Meso-substituted boron-dipyrromethene compounds: synthesis, tunable solid-state emission, and application in blue-driven LEDs

Article abstract of DOI:10.1002/bio.4111

In this work, we depict the synthesis and characterization of a series of meso-substituted boron-dipyrromethene (BODIPY) compounds. Their optical and electrochemical properties were investigated systematically. All these compounds exhibited intense absorption bands in the ultraviolet (UV) and visible regions, which arise from the π–π* transitions based on their BODIPY core segments. By comparing electron-withdrawing substituents and electron-donating substituents, we found that these compounds exhibited some similar photophysical properties but exhibited different fluorescence in the solid state. All compounds were highly emissive in dichloromethane at room temperature (λ_em = 512–523 nm, Φ_PL > 0.9). When these compounds were applied in blue-driven light-emitting diodes (LEDs) as light-emitting materials, the devices showed luminescence efficiency ranging from 1.09 to 34.13 lm/W. Their luminescence and electrochemical properties could be used for understanding the structure–property relationship of BODIPY compounds and developing functional fluorescent materials.

Full text of DOI:10.1002/bio.4111

Received: 8 March 2021
DOI: 10.1002/bio.4111
Revised: 5 June 2021
Accepted: 16 June 2021
R E S E A R C H A R T I C L E
Meso-substituted boron-dipyrromethene compounds:
synthesis, tunable solid-state emission, and application in
blue-driven LEDs
Hao Liu1 | Huan Su1 | Zhiyuan Chen1 | Senqiang Zhu1 | Rui Liu1,2
Hongjun Zhu1
|
1School of Chemistry and Molecular
Abstract
Engineering, Nanjing Tech University, Nanjing,
China
In this work, we depict the synthesis and characterization of a series of meso-
substituted boron-dipyrromethene (BODIPY) compounds. Their optical and electro-
chemical properties were investigated systematically. All these compounds exhibited
intense absorption bands in the ultraviolet (UV) and visible regions, which arise from
the π–π* transitions based on their BODIPY core segments. By comparing electron-
withdrawing substituents and electron-donating substituents, we found that these
compounds exhibited some similar photophysical properties but exhibited different
fluorescence in the solid state. All compounds were highly emissive in
dichloromethane at room temperature (λem = 512–523 nm, ΦPL > 0.9). When these
compounds were applied in blue-driven light-emitting diodes (LEDs) as light-emitting
materials, the devices showed luminescence efficiency ranging from 1.09 to
34.13 lm/W. Their luminescence and electrochemical properties could be used for
understanding the structure–property relationship of BODIPY compounds and
developing functional fluorescent materials.
2Jiangsu Greenscie Chemical Co., Ltd,
Zhenjiang, China
Correspondence
Rui Liu and Senqiang Zhu, School of Chemistry
and Molecular Engineering, Nanjing Tech
University, Nanjing 211816, China.
Email: rui.liu@njtech.edu.cn;
zhusenqiang1993@njtech.edu.cn
Funding information
Natural Science Foundation of Jiangsu
Province, Grant/Award Number:
BK20170104; Postgraduate Research &
Practice Innovation Programme of Jiangsu
Province, Grant/Award Number:
KYCX20_1032; ‘Six Talent Peaks Project’ of
Jiangsu Province, Grant/Award Number: XCL-
037; National Natural Science Foundation of
China, Grant/Award Number: 21602106
K E Y W O R D S
BODIPY, electrochemical properties, emission, light-emitting diodes, synthesis
1
|
INTRODUCTION
The optoelectronic properties of BODIPY compounds are usually
adjustable by introducing various substituents. This has led to com-
Boron-dipyrromethene (BODIPY) compounds were first reported by
Treibs and Kreuzer in 1968,[1] and have attracted great interest in the
past decades due to their potential applications in photodynamic
therapy, molecular probes, nonlinear optical materials, laser dyes, and
solar cells.[2–7] There is also a huge variety of functionalized BODIPY
dyes for studying biochemistry processes in biological systems (DNA,
proteins, lipids, and so on).[8,9] They have been widely used due to
their excellent optical properties including strong ground-state
absorption, high-fluorescence quantum yield, and high optical
stability.[10,11] In addition, adjusting their photophysical properties
through structural modifications also provided good opportunities for
various functional applications.
prehensive synthetic efforts and the minor adjustment of their optical
performance through both theoretical modeling[12] and experimental
studies.[13] BODIPYs with a meso-substitution have generated wide
research interest due to their special luminescence characteris-
tics.[14,15] Different photophysical properties are related to slight
changes with the properties of the substituents. For instance, the out-
standing fluorescence properties of BODIPY are exhibited in
meso-substituted alkyl chain cores (photoluminescence quantum yield,
ΦPL > 0.9),[16–18] but the meso-alkenyl derivatives are nearly non-
fluorescent (ΦPL ≈ 0).[19] Research has shown that their excited state
relaxation speed caused this phenomenon. The meso-phenyl deriva-
tives are different from alkyl and alkenyl derivatives, although they
Luminescence. 2021;1–9.
wileyonlinelibrary.com/journal/bio
© 2021 John Wiley & Sons Ltd.
