The synthesis of a series of fluorescent emitters and their application for dye lasing and cation sensing

Article abstract of DOI:10.1016/j.saa.2020.118978

The following paper reported four carbazole-modified fluorescent emitters. Their molecular structure and electronic transition were analyzed via their single crystal and theoretical calculation. Their photophysical parameters, including absorption, emission and quantum yield, were determined and discussed. It was found that the emission performance of benzo-thiazole-based dyes was better than that of benzo-imidazole-based dyes, owing to the electron-donating effect from the S atom. Upon the presence of metal cations, these photophysical parameters were re-measured. Benzo-thiazole-based dyes were found insensitive towards most metal cations, while benzo-imidazole-based dyes showed obvious photophysical variation upon these metal cations, including absorption red shift and emission quenching. Detailed sensing performance of a representative dye was discussed. A linear working curve with good selectivity was finally observed. Its sensing mechanism was confirmed as the coordination between metal cation and deprotonated benzo-imidazole group. Benzo-thiazole-based dyes showed amplified spontaneous emission (ASE) behavior, with threshold energy of ~220 μJ. Given the optimal condition, a highest ASE efficiency of ~2% was observed, with FWHM value of 6 nm and emission peak of 435 nm. The major novelty and advancement of these fluorescent dyes shall be the stable ASE output (dye 4) under UV excitation and the linear sensing curve with good selectivity (dye 3), which was a hard task for emission turn off sensing probes. We attributed its causation to the valent-recognizing sensing mechanism.

Full text of DOI:10.1016/j.saa.2020.118978

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular
Spectroscopy
journal homepage: www.elsevier.com/locate/saa
The synthesis of a series of fluorescent emitters and their application for
dye lasing and cation sensing
b
a
b
c
Zixuan Yanga, , Yaqiang Tian , Sheng Cang , Leiyu Zhang , Liang Liu
⁎
a
Key Laboratory of Impact and Safety Engineering, Ministry of Education, Ningbo University, Ningbo 315211, China
Key Laboratory of the Ministry of Education for Modern Metallurgy Technology, North China University of Science and Technology, Tangshan 063210, China
Green Intelligence Environmental School, Yangtze Normal University, Fuling, Chongqing 408100, China
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 June 2020
Received in revised form 16 September 2020
Accepted 17 September 2020
Available online 24 September 2020
The following paper reported four carbazole-modified fluorescent emitters. Their molecular structure and elec-
tronic transition were analyzed via their single crystal and theoretical calculation. Their photophysical parame-
ters, including absorption, emission and quantum yield, were determined and discussed. It was found that the
emission performance of benzo-thiazole-based dyes was better than that of benzo-imidazole-based dyes,
owing to the electron-donating effect from the S atom. Upon the presence of metal cations, these photophysical
parameters were re-measured. Benzo-thiazole-based dyes were found insensitive towards most metal cations,
while benzo-imidazole-based dyes showed obvious photophysical variation upon these metal cations, including
absorption red shift and emission quenching. Detailed sensing performance of a representative dye was
discussed. A linear working curve with good selectivity was finally observed. Its sensing mechanism was con-
firmed as the coordination between metal cation and deprotonated benzo-imidazole group. Benzo-thiazole-
based dyes showed amplified spontaneous emission (ASE) behavior, with threshold energy of ~220 μJ. Given
the optimal condition, a highest ASE efficiency of ~2% was observed, with FWHM value of 6 nm and emission
peak of 435 nm. The major novelty and advancement of these fluorescent dyes shall be the stable ASE output
(dye 4) under UV excitation and the linear sensing curve with good selectivity (dye 3), which was a hard task
for emission turn off sensing probes. We attributed its causation to the valent-recognizing sensing mechanism.
© 2020 Elsevier B.V. All rights reserved.
Keywords:
Laser dyes
ASE
Optical sensing
Emission quenching
1. Introduction
based on BODIPY in 2012 [8]. Depauw and cowokers reported several
fluorescent sensors containing BODIPY fluorophore [9].
Given the numerous applications in modern technique, including
scientific application, military application, surgery, optical fiber and
manufacturing, laser has gained more and more research efforts [1–3].
The emission wavelength of these laser systems, however, depends on
their working medium, which limits the variety of laser wavelength.
Such problem shall be solved by dye lasers since their working media
are fluorescent dyes with adjustable emission wavelength [3]. Desired
features of these dye lasers include highly fluorescent and stable struc-
ture, wide optical cross section and limited triplet non-radiative decay
[4]. ASE (amplified spontaneous emission) test has been developed to
evaluate the potential of a fluorescent dye to be used as dye laser [5,6].
ASE phenomenon has been found in some fluorescent dyes such as
anthracene derivatives, coumarin derivatives, oxazine derivatives and
boron dipyrromethene (BODIPY) derivatives. Recently, Jorge
Banuelos-Prieto and Rebeca Sola Lianoon have published a book about
BODIPY dyes, including their design and application in chemosensing
and ASE [7]. Boens, Leen and Dehaen reviewed fluorescent indicators
The development of fluorescent dyes extends their application in op-
tical sensing, which determines analyte by optical parameters, such as ab-
sorbance, emission intensity, lifetime and color [10,11]. Although these
organic dyes have proved their potential of being used for optical sensing,
they are generally highly fluorescent ones, which makes them emission
quenching probes. The presence of analyte quenches their emission, so
that analyte quantity can be determined by measuring emission decrease
of these probes. For a practical sample, analyte is always dispersed and
surrounded by impurities which may quench probe emission as well. As
a consequence, these probes always face a problem of unsatisfied selectiv-
ity and limited precision. To get this problem solved, sensing probes with
good selectivity are desired, which generally needs a specific coordination
site for analyte molecules, so that competing molecules cannot match
such coordination site and thus won't affect or quench probe emission,
which makes the design and synthesis of fluorescent dyes with variable
structures always a challenging task [10,11].
