Synthesis, photoluminescence, and electroluminescence characterization of double tetraphenylethene-tethered BODIPY luminogens

Article abstract of DOI:10.1002/jccs.201900163

Three double tetraphenylethene (TPE)-tethered 4-difluoro-4-bora-3a,4a-diaza-s-indance (BODIPY) fluorophores, 35TPEBODP, 88TPEBODP, and 26TPEBODP, have been synthesized and characterized. The green 35TPEBODP with deep red fluorescence shows serious thermal decomposition in the purification process of sublimation, which prohibits its test for an organic light-emitting diode (OLED) fabricated by the vacuum–thermal evaporation process. The tethered TPE is attached to BODIPY at three different positions, resulting in different photoluminescence (emission wavelength and quantum yield) and electroluminescence (EL). Different from TPE-tethered BODIPY fluorophores reported in literature, none of the BODIPY fluorophores studied here exhibits aggregation-induced emission (AIE), aggregation-induced enhanced emission (AIEE), or twisted intramolecular charge transfer (TICT) characteristics. Although solution (10^?5 M THF) photoluminescence quantum yields (?s) are relatively high at 78%, 68%, and 86% for 35TPEBODP, 88TPEBODP, and 26TPEBODP, respectively, which are all higher than 41% of PhBODP (a non-TPE-tethered BODIPY), the ? is significantly decreased to 1–6% in 5 wt% dopant polystyrene thin film or as a solid powder, except for 13% of 26TPEBODP. Therefore, due to the low ? of dopant thin film or solid powder, either dopant or nondopant OLEDs exhibit inferior external quantum efficiency (EQE) and intensity of EL. The best OLED in this study is the 26TPEBODP device, and its EQE reaches 1.3%, and the highest EL intensity is approximately 1,600 cd/m².

Full text of DOI:10.1002/jccs.201900163

Received: 27 April 2019
Revised: 9 July 2019
Accepted: 10 July 2019
DOI: 10.1002/jccs.201900163
S P E C I A L I S S U E P A P E R
Synthesis, photoluminescence, and electroluminescence
characterization of double tetraphenylethene-tethered BODIPY
luminogens
Chiun-Jen Yang1 | Jian Haur Lee2 | Chin-Ti Chen1,2
1Department of Applied Chemistry, National
Chiao Tung University, Hsinchu,
Taiwan, ROC
2Institute of Chemistry, Academia Sinica,
Taipei, Taiwan, ROC
Three double tetraphenylethene (TPE)-tethered 4-difluoro-4-bora-3a,4a-diaza-s-
indance (BODIPY) fluorophores, 35TPEBODP, 88TPEBODP, and 26TPEBODP,
have been synthesized and characterized. The green 35TPEBODP with deep red
fluorescence shows serious thermal decomposition in the purification process of
sublimation, which prohibits its test for an organic light-emitting diode (OLED)
fabricated by the vacuum–thermal evaporation process. The tethered TPE is
attached to BODIPY at three different positions, resulting in different photo-
luminescence (emission wavelength and quantum yield) and electroluminescence
(EL). Different from TPE-tethered BODIPY fluorophores reported in literature,
none of the BODIPY fluorophores studied here exhibits aggregation-induced emis-
sion (AIE), aggregation-induced enhanced emission (AIEE), or twisted intramolec-
ular charge transfer (TICT) characteristics. Although solution (10−5 M THF)
photoluminescence quantum yields (ϕs) are relatively high at 78%, 68%, and 86%
for 35TPEBODP, 88TPEBODP, and 26TPEBODP, respectively, which are all
higher than 41% of PhBODP (a non-TPE-tethered BODIPY), the ϕ is significantly
decreased to 1–6% in 5 wt% dopant polystyrene thin film or as a solid powder,
except for 13% of 26TPEBODP. Therefore, due to the low ϕ of dopant thin film
or solid powder, either dopant or nondopant OLEDs exhibit inferior external quan-
tum efficiency (EQE) and intensity of EL. The best OLED in this study is the
26TPEBODP device, and its EQE reaches 1.3%, and the highest EL intensity is
approximately 1,600 cd/m2.
Correspondence
Chin-Ti Chen, Institute of Chemistry,
Academia Sinica, Taipei 11529, Taiwan,
ROC.
Email: chintchen@gate.sinica.edu.tw
Funding information
Innovative Materials and Analysis
Technology Exploration Program, Academia
Sinica, Grant/Award Number: AS-iMATE-
108-33; Ministry of Science and
Technology, Taiwan, Grant/Award Number:
MOST 106-2113-M-001-016-MY3;
National Chiao Tung University
K E Y W O R D S
BODIPY, electroluminescence, OLED, photoluminescence, tetraphenylethene
em
λmax of 506 nm and 95% of photoluminescence quantum
1 | INTRODUCTION
yield (PLQY) in methanol solution.[1] However, these highly
fluorescent BODIPY derivatives often suffer from serious
emission quenching in solid state, causing diminishingly
small PLQY and limiting their applications in organic opto-
electronics, such as organic lighting-emitting diodes
(OLEDs).
4,4-Difluoro-4-bora-3a,4a-diaza-s-indance (BODIPY) is
boron difluoride chelated by a π-conjugated dipyrromethene,
exhibiting an intense fluorescence in solution. For example,
one of the BODIPY derivatives, PM 546 (Scheme 1), a com-
mercially available green laser dye, has been reported with a
© 2019 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J Chin Chem Soc. 2019;1–12.
http://www.jccs.wiley-vch.de
1

