Multiple fluorescence ΔcIE and ΔrGB codes for sensing volatile organic compounds with a wide range of responses

Article abstract of DOI:10.1039/c1cc13056k

A highly luminescent compound, stilbene-2,4-dimethyl-6-(1,2,2,4- tetramethyl-1,2-dihydroquinolin-6-yl)-1,3,5-s-triazine (MQT), exhibits solvent polarity-induced emission enhancement. A fluorescent sensor array was fabricated with MQT and porous polymer substrates. The colorimetric changes of the array exposed to VOCs can be readily distinguished by the naked eye. Post-processing procedures proved the high selectivity of the array for VOCs. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c1cc13056k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 10052–10054
www.rsc.org/chemcomm
COMMUNICATION
Multiple fluorescence DCIE and DRGB codes for sensing volatile organic
compounds with a wide range of responsesw
a
a
a
a
a
b
Kaijun Tian, Dehui Hu, Rui Hu, Shuangqing Wang, Shayu Li,* Yi Li* and
a
Guoqiang Yang*
Received 25th May 2011, Accepted 19th July 2011
DOI: 10.1039/c1cc13056k
A highly luminescent compound, stilbene-2,4-dimethyl-6-(1,2,2,4-
tetramethyl-1,2-dihydroquinolin-6-yl)-1,3,5-s-triazine (MQT),
exhibits solvent polarity-induced emission enhancement. A fluores-
cent sensor array was fabricated with MQT and porous polymer
substrates. The colorimetric changes of the array exposed to
VOCs can be readily distinguished by the naked eye. Post-
processing procedures proved the high selectivity of the array
for VOCs.
each kind of VOC. VOCs are discriminated between by
an absorbance-based color variation induced by the diverse
interactions between VOCs and reagents in the array.
It is well known that fluorescence is more sensitive to the
environment than absorption, which means that fluorescence-
based sensors can have low background signals and wide
5
dynamic ranges. Previous studies have demonstrated that
D–p–A intramolecular charge transfer (ICT) compounds
show an apparent positive solvatochromism. However, most
6
Volatile organic compounds (VOCs) refer to natural or man-
made organic compounds with significant vapor pressures in
the environment. As a major source of airborne contaminants,
VOCs are known to cause adverse effects on human health.
It has been confirmed that VOCs cause damage to the liver,
kidneys and central nervous system and are even known to be
of them exhibit weak emission or even non-luminescence in
7
highly polar solvents, which stops most ICT compounds from
being applied in a wide range of VOC detection methods.
In this work, a new ICT compound, stilbene-2,4-dimethyl-6-
(1,2,2,4-tetramethyl-1,2-dihydroquinolin-6-yl)-1,3,5-s-triazine
(MQT), was synthesized. The details of the synthesis and
photophysical mechanism are provided in the supporting
informationw. MQT exhibits more than 100 nm shifts of
the emission maximum from hexane to acetonitrile, from
visual blue to yellow, which is indicative of an efficient ICT
process, as shown in Fig. 1. More importantly, in contrast
with most other ICT compounds, the emission quantum
yield of MQT shows a enhancement with increasing solvent
1
a cause of cancer in humans. Hence, the determination of
VOCs is important in environmental monitoring. The most
accurate method is gas chromatography-mass spectrometry
(
GC/MS) which can give a characteristic mass spectrometry
2
fingerprint for each VOC. However, GC/MS requires a rather
complicated procedure and is expensive as a general pro-
cedure. Therefore, it is helpful to develop novel, easily fabri-
cated VOC-responsive sensors with high selectivity, a broad
and rapid response and convenient usage.
Several systems based on electrochemical, conductive, optical
or chromic changes have been proposed for VOCs detection, but
only a few of them show a response across a wide range of VOC
3
species. Among the broad response methods, the colorimetric
sensor array is particularly attractive because of its effectiveness,
4
simplicity and low cost. This method is based on a cross-
responsive principle, rather than having specific receptors for
specific analytes, producing a unique combinatorial response to
a
Beijing National Laboratory for Molecular Sciences,
Key Laboratory of Photochemistry, Institute of Chemistry,
Chinese Academy of Sciences, Beijing 100190, P. R. China.
E-mail: gqyang@iccas.ac.cn, shayuli@iccas.ac.cn;
Fax: (+86) 10-82617315
b
Key Laboratory of Photochemical Conversion and Optoelectronic
Materials, Technical Institute of Physics and Chemistry,
Chinese Academy of Sciences, Beijing 100190, P. R. China.