Bearing the above consideration in mind, in this work, four carbazole-
derived fluorescent dyes are synthesized, as shown in Scheme 1. A com-
parable data production on optical parameters of these fluorescent dyes
have been finished, including ASE effect and sensing performance. It is
⁎
Corresponding author.
E-mail address: yangzixuan@nbu.edu.cn (Z. Yang).
https://doi.org/10.1016/j.saa.2020.118978
1386-1425/© 2020 Elsevier B.V. All rights reserved.

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
Scheme 1. Synthetic route and molecular structure of dyes 1–4.
found that the free rotation of σ bond among their conjugation planes and
the active N\\H vibration compromise ASE performance. On the other
hand, such free rotation endows them with a variable structure, showing
emission turn off effect towards metal cations. Corresponding sensing
performance is discussed as well. The major novelty and advancement
of these fluorescent dyes shall be as follows. For the ASE part, the novelty
localizes at the stable ASE output (dye 4) under UV excitation, which is
also known as good photostability. For the sensing part, the major ad-
vancement shall be the linear working curve with good selectivity (dye
3), which is a hard task for most emission turn off sensing probes. We at-
tribute its causation to the valent-recognizing sensing mechanism with
binding ratio of 1:1. This is one of a few recent sensing systems applying
such sensing mechanism.
with a beam splitter equipped with a diaphragm. One of its sub-
beams was guided to a power meter and the other was shaped by a cy-
lindrical lens to form a stripe (6 mm × 0.6 mm). Each sample was
poured into a quartz sample pool and placed under this laser stripe.
Pumping beam polarization direction was perpendicular to this laser
stripe. ASE energy power was calculated against pumping power.
Sample emission quantum yield (Φ) was determined via a literature
method described with Formula 1, where ΦX and ΦST are emission
quantum yields of sample and reference, GradX and GradST are inte-
grated area values of sample and reference emission spectra, nX and
n
ST indicate solvent refractive index values of sample and reference.
We here used fluorescein (dissolved in NaOH aqueous solution,
ST = 0.95) as reference.
Φ
!
ꢀ
ꢁ
2. Experimental section
GradX
GradST
nX2
n2ST
ΦX ¼ ΦST
ð1Þ
2.1. Reagents and equipment
Starting reagents with AR grade were purchased from Alfase Chem-
ical Corporation and Shanghai Chemical Corporation. They were used as
received, including carbazole, EtBr, o-phenylenediamine and 2-
aminobenzenethiol. Standard purification procedures were applied for
solvents before synthesis. NMR and MS data were collected on a Bruker
Avance 500 spectrometer and an Agilent 1100 MS spectrometer.
Photophysical spectra were recorded using a Hitachi F-7000 fluores-
cence spectrophotometer and a Shimadzu UV-3101PC spectrophotom-
eter. Laser spectra and power intensity were determined using an Ocean
Optics Maya2000 Pro Fiber Optic spectrometer and a Newport 2936C
power meter, excited with Nd: YAG source (repetition rate = 10 Hz,
pulse duration = 10 ns).
2.3. Synthesis for starting compounds of carbazole aldehydes
Firstly, 9H-carbazole-2-carbaldehyde (carb-CHO) was synthesized
following a literature procedure [12]. DMF (30 mL) was placed under
ice bath. Then POCl3 (20 mL) was added. After stirring this solution for
20 min, carbazole solution (40 mmol in DMF 30 mL) was dropwise
added. This mixture was stirred at 100 °C for 2 h. Cold ice (200 g) was
added. The resulting mixture was extracted with CH2Cl2 and then sepa-
rated on a silica gel column to give yellow solid (5.0 g). 1H NMR (CDCl3)
δ 9.01 (s, 2H), 8.83–8.81 (d, 2H), 7.32–7.30 (m, 3H), 3.79 (s, 1H). ESI-
MS: m/z = 195.1.
9-Ethyl-9H-carbazole-2-carbaldehyde (carb-Et-CHO) was synthe-
sized following a similar procedure to that for carb-CHO. However, 9-
ethyl-9H-carbazole was firstly synthesized by the following method.
NaOH (12 g) was dissolved in DMF (150 mL), then carbazole (16.8 g)
and EtBr (25 g) were added. This solution was sealed and stirred at
25 °C for 60 h. Pure water was added. The resulting mixture was
2.2. ASE test and emission efficiency
Sample ASE performance was determined in solution. A schematic
presentation is shown as Scheme 2. First, excitation laser (Nd:YAG,
355 nm, repetition rate = 1 Hz, pulse duration = 10 ns) was treated
2

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
Scheme 2. A schematic explanation for ASE experiment.
extracted with CH2Cl2 and then separated on a silica gel column to give
white solid (14.1 g). 1H NMR (CDCl3) δ 9.01 (s, 3H), 8.85–8.82 (d, 2H),
7.31–7.29 (m, 3H). ESI-MS: m/z = 195.1. This white solid was used as
starting compound to synthesize carb-Et-CHO following a similar syn-
thetic procedure to that for carb-CHO. Yield (4.6 g). 1H NMR (CDCl3) δ
9.01 (s, 2H), 8.82–8.77 (d, 2H), 7.32–7.29 (m, 3H), 4.13–4.11 (d, 2H),
1.26–1.22 (m, 3H). ESI-MS: m/z = 223.1.
2-(9-ethyl-9H-carbazol-2-yl)benzo[d]thiazole (denoted as dye
4) was synthesized following a similar procedure to that of dye 2, except
that carb-CHO was replaced by carb-Et-CHO in this run. 1H NMR
(DMSO‑d6) δ 7.99 (d, 2H), 7.47 (s, 2H), 7.32 (t, 4H), 7.23 (m, 1H),
7.18–7.13 (m, 2H), 4.12–4.09 (d, 2H), 1.27–1.24 (m, 3H). ESI-MS: m/
z = 328.1. Its single crystal structure was deposited as CCDC 19237828.
