figure S4, Supporting Information) indicate the
lack of emission at longer wavelengths also in the solid state.
We can then hypothesize that the ECL emission of these
compounds arises from the radiative decay of the singlet excited
state together with a contribution from a side product formed
after pulsing the potential (vide supra). However, we cannot
exclude a contribution to the ECL emission by the presence of
polarons and/or bipolarons and their recombination.32,33
Further investigations in this direction, which are now beyond
the aim of this work, will be taken into consideration for future
experiments.
Figure 5. PL (black) and ECL (blue) spectra of 1 in DMF/0.1 M
TBAH.
Furthermore, we investigated the effect of the concentration
of radical ions in annihilation process. By pulsing selectively
between the oxidation peak Eox and E1,red or E2,red potential
(+1.2 and −2.3/−2.7 V for 1, and +1.1 and −2.2/−2.65 V for
2), an increasing ECL intensity occurred for all the compounds,
as shown in Figure 8. This behavior could be explained by
taking into account the stoichiometry of radical ions present in
solution.8,22a In fact, in the positive bias the radical dication is
formed, whereas while pulsing in the negative bias, the
monoanion or dianion is produced selectively. Pulsing between
the oxidation and the first reduction peak results in an excess of
radical cation in solution, thus a lower concentration of
emitting species. When the spectrum is collected with the
second reduction peak, the stoichiometry of radical ions is
balanced, and the annihilation reaction of the dication−dianion
pair produces a higher intensity ECL emission.
Figure 6. PL (black) and ECL (blue) spectra of 2 in DMF/0.1 M
TBAH.
phosphorescence spectra for 1 and 2 and clearly shows that the
phosphorescence maxima are at lower energy (λmax = 535 nm)
for both compounds than the ECL lowest energy maximum.
We can therefore conclude that the low-energy emissions
observed in the ECL are not attributed to triplet state
deactivations.
To ensure that the feature of the ECL spectrum was not due
to excimer formation,30 BPO was employed as the coreactant,
and the potential was scanned only in the negative bias.
Unexpectedly, the shape of the emission spectra for all
compounds did not show any difference. This is a very
surprising result, especially in the case of 1, because the
formation of excimers and their ECL emission was already
Figure 7. Normalized PL, ECL, and phosphorescence spectra for 1 (a) and 2 (b).
15407
dx.doi.org/10.1021/ja3054018 | J. Am. Chem. Soc. 2012, 134, 15402−15409

Journal of the American Chemical Society
Article
Figure 8. ECL spectra of (a) 1 and (b) 2 collected by pulsing between the oxidation and the first (black) or second (red) reduction peaks.
network for nanomedicine−RINAME” of MIUR is also
CONCLUSION
■
acknowledged.
We have reported electrochemical and photophysical behavior
of efficient ECL-emitting spirobifluorene- and fluorene-based
molecules bearing triphenylamine as substituents. In solution all
the compounds showed blue emission with high quantum
yields upon photoexcitation and ECL emission in the greenish
blue region, resulting in higher efficiency than that of the
standard 9,10-diphenylanthracene. The most efficient mole-
cules, i.e., 1 and 2, have been carefully analyzed from the
photophysical and electrochemical points of view in order to
understand the origin of the bands that occurred in the ECL
spectra. The coreactant approach using BPO generated more
intense ECL emission in all molecules (also in 3) compared to
the ion annihilation pathway. However, unlike observations for
many other fluorene-based compounds reported in the
literature, no change is observed in the shape of ECL spectra
produced by reduction in the presence of BPO coreactant. We
can suppose that the ECL emission arose from two
components, i.e., the singlet excited state obtained by the S-
route and a side product formed by scanning potential between
anodic and cathodic sweeps. In conclusion, the investigated
molecules demonstrate the possibility to design highly blue-
emitting systems for ECL applications.
REFERENCES
■
(1) Electrogenerated Chemiluminescence; Bard, A. J., Ed.; Marcel
Dekker: New York, 2004.
(2) Richter, M. M. Chem. Rev. 2004, 104, 3003−3036 and references
therein.
(3) Hu, L.; Xu, G. Chem. Soc. Rev. 2010, 39, 3275−3304 and
references therein. Guo, Z.; Hao, T.; Shi, L.; Gai, P.; Duan, J.; Wang,
S.; Gan, N. Food Chem. 2012, 132, 1092−1097. Joshi, T.; Barbante, G.
J.; Francis, P. S.; Hogan, C. F.; Bond, A. M.; Spiccia, L. Inorg. Chem.
2011, 50, 12172−12183. Ge, L.; Yan, J.; Song, X.; Yan, M.; Ge, S.; Yu,
J. Biomaterials 2012, 33, 1024−1031.
(4) Miao, W. Chem. Rev. 2008, 108, 2506−2553 and references
therein.
(5) Santa Cruz, T. D.; Akins, D. L.; Birke, R. L. J. Am. Chem. Soc.
1976, 98, 1677−1682.
