A ratiometric fluorescent probe with high sensitivity and selectivity for phosgene sensing in solution and gas

Article abstract of DOI:10.1002/jccs.201900484

Phosgene has attracted wide attention because of its important applications and value in modern industry, agriculture, and other fields, though it easily leaks and is difficult to detect. In this work, we designed and synthesized a naphthalimide-based fluorescent probe, which is easy to prepare, stable, and able to discriminate between phosgene, acetyl chloride, oxalyl chloride, thionyl chloride, phosphorus oxychloride, and tosyl chloride. Our results indicate that the probe can react with phosgene selectively and sensitively, showing remarkable ratiometric fluorescence changes. Furthermore, the probe can be made into test strips, which can determine phosgene in air effectively. The present work provides a novel class of naphthalimide-based derivatives with potential application in phosgene sensing in real time simply and safely with further optimization.

Full text of DOI:10.1002/jccs.201900484

Received: 12 November 2019
DOI: 10.1002/jccs.201900484
Revised: 23 December 2019
Accepted: 29 December 2019
A R T I C L E
A ratiometric fluorescent probe with high sensitivity and
selectivity for phosgene sensing in solution and gas
1
1
1,2
1
1,2
Wu Zhou
| Qian Chen | Aibin Wu
| Ying Zhang | Weichu Yu
1
School of Chemistry and Environmental
Engineering, Yangtze University,
Abstract
Jingzhou, People's Republic of China
Phosgene has attracted wide attention because of its important applications
and value in modern industry, agriculture, and other fields, though it easily
leaks and is difficult to detect. In this work, we designed and synthesized a
naphthalimide-based fluorescent probe, which is easy to prepare, stable, and
able to discriminate between phosgene, acetyl chloride, oxalyl chloride, thionyl
chloride, phosphorus oxychloride, and tosyl chloride. Our results indicate that
the probe can react with phosgene selectively and sensitively, showing remark-
able ratiometric fluorescence changes. Furthermore, the probe can be made
into test strips, which can determine phosgene in air effectively. The present
work provides a novel class of naphthalimide-based derivatives with potential
application in phosgene sensing in real time simply and safely with further
optimization.
2
Unconventional Oil and Gas
Collaborative Innovation Center, Yangtze
University, Jingzhou, People's Republic of
China
Correspondence
Aibin Wu and Weichu Yu, School of
Chemistry and Environmental
Engineering, Yangtze University,
Jingzhou 434023, People's Republic
of China.
Email: abwu@yangtzeu.edu.cn (A. W.)
and yuweichu@126.com (W. Y.)
Funding information
Natural Science Foundation of China
(
NSFC), Grant/Award Number: 21801022;
K E Y W O R D S
the National Science and
fluorescent probe, naphthalimide, phosgene, test paper
Technology Major Project of China,
Grant/Award Number:
2017ZX05049-003-007
1
| INTRODUCTION
has attracted extensive attention because of its high
[
4–9]
chemical reactivity and easy preparation.
Rapid,
Phosgene is a colorless suffocating agent. It can enter
human body through the respiratory tract and lead to
pulmonary edema or emphysema, resulting in abnormal
oxygen inhalation, carbon dioxide exhalation, and hyp-
effective, and safe detection of phosgene, therefore, has
become an urgent and significant need for researchers.
The traditional methods to detect phosgene (electro-
chemical method, gas chromatography, etc.) are usually
[1,2]
oxia.
For this reason, phosgene was used as a chemi-
time-consuming, involve tedious processes, or give
[3]
[10–13]
cal weapon during the First World War. Compared
with other chemical weapons such as Tabun, Sarin,
Soman, etc., phosgene production and use are subjected
to strict supervision according to the laws of each country
or region. However, phosgene is widely used as an impor-
tant intermediate or basic raw material in modern indus-
try, agriculture, medical, material, and other fields, and
poor real-time performance.
technology,
Fluorescence probe
[
14–16]
in contrast, has special advantages,
such as simple preparation, low cost, high sensitivity,
[17–19]
high selectivity, and fast and real-time response.