E-mail: yili@mail.ipc.ac.cn; Fax: (+86) 10-82543518
w Electronic supplementary information (ESI) available: Synthesis,
characterization and other detailed experimental data. See DOI:
Fig. 1 The molecular structure and fluorescence spectra of MQT
À6
10.1039/c1cc13056k
(3.0Â10 M) in six solvents (at 20 1C) with excitation at 400 nm.
1
0052 Chem. Commun., 2011, 47, 10052–10054
This journal is c The Royal Society of Chemistry 2011

View Article Online
Table 1 A summary of the absorption and emission data of MQT in
six different solvents
of the cooling chip. The fluorescence of the array was recorded
by a spectrometer and camera after a fixed interval of time.
Organic solvents are a major source of VOCs. Several ten
million tons of them with different physical properties are
produced annually. Six typical solvents, n-hexane (HA),
toluene (TOL), diethyl ether (DEE), ethyl acetate (EA),
chloroform (TCMA) and acetonitrile (AN) were selected to
be representatives of VOCs due to their wide range of polarity
to test the sensing capabilities of the sensor array. The flow of
dried air saturated with solvent (at 20 1C) passed over the
max
abs
max
À1
Solvent
l
/nm
l
em
/nm Stokes shift/cm
f
F /%
n-Hexane
Toluene
Ether
402
417
411
424, 448
482
483
1291, 2554
3234
3627
2.5
9.5
9.2
Ethyl acetate 417
503
524
540
4100
4064
5291
20.5
47.3
42.2
Chloroform
Acetonitrile
432
420
polarity (Table 1). The high emission quantum yield can be
attributed to the locked nitrogen atom in MQT, which
diminished the non-radiative transition by the rotation of
À1
À1
surface of sensor array at 1 L min (0.5 L min for ether and
À1
2
L min for toluene, because of their relative higher and
8
lower vapor pressures at room temperature, respectively).
Images of the MQT-adsorbed PP, PES, PES-C and PAN films
were taken before and after 60 s VOC exposure at À20 1C, as
shown in Fig. 2. The sensor array exhibits various fluorescence
responses to different solvent vapors, which are easily distin-
guishable by the naked eye. The emission is bathochromic with
an increase in the polarity of the analyte and substrate, with
only a few exceptions, such as in PP film. The PP film shows a
weaker response to acetonitrile vapor than the other lower
polarity solvents. This can be attributed to the great difference
in the cohesive energy density between PP (137 MPa) and
acetonitrile (290 MPa), which only allows a few acetonitrile
molecules to infiltrate into the PP film. Based on a similar
mechanism, the PES, PES-C and PAN films, on exposure to
toluene, are more hypsochromic than n-hexane because the
latter can infiltrate into those films more easily. With a little
red shift of the PP film in toluene, the sensor array can easily
discriminate between alkanes and aromatics, even with a
single sensor. These results are consistent with the expectation
that the fluorescent color of the sensor array is determined
by the interactions between the analyte, fluorophore and
substrate film.
the substituted amino group. The enhancement of the emis-
sion quantum yield with solvent polarity is the so-called
‘
‘negative solvatokinetic effect’’, which has been explained
by mechanisms of biradicaloid charge transfer, proximity
9
effect and conformational changes. This solvent polarity-
dependent fluorescence with high quantum yield suggests that
MQT is a good candidate for detecting VOCs in a wide range.
To increase the discrimination of one receptor responding to
many analytes, a sensor array was fabricated by adsorbing
MQT into porous polymers with different polarities. During
VOC detection, polymer substrates and VOCs will work
together to produce environment surrounding the MQT
molecules, which give a distinct combination response to each
analyte. The polymer also acts as an in situ pre-concentrator
for the analytes and has a role to suppress the aggregation of
the MQT molecules. Four commercial polymers were chosen
as the substrates because of their high surface area and solvent
resistance, including polypropylene (PP), poly(ether sulfone)
(
PES), poly(aromatic ether sulfone) (PES-C) and polyacrylo-
nitrile (PAN). The different cohesive energies of polymer
substrates affects their adsorption capacity for the same
1
0
VOCs, which may alter the interactions between the recogni-
tion reagent and the analyte molecules, resulting in a shift in
the fluorescence emission. A schematic drawing of the feasible
luminescence sensing mechanism for VOCs detection is shown
in Scheme 1.