1

2
LIU ET AL.
demonstrate a high-fluorescence quantum yield. However, the molec-
ular structure is key to controlling their luminescence.
2.8 ml (2.5 mmol) n-butyllithium was added dropwise. The reactant
was stirred for 30 min, then 1.2 g N, N-dimethylformamide was
dropped into the round-bottomed flask and agitated at ꢀ78ꢁC for
60 min. The container was placed into an ice bath and agitated
for 30 min; 15 ml saturated NH4Cl solution was added to quench the
reaction. Then, 50 ml CH2Cl2 was added, organic solution extraction
was completed by brine washing several times, and then a specified
amount of anhydrous magnesium sulfate was used to absorb excess
water in the solution. A rotary evaporator spiral was used to evapo-
rate the solvent to obtain a coarse product. The crude product was
purified by column chromatography with PE:EA = 10:1 as the devel-
oping agent. Finally, product 3, a 5-dibromobenzaldehyde white solid
Although the BODIPY compound has excellent photophysical
properties, it is still at a disadvantage for the application of light-
emitting devices. The main factor that restricts traditional BODIPYs in
luminescence displays and illumination is its weak fluorescence, and
even aggregation-caused quenching (ACQ) occurs in the solid state or
an aggregation state. The cause of ACQ is that BODIPYs with a flat
and rigid construction behave unfavourably in intermolecular interac-
tions (e.g. π–π stacking) in the solid state, which seriously affects the
application of this series of compounds in organic light-emitting mate-
rials. To reduce the ACQ effect, methods such as introducing a
sterically hindered substituent and employing a twisted and branched
molecular structure have been used in the construction of boron-
difluoride compounds or other organic luminophores.[20–24] The incor-
poration of various π-conjugated substituents at the meso-position of
BODIPY perturbs the photophysical and electrochemical properties to
a great extend. This perturbation leads to strong absorption with a
high molar extinction coefficient, and excellent luminescence proper-
ties. Moreover, bulky substituents are introduced in the BODIPY core
to keep the individual molecules apart from each other at a certain dis-
tance so that close packing of molecules is avoided in the solid state.
In this work, a series of BODIPY compounds was designed
with different electron-donating/withdrawing substituents in the
meso-position. All of the compounds 1a–1e have rotatable junctions
(e.g. 3,5-bis[30,50-dimethyl phenyl]-phenyl), three benzene rings as
rotatable junctions, and these compounds demonstrated a high-
emission efficiency in solution. It was found that this series of BODIPY
compounds can achieve tunable solid-state fluorescence emission and
could be used to make LED device applications, and finally obtain
blue-driven LEDs, which makes them promising candidates for
the development of efficient light-emitting materials in LED devices.
powder was obtained with
a
yield of 45% (1.98 g). 1H-NMR
(400 MHZ, CDCl3) δ 9.91 (s, 1H), 7.91 (d, J = 12, 3H). 13C-NMR
(101 MHz, CDCl3) δ 189.16, 139.54, 138.86, 131.18, 123.80.
2.2.2
|
Preparation of 3,5-dimethylphenylboric (2)
A volume of 50 ml of redistilled tetrahydrofuran was added to a
round-bottomed flask, and 5 g of 3,5-dimethylbrombenzene and 3 g
of trimethyl borate were added to the flask and placed into a
cryogenic coolant circulation pump cooled to ꢀ78ꢁC. Then, 3 ml
(2.5 mmol) n-butyllithium were added dropwise and the reactant was
stirred for 60 min. Then, the reaction temperature was slowly raised
to room temperature and the reaction was continued overnight. The
quenching reaction was implemented with 15 ml saturated ammo-
nium chloride solution at an ambient temperature for about 3 h. The
organic phase was separated using a separating funnel. The organic
solution extraction was completed via brine washing several times,
and then a large amount of anhydrous sodium sulfate was used to
absorb the excess water in the solution. A rotary evaporator spiral
was used to evaporate the solvent to obtain the coarse product, and
the coarse product was obtained by column chromatography with a
PE:EA = 5:1 solution as the expansion agent and for purification.
Finally, the 3,5-dimethylphenylboric white solid powder product was
obtained with a yield of 43% (2.2 g). 1H-NMR (400 MHZ, CDCl3) δ
7.88 (s, 2H), 7.27 (s, 1H), 2.48 (s, 6H). 13C-NMR (101 MHz, CDCl3)
δ 137.28, 134.39, 133.30, 131.17, 21.27.
2
|
EXPERIMENTAL
| Synthesis
2.1
All solvents and reagents were obtained from TCI, Macklin, and the
Sinopharm Chemical Reagent Co. Ltd. Tetrahydrofuran (THF) and dic-
hloromethane (CH2Cl2) were distilled over sodium ben-
zophenoneketyl. Silica gel (300–400 mesh) was obtained from the
J&K Scientific Ltd.