Especially, ratiometric fluorescent probing, which is
based on the ratio of two fluorescence emission peaks,
can provide a more reliable response range and test
result, and eliminate errors that are caused by the con-
centration and environment of the probe. Consequently,
Wu Zhou and Qian Chen contributed equally to this study.
©
2020 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J Chin Chem Soc. 2020;1–6.
http://www.jccs.wiley-vch.de
1

2
ZHOU ET AL.
ratiometric fluorescence probes have attracted great
attention and developed rapidly.
starting material, including four simple reaction steps:
(a) nitration of naphthalene anhydride; (b) reduction of
nitro group; (c) amidation of acid anhydride; and
(d) cyclization of diamino groups. All the compounds
were confirmed by H NMR, C NMR, and high-
resolution mass spectrometry (HRMS).
[20–24]
Recently, a series of fluorescent probes for the detec-
tion of phosgene have been reported (Scheme 1). These
molecules were mainly designed on the basis of the cycli-
zation reaction of phosgene with hydroxyl or amino
1
13
[25]
groups,
and the mechanism is photo-induced electron
transfer. Although these reported probes are effective,
they still have some drawbacks such as a complex struc-
ture, etc., which inevitably hinder their application in
practice.
2.2 | Spectral response of 11 to phosgene
in solution
In this work, we have designed a fluorescent sensor
,7-diamino-2-butyl-1H-benzo[de]isoquinoline-1,3(2H)-
As expected, the ultraviolet (UV)–vis absorption spectra
of a solution of 11 containing trithylamine (TEA) reveal
the conversion of 11 to 12 upon addition of triphosgene.
As shown in Figure 1a, the absorption spectra display an
evident decrease at 225 nm and a slight increase at
435 nm with an isobestic point at 408 nm. Meanwhile,
the fluorescence spectra show prominent fluorescence
maxima from 483 to 455/430 nm with an isobestic point
at 475 nm, and the intensity at 483 nm decreases drasti-
cally (Figure 1b). Consequently, the fluorescence of the
solution changes from green-yellow to blue (Figure 1b
inset).
6
dione (11) for detecting phosgene, in which diamine as
the reactive recognition site and naphthalimide as
the fluorophore moiety are coupled to one single
naphthene-fused motif (Scheme 2). Unlike the reported
o-phenylenediamine-based or diamine-based phosgene
[
22]
sensors,
where one of two amino groups is
substituted by the fluorophore unit, both amino groups
in 11 are free, benefiting their rapid and effective acyl-
ation by phosgene and resulting the formation of the
tetrahydropyrimidin-2(1H)-one ring (Scheme 2). This
greatly inhibits the original electron push-pull effect in
sensor 11, giving it a remarkable and ratiometric blue-
shift fluorescence response, sensitively and selectively.
Moreover, it was easy to prepare by a simple three-step
synthesis compared to the other reported fluorescent
sensors. Furthermore, a portable test strip immobilizing
1
1 on a filter paper by polyethylene glycol was fabri-
cated for visual on-site detection of phosgene in the
gas phase.
2
2
| RESULTS AND DISCUSSION
.1 | Synthesis of related compounds
SCHEME 2 Chemical structures of the green-yellow
fluorescent sensor 11 as well as its fluorescent reaction product
7-butylpyrido[3,4,5-gh]perimidine-2,6,8(1H,3H,7H)-trione (12) for
As shown in Scheme 3, compounds 11 and 12 were syn-
thesized using 4-nitro-1,8-naphthalene anhydride as the
detecting phosgene. Inset shows the corresponding fluorescence
images under 365 nm light
SCHEME 1 Selected
recently reported fluorescent
sensors for phosgene

ZHOU ET AL.
3
SCHEME 3 Synthetic route for
sensor 11. Reagents and conditions: a)
fuming nitric acid and concentrated
ꢀ
sulfuric, 0–70 C, 2 hr; b) SnCl
2
,
concentrated hydrochloric acid, reflux
3
7
hr; c) CH
3
CH
2
CH
2
CH
2
NH
2
, DMF,
ꢀ
0 C 3 hr; d) triphosgene, TEA,
2 2
CH Cl (DCM), r.t. DCM,
dichloromethane; DMF, N,N-
dimethylformamide
FIGURE 1 UV–vis
absorption (a) and fluorescence
(
b) spectra of compound 11
5 μM) in CH Cl (DCM) upon
(
2
2
addition of different
concentrations of triphosgene
(
0–300 μM)/TEA (1:5). Photos of
above-mentioned solutions
under 365 nm wavelength.
