A novel coumarin-derived dithioacetal chemosensor for trace detection of Hg2+ in real water samples

Article abstract of DOI:10.1177/1747519819890561

A novel coumarin-derived dithioacetal chemosensor, 8-(1,3-dithian-2-yl)-7-hydroxy-4-methylcoumarin (LS), has been designed and synthesized. The sensor LS showed highly selective fluorescent sensing for Hg²⁺ with a low detection limit of 0.81 nM in the pH range from 6.15 to 9.96 in ethanol/water (1:1, v/v) solution. The sensing mechanism of LS toward Hg²⁺ was proposed and verified by ¹H nuclear magnetic resonance spectroscopy studies. Under an ultraviolet lamp, the fluorescence color changes could be easily detected by the naked eye. In addition, the sensor LS has been applied in the trace detection of Hg²⁺ in real water samples.

Full text of DOI:10.1177/1747519819890561

8
research-arti
9cle2019 0561CHL0010.1177/1747519819890561Journal of Chemical ResearchKe et al.
Research Paper
Journal of Chemical Research
–6
1
A novel coumarin-derived dithioacetal
chemosensor for trace detection
of Hg in real water samples
© The Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
htt pDs : O/ /d Io: i 1. o 0r .g 1/ 11 07 . 17 1/ 17 77 /41 77 45 71 59 1 89 18 91 98 89 90 0 55 66 11
journals.sagepub.com/home/chl
2+
1
1,2
1
3
Xing Ke , Yilei Fan
, Jing Zhou and Zhongping Huang
Abstract
A novel coumarin-derived dithioacetal chemosensor, 8-(1,3-dithian-2-yl)-7-hydroxy-4-methylcoumarin (LS), has been
designed and synthesized. The sensor LS showed highly selective fluorescent sensing for Hg2+ with a low detection limit
of 0.81nM in the pH range from 6.15 to 9.96 in ethanol/water (1:1, v/v) solution. The sensing mechanism of LS toward
2
+
1
Hg was proposed and verified by H nuclear magnetic resonance spectroscopy studies. Under an ultraviolet lamp, the
fluorescence color changes could be easily detected by the naked eye. In addition, the sensor LS has been applied in the
2
+
trace detection of Hg in real water samples.
Keywords
chemosensor, coumarin, desulfurization, dithioacetal, mercury(II) ions
Date received: 12 August 2019; accepted: 1 November 2019
of methods for efficiently detecting Hg2 is of great
significance.
+
Introduction
Mercury is a natural component of our environment; it con-
stitutes 0.5 parts per million of the Earth’s crust and is one
of the most toxic metals. Mercury species are broadly clas-
sified into three different chemical categories (elemental
forms, divalent inorganic forms, and organic mercury),
which have different toxic-kinetic properties. Hg (inor-
ganic form) is highly soluble, is the most stable form of
inorganic mercury in the aquatic environment; it is carcino-
In the last few decades, significant efforts have been
made to quantify mercury species, and many methods have
been reported, such as atomic absorption-emission spectros-
1
8
9
copy, polarography, reversed-phase high-performance
2
2+
1Key Laboratory of Drug Prevention and Control Technology of
Zhejiang Province, Department of Criminal Science and Technology,
Zhejiang Police College, Hangzhou, P.R. China
College of Pharmaceutical Science & Green Pharmaceutical
Collaborative Innovation Center of Yangtze River Delta Region,
Zhejiang University of Technology, Hangzhou, P.R. China
3
genic and displays high cellular toxicity. The WTO explic-
itly defined that the concentration of Hg in drinking water
2
2
+
−
1
4
must be less than 6µgL (30nM). Mercury can be
absorbed by microorganisms and converted into methyl-
mercury, which can enter into the food chain and cause
various grave disorders and illness. Long-lasting uptake of
high levels of mercury can lead to serious health problems,
such as neurological damage, effects on the immune sys-
tem, Minamata disease, muscle weakness, motion disor-
3
College of Chemical Engineering, Zhejiang University of Technology,
Hangzhou, P.R. China
Corresponding author:
Yilei Fan, Key Laboratory of Drug Prevention and Control Technology
of Zhejiang Province, Department of Criminal Science and Technology,
Zhejiang Police College, Hangzhou 310053, P.R. China.
5
–7
ders, and chronic diseases. Therefore, the development Email: fanyilei@zjjcxy.cn