2
YANG ET AL.
deposition process. We try to study the electroluminescence
(EL) and understand or clarify the photoluminescence
(PL) of these BODIPYs, of which double-tethered TPE are
at three different positions, that is, 8 and 8, 3 and 5, and
2 and 6.
2 | RESULTS AND DISCUSSIONS
2.1 | Synthesis
Previously known PhBODP and 26TPEBODP were readily
prepared based on procedures described in the litera-
ture.[5,7,11] The synthesis of 88TPEBODP is shown in
Scheme 2. A key intermedia, 3,5-bis[1,2,2-triphenylvinyl]
benzaldehyde, an unknown new compound, was synthesized
by the Suzuki cross coupling of bromotriphenylethylene and
3,5-bis(pinacolboranyl)benzaldehyde, which was synthe-
sized following a conventional procedure of a two-step reac-
tion from tribrobenzene in moderate yields.
The synthesis of 35TPEBODP is shown in Scheme 3. It
was prepared using microwave heating via the Knövenagel
condensation of PhBODP and 4-(1,2,2-triphenylvinyl)benz-
aldehyde, which was facilely synthesized following modi-
fied procedures described in the literature,[12,13] and its
structural characterization was consistent with that given in
the literature.[12]
SCHEME 1 Chemical structure of BODIPY luminogens in this
study
Tetraphenylethene (TPE) has been known as one of the
most potent chemical moieties to alleviate emission
quenching in solid state through the so-called effect of
aggregation-induced emission (AIE).[2,3] In light of the AIE,
a chemical structure hybrid of TPE and BODIPY seems to
be a logical approach to overcome the problem of emission
quenching in solid state. In fact, there are already several
reports of TPE-decorated BOPIPY luminogens, including
26TPEBODP (Scheme 1), in the literature.[4–9] However,
these TPE-decorated BOPIPY luminogens show moderate to
intense emission in organic solutions without water. There-
fore, instead of AIE, aggregation-induced enhanced emis-
sion (AIEE) or even aggregation-caused quenching (ACQ)
is probably a better term for characterizing such TPE-
containing luminogens. Furthermore, compared with TPE-
undecorated BODIPY, some of these TPE-containing
BODIPY luminogens exhibit color-shifted red fluorescence
in solid state, simply due to the ππ interaction of molecular
stacking, or it is ascribed to a twisted intramolecular charge
transfer (TICT).[8,10] AIEE or ACQ, as well as ππ interaction
or TICT, of TPE-decorated BOPIPY luminogens hinges on
the position (and maybe the number) of the decorated
(or “tethered” used in this study) TPE moiety. Nevertheless,
except for one case, none of them has been applied for
OLEDs. Only the triple TPE-tethered BODIPYs have been
fabricated as a polymer-blended OLED, a dopant-based
device, from a solution process,[9] rather than a vacuum–
thermal deposition one.
SCHEME 2 Synthesis of 88TPEBODP
Accordingly, in this study, we have synthesized and char-
acterized two new double TPE-decorated BOPIPY
luminogens, 88TPEBODP and 35TPEBODP (Scheme 1),
together with previously known 26TPEBODP, which is
also a double TPE-tethered BODIPY.[5,7] In addition, for the
purpose of comparison, PhBODP, a non-TPE-tethered
BODIPY, was also adopted in the dopant and nondopant
OLEDs, which were all fabricated by the vacuum–thermal
SCHEME 3 Synthesis of 35TPEBODP

YANG ET AL.
3
decomposition because the intact 35TPEBODP should
show a dark green color (see Figure 2). Such a result mostly
prohibits its OLED fabrication with a vacuum–thermal depo-
sition process. We infer that the more extended molecular
shape and the flexible vinyl connection of 35TPEBODP
reduces the volatility of the material, which needs a higher
temperature to be sublimed and causes extensive decomposi-
tion of the material. The extended molecular shape of
26TPEBODP also hinders the evaporation process of subli-
mation. A time period almost three to four times longer was
required to complete the purification process. This also
resulted in a prolonged OLED fabrication process, and a par-
tial thermal decomposition was also observed for
26TPEBODP.
2.2 | Thermal stability
Thermal stability is one of the vital factors that validate the
feasibility of vacuum–thermal deposition in device fabrica-
tion. Figure 1 illustrates the TGA thermograms of three
TPE-tethered BODIPY compounds in this study.
Although the TGA results show similar thermal stability
of the three BODIPY compounds, the purification process of
gradient sublimation afforded 35TPEBODP with serious
2.3 | Photoluminescence characterization
2.3.1 | 35TPEBODP
Solution PL of 35TPEBODP is concentration dependent.
First, compared with 506 nm of PM546 or 519 nm of
PhBODP, 35TPEBODP exhibits a significantly red-shifted
em
λmax of 675 and 725 nm in THF solution of a concentra-
tion of 10−5 and 10−3 M, respectively. This is mainly due to
the effective extension of the π conjugation of the BODIPY
core via a vinyl group connected to TPE units. At a high
concentration of 10−3 M, the 35TPEBODP THF solution
exhibits a simple and broad deep red emission spectra with a
FIGURE 1 TGA thermograms of three TPE-tethered BODIPY
compounds
em
λmax of around 725 nm at a water volume fraction (vol%)
of 0%, 20%, and 40% and slightly red-shifted to approxi-
mately 750 nm at water vol% of 60, 70, and 80%, respec-
tively (Figure 3). Regarding fluorescence intensity, 60% is
the common water vol% at which the brightness of the fluo-
rescence photograph and the intensity of the fluorescence
spectra become significantly decreased. We can infer that
60% is the water vol% at the beginning of the formation of
the aggregate of 35TPEBODP in THF solution. Such a PL
behavior is a typical ACQ, which is neither AIE nor AIEE.
At a low concentration of 10−5 M, spectra become quite
different (Figure 4). First, there is an emission band with a
em
λmax of around 735–740 nm, which is not observed in the
solution with a high concentration of 10−3 M. Based on a
vibronic shoulder band observed in the absorption spectra
(not shown here), this emission band can be attributed to the
em
vibronic emission of the main emission with λmax of
around 677 nm at water vol% of 0, 20, and 40%. These are
all blue-shifted spectra compared with those of high concen-
tration (10−3 M). Unlike those of high concentrations of
10−3 M, these fluorescence intensities (water vol% of 0, 20,
and 40%) are almost same. The more diluted concentration
prevents the molecules of 35TPEBODP from ππ interaction,
rendering better-resolved, similar intensity, and blue-shifted
spectra. Nevertheless, similar to that of a high concentration
FIGURE 2 Photograph (under visible light and 366 nm UV
light) of gradient sublimation of PhBODP and three TPE-tethered
BODIPY compounds. Color pictures on the right are the corresponding
5 wt% dopant samples in polystyrene thin film