3. Results and discussion
2.4. Synthesis for fluorescent dyes 1–4
3.1. Molecular structure and electronic transition of dyes 1–4
2-(1H-benzo[d]imidazol-2-yl)-9H-carbazole (denoted as dye 1) was
synthesized as follows. A solution of carb-CHO (10 mmol in MeOH
10 mL) was dropwise added into a solution of o-phenylenediamine
(10 mmol in MeOH 20 mL) at room temperature. This mixture was
heated at 85 °C for 6 h. Pure water was added. The resulting solid prod-
3.1.1. Single crystals
As shown in Scheme 1, there are two major structural components
in each dye, carbazole group and benzo-imidazole (or benzo-thiazole)
group. These two components are bonded via an σ bond, forming a con-
jugation plane, which facilitates crystal packing and growth. Luckily,
single crystals of all four dyes are obtained (CCDC 19237825-
19237828) and shown in Fig. 1. Their selected geometric parameters
are listed in Table 1. These four single crystals are found rather similar
to a previously reported single crystal (CCDC 937825). All four mole-
cules adopt a coplanar structure, with the dihedral angle between car-
bazole and benzo-imidazole (or benzo-thiazole) planes varying from
4° to 15°, so that a donor-acceptor (D-A) structure can be formed. It
has been mentioned that such D-A structure will provide an onset elec-
tronic transition of internal charge transfer (ICT), which leads to prom-
ising photophysical performance, such as strong emission, high
quantum yield and even ASE effect [7,13].
uct was purified on a silica gel column to give white solid (1.5 g). 1
H
NMR (DMSO‑d6) δ 12.71 (s, 1H), 9.77 (s, 1H), 8.02 (d, J = 8.7 Hz, 2H),
7.51 (s, 2H), 7.35 (t, J = 7.9 Hz, 4H), 7.29 (m, 1H), 7.21–7.18 (m, 2H).
ESI-MS: m/z = 283.1. Its single crystal structure was deposited as
CCDC 19237825.
2-(9H-carbazol-2-yl)benzo[d]thiazole (denoted as dye 2) was syn-
thesized as follows.
A solution of carb-CHO (10 mmol in DMSO 10 mL) was dropwise
added into a solution of 2-aminobenzenethiol (10 mmol in DMSO
20 mL) at room temperature. This mixture was heated at 125 °C for
6 h. Pure water was added. Organic product was extracted with
CH2Cl2. The crude product was further purified on a silica gel column
to give white solid (1.7 g). 1H NMR (DMSO‑d6) δ 12.70 (s, 1H), 7.98 (d,
J = 8.7 Hz, 2H), 7.50 (s, 2H), 7.33 (t, J = 7.9 Hz, 4H), 7.27 (m, 1H),
7.20–7.17 (m, 2H). ESI-MS: m/z = 300.1. Its single crystal structure
was deposited as CCDC 19237826.
2-(1H-benzo[d]imidazol-2-yl)-9-ethyl-9H-carbazole (denoted as
dye 3) was synthesized following a similar procedure to that of dye 1,
except that carb-CHO was replaced by carb-Et-CHO in this run. 1H
NMR (DMSO‑d6) δ 9.75 (s, 1H), 8.01 (d, J = 3.1 Hz, 2H), 7.50 (s, 2H),
7.34 (t, 4H), 7.27 (m, 1H), 7.20–7.17 (m, 2H), 4.13–4.09 (d, 2H),
1.28–1.24 (m, 3H). ESI-MS: m/z = 311.1. Its single crystal structure
was deposited as CCDC 19237827.
3.1.2. Electronic transition
For a further understanding on the onset electronic transition of
these dyes, time dependent density functional theory (TD-DFT) calcula-
tion is performed on them. The energy values of their first ten singlet
electronic transitions are listed in Table 2. It is observed that sulfur-
substitution and ethyl-substitution both decrease transition energy. A
graphical presentation for frontier molecular orbitals (FMOs) of these
four dyes is shown as Fig. 2. It is clear that the FMO electron distributions
of these four dyes are rather similar to each other, owing to their similar
molecular structures. The benzo-thiazole group in dye 2 and dye 4 con-
tributes slightly to HOMO, while the benzo-imidazole group in dye 1
3

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
Fig. 1. Single crystal structures of dye 1 (a), dye 2 (b), dye 3 (c) and dye 4 (d).
and dye 3 contributes a lot to HOMO. The LUMOs of these dyes are
mainly composed of carbazole group, and thus found nearly identical
to each other. The electronic transition between HOMO and LUMO
thus has a nature of ICT, which allows promising photophysical perfor-
mance [14].
as 100 nm. No vibronic progressions are detected. It is thus concluded
that these emission spectra come from a charge-transfer (CT) based
emissive center, and this assumption is consistent with the ICT transi-
tion nature in these four dyes. Generally, the FWHM values of benzo-
thiazole-based dyes are larger than those of benzo-imidazole-based
dyes, which means that the electron-donating sulfur atom in the thia-
zole group broadens electronic distribution of excited state, leading to
a broadened emission spectrum. In addition, the emission intensity of
dye 2 and dye 4 is obviously stronger than that of dye 1 and dye 3. For
a better understanding on these dyes, their emission quantum yields
(Φ) are determined and listed in Table 3. Φ values of benzo-thiazole-
based dyes (~0.89) are found 2-fold higher than those of benzo-
imidazole-based dyes (~0.45), owing to the electron-donating effect
from the sulfur atom. These high Φ values make them efficient emitters
for laser application.