2.2.3
|
Preparation of 3,5-bis(30,50-dimethylphenyl)-
benzaldehyde (3)
1,4-Dioxane, water, and potassium carbonate solutions were
2.2
|
General procedure for the synthesis of the
added to
2 g 3,5-dimethylphenylboric
3,5-dibromobenzaldehyde and
a double-necked circular-bottomed flask containing
BODIPY compound
acid
and
then
1.76 g
608 mg
tetraphenylphosphine
2.2.1
|
Preparation of 3,5-dibromobenzaldehyde (1)
palladium were added. The reactants were magnetically stirred for a
reflux reaction at 90ꢁC for 12 h in a nitrogen atmosphere. The organic
phase was separated using a separating funnel and the water phase
was extracted several times using CH2Cl2, and then a large amount of
anhydrous sodium sulfate was used to absorb the excess water in the
Here, 4 g of 1,3,5-tribromobenzene was added into a round-bottomed
flask containing anhydrous THF and placed into a cryogenic coolant
circulation pump cooled to ꢀ78ꢁC. After the temperature was stable,

LIU ET AL.
3
extracted organic solution. A rotary evaporator was used to concen-
trate the solvent. The crude product was used for column chromatog-
raphy PE:CH2Cl2 = 15:1, PE:CH2Cl2 = 13:1, PE:CH2Cl2 = 10:1;
500 ml was purified as the eluent, and 3,5-bis(30,50-dimethylphenyl)-
benzaldehyde with a yield of 89.4% (1.87 g) was obtained. 1H-NMR
(400 MHZ, CDCl3) δ 10.17 (s, 1H), 8.07 (s, 3H), 7.33 (s, 4H), 7.10 (s,
2H), 2.42 (s, 12H). 13C-NMR (101 MHz, CDCl3) δ 192.49, 142.88,
139.79, 138.61, 137.29, 131.85, 129.71, 127.05, 125.15, 21.43.
2.59 (s, 6H), 1.55 (s, 6H), 1.39 (s, 36H). 13C-NMR (101 MHz, CDCl3) δ
164.60, 162.16, 156.32, 142.86, 140.96, 139.97, 136.75, 131.00,
126.06, 121.63, 110.21, 109.95, 103.84, 103.19, 14.67. 19F-NMR
(376 MHz, CDCl3) δ ꢀ146.17 (q, JB, F = 67.68 Hz). HRMS (ESI) m/z
+
calcd for
C
47H59BFN2 [M + H]+ 681.4749, found 681.4751
(Figure S3).
2.2.7
|
Compound 1d
2.2.4
|
Compound 1a
A red solid was obtained, yielding 63%. 1H-NMR (400 MHZ, CDCl3) δ
7.84 (s, 1H), 7.56 (d, J = 4.0, 2H), 7.15 (dd, J = 8.0, 4H), 6.84–6.89 (m,
J = 4.0, 2H), 6.03 (s, 2H), 2.58 (s, 6H), 1.48 (s, 6H). 13C-NMR
(101 MHz, CDCl3) δ 155.57, 151.64, 143.56, 143.29, 141.77, 139.47,
135.75, 131.51, 126.86, 125.22, 122.29, 121.41, 121.33, 35.07,
In total, 1.29 g 2,4-dimethylpyrrole and 1.7 g 3,5-bis(30,50-
dimethylphenyl)-benzaldehyde were dissolved in anhydrous CH2Cl2
(100 ml) in a N2 atmosphere. Eight drops of trifluoroacetic acid (TFA)
were added to the reaction, and the mixture was stirred overnight.
After the reaction was completed through TLC, 1.35 g 2,3-dichloro-
5,6-dicyano-1,4-benzoquinone (DDQ) were mixed with redistilled dic-
hloromethane and added to the reaction, which continued for 12 h.
Triethylamine (10 ml) was added to the reaction mixture and stirred
for 15 min, BF3ꢂEt2O (10 ml) was added. The reaction mixture was
stirred for another 5 h, the mixture was washed several times with
water and then extracted with CH2Cl2. The organic phase was dried
over anhydrous sodium sulfate, the coarse product was purified by
column chromatography using PE:CH2Cl2 = 6:1, and 2.5 L was puri-
31.53, 14.79. 19F-NMR (376 MHz, CDCl3)
δ
ꢀ146.22 (q, JB,
+
F = 67.68 Hz), ꢀ108.61 (s). HRMS (ESI) m/z calcd for C31H23BF5N2
[M + H]+ 529.1869, found 529.1865 (Figure S4).