4
YANG ET AL.
FIGURE 4 Fluorescence spectra (top) and fluorescence
photograph (bottom) of 35TPEBODP in THF solution (10−5 M) with
various water volume fractions. Fluorescence spectra were recorded
with a 600 nm excitation wavelength. Fluorescence photographs were
taken under a UV illumination at 366 nm
FIGURE 3 Fluorescence spectra (top) and fluorescence
photograph (bottom) of 35TPEBODP in THF solution (10−3 M) with
various water volume fractions. Fluorescence spectra were recorded
with a 600 nm excitation wavelength. Fluorescence photographs were
taken under a UV illumination at 366 nm
water vol% (Figure 5), of which fluorescence intensity is dis-
played in a log scale for clarity. However, at high water vol
%, a new emission band with a longer wavelength of around
625 nm, as well as a new emission band with shorter wave-
length between 510 and 525 nm, appears. Different from the
literature precedents,[8,10] we attribute such new emission
bands to a J-type and H-type aggregate of 88TPEBODP for
the longer and shorter wavelengths, respectively. Particularly
for water vol% of 70 and 80%, the emerging new emission
solution, the intensity of the emission decreases significantly
when the water vol% increases to 60%, although the wave-
length is slightly blue-shifted. The intensity decreases fur-
em
ther, and the wavelength of λmax
is red-shifted to
approximately 699 nm with a higher water vol% of 70%.
However, at water vol% of 80%, such a red emission is blue-
shifted slightly to approximately 690 nm (in terms of energy,
the difference between 690 and 699 nm is rather small), and
the intensity of the emission becomes moderately stronger
(Figure 4). Similar to that of high concentration, we suggest
that the solution PL of 35TPEBODP exhibits ACQ even at
diluted concentrations of 10−5 M. The two vinyl-connected
TPE units at the 3, 5 position cannot overcome the molecu-
lar ππ interaction that quenches the fluorescence of
35TPEBODP.
em
band with a λmax of approximately 620 nm, together with
an intensity-declined fluorescence of around 550 nm, results
in a visible fluorescence color change from green to orange
(Figure 5).
At a low concentration of 10−5 M, the 88TPEBODP
THF solution exhibits much simpler green emission spectra,
em
showing a λmax of around 516 nm at water vol% of 0, 20,
40, and 60%. A weaker emission band with λmaxem of around
550 nm can be assigned to the vibronic emission side-band
of λmaxem of around 516 nm. All these four emission spectra
have a rather similar intensity. At higher water vol% of 70%
and 80%, this green emission band slightly red-shifts to
522 nm, but its intensity decreases significantly. The
vibronic emission side-band is still recognizable at around
560 nm in these spectra. However, a new emission band
2.3.2 | 88TPEBODP
The solution PL of 88TPEBODP is also concentration
dependent. At
a
high concentration of 10−3 M,
em
88TPEBODP THF solution exhibits green emission λmax
of around 543 nm with 0, 20, and 40% of water vol% and
progressively red-shifts to 550 nm with 60, 70, and 80%

YANG ET AL.
5
FIGURE 5 Fluorescence spectra (top) and fluorescence
photograph (bottom) of 88TPEBODP in THF solution (10−3 M) with
various water volume fractions. Fluorescence spectra were recorded
with a 480 nm excitation wavelength. Fluorescence photographs were
taken under a UV illumination at 366 nm
FIGURE 6 Fluorescence spectra (top) and fluorescence
photograph (bottom) of 88TPEBODP in THF solution (10−5 M) with
various water volume fractions. Fluorescence spectra were recorded
with a 480 nm excitation wavelength. Fluorescence photographs were
taken under a UV illumination at 366 nm
emerges around 605 nm (Figure 5). Similarly, we attribute
the newly emerging emission band to the J-type aggregate of
88TPEBODP. Presumably, there should be an emission
band of around 500–515 nm due to the H-type aggregate of
88TPEBODP, which is barely distinguishable in the spec-
solution simply rule out the possibility of TICT, which was
used to explain the color change of PL of TPE-containing
BODIPY derivatives.[8]
em
tra. As the emission spectra are predominated by the λmax
of 522 nm, a greenish color remains in the fluorescence pho-
tograph regardless of water vol%, except for a slight yellow-
ish hue of the THF solution with 70% and 80% of water vol
2.3.3 | 26TPEBODP
Unlike those of 35TPEBODP or 88TPEBODP, the fluores-
cence spectra are much simpler at either high or low concen-
trations. First, with increasing water vol%, the shifting of
spectroscopic wavelength (λmax
% (Figure 6).
From either fluorescence photograph or the emission
spectra, 60% (10−3 M) or 70% (10−5 M) seems to be a com-
em
)
is very small,
mon water vol% at which the brightness of the photograph
and the intensity of spectra decrease. Similar to what hap-
pens for 35TPEBODP, we infer that this is the water vol%
that initiates the formation of a substantial aggregate of
88TPEBODP in THF solution. Similarly, based on the PL
study described here, 88TPEBODP is an ACQ (not AIE nor
AIEE) luminogen. From the solution PL study of
88TPEBODP, we have observed a color change, which is
clearly visible at a high concentration of 10−3 M but is less
clear at 10−5 M. Finally, the multiple emission bands
observed for 88TPEBODP in high water-containing THF
606–610 nm and 577–583 nm for a high concentration of
10−3 M and a low concentration of 10−5 M, respectively
(Figures 7 and 8). This is consistent with the yellowish
orange color in THF solution regardless of water content
(vol%). Second, the H-type or J-type aggregate mentioned
above for 88TPEBODP simply does not occur based on the
profile of recorded fluorescence spectra. However, the varia-
tion of fluorescence intensity is more subtle and requires fur-
ther elaboration.
For a high-concentration (10−3 M) THF solution
(Figure 7), the florescence intensity first increases