3.2. Photophysical performance of dyes 1–4
3.2.1. Absorption spectra
The absorption spectra of dyes 1–4 are recorded in MeCN and shown
as Fig. 3. Key photophysical parameters are listed in Table 3. There are
two major absorption bands for each dye, a strong one peaking at
~350 nm (for dyes 1,2) or 370 nm (for dyes 3,4), and a moderate one
peaking at ~295 nm, respectively. After comparing these absorption
spectra, it is found that ethyl modification on carbazole group red shifts
absorption by ~3 nm. The former absorption band of dyes 3 and 4 shows
red shift of ~20 nm compared to that of dyes 1 and 2. While the latter
one is localized at ~295 nm with minor spectral shift ( 2 nm). Taking
the above DFT calculation result into account, the former absorption
band is assigned as the ICT transition absorption from HOMO (benzo-
thiazole group or benzo-imidazole group) to LUMO (carbazole group),
while the latter absorption band is assigned as the inner-molecular
transition absorption between π and π* of carbazole group. As above
mentioned, an ICT transition generally has a large dipole moment
owing to the distance between its separated σ+ and σ− centers,
which endows itself with promising photophysical performance [12].
3.3. Photophysical variation of dyes 1–4 caused by metal cations
3.3.1. Absorption spectra
Some common metal cations are added into solutions of dyes 1–4, so
that photophysical variation of dyes 1–4 is observed. Corresponding ab-
sorption spectra are shown in Fig. 4. It is observed that dye 1 ICT absorp-
tion (at 350 nm) is influenced by several metal cations, including Fe2+
,
Fe3+, Zn2+ and Hg2+, resulting in absorption red shift and enhanced ab-
sorption intensity (Fe2+ and Fe3+). On the other hand, the π-π* absorp-
tion at 295 nm is generally preserved in the presence of these metal
cations. As for dye 3, its molecular structure is similar to that of dye 1
but the active hydrogen in carbazole group has been replaced by an
ethyl group. As a result, dye 3 absorption is only affected by Fe3+
3.2.2. Emission spectra and quantum yields
The emission spectra of dyes 1–4 are shown in Fig. 3. Gaussian-liked
emission bands are observed for all four dyes. They are broad ones with
FWHM (full width at half maximum) values of 80 4 nm. The Stokes
shift (Δλ) values between absorption and emission peaks are as large
Table 2
Singlet transitions of dyes 1–4 calculated at RB3LYP/6-31G(d) level.
Transition (nm)
Dye 1
Dye 2
Dye 3
Dye 4
Table 1
S0 → S1
S0 → S2
S0 → S3
S0 → S4
S0 → S5
S0 → S6
S0 → S7
S0 → S8
S0 → S9
S0 → S10
323.09
318.89
285.84
271.56
262.67
250.97
247.37
244.96
240.23
238.08
336.70
319.72
291.48
269.57
264.86
262.09
259.70
248.24
245.48
240.09
345.44
335.25
293.33
276.12
268.35
257.04
254.08
249.41
246.26
239.76
365.31
341.63
304.28
275.65
271.38
266.42
264.13
255.14
247.87
242.90
Selected geometric parameters of dyes 1–4.
Bond length (Å)
Dye 1
Dye 2
Dye 3
Dye 4
N1-C6
N1-C7
1.392
1.300
1.367
1.427
6.50
1.390
1.301
1.730
1.758
6.97
1.396
1.303
1.361
1.425
3.51
1.395
1.303
1.730
1.758
15.07
N3-C1 (dyes 1, 3) or S1-C1 (dyes 2, 4)
N3-C7 (dyes 1, 3) or S1-C7 (dyes 2, 4)
Dihedral anglea (°)
a
Dihedral angle between carbazole and benzo-imidazole (or benzo-thiazole) planes.
4

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
Fig. 2. HOMO (up row) and LUMO (down row) of dye 1 (a), dye 2 (b), dye 3 (c) and dye 4 (d).
owing to its strong electron affinity. The presence of other metal cations
has slim effect on both ICT and π-π* absorption bands of dye 3. This good
selectivity makes dye 3 a potential sensor for Fe3+, which will be
discussed later. As for dye 2 and dye 4, their ICT absorption band re-
mains nearly constant in the presence of these metal cations. Their π-
π* absorption band is enhanced or weakened by these metal cations,
with no obvious spectral shift, showing even better selectivity. This
fact suggests that the ICT absorption is vulnerable to metal cations but
the π-π* absorption is insensitive to metal cations. The above DFT calcu-
lation result has suggested that the ICT absorption is originated from
HOMO (benzo-thiazole group or benzo-imidazole group) to LUMO (car-
bazole group), while the π-π* absorption is the inner-molecular
transition absorption between π and π* of carbazole group. It is rational
to assume that metal cations tend to bond with benzo-thiazole group or
benzo-imidazole group and replace the active hydrogen atom. The car-
bazole group in dyes 1–4 has no direct interaction with metal cations
since their absorption shows no obvious spectral shift. The intensity var-
iation should be explained by structural distortion of these dyes when
coordinating with metal cation. A detailed sensing mechanism will be
discussed and confirmed later.
3.3.2. Emission spectra
Emission spectra of dyes 1–4 in the presence of metal cations are
shown in Fig. 5. Similar to the case for absorption, dye 1 responds to
5

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
Fig. 3. Absorption (a) and emission (b) of dyes 1–4 in MeCN (1 μM).
Table 3
which will be confirmed later. As for dye 3, with its active hydrogen
atom replaced by an ethyl group, its emission becomes insensitive to-
wards most metal cations, expect for Fe3+. There is no obvious spectral
shift or new emission band. Fe3+ still quenches dye 3 emission intensity
efficiently (intensity ratio = 7.8%). As for dye 2 and dye 4, their emission
is generally insensitive to all above mentioned metal cations. Fe3+ can
slightly quench their emission with intensity ratio of ~82%. An obvious
emission red shift (5 nm) is observed upon adding these metal cations,
peaking at 495 nm for dye 2 and 500 nm for dye 4, respectively. Such
emission red shift suggests that dye 2 and dye 4 have adjusted their mo-
lecular structure after meeting these metal cations. It is assumed that
these electron-deficient metal cations are attached to the π electrons
of benzo-thiazole group in dye 2 and dye 4, which makes dye 2 and
dye 4 take a coplanar conjugation structure, leading to the emission
red shift. Dye 1 and dye 3 fail to adopt such coplanar conjugation struc-
ture since their benzo-imidazole group and π electrons may have just a
weak interaction with metal cations.