2.2.8
|
Compound 1e
A red solid was obtained, yielding 62%. 1H-NMR (400 MHZ, CDCl3) δ
8.06 (s, 4H), 7.96 (s, 2H), 7.93 (s, 1H), 7.70 (s, 2H), 6.06 (s, 2H), 2.59 (s,
6H), 1.51 (s, 6H). 13C-NMR (101 MHz, CDCl3) δ 156.59, 142.47,
141.48, 140.63, 139.08, 137.66, 133.25, 132.92, 132.59, 132.19,
131.18, 127.48, 127.25, 126.46, 124.49, 122.15, 121.82,
46.67, 15.03, 14.67, 8.72. 19F-NMR (376 MHz, CDCl3) δ ꢀ146.20
fied as an eluent.
A
difluoro-boron compound of 3,5-bis(30,50-
dimethylphenyl)-phenyl (1a) was obtained as a red powder with a
yield of 58% (986 mg). 1H-NMR (400 MHZ, CDCl3): δ 7.91 (s, 1H),
7.51 (s, 2H), 7.04 (s, 2H), 6.01 (s, 2H), 2.57 (s, 6H), 2.39 (s, 12H), 1.52
(s, 6H). 13C-NMR (101 MHz, CDCl3) δ 155.58, 143.15, 142.81,
141.66, 140.09, 138.61, 135.74, 131.45, 129.58, 126.20, 124.95,
121.26, 21.42, 14.79. 19F-NMR (376 MHz, CDCl3) δ ꢀ146.22 (q, JB,
(q, JB,
= 63.92 Hz), ꢀ62.66 (s). HRMS (ESI) m/z calcd for
F
C
35H24BF14N2+ [M + H]+ 749.1803, found 749.1808 (Figure S5).
F = 67.68 Hz). HRMS (ESI): m/z calcd for C35H35BF2N2 [M + H]+
2.3
|
DFT calculations
+
533.2961, found 533.2997 (Figure S1).
Molecular calculations were implemented in the Gaussian 09 pro-
gram.[28,29] The ground states of the compounds 1a–1e were fully
optimized by a functional (B3LYP) and basis set 6-31G*.
2.2.5
|
Compound 1b
A red solid was obtained, yielding 69% 1H-NMR (400 MHZ, CDCl3) δ
7.90 (s, 1H), 7.52 (d, J = 4.0, 2H), 6.77 (d, J = 4.0, 4H), 6.51 (t, J = 4.0,
2H), 6.01 (s, 1H), 3.86 (s, 12H), 2.58 (s, 6H), 1.51 (s, 6H). 13C-NMR
(101 MHz, CDCl3) δ 161.31, 155.87, 143.07, 142.52, 142.00, 141.29,
135.77, 131.16, 126.37, 125.93, 121.70, 105.35, 99.59, 55.55, 14.86.
19F-NMR (376 MHz, CDCl3) δ ꢀ146.23 (q, JB, F = 60.16 Hz). HRMS
2.4
|
Fabrication of blue-driven LEDs
The chip emits a blue light at around 470 nm. The filling ratio (C:com-
pound) of the compound in the glue is A:B:C = 100:10:40 (mg). The
mixture was stirred in the same direction, and air bubbles were
removed by vacuum for 15 min, and the mixture then deposited on
the LED chip. Electroluminescence spectra of the LEDs were
implemented at 125.9ꢁC using an Everfine HAAS-2000 instrument,
the LED with a forward current range from 300 to 350 mA. To gather
light, the LED was placed in an integrating sphere with a diameter of
30 cm, and coupled to a high-precision array with a spectroradiometer
(wavelength accuracy < 0.3 nm) and a programmable test power
supply LED 300.
(ESI) m/z calcd for C35H35BFN2O4 [M + H]+ 577.2674, found
+
577.2624 (Figure S2).
2.2.6
|
Compound 1c
An orange solid was obtained, yielding 65%. 1H-NMR (400 MHZ,
CDCl3) δ 7.97 (s, 1H), 7.53 (d, J = 4.0, 2H), 7.51 (s, 6H), 6.02 (s, 2H),

4
LIU ET AL.
3
|
RESULTS AND DISCUSSION
| Synthesis and characterization
were measured in five solvents with different polarities: acetonitrile
(ACN), tetrahydrofuran (THF), dichloromethane (DCM), toluene (TL),
and hexane (HEX) (Figure S6). A similar phenomenon was observed
for compounds 1a–1e, and all of the compounds exhibited
minor solvatochromic effects, as shown in Figure S6. This
minor solvatochromic effect also supported the 1π–π* distribution of
their absorption bands. In addition, compared with their absorption
spectra in solution, the electron-donating/withdrawing substituents
connected to BODIPY had a minor influence on their electronic
absorption. The absorption properties of compounds 1a–1e were also
measured in the solid state (Figure 1b), which showed strong
absorption bands ranging from 430 to 550 nm. Compared with the
UV–vis absorption measured in solution, the maximum absorption
wavelengths (504–525 nm) were red-shifted, and the absorption
bands were broader than those obtained in solution. Compared with
1a, the solid-state absorptions of compound 1c with t-Bu groups and
1e with trifluoromethyl groups exhibited obvious blue shifts. This
could be attributed to the greater steric hindrance of these
substituents than methyl groups. Therefore, the π–π stacking effect
among the molecules could be reduced, which produced the blue shift
of their solid-state absorptions.