6
YANG ET AL.
FIGURE 8 Fluorescence spectra (top) and fluorescence
FIGURE 7 Fluorescence spectra (top) and fluorescence
photograph (bottom) of 26TPEBODP in THF solution (10−3 M) with
various water volume fractions. Fluorescence spectra were recorded
with a 480 nm excitation wavelength. Fluorescence photographs were
taken under a UV illumination at 366 nm
photograph (bottom) of 26TPEBODP in THF solution (10−5 M) with
various water volume fractions. Fluorescence spectra were recorded
with a 480 nm excitation wavelength. Fluorescence photographs were
taken under a UV illumination at 366 nm
substantially with 20% of water vol%. This can be explained
by the AIEE characteristic of 26TPEBODP. However,
when the water vol% further increases to 40%, the flores-
cence intensity decreases significantly. At first, we infer that
ACQ surpasses AIEE at the water vol% of 40%. Neverthe-
less, the florescence intensity further decreases at water vol
% of 60% and 70% but bounces back and beyond at water
vol% of 80%. Based on a pertinent literature report,[6] the
aggregate formation at the early stage, including water vol%
of 40%, 60%, and 70%, has a crystalline nature, and the for-
mation of an amorphous aggregate takes place at a higher
water vol% of 80%.
10−3 M disappears when the solution concentration is lower
at 10−5 M. Therefore, once again, the solution PL of
26TPEBODP is concentration dependent, similar to that of
35TPEBODP and 88TPEBODP.
2.4 | Fluorescence quantum yields
With the correction of refractive index difference (refractive
index of THF solution containing 80 vol% water is 1.350),
the solution fluorescence quantum yields were determined
relative to that of PM546 (ϕ = 78% in THF)[1] for PhBODP
and 88TPEBODP, of H2TPP (ϕ = 11% in benzene)[14] for
35TPEBODP, and of rubrene (ϕ = 61% in toluene)[15] for
26TPEBODP. Table 1 summarizes the fluorescence quan-
tum yield (ϕ) data of BODIPY compounds in solution. For
BODIPY compounds in solid state, as powder, or as 5 wt%
dopant in PS-hosted thin film, ϕ data were measured by the
integrating sphere equipped in a Hitachi fluorescence spec-
trophotometer F-7000.
Such a variation of fluorescence intensity is somewhat
different when the concentration is lower at 10−5
M
(Figure 8). The first reduced and then bounced back fluores-
cence intensity is still there but occurs at a higher water vol
% of 60–80% due to a more diluted 26TPEBODP in the
solution. A greater difference was found for high- and low-
concentration solutions at low water vol%. Very different
from those of high-concentration solutions, the fluorescence
intensity of the low-concentration solution remains nearly
unchanged at water vol% of 0, 20, and 40%. The plausible
AIEE observed for 26TPEBODP at a high concentration of
First, high-concentration solutions always show lower ϕ
than low-concentration solutions, either with or without teth-
ered TPE, regardless the position of tethered-TPE. Such
results indicate that the TPE tethering cannot prevent the

YANG ET AL.
7
TABLE 1 Fluorescence quantum yields of PhBODP and three
TPEBODP compounds in this study
ϕsolutiona (%)
ϕsolutionb (%)
ϕsolidc (%)
THF solution
80% H2O vol%
THF solution
80% H2O vol%
5 wt%
100 wt%
88TPEBODP
35TPEBODP
26TPEBODP
PhBODP
68
33
78
8
59
6
5
2
30
1
5
1
86
44
41
8
24
12
28
6
13
6
3
1
FIGURE 9 Energy level diagram of four BODIPY compounds
and materials used in device fabrication
a1 × 10−5 M.
b1 × 10−3 M.
cIn polystyrene (PS) thin film.
layer of NPB and TPBI, even though the choice of CBP as
the host material seems fine in the exciton confinement of
BODIPY fluorophores.
First, comparing EL spectra of the three BODIPY
OLEDs, a dual EL has been observed for PhBODP OLEDs,
having either 50 or 100% dopant concentration, whereas
dual EL can be observed only at 100% concentration of
88TPEBODP, that is, nondopant device (see EL spectra
shown in Figure 10). Moreover, a dual EL observed only at
100% dopant concentration of 88TPEBODP implies that
double-tethered TPE on PhBODP effectively reduces the
aggregation of PhBODP and hence the possible excimer
emission (at long wavelength), which is consistent with
higher ϕ obtained for 88TPEBODP than PhBODP
(Table 1). Accordingly, regardless of dopant concentration,
88TPEBODP OLEDs always show higher external quan-
tum efficiencies (EQEs) than PhBODP OLEDs, and the for-
mer devices always exhibit much brighter EL
(electroluminescence, L) than the latter devices (Table 2 and
Figure 10). In addition, the very high-driving voltages of
PhBODP OLEDs have been significantly reduced in
88TPEBODP OLEDs.
fluorescence of BODIPY from concentration quenching.
However, in a THF solution of 1 × 10−5 M, all three TPE-
tethered BODIPY fluorophores show higher
ϕ than
PhBODP. It is more or less the same in THF solution with a
higher concentration of 1 × 10−3 M. Based on these results,
TPE tethering seems to be effective in promoting the fluo-
rescence intensity of BODIPY compounds. In THF solution
with 80 vol% of water, except for one case, 35TPEBODP
with a concentration of 1 × 10−3 M, all TPE-tethered
BODIPY fluorophores exhibit ϕ greater than or similar to
the ϕ of PhBODP. In this regard, once again, TPE tethering
seems to be effective (for enhancing ϕ).
2.5 | Electroluminescence characterization
A series of BODIPY (PhBODP, 88TPEBODP, and
26TPEBODP) OLEDs have been fabricated based on
device configuration of ITO/NPB(35 nm)/TAPC(5 nm)/
CBP:BODIPY(x%, 30 nm)/BCP(10 nm)/TPBI(30 nm)/LiF
(1 nm)/Al(100 nm). Here, we use NPB, TAPC, CBP, BCP,
and TPBI for the hole-transporting layer and electron-
blocking layer, host material (for dopant devices), hole-
blocking layer, and electron-transporting layer, respectively.
The HOMO and LUMO energy levels were estimated by a
low-energy photoelectron spectrometer (Riken-Keiki AC-2)
and the solid-state onset absorption wavelength (as optical
band-gap energy), respectively. Including those of other
materials used in OLED fabrication, HOMO and LUMO
energy levelsof PhBODP, 88TPEBODP,35TPEBODP, and
26TPEBODP are summarized in Figure 9. The deployment
of those blocking layers, TAPC and BCP, in the device is
for the purpose of eliminating the EL from the transporting
Regarding 26TPEBODP OLEDs, devices were only
observed for a single EL regardless of the dopant concentra-
tion. The different EL behavior of dopant concentration
dependence is quite consistent with what we observed in the
PL study. Furthermore, such a single emission band of EL
has
a
longer wavelength compared with that of
88TPEBODP or PhBODP, and the λmax of EL varies little
with different dopant concentrations of 1, 50, and 100 wt%,
although the EL bandwidth becomes larger (Figure 10). Sim-
ilar to the PL, the EL of 26TPEBODP is rather different
from that of 88TPEBODP. As the determined ϕ of
26TPEBODP is the highest (86% in diluted THF, 13% in
5 wt% PS thin film, and 6% as solid powder) compared with