2
Journal of Chemical Research 00(0)
Scheme 1. Synthetic protocol for sensor LS.
liquid chromatography,10 and infrared spectroscopy.11,12
Unfortunately, most of these methods are not appropriate for
real-time and on-site assays and require expensive devices
and complicated sample preparation. In recent years, chem-
osensors for detecting mercury ions have attracted great
attention due to their high selectivity and excellent sensitiv-
ity. In addition, fluorescent chemosensors which can directly
measure mercury ions in water samples are advantageous in
Figure 1. Absorption spectra of sensor LS (100μM) in ethanol
and water (1:1, v/v) solution obtained by adding aliquots of
2+
[
Hg ] (0, 0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45, 0.50,
.55, 0.60, 0.65, 0.70, and 0.75equiv.).
0
13
terms of time savings, lower costs, and simplicity.
There are two types of fluorescent chemosensors for Hg
detection that have been reported: coordination-based chem-
2+
1
4–16
17–26
osensors
and reaction-based chemosensors.
For reac-
2+
tion-based sensors, the determination of Hg is achieved by
specific chemical reactions between receptors and Hg . This
type of chemosensor usually exhibits excellent selectivity and
2+
2+
high sensitivity toward Hg due to their specific selective
2+ 13
reactivity to Hg . Therefore, research of reaction-based
chemosensors has gained more attention in recent years.
With large molar extinction coefficients, relatively long
excitation and emission wavelengths and high quantum
efficiencies, coumarin derivatives have been widely used
as fluorescent chemosensors. It has been reported that the
thioacetal group can be selectively desulfurized by Hg ,
resulting in the formation of the corresponding aldehyde
2
7
2
+
Figure 2. Fluorescent spectra of sensor LS (2μM) with
and changes of the fluorescent intensity. However, most of 1.0equiv. of various metal ions in ethanol and water (1:1, v/v).
these chemosensors have drawbacks of relatively high fluo- Inset: The colors of sensor LS upon addition of various metal
2
8
ions as viewed by the naked eye under a 365-nm UV lamp.
rescent background signals, limits of detection, and poor
2
9
solubility in aqueous solution. Herein, we have designed
and synthesized a novel “on-off” fluorescent chemosensor,
-(1,3-dithian-2-yl)-7-hydroxy-4-methylcoumarin
Scheme 1), in which a thioacetal group has been added to
a coumarin derivative. The sensor LS showed highly selec-
tive fluorescent sensing for Hg with a low detection limit
of 0.81nM in the pH range from 6.15 to 9.96 in ethanol/
water (1:1, v/v) solution. The recognition mechanism of
The fluorescence emissions of sensor LS (2μM) on add-
ing different metal ions (1.0equiv. of LS) including Ag ,
8
(LS)
+
(
3+
2+
2+
2+
2+
2+
3+
3+
2+
Al , Ba , Ca , Cd , Co , Cu , Cr , Fe , Hg ,
2+
2+
2+
2+
2+
Mg , Mn , Ni , Pb , and Zn were determined in etha-
nol/water (1:1, v/v). After excitation at 369nm, the sensor
LS exhibited very strong fluorescence emission at 451nm.
2
+
2+
Upon the addition of Hg (2μM), the fluorescence intensity
2
+
1
sensor LS toward Hg is proposed and confirmed by H
nuclear magnetic resonance spectroscopy (NMR) studies.
was quenched immediately (Figure 2). In contrast, no sig-
nificant fluorescence change was detected after addition of
other metal ions under the same conditions, except for with
2
+
3+
3+
Cu , Fe , and Cr . These common fluorescence-quench-
Results and discussion
2+
3+
3+
ing ions (Cu , Fe , and Cr ) quenched about 23%–36%
The ultraviolet–visible (UV–Vis) titration spectra of sensor of the fluorescence compared with LS alone. Under the 365-
2
+
LS upon addition of various concentrations of Hg were nm UV lamp, the remarkable light-blue fluorescence emis-
2+
carried out in ethanol and water (1:1, v/v) at room tempera- sion of sensor LS disappeared after adding Hg , but no
ture. As shown in Figure 1, the absorption band of receptor significant fluorescence change was detected for LS solution
LS in the UV–Vis spectrum appears at 272nm. Upon addi- with other metal ions. These results indicate that sensor LS
2
+
2+
tion of increasing concentrations of Hg (0–0.75equiv.) to exhibited high selectivity for Hg over other metal ions.
the solution of sensor LS (100μM), the absorption band at
The quantitative sensing abilities of sensor LS (2μM)
72nm was enhanced gradually, which clearly suggests toward Hg were studied and a working curve was obtained
2+
2
2
+
that LS participates in a reaction with Hg .
(Figure 3). With addition of increasing concentrations of