3.1
The synthetic routes of compounds 1a–1e are illustrated in Scheme 1.
To adjust the photophysical properties of the compounds, varied
electron-donating/withdrawing substituents were connected to the
BODIPY π-conjugation cores via phenyl units, which produced good
yields (over 60%). All target compounds were soluble in CH2Cl2,
tetrahydrofuran (THF), toluene, N,N-dimethylformamide (DMF) and
air stable. 1H-NMR, 13C-NMR, 19F-NMR, and mass spectroscopy
confirmed the proposed structures for compounds 1a–1e.
3.2
solid state
|
Electronic absorption in solution and
The UV–vis absorption of compounds 1a–1e obeys the Lambert–
Beers law in the concentration range of this study (1 ꢃ 10–6 to
1 ꢃ 10–4 mol L–1) in CH2Cl2 (Figure S8), illustrating that no aggrega-
tion occurred at this concentration range. The optical data measured
in different solvents are listed in Table 1 and the UV–vis absorption
spectra of 1a–1e in CH2Cl2 are presented in Figure 1. All these com-
pounds showed strong and sharp absorption bands between 420 to
530 nm, which were attributed to the 1π–π* transitions localized on
the BODIPY aromatic rings.[25] Compounds 1d and 1e connect with-
drawing substituents that show a slight bathochromic shift and can be
ignored. In addition, the photophysical properties of these BODIPYs
3.3
|
Photoluminescence in solution and solid state
The photoluminescence properties of compounds 1a–1e in different
solvents were tested (Figure S7), and their normalized emission spec-
tra in CH2Cl2 (1 ꢃ 10–5 mol L–1) are shown in Figure 2. The emission
SCHEME 1 The general synthetic route of the BODIPYs
TABLE 1 Optical data measured in different solvents at room temperature
1a
1b
1c
1d
1e
λab
(nm)
λem
(nm)
λab
(nm)
λem
(nm)
λab
(nm)
λem
(nm)
λab
(nm)
λem
(nm)
λab
(nm)
λem
(nm)
a
a
a
a
a
Solvent
Hexane
Toluene
DCM
ΦF
ΦF
ΦF
ΦF
ΦF
501
504
502
501
498
514
519
516
515
512
0.94
0.95
0.96
0.98
0.92
501
505
502
501
498
515
520
517
515
512
0.97
0.95
0.94
0.95
0.92
501
502
501
501
498
514
519
516
514
512
0.96
0.91
0.98
0.98
0.96
503
506
503
502
499
517
521
518
516
513
0.89
0.98
0.95
0.97
0.88
504
508
504
503
499
518
523
519
517
513
0.97
0.99
0.98
0.96
0.93
THF
Acetonitrile
aΦF (absolute fluorescence quantum yield) was determined in different solvents. Abbreviations: λab = absorption maximum, λem = emission maximum,
were determined in different solvents (1 ꢃ 10–5 M) at room temperature.

LIU ET AL.
5
FIGURE 1 (a) Normalized UV–vis absorption spectra of 1a–1e in CH2Cl2 (1 ꢃ 10–5 M) at room temperature. (b) Normalized UV–vis
absorption spectra of 1a–1e in the solid state
FIGURE 2 (a) Normalized emission spectra of 1a–1e in CH2Cl2 (1 ꢃ 10–5 M). (b) Normalized emission spectra of 1a–1e in the solid state
TABLE 2 Experimental data of
photophysical properties of compounds
Compound
λ
ab nm
λ
em nm
Φf %
18.4
5.3
τav ns
2.8
kra ns–1
0.066
0.020
0.006
0.0005
0.005
knrb ns–1
0.291
0.364
0.410
0.454
0.429
1a
1b
1c
1d
1e
518
517
504
520
512
620
619
575
640
645
1a–1e in the solid state
2.6
1.5
2.4
0.1
2.2
1.2
2.3
aRadiative rate constant kr = Φf/τav
bNonradiative rate constant knr = (1-Φf)/τav
.
.
band maxima and photoluminescence quantum yields are listed in
Table 1. All compounds shared different photoluminescence
quantum yields in solution (ΦPL = 88–99%) and the solid state
(Φf = 0.1–18.4%) (Table 2). As shown in Figure 2a, compounds 1a–1e
emitted a green light in a CH2Cl2 solution and exhibited quite small
Stokes’ shifts (ꢄ15 nm). Considering these characteristics, we could
attribute the emission to the 1π–π* transition. The compounds 1d and
1e exhibited slight bathochromic shifts (<5 nm), which connected
electron-withdrawing substituents (F and CF3). Moreover, the
bathochromic shifts of emissions could be explained by the decreased
energy band gap (Table 3). The compounds 1a–1c connecting
electron-donating substituents (such as CH3, OCH3, and t-Bu) showed
a similar emission performance, and this was almost identical to the
UV–vis absorption spectra that we obtained.