DCM, dichloromethane
The key factor of an effective fluorescent sensor is its
response time. Because of the inconvenience of handling
and toxicity of phosgene, triphosgene is employed as a
substitute in solution detection. Consequently, the
response time can be longer than that of direct phosgene-
induced reaction because triphosgene needs to decom-
pose to phosgene in the presence of a tertiary amine.
Time-dependent fluorescence intensities of 5 μM of com-
pound 11 solution at 430 nm after the addition of
triphosgene and TEA were recorded for estimating the
response time. As shown in Figure 2a, there is hardly any
response when triphosgene was added to the solution of
2.3 | The selectivity of 11 toward
phosgene in solution
In order to evaluate the sensing specificity of 11 toward
phosgene, the selectivity experiment was performed by
recording the fluorescence spectra of 11 with phosgene
and some other analytes, namely acetyl chloride
(CH COCl), oxalyl chloride ((COCl) ), thionyl chloride
3
2
(SOCl ), phosphorus oxychloride (POCl ), and tosyl chlo-
2
3
ride (p-CH C H SO Cl).
3
6
4
2
As shown in Figure 3a, a significant blue shift occurs
for triphosgene/TEA and slight change also occurs for
triphosgene (It can be explained by the partial decompo-
1
1 after 20 min. However, a rapid response was observed
when TEA was added to the mixture. The process lasted
for about 20 min before reaching its peak, which
depended on the amount of TEA and the ratio of
triphosgene to TEA. This phenomenon is ascribed to the
TEA-induced decomposition of triphosgene to phosgene
according to the reported literature.
sition of triphosgene). Except for (COCl) , which almost
completely quenches the fluorescence, no evident
changes are observed for the other analytes. The reason
2
for this peculiar phenomenon caused by (COCl) is being
2
explored in our lab. Anyway, only phosgene can induce
the fluorescent color change from green-yellow to blue
under 365 nm light. These results show the higher selec-
tivity of 11 toward phosgene compared to the other
analytes.
Based on fluorescence titration, the fluorescence
intensity ratio (F430/F483) and the concentration of
triphosgene were linearly fitted (Figure 2b). The limit of
detection (LOD) was determined as 2.25 μM according to
[26]
the equation LOD = 3σ/k,
where σ is the standard
deviation of blank measurement (n
k = 0.00686 (R = 0.99851) is the slope of the fitting
=
15) and
2.4 | Detection of phosgene by test strip
embedded with 11
2
straight line (7.2–270 μM).
Hence, such sensitive and rapid response to phosgene
makes 11 a very promising candidate for the detection of
phosgene in practical applications.
To develop a facile, fast, and reliable method for the
detection of phosgene in gas phase, test strips embedded
with 11 were prepared.

4
ZHOU ET AL.
FIGURE 2 (a) Plots of
time-dependent fluorescence
intensity of compound 11
(
5 μM), after addition of
triphosgene (100 μM) after
0 min, and after addition of
TEA (200 μM) after 30 min.
b) Plots of fluorescence
1
(
intensity change (I430/I483
versus concentration of
triphosgene (0–300 μM)
)
FIGURE 3 (a) Fluorescence spectra and (b) relative intensity F430/F483 of 11 (5 μM) in the presence of various species (300 μM) in
DCM containing TEA (600 μM) after 20 min: (1) SOCl , (2) (COCl) , (3) POCl , (4) CH COCl, (5) Tos-Cl, (6) solid triphosgene (200 μM), no
2
2
3
3
TEA, (7) triphosgene (200 μM) + TEA (600 μM), and (8) blank. Inset: photos of the above-mentioned solutions under 365 nm light. DCM,
dichloromethane
1
3
As shown in Figure 4a, the test strips emit green-
yellow fluorescence under 365 nm light. After exposure
to various analytes (vapor) for 20 min, the fluorescence
color of the test strips gradually changed from green-
yellow to blue, which was easily observed under 365 nm
light. The change in fluorescence of the test strips was
also examined after exposure to various analytes
C NMR); the chemical shifts are reported in ppm by
using tetramethylsilane as the internal standard. UV–vis
absorption and fluorescence spectra were recorded at
room temperature on a Shimadzu UV-2450 UV–vis spec-
trometer and a Shimadzu RF-5301PC spectrofluo-
rophotometer, respectively. Mass spectra were obtained
with a Thermo LTQ Orbitrap mass spectrometer.