For a better understanding of the colorimetric change
reflected by the photographs, two kinds of data processing
were applied to the results. The change in the Commission
Internationale de L’Eclairage (CIE) 1931 space is convenient
for direct observation with eyes and a colorimeter, while the
The porous polymer films were immersed into the solution
of MQT and then the solvent was removed. The concentration
of the solution was optimized to avoid aggregation of MQT in
the polymer films. A semiconductor cooling chip was induced
to provide low temperature conditions to accelerate the detec-
tion because the adsorption process of a vapor is generally
1
1
exothermic and spontaneous. A flow of the VOC vapor was
passed into a cage, where the array was placed on the surface
Fig. 2 Photographs of the MQT-embedded porous PP, PES, PES-C
Scheme 1 The luminescence sensing mechanism for VOCs detection.
This journal is c The Royal Society of Chemistry 2011
and PAN before and after exposure to six solvent vapors.
Chem. Commun., 2011, 47, 10052–10054 10053

View Article Online
Table 2 A summary of the digital database of DCIE coordinates for VOCs
PP
PES
PES-C
PAN
HA
(À0.010, À0.008)
(À0.022, À0.008)
(À0.001, 0.016)
(0.040, 0.068)
(0.148, 0.100)
(0.061, 0.033)
(À0.009, 0.002)
(0.028, 0.008)
(À0.034, À0.049)
(0.021, 0.045)
(0.131, 0.109)
(0.087, 0.080)
(À0.020, À0.048)
(À0.061, À0.083)
(À0.041, À0.024)
(0.033, 0.095)
(0.093, 0.120)
(0.128, 0.061)
TOL
DEE
EA
TCMA
AN
(0.009, 0.175)
(0.025, 0.167)
(0.066, 0.202)
(0.156, 0.382)
(0.043, 0.087)
RGB (red, green, blue) color space is suitable for camera
recording and computer processing.
VOCs, as long as the appropriate and sufficient polymers are
selected.
In the DCIE recognition procedure, all of the emission
spectra of the sensor array shown in Fig. 2 were measured
In summary, we developed a simple and universal method
using only one fluorophore to fabricate a fluorescence sensor
array for VOCs detection. Fluorescent modulations are due to
the specific interactions between the VOCs, polymer substrates
and fluorescence dye. Using the sensor array, the vapors are
identified by the solo colorimetric changes, DCIE or DRGB.
In principle, most of VOCs can be detected by using the array,
as long as the polymer substrates are able to adsorb them.
We are grateful to the National Natural Science Foundation
of China (Grant Nos. 21073206 20703049, 20733007,
20873165, 50973118) and the National Basic Research
Program (2007CB808004, 2009CB930802) and Chinese
Academy of Sciences.
(
see Fig. S3 in the supporting informationw) and transformed
into CIE coordinates (see Table S4 in the supporting informa-
tionw). The combination of the CIE coordinates of the sensor
array is unique for each solvent vapor and this can be used
to constitute a fingerprint database to each VOC. A series of
two-dimensional vectors (Dx, Dy) were obtained by subtract-
ing the CIE coordinates of the image before exposure from
those after exposure and the data are listed in Table 2. The
resulting vectors represent the amount of influence on the
detection devices from VOCs. Most of the vectors are much
larger than the resolution (0.001) of a commercial colorimeter
and can be readily measured. This result indicates that, with
the DCIE procedure, the sensor array is highly selective for the
solvent vapors.
Notes and references
1
M. L. Boeglin, D. Wessels and D. Henshel, Environ. Res., 2006,
00, 242.