6
LIU ET AL.
As shown in Figure 2b, the emission bands of the solid fluores-
state fluorescence quantum yields of these complexes were less
than 20%. This could be attributed to the strong π–π interaction
between the molecules, which produced excimer complexes that
were not conducive to luminescence, resulting in a decrease in
fluorescence quantum yield.[26] The complexes with electron-donating
groups were slightly beneficial for the enhancement of solid-state
fluorescence.
cence of compounds 1a–1e were broader than those in solution.
Compared with 1a and 1b, compounds 1d and 1e bearing electron-
withdrawing groups showed minor bathochromic shifts (13 and
16 nm). This phenomenon was similar to the photoluminescence
in CH2Cl2. However, the emission of compound 1c was quite
blue-shifted compared with that of other compounds, which is
also consistent with the observation of their solid-state absorption.
In addition, 1c showed a broader emission band; the steric hinderance
of the molecule increased due to the bulky tert-butyl groups
and decreased the intermolecular π–π stacking interaction. This
made it difficult to delocalize π-electrons, therefore 1c required more
energy for emission in the accumulation and showed a hypsochromic
shift. Compared with the emission spectra in solution, broader
emission bands with considerable bathochromic shifts (except 1c)
were observed in the solid state, which indicated that the π–π
stacking of the planar configurations of the BODIPY compound in
solid state led to an increase in the π-conjugation length. The solid-
3.4
|
Theoretical calculations and electrochemistry
The ground-state electron density distribution of the highest occupied
molecular orbital (HOMO) and lowest unoccupied molecular orbital
(LUMO) are illustrated in Figure 3. The HOMO–LUMO energy differ-
ences (energy band gaps, calculated Egcal) are presented in Table 2.
Compounds 1a–1c possessed almost the same values for the HOMO
(ꢀ2.35/ꢀ2.36 eV) and LUMO (ꢀ5.40/ꢀ5.41 eV), while the band gap
of compounds 1d and 1e were the largest with a HOMO of (ꢀ2.60/
ꢀ2.75 eV) and a LUMO of (ꢀ5.63/ꢀ5.76 eV). Due to the coplanar
configuration, the π-electrons in the HOMO of compounds 1a–1e
were distributed on the BODIPY cores, Strong orbital interactions
could be generated among the stacked π-systems. While their LUMOs
showed almost no change compared with the HOMOs, the
π-electrons in the LUMO of compounds 1d and 1e had a tendency to
shift to meso-position substituents. Furthermore, observation of the
ground-state optimized structures through the compounds 1a–1e,
the dihedral angles between the aromatic rings, and the cores were
studied. For instance, the dihedral angle between the aromatic ring of
1a and its BODIPY core unit was 90ꢁ, as illustrated in Figure 4. This
TABLE 3 DFT calculation results of HOMO–LUMO energy levels
Compound
LUMO (eV)
–2.36
HOMO (eV)
–5.41
Egcal (eV)a
3.05
1a
1b
1c
1d
1e
–2.35
–5.40
3.05
–2.35
–5.40
3.05
–2.60
–5.63
3.03
–2.75
–5.76
3.01
aCarried out at the B3LYP/6-31G* level of theory.
FIGURE 3 Molecular orbital surfaces of the HOMOs and LUMOs for 1a–1e obtained at the B3LYP/6-31G* level

LIU ET AL.
7
large dihedral angle made it difficult for electrons in molecules to form
an effective delocalization. The above results indicated that the inter-
mediate bridging group, either the strong electron donor or the strong
electron acceptor, could not produce a significant intramolecular
charge transfer for the whole molecule. Therefore, there was no
significant bathochromic shift in the absorption spectrum. Aryl
substituents introduced onto the meso-carbon generally produce a
slight effect on the HOMO and LUMO energy levels.[27] In addition,
the electrochemical properties of compounds 1a–1e were investi-
gated using cyclic voltammetry. The electrochemical potentials and
energy levels of these compounds are shown in Table 4, and although
3.5
|
LED devices
Compounds 1a–1e were applied to LEDs actuated by a blue LED chip.
The electroluminescence measurement results are listed in Table 5.
The colour rendering index (CRI) values of these LEDs ranged from
33 to 66, which confirmed that different emissions could be achieved
via the fabrication of LEDs based on compounds 1a–1e (Figure 5).
Among these, the CRI value of the LED device constructed by com-
pound 1e exceeded 60, therefore exhibiting a blue emission property.