Ke et al.
3
Hg2+ ions (0.2, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, 1.8, 2.0, 2.2,
According to the 3σ method (limit of detection
LOD)=3σ/K, σ represents the standard deviation of a
blank solution and K represents the slope of the calibration
(
2.4, 2.6, 2.8, 3.0, 3.2, and 3.4μM), the fluorescence inten-
sity of LS at 451nm decreased significantly. Besides, a
2
curve in Supplemental Figure S1), the LOD of chemosen-
good linear correlation (R =0.9937) between the emission
2
+
2+
sor LS for Hg reached 0.81nM, which is much lower for
intensity of LS and the concentration of Hg was observed
2
+
2+
2+
the detection of Hg than other published Hg chemosen-
sors (Table 1).
on addition of 0–1.0equiv. of Hg ions.
3
0–37
These results demonstrate that sensor
LS could be potentially used for selective detection of trace
2
+
amounts of Hg in analytical chemistry.
2
+
To determine the utility of sensor LS (2μM) as a Hg -
selective receptor in the complex background of competing
species, the fluorescence emissions of LS-M solutions (M
represents different metal ions, 1equiv. of LS) were exam-
2
+
ined in the presence of Hg (Figure 4). LS-M solutions
exhibited strong fluorescence emissions, even with com-
2
+
3+
mon fluorescence-quenching ions (such as Cu , Cr , and
3
+
2+
Fe ). Upon addition of Hg (1μM), the fluorescence
emissions of the LS-M solutions were quenched almost
immediately. This result demonstrated that sensor LS has
2
+
the potential to efficiently detect Hg in complex environ-
ments and is not interfered with other competing species.
The fluorescence emission time-course of sensor LS
with Hg2 in ethanol/water (1:1, v/v) was studied (see
+
Figure 3. Fluorescence titrations of 2μM LS (λ =369nm)
ex
2+
Supplemental Figure S2). After adding Hg to the LS
in ethanol and water (1:1, v/v) in the presence of different
2
+
solution, the fluorescence emission of LS-Hg was sig-
2
+
equivalents of Hg ions (0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8,
.9, 1.0, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, and 1.7equiv.). Inset: Graph
of the fluorescence intensity at 451nm as a function of the
nificantly decreased in 2min. The minimum of fluores-
cence intensity of LS-Hg was reached after 6min and
then remained stable. This result demonstrated that the
0
2
+
2
+
concentration ratio of Hg and LS.
Table 1. Comparison of LS-Hg2+ sensing with other dithioacetal-based Hg2+ fluorescent sensors.
Sensor
λ /λ (nm)
Sensing method
Fluorescence31
Detection limit
pH tolerance
7.4
ex em
OH
OH
–9
N
S
S
390/509
1.16×10 M
N
H
S
CH
2
OH
Fluorescence32
1.03×10 M
–9
7.4
3
95/457
S
2
CH OH
N
S
S
Fluorescence33
Fluorescence34
Fluorescence35
–8
4
49/552
00/412
1.59×10 M
7.4
–a
S
HO
S
–6
3
5.0×10 M
S
MeO
O
N
–8
NH
420/538
2.04×10 M
7–12
OH
HO
O
S
S
O
OH
CO
2
H
–
8
4
3
77/502
20/503
Colorimetric
fluorescence
2.4×10 M
7.4
–a
S
CO
2
H
36
S
Et
2
N
O
O
EtS
SEt
Fluorescence37
7.27×10 M
–8
BMe
2
NMe
2
N
OH
S
Fluorescence38
–7
3
49/526
2.0×10 M
5.0–9.0
S
S
This work
369/451
Fluorescence
8.1×10–10
M
6.15–9.96
aNot mentioned.