(Figure 4b). Obviously, these changes are in accordance
with our above statement. Therefore, the specific
response to phosgene exhibited by test strip indicates that
it can be considered as a potential tool for the facile
detection of phosgene in the gas phase.
3.2 | Synthesis and characterization of
compounds 9–12 (Scheme 3)
3.2.1 | 6,7-Dinitro-1H,3H-benzo[de]
isochromene-1,3-dione (9)
3
3
| EXPERIMENTAL
Twenty-five milliliters of fuming nitric acid was added
dropwise to a solution of 6,7-dinitro-1H,3H-benzo[de]
isochromene-1,3-dione (8) (6 g, 24.7 mmol) in 20 ml con-
.1 | Materials and methods
ꢀ
All reagents for synthesis were purchased from commer-
cial suppliers and used without further purification.
centrated sulfuric acid at 0 C, and then the mixture was
ꢀ
heated to 70 C and reacted for 2 hr. After that, the tem-
NMR spectra were recorded in DMSO-d on a Bruker AV
spectrometer (400 MHz for H NMR and 100 MHz for
perature was decreased to room temperature, and the
mixture was poured into 500 ml of ice water. The
6
1

ZHOU ET AL.
5
2
FIGURE 4 (a) Photograph of test strips embedded with 11 upon exposure to various analytes (vapor) after 15 min: (1) SOCl ,
(
2) (COCl)
2 3 3
, (3) POCl , (4) CH COCl, (5) TosCl, (6) triphosgene + TEA, (7) triphosgene, and (8) blank. (b) Photograph of test strips
embedded with 11 upon exposure to gas containing various amounts of phosgene from 1 to 10 mg/L under 365 nm light
+
precipitate was filtered and recrystallized from ethyl ace-
tate and toluene (1:1, vol/vol) to give a yellow product:
20.34, 14.28 ppm. HRMS (ES+) for C H N O [M + H]
16 17 3 2
calcd: 283.1321, found: 283.1215.
ꢀ
1 2
1
3
.4 g, yield 47.8%, m.p. 226–228 C. H NMR (400 MHz,
DMSO-d ) δ 8.76 (dd, J = 20.6 Hz, J = 8.0 Hz, 4H) ppm.
6
13
C NMR (100 MHz, DMSO-d ) δ 159.24, 147.89, 132.93,
3.2.4 | 7-Butylpyrido[3,4,5-gh]perimidine-
2,6,8(1H,3H,7H)-trione (12)
6
1
31.99, 128.33, 125.28, 114.56 ppm.
TEA (65 mg, 0.65 mmol) was added dropwise to a solu-
3
.2.2 | 6,7-Diamino-1H,3H-benzo[de]
tion of probe 11 (100 mg, 0.35 mmol) in 20 ml CH Cl .
2
2
isochromene-1,3-dione (10)
After that, triphosgene solution (274 mg, 0.92 mmol
triphosgene in 20 ml CH Cl ) was added. The mixture
2
2
ꢀ
Compound 9 (2.5940 g, 9 mmol) was dissolved in a solu-
tion of 50 ml of absolute ethanol and 21 ml of concen-
trated hydrochloric acid, and then stannous chloride
was stirred at 25 C for 10 min, and the solvent was
removed. The residue was subjected to column chroma-
tography on silica gel using CH Cl /CH OH 8:1 (vol/vol)
2
2
3
1
(10.1543 g, 45 mmol) was added. The mixture was
as eluent, giving 85 mg orange solid: yield 82%. H NMR
refluxed for 3 hr, then cooled to room temperature, and
filtered. The precipitate was washed with 10 ml anhy-
(400 MHz, DMSO-d ) δ 11.44 (s, 2H), 8.31 (d, J = 7.3 Hz,
6
2H), 6.91 (d, J = 7.4 Hz, 2H), 4.03 (s, 2H), 1.59 (s, 2H),
13
drous ether, 20 ml iced ethanol, dried in vacuum, giving
1.36 (s, 2H), 0.94 (s, 3H) ppm. C NMR (100 MHz,
1
1
.7972 g yellow powder, yield 69.3%. H NMR (400 MHz,
DMSO-d ) δ 161.11, 135.88, 132.95, 131.85, 130.20,
6
+
DMSO-d ) δ 8.50 (d, J = 7.5 Hz, 2H), 6.84 (s, 4H), 7.20 (d,
128.05, 119.51. HRMS (ES+) for C H N O [M + H]
6
17 16 3 3
13
J = 7.5 Hz, 2H), 5.40 (m, 4H) ppm. C NMR (100 MHz,
calcd: 310.1192, found: 310.1373.