Compound 1d showed the highest luminescence efficiency (up to
34 lm/W) compared with other compounds. The dominant wave-
length of each device on the electroluminescence spectrum was in
the range of 469–605 nm, which could change the colour of the
device from blue to orange by blue-driven LEDs. In addition, the cor-
related colour temperature (CCT) of these LEDs could be revised from
1130 K to 83 060 K, which realized the control of cold and warm col-
our temperatures. The CIE coordinate plot are shown in Figure S9,
and accurate values are listed in Table 5. Five devices were made from
this series of compounds, their CIE coordinate values not only indi-
cated the luminescence colour, but also were related to the lumines-
cence efficiency. When the X value of the CIE coordinate was less
than 0.3, the luminescence efficiency of the other four devices
(except device 1b) was greater than 10%. Furthermore, the Y value of
the four devices was greater than 0.3, and the luminescence efficiency
was further increased. By comparison, it was found that the X value
of the CIE coordinates was less than 0.3 and the Y value was greater
than 0.3, therefore the device emitted high luminescence efficiently
under a low loading amount. These results also provide a basis for
designing BODIPY complexes rationally with intensive emission
for light-emitting applications.
the theoretically calculated Egcal value (3.01–3.05 eV) was significantly
opt
higher than the energy gap Eg
(2.32–2.34 eV) obtained from the
ground-state absorption, the changing trend of the compound energy
gap remained consistent.
FIGURE 4 Dihedral angle of the ring between the aromatic ring
and the core according to the ground-state optimized structure of 1a
TABLE 4 Electrochemical potentials
and energy levels of compounds 1a–1e
Compound
Eox (V)a
Ered (V)a
–0.51
–0.56
–0.22
–0.40
─
HOMO (eV)b
5.34
LUMO (eV)c
3.00
Eg (eV)d
2.34
1a
1b
1c
1d
1e
0.54
0.59
1.18
0.58
0.95
5.39
3.06
2.33
5.98
3.64
2.34
5.38
3.06
2.32
5.75
3.43
2.32
aPotentials determined by cyclic voltammetry in a deoxygenated solution of THF, containing 0.1 M
TBAP, at a concentration of 10–3 M, at room temperature. Potentials were standardized with a
ferrocene-ferrocenium (Fc/Fc+) couple as the internal reference vs Ag/AgNO3. Scan rate 0.1 V/s.
Working electrode: glassy carbon, counter electrode: Pt wire, reference electrode: Ag/AgNO3. b
HOMO = Eox + 4.8 eV. c LUMO = HOMO – Eg. d Eg estimated from the UV–vis absorption spectra.
TABLE 5 Electroluminescence data of LEDs based on compounds 1a–1e
Device
1a
1b
1c
1d
1e
CIE
(0.29, 0.37)
(0.42, 0.36)
(0.23, 0.38)
(0.19, 0.36)
(0.20, 0.26)
CCT (K)
2150
33
1136
48
3163
33
4434
43
83 060
66
CRI
Luminescence efficiency (lm/W)
Photoelectric efficiency (%)
16.46
4.273
1.09
0.635
21.19
4.463
34.13
7.730
10.19
4.180

8
LIU ET AL.
FIGURE
Electroluminescence spectra
5
of LED devices 1a–1e (a-e). Inset:
Photographs of devices 1a–1e
with forward current
4
|
CONCLUSION
ACKNOWLEDGEMENTS
The authors greatly acknowledge the Natural Science Foundation of
Jiangsu Province (BK20170104), the National Natural Science Foun-
dation of China (21602106), ‘Six Talent Peaks Project’ of Jiangsu
Province (XCL-037) and Postgraduate Research & Practice Innovation
Programme of Jiangsu Province (KYCX20_1032) for financial support.
A series of BODIPY compounds (1a–1e) with different electron-
donating/withdrawing aromatic substituents was designed and
synthesized. Their absorption and emission spectra were recorded
in different solvents and in the solid state. Compounds 1a–1e
exhibited strong 1π–π* absorption in the visible spectral region and
exhibited intense green 1π–π* emissions with relatively high photo-
ORCID
Rui Liu
https://orcid.org/0000-0002-8227-6064
luminescence quantum yields (>0.9). Moreover, these compounds
cal
showed a minor difference for their Eg
value (3.01–3.05 eV),
REFERENCES
which was caused by the large dihedral angles between the
aromatic substituents and the cores. However, when these com-
pounds were applied to blue-driven LEDs as light-emitting
materials, they exhibited different solid-state fluorescence proper-
ties and produced blue to orange emissions. These results provide
a good guide that is significant not only for the synthesis of
efficient fluorescent materials by introducing twisted groups at the
meso-position of BODIPY, but also for the design and potential
construction of colour-tunable LEDs.
[1] A. Treibs, F. H. Kreuzer, Liebigs Ann Chem. 1968, 718, 208.
[2] A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891. https://doi.org/
10.1021/cr078381n
[3] T. M. Sudesh, K. Jin, L. Yunho, G. C. David, Org. Lett. 2014, 16, 520.
https://doi.org/10.1021/ol403405n
[4] W. Ryan, P. Candace, A. Jeffrey, J. Org. Chem. 2010, 75, 2883.
https://doi.org/10.1021/jo100095n
[5] B. D. Ana, H. J. Xu, Z. L. Xue, H. Katrin, Z. Shen, G. W. Michael, X.-Z.