8
YANG ET AL.
FIGURE 10 Electrolumine-
scence characteristics of PhBODP
(top row figures), 88TPEBODP
(middle row figures), and
26TPEBODP (bottom row
figures)
TABLE 2 Electroluminescence characteristics of OLEDs fabricated with four BODIPY compounds in this study
Device
DCa (%)
1
EQE, voltageb (%, V)
0.32 (23.0)
0.38 (20.0)
0.20 (20.0)
0.15 (10.2)
0.67 (13.0)
0.51 (9.0)
L, voltagec (cd/m2, V)
1.1, 9 (35.0)
λELd (nm)
526
CIEd (x, y)
(0.27, 0.67)
(0.31, 0.65)
(0.39, 0.59)
(0.43, 0.55)
(0.29, 0.64)
(0.30, 0.65)
(0.42, 0.56)
(0.50, 0.48)
(0.50, 0.44)
(0.55, 0.43)
(0.56, 0.42)
(0.58, 0.41)
PhBODP
10
1.2, 24 (35.0)
1.1, 77 (31.5)
1.7, 108(17.0)
517, 1,359 (15.0)
40, 268 (15.0)
61, 569 (15.0)
3, 140 (12.0)
528
50
536, 622
530, 650
524
100
1
88TPE BODP
26TPE BODP
10
526
50
0.67 (10.0)
0.24 (7.5)
558
100
1
552, 622
582
1.30 (10.5)
0.27 (14.0)
0.16 (13.5)
0.10, (9.5)
231, 1,603 (15.0)
4, 15 (17.0)
10
588
50
7, 256 (25.5)
592
100
1, 48 (21.0)
594
aDopant concentration.
bMaximum EQE, driving voltage.
cElectroluminance at maximum EQE, maximum L, and driving voltage.
dAt maximum L.

YANG ET AL.
9
those of either PhBODP or 88TPEBODP (Table 1),
26TPEBODP OLEDs with 1 wt% of dopant concentration
display the best EQE of 1.3% and the most intense EL of
approximately 1,600 cd/m2 (Table 2 and Figure 10).
reasonably surmise that such an opposite dependence of
deposition rate must be due to the chemical structure of teth-
ered TPE, which alters the molecule-stacking interaction of
BODIPY fluorophores.
During the nondopant device fabrication, we have
observed that the deposition rate of PhBODP and
88TPEBODP fluorophores has a different effect on the pro-
motion of molecular aggregation and hence the long wave-
length excimer emission (Figure 11 and Table 3). Although
a slower deposition rate promotes the long wavelength EL
of 88TPEBODP, it is the faster deposition rate that
enhances the long wavelength EL of PhBODP. We may
3 | CONCLUSIONS
We have comprehensively studied both PL and EL of
three double TPE-tethered PhBODP fluorophores. We
have found that the solution PL (wavelength and intensity)
of these TPE-tethered BODIPY depends on the solution
concentration and the chemical position of these tethered
FIGURE 11 Electrolumine-
scence characteristics of
88TPEBODP (top row figures)
and PhBODP (bottom row
figures) nondopant OLED, with
variable deposition rates, 0.01,
0.6, and 0.15 Å/s
TABLE 3 Electroluminescence characteristics of OLEDs fabricated with nondopant 88TPEBODP and PhBODP with variable deposition
rates
Device
DRa (Å/s)
0.03
0.6
EQE,b voltage (%, V)
0.15 (7.5)
Lc (cd/m2, V)
2.5, 140 (12.0)
1.3, 288 (12.0)
1.6, 218 (12.0)
1.7, 108 (17.0)
1.5, 16 (20.5)
1.4, 26 (21.0)
λELd (nm)
552, 622
546, 620
550, 620
530, 650
538, 638
544, 648
CIEd (x, y)
(0.50, 0.48)
(0.46, 0.52)
(0.46, 0.52)
(0.42, 0.55)
(0.46, 0.51)
(0.50, 0.48)
88TPE BODP
0.21 (6.0)
1.5
0.19 (6.0)
Ph BODP
0.03
0.6
0.15 (7.5)
0.12 (13.0)
0.43 (12.0)
1.5
aDeposition rate.
bMaximum EQE, driving voltage in parenthesis.
cElectroluminance at maximum EQE, maximum L, and driving voltage in parenthesis.
dAt maximum L.