(6) (a) Bezman, R.; Faulkner, L. R. J. Am. Chem. Soc. 1972, 94,
6317−6323. (b) Wallace, W. L.; Bard, A. J. J. Phys. Chem. 1979, 83,
1350−1357.
(7) For example: (a) Omer, K. M.; Ku, S.-Y.; Wong, K.-T.; Bard, A.
J. Angew. Chem., Int. Ed. 2009, 48, 9300−9303. (b) Maness, K. M.;
Wightman, R. M. J. Electroanal. Chem. 1995, 396, 85−95. (c) Booker,
C.; Wang, X.; Haroun, S.; Zhou, J.; Jennings, M.; Pagenkopf, B. L.;
Ding, Z. Angew. Chem., Int. Ed. 2008, 47, 7731−7735.
(8) Shen, M.; Rodríguez-Lopez, J.; Huang, J.; Liu, Q.; Zhu, X.-H.;
́
Bard, A. J. J. Am. Chem. Soc. 2010, 132, 13453−13461.
ASSOCIATED CONTENT
■
S
* Supporting Information
(9) Zanarini, S.; Felici, M.; Valenti, G.; Marcaccio, M.; Prodi, L.;
Bonacchi, S.; Contreras-Carballada, P.; Williams, R. M.; Feiters, M. C.;
Nolte, R. J. M.; De Cola, L.; Paolucci, F. Chem.Eur. J. 2011, 17,
4640−4647.
(10) For example: (a) Suk, J.; Cheng, J.-Z.; Wong, K.-T.; Bard, A. J. J.
Phys. Chem. C 2011, 115, 14960−14968. (b) Omer, K. M.; Ku, S.-Y.;
Cheng, J.-Z.; Chou, S.-H.; Wong, K.-T.; Bard, A. J. J. Am. Chem. Soc.
2011, 133, 5492−5499. (c) Omer, K. M.; Ku, S.-Y.; Wong, K.-T.;
Bard, A. J. J. Am. Chem. Soc. 2009, 131, 10733−10741.
(11) Rizzo, F.; Cavazzini, M.; Righetto, S.; De Angelis, F.; Fantacci,
S.; Quici, S. Eur. J. Org. Chem. 2010, 4004−4016.
(12) (a) Saragi, T. P. I.; Spehr, T.; Siebert, A.; Fuhrmann-Lieker, T.;
Salbeck, J. Chem. Rev. 2007, 107, 1011−1065 and references therein.
(b) Liao, Y.-L.; Lin, C.-Y.; Wong, K.-T.; Hou, T.-H.; Hung, W.-Y. Org.
Lett. 2007, 9, 4511−4514. (c) Choi, I. W., ; Kim, C. S.; Kwon, H.-J.;
Cho, Y.-J.; Kim, B.-O.; Kim, S.-M.; Yoon, S. S., Gracel Display Inc.
Luminescent Compounds and Electroluminescent Device Using the
Same. WIPO Patent 2009066815, May 28, 2009. (d) Poriel, C.; Rault-
Detailed synthetic procedure for all compounds; ECL photo-
current of 2; coreactant ECL emission of 3; and PL spectra of 2
in highly concentrated solution excited at different wavelengths.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
federico.polo@unipd.it; fabio.rizzo@istm.cnr.it
Present Address
∥Dipartimento di Scienze Chimiche, Universita
Padova, Via Marzolo 1, I-35131 Padova, Italy
̀
degli Studi di
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Berthelot, J.; Thirion, D.; Barrier
17, 14031−14046.
(13) (a) Heredia, D.; Natera, J.; Gervaldo, M.; Otero, L.; Fungo, F.;
̀
e, F.; Vignau, L. Chem.Eur. J. 2011,
The authors thank Dr. Gregorio Bottaro of ISTM−CNR of
Padova for the PL measurements of concentrated solutions.
Financial support by FIRB project RBAP114AMK “Integrated
Lin, C.-Y.; Wong, K. T. Org. Lett. 2010, 12, 12−15. (b) Cho, N.; Choi,
15408
dx.doi.org/10.1021/ja3054018 | J. Am. Chem. Soc. 2012, 134, 15402−15409

Journal of the American Chemical Society
Article
H.; Kim, D.; Song, K.; Kang, M.-S.; Kang, S. O.; Ko, J. Tetrahedron
2009, 65, 6236−6243.
(14) (a) Wu, C. C.; Lin, Y. T.; Chiang, H. H.; Cho, T. Y.; Chen, C.
W.; Wong, K. T.; Liao, Y. L.; Lee, G. H.; Peng, S. M. Appl. Phys. Lett.
2002, 81, 577−579. (b) Schwartz, G.; Fehse, K.; Pfeiffer, M.; Walzer,
K.; Leo, K. Appl. Phys. Lett. 2006, 89, 083509−511. (c) Chiang, C.-L.;
Wu, M.-F.; Dai, D.-C.; Wen, Y.-S.; Wang, J.-K.; Chen, C.-T. Adv. Funct.