(a) C. Lindinger, P. Pollien, S. Ali, C. Yeretzian, I. Blank and
T. Mark, Anal. Chem., 2005, 77, 4117; (b) J. de Gouw and
In the DRGB procedure, a simple algorithm was used to
process the images shown in Fig. 2 to obtain clearer RGB
color-change profiles. The RGB values of the images before
exposure were first set to 128 (mid-grey) and, then, the changes
of the three color channel (listed in Table S5w) after vapor
exposure were added to get the final values. Consequently,
new images of the RGB changes before and after exposure
are shown in Figure S4w, which enables an easier visualization
of the influence of vapors on the sensor array. These more
intuitive images establish a clear connection between the
fluorescence change and solvent species. To better understand
the relationships illustrated by the images, a principal compo-
1
2
¨
C. Warneke, Mass Spectrom. Rev., 2007, 26, 223.
3
(a) N. A. Rakow, M. S. Wendland, J. E. Trend, R. J. Poirier,
D. M. Paolucci, S. P. Maki, C. S. Lyons and M. J. Swierczek,
Langmuir, 2010, 26, 3767; (b) J. Yoon, S. K. Chae and
Jong-Man Kim, J. Am. Chem. Soc., 2007, 129, 3038;
(
c) Y. Takashima, V. M. Martınez, S. Furukawa, M. Kondo,
´
S. Shimomura, H. Uehara, M. Nakahama, K. Sugimoto and
S. Kitagawa, Nat. Commun., 2010, 2, 168.
4 (a) N. A. Rakow and K. S. Suslick, Nature, 2000, 406, 710;
(b) N. A. Rakow, A. Sen, M. C. Janzen, J. B. Ponder and K. S.
Suslick, Angew. Chem., Int. Ed., 2005, 44, 4528; (c) L. Feng,
C. J. Musto and K. S. Suslick, J. Am. Chem. Soc., 2010, 132, 4046.
1
2
nent analysis (PCA) on the DRGB data was performed and a
two-dimensional plot (second principal component as a func-
tion of the first principal component) is obtained, as shown in
Fig. S5.w The results reveal that the combinational array of the
four sensors shows a better recognition ability for the vapors
than a single sensor.
5
(a) J. R. Lakowicz, Principles of Fluorescence Spectroscopy,
ed. Springer Press, New York, 3rd edn, 2006; (b) F. Samain,
S. Ghosh, Y. N. Teo and E. T. Kool, Angew. Chem., Int. Ed.,
2
010, 49, 7025.
6
A. P. Kulkarni, P. T. Wu, T. W. Kwon and S. A. Jenekhe, J. Phys.
Chem. B, 2005, 109, 19584.
7 (a) Y. Ren, X. Q. Yu, D. J. Zhang, D. Wang, M. L. Zhang,
G. B. Xu, X. Zhao, Y. P. Tian, Z. S. Shao and M. H. Jiang,
J. Mater. Chem., 2002, 12, 3431; (b) Y. Z. Cui, Q. Fang,
Z. L. Huang, G. Xue, G. B. Xu and W. T. Yu, J. Mater. Chem.,
2004, 14, 2443.
From the results above, the fluorescence sensor array shows
a broad response to diverse VOC vapors and is significant due
to multiple aspects. Firstly, it proves that the ICT compound
MQT, with a large polarity-dependent wavelength shift and
high quantum efficiency in polar solvents, can be used as a
vapor sensor with a broad response. Secondly, the sensor
array is fabricated simply with a single fluorescent compound
and various porous polymers as the substrate, which differs
from other arrays that require several kinds of compounds.
The optimization of flow chambers and polymer films can
reduce the size of the array, making the system more sensitive.
Thirdly, the principle of the fluorescence response is based on
the combinational effects of both the polymer and VOCS,
Thus, fluorescent sensing can be extended to the majority of
8
J. X. Liu, Z. J. Diwu, W. Y. Leung, Y. X. Lu, B. Patch and
R. P. Haugland, Tetrahedron Lett., 2003, 44, 4355.
(a) K. Rurack, M. L. Dekhtyar, J. L. Bricks, U. Resch-Genger and
W. Rettig, J. Phys. Chem. A, 1999, 103, 9626; (b) X. P. Li, Y. Qian,
S. Q. Wang, S. Y. Li and G. Q. Yang, J. Phys. Chem. C, 2009,
9
1
13, 3862.
1
0 S. F. Grebennikov, Y. S. Chulkova and E. I. Zaitseva, Prot. Met.
Phys. Chem. Surf., 2009, 45, 177.
11 K. Banerjee, P. N. Cheremisinoff and S. L. Cheng, Water Res.,
997, 31, 249.
2 P. C. Jurs, G. A. Bakken and H. E. McClelland, Chem. Rev., 2000,
00, 2649.
1
1
1
1
3 F. G. Ince, H. Goktas, Z. Ozbek, R. Capan and F. Davis, Mol.
Cryst. Liq. Cryst., 2010, 521, 104.
1
0054 Chem. Commun., 2011, 47, 10052–10054
This journal is c The Royal Society of Chemistry 2011