10
YANG ET AL.
TPE. Nevertheless, through the determination of PL quan-
tum yields of THF and water (80 vol%)–THF solution, we
have demonstrated that none of these TPE-tethered
BODIPY fluorophores exhibits AIE or AIEE. However,
the double-tethered TPE BODIPY fluorophores, that is,
88TPEBODP, 35TPEBODP, and 26TPEBODP, do have
an enhanced ϕ in diluted THF solution and in 5 wt% PS
thin film compared with those of PhBODP. Therefore, we
have observed that 1 wt% dopant OLEDs of 88TPEBODP
and 26TPEBODP all show higher EQE and brighter EL
than those of PhBODP OLEDs. Finally, due to the low ϕ
of these BODIPY fluorophores in solid state or dopant PS
thin film, none of the OLEDs fabricated in this study
exhibits satisfactory EQE.
warmed up to 0ꢀC. The reaction was quenched with a 2 N
hydrochloric acid aqueous solution, extracted with chloro-
form, dried over magnesium sulfate, and subjected to puri-
fication by column chromatography (silica gel, hexanes).
A white solid was obtained with isolated yields of 60%
(1.01 g). 1H NMR (400 MHz, CDCl3): δ 9.878 (s, 1H),
7.915 (s, 2H), 7.896 (s, 1H). 13C NMR (100 MHz,
CDCl3): δ 189.12, 139.57, 138.92, 131.19, 123.94. MS
(EI) m/z 261.9 (M+).
4.3 | Synthesis of 3,5-bis
(2,4,4,5,5-pentamethyl-1,3,2-dioxaborolan-2-yl)
benzaldehyde
A flask charged with 3,5-dibromobenzaldehyde (0.50 g,
1.90 mmol), bis(pinacolato)diboron (1.10 g, 4.18 mmol),
PdCl2(dppf) (0.07 g, 0.095 mmol), and KOAc (1.20 g,
12.5 mmol) in 1,4-dioxane (16 mL) was stirred at a
refluxing temperature for 21 h under a nitrogen atmo-
sphere. After cooling, the solution mixture was filtered
through a short pad of celite, evaporated under reduced
pressure. The residue was redissolved in a mixture of ethyl
acetate and water and then extracted three times with
water. After drying over magnesium sulfate, the organic
solution was subjected to purification by column chro-
matograph with silica gel (dichloromethane/hexanes, 1/5)
to afford a green solid with yields of 88% (0.60 g). 1H
NMR (400 MHz, CDCl3): δ 10.03 (s, 1H), 8.46 (s, 1H),
8.37 (s, 2H), 1.33 (s, 24H). 13C NMR (100 MHz, CDCl3):
δ 192.61, 146.83, 138.79, 135.10, 84.21, 24.87. MS
(EI) m/z 359.3 (M + H+).
4 | EXPERIMENTAL
4.1 | General
1H and 13C NMR spectra were measured on a Bruker
AV-400 MHz NMR Fourier transform spectrometer at
room temperature. Electron ionization (EI) or fast atom
bombardment (FAB) mass spectroscopy (MS) were per-
formed by the Mass Spectroscopic Laboratory, an in-house
service of the Institute of Chemistry, Academic Sinica.
Elemental analyses were performed by the Instrumentation
Center, National Taiwan University. Thermal decomposi-
tion temperatures (Td's) of the BODIPY compounds were
measured by thermogravimetric analysis (TGA) using
Perkin-Elmer TGA-7 analyzer systems. UV–visible absorp-
tion spectra were recorded on a Hewlett-Packard 8453
diode array spectrophotometer. Room temperature fluores-
cence spectra were recorded on a Hitachi fluorescence
spectrophotometer F-7000. The measurement of solution
and solid-state photoluminescence quantum yields
(PLQYs) was also carried out by F-7000. The ionization
potentials (or HOMO energy levels) of the BODIPY com-
pound in this study were determined using a low-energy
photoelectron spectrometer (Riken-Keiki AC-2). A micro-
wave synthesizer (CEM Discover SP) with an irradiation
power of 150 W was set to 140ꢀC on a reaction vessel of
about 20 mL.
4.4 | Synthesis of 3,5-bis(1,2,2-triphenylvinyl)
benzaldehyde
To
a
mixture
of
3,5-bis(2,4,4,5,5-pentamethyl-
1,3,2-dioxaborolan-2-yl)benzaldehyde (0.40 g, 1.12 mmol)
and
bromotriphenylethylene
(1.20 g,
3.58 mmol),
mixture of
PdCl2(dppf) (0.040 g, 0.056 mmol),
a
1,4-dioxane (15.6 mL) and K3PO4 (1.80 g, 8.40 mmol) was
added. The reaction mixture was stirred at 80ꢀC for 16 hr
under a nitrogen atmosphere. After cooling, the solution
mixture was filtered through a short pad of celite, evaporated
under reduced pressure. The residue was redissolved in a
mixture of dichloromethane and water and then extracted
three times with water. After drying over magnesium sulfate,
the organic solution was subjected to purification by column
chromatograph with silica gel (dichloromethane/hexanes,
1/3) to afford a pale yellow solid with yields of 36%
4.2 | Synthesis of 3,5-dibromobenzaldehyde
At −78ꢀC, to a solution of 1,3,5-tribromobenzene (2 g,
6.4 mmol) in diethyl ether (33 mL), n-butyllithium (4 mL,
6.4 mmol, 1.6 M in hexane) was slowly added. The mix-
ture was stirred for 30 min under nitrogen atmosphere.
After the slow addition of a mixture of DMF (1 mL,
12.8 mmol) and diethyl ether (1.35 mL), the reaction solu-
tion was stirred for a further 60 min and then gradually
1
(0.24 g). H NMR (400 MHz, CDCl3): δ 9.61 (s, 1H), 7.25
(d, J = 1.6 Hz, 2H), 7.13–7.03 (m, 17H), 6.97 (t, J = 1.6 Hz,
1H), 6.93–6.89 (m, 8H), 6.75–6.72 (m, 4H).13C NMR