DMSO-d ) δ 151.2, 147.3, 128.6, 125.9, 115.7, 113.1,
6
1
05.2 ppm.
3.3 | Preparation of test strips embedded
with 11
3.2.3 | 6,7-Diamino-2-butyl-1H-benzo[de]
isoquinoline-1,3(2H)-dione (11)
Polyethylene glycol (2 g) was dissolved in 50 ml dic-
hloromethane (DCM) to obtain a sticky, homogenous
solution, and then 11 (2 mg) was added with stirring
using a magnetic stir bar. A filter paper was cut into
strips, and then tied with a copper wire. Then, the strips
were immersed in the solution and then taken out and
immediately dried in air. Finally, the test strips with 11
were made simply to detect phosgene in the gas phase.
Compound 10 (0.3534 g, 1.55 mmol) was dissolved in 2 ml
N,N-dimethylformamide (DMF), and 0.096 g n-
butylamine (1.63 mmol) was added. The mixture was
ꢀ
heated to 70 C and reacted for 3 hr. DMF was removed in
vacuum, and then the residue was subjected to column
chromatography on silica gel. Compound 11 was separated
with CH Cl /CH OH 10:1 (vol/vol) as an orange solids:
2
2
3
1
0
8
2
1
3
1
.2513 g, yield 71.11%. H NMR (400 MHz, DMSO-d ) δ
6
.10 (d, J = 8.3 Hz, 2H), 6.84 (s, 4H), 6.78 (d, J = 8.4 Hz,
3.4 | Detection of gaseous phosgene and
other analytes
1
2
H), 3.93 (m, 2H), 1.54 (dt, J = 14.8 Hz, J = 7.5 Hz, 2H),
1
2
.30 (dd, J = 14.9 Hz, J = 7.4 Hz, 2H), 0.90 (t, J = 7.3 Hz,
13
H) ppm. C NMR (100 MHz, DMSO-d ) δ 163.72,
Different concentrations of phosgene gas were prepared
by adding TEA solutions at different concentrations
6
54.93, 133.74, 133.13, 110.70, 109.30, 109.09, 39.02, 30.40,

6
ZHOU ET AL.
(
using DCM as solvent) and triphosgene solutions at dif-
[8] N. K. Gupta, B. Peng, G. L. Haller, E. E. Ember, J. A. Lercher,
ACS Catal. 2016, 6, 5843.
ferent concentrations (using DCM as solvent) into a
0-ml bottle in succession. Solutions of the other relevant
[
9] N. K. Gupta, A. Pashigreva, E. A. Pidko, E. J. M. Hensen,
L. Mleczko, S. Roggan, E. E. Ember, J. A. Lercher, Angew.
Chem. Int. Ed. 2016, 55, 1728.
2
analytes (acetyl chloride, oxalyl chloride, thionyl chlo-
ride, phosphorus oxychloride and tosyl chloride) were
prepared with DCM as the solvent.
[
[
[
[
10] J. Kim, O. G. Tsay, D. A. Atwood, D. G. Churchill, Chem. Rev.
2011, 111, 5345.
11] J. M. Flaud, F. K. Tchana, W. J. Lafferty, L. Manceron,
M. Ndao, J. Mol. Spectrosc. 2016, 326, 87.
12] H. Gao, J. Wu, Y. Zhu, L. Guo, J. Xie, J. Raman. Spectrosc.