Photophysical parameters of dyes 1–4.
Compound
λ
abs (nm)
λ
em (nm)
Δλ (nm)
FWHM (nm)
Φ
Dye 1
Dye 2
Dye 3
Dye 4
296, 350
293, 370
298, 351
299, 375
450
490
455
496
100
120
104
121
76
84
81
84
0.45
0.89
0.46
0.89
several metal cations, such as Fe2+, Fe3+, Zn2+, Ag+ and Hg2+, showing
emission quenching effect. A strong emission quenching (intensity
ratio = 7.7%) is observed for Fe3+, which is consistent with the absorp-
tion case mentioned above. Regardless of the decreasing emission in-
tensity, no obvious spectral shift or new emission band is observed,
indicating that the emission quenching is finished via an energy transfer
mechanism, instead of chemical reaction or photochemical reaction,
Fig. 4. Absorption spectra of dye 1 (a), dye 2 (b), dye 3 (c) and dye 4 (d) in the presence of metal cations (10 μM:10 μM) in MeCN.
6

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
Fig. 5. Emission spectra of dye 1 (a), dye 2 (b), dye 3 (c) and dye 4 (d) in the presence of metal cations (10 μM:10 μM) in MeCN.
3.4. Sensing performance of dye 3
between the absorbance at 373 nm and Fe3+ concentration, as shown
by the inset of Fig. 6. At Fe3+ concentration of 11 μM, the absorption in-
tensity at 373 nm is nearly identical to that with Fe3+ concentration of
3.4.1. Absorption response towards Fe3+ and sensing mechanism
The above discussion has revealed that Fe3+ quenches the emission
of dye 1 and dye 3 efficiently. Especially, dye 3 shows good selectivity for
Fe3+ over other competing metal cations. As a consequence, its sensing
performance is further analyzed below. The absorption spectra of dye 3
upon increasing Fe3+ concentration from 0 to 11 μM are shown in Fig. 6.
A new absorption band peaking at 373 nm is observed, and its intensity
is increased with increasing Fe3+ concentrations. While the π-π* ab-
sorption of dye 3 is blue shifted to 287 nm and enhanced with increas-
ing Fe3+ concentrations. The new absorption band should be originated
from the adduct between Fe3+ and dye 3. This adduct overlaps its ab-
sorption with dye 3 absorption, which leads to a non-linear relationship
10 μM, suggesting that all dye 3 molecules have been bonded with Fe3+
.
For a better understanding on the adduct between Fe3+ and dye 3,
Job's titration experiment is performed. Here the total concentration
of Fe3+ and dye 3 {[Fe3+] + [dye 3]} is fixed, with [Fe3+] molar fraction
increased from 0.1 to 0.9. Similarly, the ICT absorption is red shifted
from 355 nm ([Fe3+]/[dye 3] = 0.1) to 380 nm ([Fe3+]/[dye 3] = 0.9),
while the π-π* absorption of dye 3 is blue shifted from 298 nm
([Fe3+]/[dye 3] = 0.1) to 287 nm ([Fe3+]/[dye 3] = 0.9), as shown in
Fig. 7. The causation for spectral shift will be well explained by the sens-
ing mechanism which will be discussed below. First, A380/A287 vs
[Fe3+]/{[Fe3+] + [dye 3]} is plotted as Fig. 7 inset, where A380 and
A287 stand for the absorption values at 380 nm and 287 nm, respec-
tively. It is observed that A380/A287 reaches a maximum value at [Fe3+
]
molar fraction of 0.5, indicating that the adduct between Fe3+ and dye
3 adopts a molar ratio of 1:1.
For a better understanding on the adduct between Fe3+ and dye 3,
MS spectrum of a mixture between Fe3+ and dye 3 (1:1) is recorded
and shown as Fig. 8. A strong m/z peak of 367.1 is observed, which ag-
gress well with the mass sum of Fe3+ and dye 3 (m/z = 367.1). The
1H NMR spectra of dye 3 are recorded in the presence and absence of
Fe3+, and shown as Fig. 8. The –NH (benzo-imidazole) peak at δ 9.75
disappears in the presence of Fe3+. With the above data on hand, a pos-
sible sensing mechanism is proposed as Scheme 3. Dye 3 moves its ac-
tive hydrogen atom of –NH (benzo-imidazole) group to the N atom of
carbazole group, resulting in an electron-rich benzo-imidazole ring
(σ−) to form an adduct with Fe3+. The formation of this σ− ring ex-
plains why the ICT absorption of dye 3 is red shifted upon Fe3+. As for
the carbazole group, after receiving the active hydrogen atom, an
electron-deficient ring (σ+) is resulted, which explains why the π-π*
absorption of dye 3 is blue shifted upon Fe3+. According to the sensing
mechanism shown in Scheme 3, Fe3+ and dye 3 form an adduct with
molar ratio of 1:1. Other metal cations, even Fe2+, cannot form similar
Fig. 6. Absorption spectra of dye 3 upon increasing Fe3+ concentrations from 0 to 11 μM in
MeCN. Inset: absorption intensity variation at 373 nm.