YANG ET AL.
11
(100 MHz, CDCl3): δ 192.20, 144.55, 143.06, 142.77,
142.28, 141.94, 140.17, 139.41, 135.89, 131.26, 131.01,
130.90, 130.36, 127.72, 127.65, 126.65, 126.60. HRMS
(FAB) m/z calcd for C47H34O 614.2610, found 614.2613
(M+). Anal. calcd for C47H34O: C 91.82, H 5.57; found: C
90.83, H 5.48.
water and then extracted three times with water. After drying
over magnesium sulfate, the organic solution was subjected
to purification by column chromatograph with silica gel
(dichloromethane/hexanes, 1/2) to afford a dark green solid
with yields of 36% (0.11 g). 1H NMR (400 MHz, CDCl3): δ
7.62 (d, J = 16.0 Hz, 2H), 7.47–7.45 (m, 3H), 7.34–7.29 (m,
6H), 7.13–7.00 (m, 36H), 6.56 (s, 2H), 1.40 (s, 6H). 13C
NMR (100 MHz, CDCl3): δ 152.57, 144.68, 143.68,
143.60, 143.54, 141.96, 141.51, 140.55, 138.56, 135.85,
135.19, 134.69, 133.45, 131.80, 131.42, 131.34, 129.04,
128.93, 128.48, 127.85, 127.71, 127.61, 126.91, 126.72,
126.53, 126.47, 119.00, 117.75, 14.56. HRMS (FAB) m/z
calcd for C73H55BF2N2 1,008.4426, found 1,008.4436 (M+).
Anal. calcd for C29H27BF2N2: C 86.89, H 2.78, N 5.49;
found: C 85.80, H 5.38, N 2.78.
4.5 | Synthesis of 8-(3,5-bis
((1,2,2-triphenylvinyl)phenyl)-4,4-difluoro-
1,3,5,7-tetramethyl-4-bora-3a,4a-diaza-s-
indance, 88TPEBODP
To a deoxygenated solution of 3,5-bis(1,2,2-triphenylvinyl)
benzaldehyde (0.60 g, 0.97 mmol) and 2,4-dimethylpyrrole
(0.20 g, 2.21 mmol) in dichloromethane (250 mL),
trifluoroacetic acid was added (one drop). The mixture
was stirred under a nitrogen atmosphere at room tempera-
ture for 8 hr. The resulting dark orange solution was
treated with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(0.33 g, 1.36 mmol) and stirred at room temperature for
2 hr. Then, triethylamine (5.3 mL) and boron trifluoride
diethyl etherate (5.3 mL) were added, and the mixture was
stirred for another hour. After washing with water, the
organic phase was separated, dried over magnesium sul-
fate, and subjected to purification by column chromatogra-
phy (silica gel, ethyl acetate/hexanes: 1/7). An orange
solid was obtained with isolated yields of 45% (0.36 g).
1H NMR (400 MHz, CDCl3): δ 7.16–7.14 (m, 6H),
7.03–7.00 (m, 16H), 6.89–6.88 (m, 4H), 6.83 (s, 1H),
6.72–6.71 (d, J = 1.2 Hz, 2H), 6.68–6.66 (m, 4H), 5.86
(s, 2H), 2.45 (s, 6H), 1.25 (s, 6H). 13C NMR (100 MHz,
4.7 | OLED fabrication and EL
characterization
OLED devices were fabricated by vacuum–thermal deposi-
tion using Sumimoto Cryogenics at a chamber pressure of
10−6 Torr. The ITO substrate was purchased from Ruilong
with sheet resistance of around 30 Ω/sq. Host material CBP
(4,40-bis[N-carbazolyl]-2,20-biphenyl), TAPC (di-[4-[N,N-
ditolylamino]phenyl]cyclohexane), and BCP (2,9-dimethyl-
4,7-diphenyl-1,10-phenanthroline) were purchased from
Lumtec. NPB (1,4-bis[1-naphthylphenylamino]biphenyl)
and TPBI (2,20,200-[1,3,5-phenylene]tris[1-phenyl-1H-benz-
imidazole]) were prepared via published methods. For the
cathode of devices, an ultrathin LiF (1 nm) interfacial layer
and then aluminum (100 nm) were vacuum–thermal depos-
ited. The effective size of the emitting diode was 3.14 mm2.
The current density and voltage characteristics were mea-
sured by a dc current/voltage source meter (Keithley 2400),
and the device brightness (or electroluminescence, cd/m2)
and EL spectra were monitored and recorded with a spectro-
photometer (PR670; Photo Research).
CDCl3):
δ 155.25, 144.66, 143.39, 143.04, 142.75,
142.64, 141.56, 141.43, 140.15, 135.04, 134.55, 131.30,
130.95, 130.81, 128.49, 127.92, 127.66, 127.61, 126.55,
120.83, 15.57, 14,47. HRMS (FAB) m/z calcd for
C29H27BF2N2O 832.3800, found 832.3807 (M+). Anal.
calcd for C29H27BF2N2: C 85.09, H 5.69, N 3.36; found:
C 84.28, H 5.60, N 3.30.
ACKNOWLEDGMENTS
4.6 | Synthesis of 3,5-bis(1,2,2-triphenylvinyl)-
4,4-difluoro-1,7-dimethyl-8-phenyl-4-bora-
3a,4a-diaza-s-indance, 35TPEBODP
We thank the Ministry of Science and Technology (MOST
106-2113-M-001-016-MY3), iMATE Program of Academia
Sinica, and National Chiao Tung University.
To
a solution of 4-(1,2,2-triphenylvinyl)benzaldehyde
(0.44 g, 1.23 mmol) and PhBODP (0.10 g, 0.31 mmol) in
DMF (7.7 mL), six drops of acetic acid and piperidine
trifluoroacetic acid were added. The solution was deoxygen-
ated by purging in an argon atmosphere for 30 min. The
reaction of the solution was carried out by microwave
heating (150 W) set at 140ꢀC for 5 min. The solvent and
volatile organics were removed under reduced pressure. The
residue was redissolved in a mixture of dichloromethane and
REFERENCES
[1] F. L. Arbeloa, T. L. Arbeloa, I. L. Arbeloa, J. Photochem. Photo-
biol. 1999, 121, 177.
[2] Z. Zhao, J. W. Y. Lam, B. Z. Tang, Curr. Org. Chem. 2010, 14,
2109.
[3] Z. Zhao, J. W. Y. Lam, B. Z. Tang, J. Mater. Chem. 2012, 22,
23726.