4
| CONCLUSIONS
2016, 47, 233.
13] M. T. Baei, A. Soltani, S. Hashemian, S. Mohammadian, Can.
J. Chem. 2014, 92, 605.
14] W. K. Tsai, Y. H. Chan, J. Chin. Chem. Soc. 2019, 66, 9.
15] W. J. Zhu, M. Fang, Q. Zhang, Y. Xu, Z. Y. Wu, L. Cun,
J. Chin. Chem. Soc. 2018, 65, 597.
[16] S. O. Aderinto, J. Chin. Chem. Soc. 2017, 64, 1432.
[17] L. T. Zeng, H. Y. Zeng, S. F. Wang, S. Wang, J. T. Hou,
J. Yoon, Chem. Commun. 2019, 55, 13753.
In summary, we developed a ratiometric fluorescent sen-
sor 11 for the selective and rapid detection of phosgene.
Compound 11 was successfully designed with active
diamine as the reactive recognition site, which was
coupled to a fluorophore-fused motif. The design strategy
not only makes the synthetic process easier but also gives
the molecule better performance. Importantly, test strip
embedded with 11 demonstrated its potential value in
practical application.
[
[
[
[
[
18] Q. Hu, C. Duan, J. J. Wu, D. D, L. T. Z. Su, R. L. Sheng, Anal.
Chem. 2018, 90, 8686.
19] L. T. Zeng, H. Y. Zeng, L. R. Jiang, S. Wang, J. T. Hou,
J. Yoon, Anal. Chem. 2019, 91, 12070.
ACKNOWLEDGMENTS
20] D. Feng, Y. Zhang, W. Shi, X. Li, H. Ma, Chem. Commun.
This work was supported by the National Science and
Technology Major Project of China (No. 2017ZX05049-
2
010, 46, 9203.
[21] X. Zhou, Y. Zeng, L. Chen, X. Wu, J. Yoon, Angew. Chem. Int.
Ed. 2016, 55, 4729.
22] H. C. Xia, X. H. Xu, Q. H. Song, ACS Sens. 2017, 2, 178.
0
03-007) and the National Natural Science Foundation of
[
[
China (No. 21801022).
23] S. L. Wang, L. Zhong, Q. H. Song, Chem. Commun. 2017, 53,
1
530.
ORCID
[
24] Y. Zhang, A. Peng, X. Jie, Y. Lv, X. Wang, Z. Tian, ACS Appl.
Mater. Interfaces 2017, 9, 13920.
Wu Zhou https://orcid.org/0000-0001-5529-1405
[25] W. Q. Zhang, K. Cheng, X. Y. Yang, Q. Y. Li, H. Zhang, Z. Ma,
H. Lu, H. Wu, X. J. Wang, Org. Chem. Front. 2017, 4, 1719.
REFERENCES
[
[
[
1] D. E. Rowe, R. J. Carroll, C. L. Day, J. Am. Acad. Dermatol.
992, 26, 976.
2] D. S. Olivera, H. Hoard-Fruchey, A. M. Sciuto, Toxicol. Mech.
Methods 2017, 27, 45.
3] S. Rose, D. Pavett, Chemical and Biological Warfare, Beason
Press, Boston, MA 1968.
[26] S. L. Wang, C. L. Zhang, Q. H. Song, J. Mater. Chem. C. 2019,
7, 1510.
1
How to cite this article: Zhou W, Chen Q, Wu A,
Zhang Y, Yu W. A ratiometric fluorescent probe
with high sensitivity and selectivity for phosgene
sensing in solution and gas. J Chin Chem Soc. 2020;
[
[
[
4] D. R. Da Rocha, Synlett. 2007, 7, 1172.
5] R. Mullin, Chem. Eng. News 2014, 92, 7.
6] M. Markovic, P. Lopatka, P. Koos, T. Gracza, Org. Lett. 2015,
1
7, 5618.
7] C. Boersch, E. Merkul, T. J. J. Mueller, Angew. Chem. Int. Ed.
011, 50, 10448.
1–6. https://doi.org/10.1002/jccs.201900484
[
2