Mater. 2005, 15, 231−238.
(15) (a) Sartin, M. M.; Shu, C.; Bard, A. J. J. Am. Chem. Soc. 2008,
130, 5354−5360. (b) Rashidnadimi, S.; Hung, T. H.; Wong, K.-T.;
Bard, A. J. J. Am. Chem. Soc. 2008, 130, 634−639.
(16) Hosokawa, C.; Funehashi, M.; Kawamura, H.; Arai, H.; Koga,
H.; Ikeda, H., Idemitsu Kosan Company Ltd. Organic Electro-
luminescent Element. Eur. Pat. Appl. 1061112, 2000.
(17) (a) So, H.; Watanabe, H.; Yahiro, M.; Yang, Y.; Oki, Y.; Adachi,
C. Opt. Mater. 2011, 33, 755−758. (b) Komino, T.; Nomura, H.;
Yahiro, M.; Endo, K.; Adachi, C. J. Phys. Chem. C 2011, 115, 19890−
19896.
(18) Togushi, S.; Oda, A.; Ishikawa, H., NEC Corp. Organic Thin
Film Transistor. U.S. Patent 2003111692, June 19, 2003.
(19) Saroja, G.; Pingzhu, Z.; Ernsting, N. P.; Liebscher, J. J. Org.
Chem. 2004, 69, 987−990.
(20) Yu, W.-L.; Pei, J.; Huang, W.; Heeger, A. J. Adv. Mater. 2000, 12,
828−831.
(21) Pei, J.; Ni, J.; Zhou, X.-H.; Cao, X.-Y.; Lai, Y.-H. J. Org. Chem.
2002, 67, 4924−4938.
́
(22) (a) Shen, M.; Rodríguez-Lopez, J.; Lee, Y.-T.; Chen, C.-T.; Fan,
F.-R. F.; Bard, A. J. J. Phys. Chem. C 2010, 114, 9772−9780. (b) Omer,
K. M.; Ku, S.-Y.; Chen, Y.-C.; Wong, K.-T.; Bard, A. J. J. Am. Chem.
Soc. 2010, 132, 10944−10952.
(23) Vonlanthen, D.; Rudnev, A.; Mishchenko, A.; Kaslin, A.;
̈
Rotzler, J.; Neuburger, M.; Wandlowski, T.; Mayor, M. Chem.Eur. J.
2011, 17, 7236−7250.
(24) Bard, A. J.; Faulkner, L. R. Electrochemical Methods,
Fundamentals and Applications, 2nd ed.; John Wiley & Sons: New
York, 2001.
(25) Valeur, B. Molecular Fluorescence: Principles and Applications;
Wiley-VHC: Weinheim, Germany, 2001.
(26) Promarak, V.; Ichikawa, M.; Sudyoadsuk, T.; Saengsuwan, S.;
Keawin, T. Opt. Mater. 2007, 30, 364−369.
(27) Fungo, F.; Wong, K.-T.; Ku, S.-Y.; Hung, Y.-Y.; Bard, A. J. J.
Phys. Chem. B 2005, 109, 3984−3989.
(28) Maness, K. M.; Bartelt, J. E.; Wightman, R. M. J. Phys. Chem.
1994, 98, 3993−3998.
(29) (a) Seo, E. T.; Nelson, R. F.; Fritsch, J. M.; Marcoux, L. S.;
Leedy, D. W.; Adams, R. N. J. Am. Chem. Soc. 1968, 90, 3498−3503.
(b) Nelson, R. F.; Adams, R. N. J. Am. Chem. Soc. 1968, 90, 3925−
3930. (c) Otero, L.; Sereno, L.; Fungo, F.; Liao, Y.-L.; Lin, C.-Y.;
Wong, K.-T. Chem. Mater. 2006, 18, 3495−3502.
(30) Choi, J.-P.; Wong, K.-T.; Chen, Y.-M.; Yu, J.-K.; Chou, P.-T.;
Bard, A. J. J. Phys. Chem. B 2003, 107, 14407−14413.
(31) Prieto, I.; Teetsov, J.; Fox, M. A.; Vanden Bout, D. A.; Bard, A. J.
J. Phys. Chem. A 2001, 105, 520−523.
(32) Burrows, H. D.; Seixas de Melo, J.; Forster, M.; Guntner, R.;
̈
Scherf, U.; Monkman, A. P.; Navaratnam, S. Chem. Phys. Lett. 2004,
385, 105−110.
(33) Zaikowski, L.; Kaur, P.; Gelfond, C.; Selvaggio, E.; Asaoka, S.;
Wu, Q.; Chen, H.-C.; Takeda, N.; Cook, A. R.; Yang, A.; Rosanelli, J.;
Miller, J. R. J. Am. Chem. Soc. 2012, 134, 10852−10863.
15409
dx.doi.org/10.1021/ja3054018 | J. Am. Chem. Soc. 2012, 134, 15402−15409