7

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
imidazole ring (σ−, Scheme 3), which decreases emissive energy. At
Fe3+ concentration of 11 μM, dye 3 emission intensity is similar to
that with Fe3+ concentration of 10 μM, indicating that all dye 3 mole-
cules have been coordinated with Fe3+. For comparison convenience,
these emission spectra are analyzed by Stern-Volmer equation, as
shown by Formula 2. Here F0 means the emission intensity without
any emission quencher, F means emission intensity, C denotes a general
constant, Ksv is Stern-Volmer constant and [Fe3+] means the concentra-
tion of Fe3+
.
h
i
F0=F ¼ C þ Ksv Fe3þ
ð2Þ
With these emission spectra on hand, F0/F vs [Fe3+] is plotted as
Fig. inset. linear working curve is observed as F0/F
9
A
=
0.9802 + 0.4055*[Fe3+], R2 = 0.9990. The limit of detection (LOD)
value is determined via a traditional method (3σ/N) as 0.55 μM [12].
This working curve shows a down-bending tendency at Fe3+ concentra-
tion of 11 μM, and its working region should be 0–10 μM. There have
been several sensing systems applying valent-recognizing sensing
mechanism [15,16]. A recent literature has reported a similar valent-
recognizing sensing mechanism towards Fe3+, Cr3+ and Al3+ based
on a Zn compound [16]. A linear-liked working curve is observed with
LOD of ≈2.5 μM and R2 < 0.98. Our result is comparable to or even bet-
ter than this literature result. We attribute the causation for our linear
working curve with good selectivity to the valent-recognizing sensing
mechanism with binding ratio of 1:1. This is one of a few fresh sensing
systems applying such sensing mechanism [15,16].
Fig. 7. Absorption spectra of Fe3+ and dye 3 mixture in MeCN with Fe3+ molar fraction
increased from 0.1 to 0.9. Inset: A380/A287 vs [Fe3+]/{[Fe3+] + [dye 3]}.
adduct. This is because Fe3+ is a trivalent cation and thus has strong
ionic potential, which can stabilize the adduct (σ−). The other metal
cations, suffering from low ionic charge number or large ionic radius,
fail to form a stable adduct with dye 3, which explains the sensing selec-
tivity of dye 3 towards Fe3+. To further confirm this statement, two
more trivalent metal cations, Al3+ and Cr3+ are added as interfering re-
agents. It is observed from Fig. S1 (Supporting information) that they
can quench dye 3 emission as well, but their quenching effect is rather
inefficient, compared to Fe3+ quenching effect. This is because the
ionic potential of Al3+ and Cr3+ is weaker than that of Fe3+. In addition,
the fluorescence decay lifetime of dye 3 is measured as 10.0 ns, and de-
creased to 1.7 ns, 5.8 ns and 8.4 ns in the presence of Fe3+, Cr3+ and
3.5. ASE performance of dyes 1–4
The above discussion has confirmed that dyes 1–4 are efficient emit-
ters, which endows them with possible ASE effect. For a tentative eval-
uation on their ASE performance, threshold energy (EA) values of these
four dyes in various solvents are measured and listed in Table 4, includ-
ing hexane, toluene, CH2Cl2 and MeCN. For each dye, its emission wave-
length is red shifted with increasing solvent polarity. This fact is
explained by the ICT nature of emissive state since the ICT-based emis-
sive center has a high dipole moment and thus can be stabilized by po-
larized solvent molecules, leading to emission red shift. As for dye 1 and
dye 3, no valid ASE signal is observed for all tested solvents even with
laser pumping energy of 3500 μJ. It is observed from Table 4 that their
emission quantum yields are all lower than 50%. Although efficient fluo-
rescence can be generated from dye 1 and dye 3 upon Xe lamp excita-
tion, they fail to give ASE effect upon laser excitation. As for dye 2 and
dye 4, their emission quantum yields are ~90% in solvents of high polar-
ity, such as CH2Cl2 and MeCN. No valid ASE effect is observed in solvents
of high polarity. While, in solvents of low polarity, such as hexane and
Al3+
, respectively. These decreased lifetime values and their
monoexponential decay mode indicate a static quenching between
dye 3 and metal cations. This valent-recognizing sensing mechanism al-
lows further sensing application of dye 3.
3.4.2. Emission response towards Fe3+ and working curve
The emission spectra of dye 3 upon increasing Fe3+ concentration
from 0 to 11 μM are shown in Fig. 9. It is observed that dye 3 emission
intensity is gradually quenched by increasing Fe3+ concentrations, as
shown by Fig. S2 (Supporting information). A slight emission red shift
(10 nm) is observed at Fe3+ concentration of 10 μM. This emission red
shift can be explained by the formation of an electron-rich benzo-
Fig. 8. MS spectra (a) and 1H NMR spectra (b) of Fe3+ and dye 3 (1:1).
8

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
Scheme 3. A possible sensing mechanism of dye 3 for Fe3+
.
toluene, their emission quantum yields are higher than 95%, and their
ASE effect is observed. On the other hand, it should be pointed out
that directly comparing the emission quantum yield to ASE perfor-
mance is inappropriate. According to Jagtap and Ray, ASE effect and
laser efficiency of dye solution depend on several factors, including
the dynamic gain characteristics in laser cavity depending on pumping
energy, effective stimulated emission (ESA) cross section, and ground
state absorption (GSA) cross section [17,18]. The absence of ASE effect
in dye 1 and dye 3 should be attributed to at least one of the above
factors.
In hexane, ASE effect is observed for both dye 2 and dye 4. But the
threshold energy values are high, which are 2120 μJ for dye 2 and
1080 μJ for dye 4, respectively. This is because dye 2 and dye 4 have
low solubility in hexane. Their dilute solution in hexane thus needs a
high threshold energy. As for solvent toluene, with an improved solubil-
ity, the threshold energy values are decreased to 870 μJ for dye 2 and
220 μJ for dye 4, respectively. Dye 4 serves as a superior emitter over
dye 2 due to its higher emission quantum yield and lower threshold en-
ergy. Dye 4 is thus selected for later ASE experiment.
Various concentrations of dye 4 in toluene are tried upon laser
pumping energy of 3000 μJ, with ASE efficiency values shown as
Fig. 10 inset, so that the optimal doping concentration can be deter-
mined. It is observed that when dye 4 concentration is lower than
4 mM, ASE efficiency is increased with increasing dye 4 concentration.