12
YANG ET AL.
[4] R. Hu, C. F. A. Gómez-Durán, J. W. Y. Lam, J. L. Belmonte-
Mr Jian Haur Lee was born in Negeri
Sembilan, Malaysia. He received his
MS degree in chemistry from National
Vázquez, C. Deng, S. Chen, R. Ye, E. Peňa-Cabrera, Y. Zhong,
K. S. Wong, B. Z. Tang, Chem. Commun. 2012, 48, 10099.
[5] Z. Li, Y. Chen, X. Lv, W.-F. Fu, New J. Chem. 2013, 37,
3755.
Chiao Tung University in 2018. He is
[6] M. Baglan, S. Ozturk, B. Gür, K. Meral, U. Bozkaya,
O. A. Bozdemir, S. Atilgan, RSC Adv. 2013, 3, 15866.
[7] Z. Zhao, B. Chen, J. Geng, Z. Chang, L. Aparicio-lxta, H. Nie,
C. C. Goh, L. G. Ng, A. Qin, G. Ramos-Ortiz, B. Liu, B. Z. Tang,
Part. Part. Syst. Charact. 2014, 31, 481.
[8] C. F. A. Gomez-Duran, R. Hu, G. Feng, T. Li, F. Bu,
M. Arseneault, B. Liu, E. Peňa-Cabrera, B. Z. Tang, ACS Appl.
Mater. Interfaces 2015, 7, 15168.
[9] S. Baysec, A. Minotto, P. Klein, S. Poddi, A. Zampetti,
S. Allard, F. Cacialli, U. Scherf, Sci. China Chem. 2018,
61, 932.
[10] R. Hu, E. Lager, A. Aguilar-Aquilar, J. Liu, J. W. Y. Lam,
H. H. Y. Sung, I. D. Williams, Y. Zhong, K. S. Wong, E. Peňa-
Cabrera, B. Z. Tang, J. Phys. Chem. C 2009, 113, 15845.
[11] A. Vázquez-Romero, N. Kielland, M. J. Arévalo, S. Preciado,
R. J. Mellanby, Y. Feng, R. Lavilla, M. Vendrell, J. Am. Chem.
Soc. 2013, 135, 16018.
currently a research assistant in Chen's
group at the Institute of Chemistry,
Academia Sinica. His research mainly
focused on the material design and synthesis of organic
light-emitting diodes, including host materials and lumi-
nescence materials. His interest also includes the discus-
sion on the possibility of fabricating the materials into
optoelectronics devices.
Since 2003, Dr Chin-Ti Chen has been
a Research Fellow at the Institute of
Chemistry, Academia Sinica. He
received his PhD from the University
of Illinois at Urbana-Champaign in
1992. Dr Chen has worked as a Post-
[12] B. Xu, W. Li, J. He, S. Wu, Q. Zhu, Z. Yang, Y.-C. Wu,
Y. Zhang, C. Jin, P.-Y. Lu, Z. Chi, S. Liu, J. Xu, M. R. Bryce,
Chem. Sci. 2016, 7, 5307.
[13] T. Yoshida, L. Ilies, E. Nakamura, ACS Catal. 2017, 7, 3199.
[14] P. G. Seybold, M. Gouterman, J Molec Spectr 1969, 31, 1.
[15] G. Laurent, N. T. Ha-Duong, R. Meallet-Renault, R. B. Pansu,
Handai Nanophoton 2004, 1, 89.
doctoral Research Fellow at the Cali-
fornia Institute of Technology in 1992–1995. In 1995, he
moved to the Institute of Chemistry, Academia Sinica as
an Assistant Research Fellow, promoted to Associate
Research Fellow in 2000. His research interests include
organic materials for optoelectronics applications, such
as perovskite solar cells (PVSCs), organic photovoltaic
(OPVs), and organic light-emitting diodes (OLEDs).
AUTHOR BIOGRAPHIES
Mr Chiun-Jen Yang graduated from the
Department of Chemistry, Fu Jen Cath-
olic University, with a BS in 2017. He
then entered the Department of Applied
Chemistry, National Chiao Tung Uni-
versity, as an MS student. Chiun-Jen's
research focuses on the organic synthe-
sis and physical characterization of fluorescence mate-
rials for organic light-emitting diodes or bioimaging
application.
How to cite this article: Yang C-J, Lee JH,
Chen C-T. Synthesis, photoluminescence, and
electroluminescence characterization of double
tetraphenylethene-tethered BODIPY luminogens. J
Chin Chem Soc. 2019;1–12. https://doi.org/10.1002/
jccs.201900163