You, K. Rurack, Org. Lett. 2008, 10, 1581. https://doi.org/10.1021/
ol800271e

LIU ET AL.
9
[6] K. Syam K, B. Christoffer, J. A. D. Good, M. Shamil, C. Erik, B. A.
Lennart, F. Almqvist, J. Org. Chem. 2013, 78, 12207. https://doi.org/
10.1021/jo401844y
[21] X. F. Mei, J. W. Wang, Z. G. Zhou, S. Y. Wu, L. M. Huang, Z. H. Lin, Q.
Ling, J. Mater. Chem. C 2017, 5, 2135. https://doi.org/10.1039/
C6TC05519B
[7] W. J. Shi, P. C. Lo, S. Anu, L. R. Isabelle, Tetrahedron 2012, 68, 8712.
https://doi.org/10.1016/j.tet.2012.08.033
[22] J. P. Wang, Y. Qi, X. L. Niu, H. Tang, J. Nutr. Biochem. 2018, 54, 130.
https://doi.org/10.1016/j.jnutbio.2017.12.004
[23] Y. Wan, J. Li, X. Peng, RSC Adv. 2017, 7, 35543. https://doi.org/10.
1039/c7ra06237k
[8] L. M. Michael, J. Lu, A. G. Richard, Science 1996, 271, 1420. https://
doi.org/10.1126/science.271.5254.1420
[9] A. F. Steven, P. Michael, S. Y. Ho, D. J. Lain, S. W. Daniel, Science
2001, 292, 1385. https://doi.org/10.1126/science.1060418
[10] U. Gilles, Z. Raymond, H. Anthony, Angew. Chem., Int. Ed. 2008, 47,
1184.
[24] J. Li, C. Yang, X. Peng, J. Mater. Chem. C 2018, 6, 19. https://doi.org/
10.1039/c7tc03780e
[25] H. Lu, M. John, Y. C. Yang, Z. Shen, Chem. Soc. Rev. 2014, 43, 4778.
https://doi.org/10.1039/C4CS00030G
[11] Z. Raymond, U. Gilles, H. Anthony, New J. Chem. 2007, 31, 496.
https://doi.org/10.1039/B617972J
[26] W. Christoph, S. W. Mark, Macromolecules 1996, 29, 5157. https://
doi.org/10.1021/ma950902y
[12] S. Makoto, N. Takashi, U. Junji, Y. Masaki, Chem. Commun. 2016, 52,
10727. https://doi.org/10.1039/C6CC05439K
[27] M. Josef, J. Am. Chem. Soc. 1978, 100, 6801. https://doi.org/10.
1021/ja00490a001
[13] L. G. Boris, M. Olivier, J. Denis, Phys. Chem. Chem. Phys. 2012, 14,
157. https://doi.org/10.1039/C1CP22396H
[28] D. B. Axel, J. Chem. Phys. 1992, 96, 2155. https://doi.org/10.1063/1.
462066
[14] R. Hu, L. Erik, A.-A. Ange'lica, J. Liu, J. W. Y. Lam, D. W. Ian, Y. Zhong,
K. S. Wong, E. Peña-Cabrera, B. Z. Tang, J. Phys. Chem. C 2009, 113,
15845. https://doi.org/10.1021/jp902962h
[29] M. F. Michelle, J. P. William, J. H. Warren, J. Chem. Phys. 1982, 77,
3654. https://doi.org/10.1063/1.444267
[15] K. Katerina, C. Gonzalo, J. Am. Chem. Soc. 2010, 132, 17560. https://
doi.org/10.1021/ja1075663
SUPPORTING INFORMATION
[16] P. Antonio, F. Alberto, C. Clemence, Phys. Chem. Chem. Phys. 2016,
18, 32668. https://doi.org/10.1039/C6CP06799A
Additional supporting information may be found online in the
Supporting Information section at the end of this article.
[17] B. Jorge, J. A. Ismael, V. E. Ismael, A. H. Alejandro, RSC Adv. 2011, 1,
677. https://doi.org/10.1039/C1RA00020A
[18] J. A. Ismael, R. Hu, B. Z. Tang, I. L. Fabiola, P. C. Eduarodo, Tetrahe-
dron 2011, 67, 7244. https://doi.org/10.1016/j.tet.2011.07.067
[19] L. Y. Jiao, C. J. Yu, M. Liu, Y. Wu, K. B. Cong, J. Org. Chem. 2010, 75,
6035. https://doi.org/10.1021/jo101164a
How to cite this article: H. Liu, H. Su, Z. Chen, S. Zhu, R. Liu,
H. Zhu, Luminescence 2021, 1. https://doi.org/10.1002/bio.
4111
[20] T. P. I. Saragi, S. Till, S. Achim, F. L. Thomas, S. Josef, Chem. Rev.
2007, 107, 1011. https://doi.org/10.1021/cr0501341