This is because a dilute solution generally fails to intercept and absorb
enough excitation photons, compromising population inversion and
consequently ASE efficiency. When dye 4 concentration is higher than
4 mM, re-absorption and self-aggregation effect become sever and
thus compromise ASE process as well. Given the optimal doping con-
centration of 4 mM, the highest ASE efficiency of ~1.4% is observed.
With this optimal doping concentration on hand, ASE output energy
against pumping energy is plotted and shown as Fig. 10. Corresponding
ASE efficiency values are given as well. No surprise, ASE output energy
increases with increasing pumping energy, from 250 μJ to 3500 μJ. ASE
efficiency firstly increases to ~2% at pumping energy of 280 μJ, then de-
creases to ~1% at 1000 μJ and remains. At pumping energy higher than
2000 μJ, ASE efficiency is finally increased to ~1.4%. Thermal effect and
re-absorption process should be deleterious paths for ASE action.
Fig. 11 shows emission and ASE spectra of dye 4 in toluene (4 mM)
upon various pumping energy values. When pumping energy is as low
as 50 μJ, a normal fluorescence spectrum is observed from dye 4, with
FWHM value of 35 nm. At pumping energy of 200 μJ which is close to
the threshold energy of dye 4 (220 μJ), narrowed ASE spectrum is gen-
erated, with FWHM value of 11 nm and emission peak of 435 nm. Given
higher pumping energy values (500 μJ, 1000 μJ and 2000 μJ), standard
ASE spectra are observed, showing FWHM value of 6 nm and emission
peak of 435 nm. A representative photo is given as Fig. 11 inset, where
the ASE beam is clearly identified.
There is still one more point to be discussed though. Photostability of
laser dyes, especially those excited by UV radiation, is imperative for
laser application since their excitation involves a higher singlet state
(S2). ASE output of dye 4 is monitored upon laser pumping energy of
1000 μJ. It is observed from Fig. S3 (Supporting information) that ASE
output of dye 4 is gradually decreased with increasing radiation time.
After 280 min, 91.8% of initial ASE efficiency is preserved, indicating a
good photostability of dye 4. This result is found better than that from
a recent literature (53% of initial value after 120 min) [19]. This is be-
cause dye 4 has a rigid coplanar structure, compared to the fluorescent
dye reported by Sakr and coworkers [19].
4. Conclusion
As a conclusion, four fluorescent dyes were designed with carbazole
as a structural component. Based on their single crystals, their molecular
structure and electronic transition were investigated. There were two
kinds of electronic transitions in these dyes, which were ICT transition
Table 4
ASE parameters of dyes 1–4 under excitation of 355 nm laser.
c
Hexanea
Tolueneb
CH2Cl2
MeCNd
Dye 1, λem (nm)
Dye 1, Φ
Dye 1, EA (μJ)
Dye 2, λem (nm)
Dye 2, Φ
Dye 2, EA (μJ)
Dye 3, λem (nm)
Dye 3, Φ
Dye 3, EA (μJ)
Dye 4, λem (nm)
Dye 4, Φ
428
0.45
∞
433
0.94
2120
429
0.45
∞
429
0.47
∞
433
0.95
870
430
0.49
∞
442
0.46
∞
459
0.90
∞
445
0.46
∞
460
0.91
∞
450
0.45
∞
490
0.89
∞
455
0.46
∞
496
0.89
∞
435
0.96
1080
435
0.97
220
Dye 4, EA (μJ)
∞ means no valid ASE signal detected with laser pumping energy at 3500 μJ (equipment
limit).
a
Dyes 1–4 have low solubility in hexane (2 mM).
5 mM.
5 mM.
5 mM.
b
c
Fig. 9. Emission spectra of dye 3 upon increasing Fe3+ concentrations from 0 to 11 μM in
d
MeCN. Inset: F0/F vs [Fe3+].
9

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
sensing performance were rather different. We tentatively discussed
the correlation between ASE/sensing performance and molecular struc-
ture. This result can be useful when designing laser dyes and sensing
probes for metal cations in the future. For further research, emission
quantum yield should be increased since it plays an important role in
controlling ASE effect. In addition, sensing performance should be im-
proved as well, especially sensitivity.
CRediT authorship contribution statement
Zixuan Yang: Writing - review & editing. Yaqiang Tian: Data
curation, Formal analysis. Sheng Cang: Investigation. Leiyu Zhang: Su-
pervision. Liang Liu: Writing - original draft.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Fig. 10. ASE output energy and efficiency against pumping energy. Inset: ASE efficiency
values of dye 4 in toluene with various concentrations upon laser pumping energy of
3000 μJ.
Acknowledgements
and π-π* transition, respectively. Photophysical performance compari-
son between these four dyes suggested that benzo-thiazole-based
dyes showed better emission performance than benzo-imidazole-
based dyes, owing to the electron-donating effect from the sulfur
atom. After meeting metal cations, benzo-thiazole-based fluorescent
dyes showed emission quenching effect and absorption red shift,
while benzo-thiazole-based fluorescent dyes were insensitive towards
metal cations. Dye 3 showed good selectivity for Fe3+, and its sensing
mechanism was found as the binding between Fe3+ and deprotonated
benzo-imidazole group with molar ratio of 1:1, resulting in emission
quenching and absorption red shift. A linear working curve with good
selectivity was finally observed. Owing to their high emission quantum
yields, benzo-thiazole-based dyes showed ASE behavior, with threshold
energy of ~220 μJ. Given the optimal condition, a highest ASE efficiency
of ~2% was observed for dye 4, with FWHM value of 6 nm and emission
peak of 435 nm. It should be admitted that the above mentioned sensing
performance and ASE performance were not shining ones, compared to
up-to-date result. On the other hand, this work was mainly focused on
the comparison between these four dyes. Although their molecular
structures were similar to each other, their ASE performance and
This project was supported by the National Natural Science Founda-
tion of Ningbo under grant number 2019A610178, the Special research
funding from the Marine Biotechnology and Marine Engineering Disci-
pline Group in Ningbo University.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.saa.2020.118978.
References
[1] T.H. Maiman, Stimulated optical radiation in ruby [J], Nature 187 (1960) 493–494.
[2] Z. Dong, X. Tian, Y. Chen, J. Hou, Y. Guo, J. Sun, J. Ma, A highly selective fluorescent
chemosensor for Hg2+ based on rhodamine B and its application as a molecular
logic gate [J], Dyes Pigments 97 (2) (2013) 324–329.
[3] P.P. Sorokin, J.R. Lankard, E.C. Hammond, V.L. Moruzzi, Laser-pumped stimulated
emission from organic dyes: experimental studies and analytical comparisons [J],
IBM J. Res. Dev. 11 (2) (1967) 130–148.
[4] M.D. McGehee, A.J. Heeger, Semiconducting (conjugated) polymers as materials for
solid-state lasers [J], Adv. Mater. 12 (22) (2000) 1655–1668.
[5] A. Loudet, K. Burgess, BODIPY dyes and their derivatives: syntheses and spectro-
scopic properties [J], Chem. Rev. 107 (11) (2007) 4891–4932.
[6] T.L. Arbeloa, F.L. Arbeloa, I.L. Arbeloa, Correlations between photophysics and
lasingproperties of dipyrromethene-BF2 dyes in solution [J], Chem. Phys. Lett. 299
(3–4) (1999) 315–321.
[7] R. Sola-Llano, J. Banuelos, BODIPY Dyes - A Privilege Molecular Scaffold With Tun-
able Properties [B], 2018https://doi.org/10.5772/intechopen.82682.
[8] N. Boens, V. Leen, W. Dehaen, Fluorescent indicators based on BODIPY [J], Chem. Soc.
Rev. 41 (2012) 1130–1172.
[9] A. Depauw, N. Kumar, M.-H. Ha-Thi, I. Leray, Calixarene-based fluorescent sensors
for cesium cations containing BODIPY fluorophore [J], J. Phys. Chem. A 119 (2015)
6065–6073.
[10] S. Qu, Y. Li, L. Wang, Q. Lu, X. Liu, Waveguide and ultralow-threshold amplified
spontaneous emission in an aligned ordered solid state based on a highly fluores-
cent twin-tapered bi-1,3,4-oxadiazole derivative [J], Chem. Commun. 47 (14)
(2011) 4207–4209.
[11] S. Qu, H. Wang, W. Zhu, J. Luo, Y. Fan, L. Song, H.-X. Zhang, X. Liu, Toward highly fluo-
rescence and ultralow-threshold amplified spontaneous emission in ordered solid
state from twin-tapered bi-1,3,4-oxadiazole derivatives [J], J. Mater. Chem. 22 (9)
(2012) 3875–3881.
[12] Y. Yang, B. Li, L. Zhang, Design and synthesis of triphenylamine-malonitrile deriva-
tives as solvatochromic fluorescent dyes [J], Sensors Actuators B Chem. 183
(2013) 46–51.
[13] N.A. Weires, J. Boster, J. Magolan, Combined Pd/C and montmorillonite catalysis for
one-pot synthesis of benzimidazoles [J], Eur. J. Org. Chem. (2006) 3938–3946.
[14] Y. Kim, M.R. Kumar, N. Park, Y. Heo, S. Lee, Copper-catalyzed, one pot, three-
component synthesis of benzimidazoles by condensation and C-N bond formation
[J], J. Org. Chem. 76 (23) (2011) 9577–9583.
[15] L. Wu, C. Huang, B.P. Emery, A.C. Sedgwick, S.D. Bull, X.P. He, H. Tian, J. Yoon, J.L. Sessler,
T.D. James, Forster resonance energy transfer (FRET)-based small-molecule sensors
and imaging agents [J], Chem. Soc. Rev. 49 (15) (2020) 5110–5139.
Fig. 11. Emission and ASE spectra of dye 4 in toluene (4 mM) upon various pumping
energy values. Inset: a photo of ASE beam (a screen shot from the video file for the ASE
experiment recorded with a high-resolution digital camera).
10

Z. Yang, Y. Tian, S. Cang et al.
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 246 (2021) 118978
[16] B. Dutta, S. Dey, K. Pal, S. Bera, S. Naaz, K. Jana, C. Sinha, M.H. Mir, Supramolecular
assembly of a 4-(1-naphthylvinyl)-pyridine-appended Zn(II) coordination com-
[18] A.K. Ray, S. Kundu, S. Sasikumar, C.S. Rao, S. Mula, S. Sinha, K. Dasgupta, Comparative
laser performances of Pyrromethene 567 and Rhodamine 6G dyes in copper vapour
laser pumped dye lasers [J], Appl. Phys. B 87 (3) (2007) 483–488.
[19] M.E.M. Sakr, M.T.H.A. Kana, A.H.M. Elwahy, H.M. Kandel, M.S. Abdelwahed, S.A. El-
Daly, E.M. Ebeid, Optical, photo physical parameters and photo stability of 6-
substituted-1, 2, 4-triazine mono glucosyl derivative to act as a laser dye in various
solvents [J], Spectrochim. Acta A 232 (2020) 118145-118145-7.
pound for the turn-on fluorescence sensing of trivalent metal ions (Fe3+, Al3+
,
and Cr3+
13163–13171.
)
and cell imaging application [J], New J. Chem. 44 (30) (2020)
[17] K.K. Jagtap, D.K. Maity, A.K. Ray, K. Dasgupta, S.K. Ghosh, High efficiency dye laser
with low fluorescence yield pyrromethene dyes: experimental and theoretical stud-
ies [J], Appl. Phys. B 103 (4) (2011) 917–924.
11