4
Journal of Chemical Research 00(0)
Additional evidence was given by NMR studies of the
2
+
1
LS and LS-Hg mixture. The H NMR spectra of LS in
2
+
the absence and presence of Hg were recorded in CDCl3.
Significant spectral changes were observed (Figure 5).
Signals for the protons of the methylene (3.16, 2.91, 2.23,
and 1.93ppm) and methine groups (6.26ppm) disappeared
in the presence of Hg , and a new resonance at 10.62ppm
formed being characteristic of an aldehyde proton. These
phenomena indicated that the reaction between Hg and
LS causes desulfurization and formation of the correspond-
ing aldehyde.
2
+
2
+
38,39
By utilizing a previously reported method,
sensor LS
2+
was used to measure the Hg content in actual water sam-
ples, including tap water (from our laboratory), domestic
sewage (from student residences at our university), and
industrial sewage (from industrial areas in Hangzhou City).
All the water samples were filtered through a 0.2-mm filter
membrane to remove large particular impurities, followed by
the removal of remaining organics by extraction processes.
The resulting samples were diluted with ethanol and water
Figure 4. Fluorescence intensities of LS-M (2μM) solutions
2
in the absence and presence of Hg + (1μM). (λ =369nm,
ex
λem =451nm).
(1:1, v/v) in a 10.0-mL volumetric flask. Table 2 shows the
results acquired using sensor LS with the appropriate con-
2+
centration gradient of Hg added. The results indicate that
sensor LS had good recovery and demonstrated high accu-
2+
racy in the analysis of Hg . Therefore, sensor LS can meas-
2+
ure the concentration of Hg in real water samples and has
practical value in the environmental analysis.
Conclusion
Scheme 2. Proposed sensing mechanism between sensor LS
In summary, a highly selective and sensitive coumarin-derived
dithioacetal chemosensor, 8-(1,3-dithian-2-yl)-7-hydroxy-
2
+
and Hg
.
4
-methylcoumarin (LS), has been designed and synthesized.
reaction between LS and Hg2 proceeded quickly, and sen- LS shows significant “on-off” fluorescence behavior with
+
2+
sor LS could be practically used for the practical and rapid Hg in ethanol/water (1:1, v/v) solution, with strong light-
2
+
detection of Hg .
blue fluorescent emission quenching. The detection limit was
–10
–1
To investigate the applicability of sensor LS for detecting calculated to be 8.1×10 molL and remained stable,
Hg2 in different environments, the pH effects were studied detecting Hg in the pH range of 6.15–9.96. In addition, the
+
2+
2+
(
see Supplemental Figure S3). The fluorescence emissions sensor LS exhibits satisfactory results for Hg detection in
of LS (2μM) in the absence and presence of Hg (2 μM) the analysis of real water samples and can be further used in
2+
were detected in various buffer solutions (see Supplemental potential applications for the detection of nanomolar concen-
2+
Table S1). The free sensor LS exhibited relatively strong trations of Hg in chemical and environmental systems.
fluorescence emissions in the pH range of 6.15–9.96. Upon
2+
addition of Hg , the fluorescence emissions of the resulting
solutions were quenched efficiently. These results demon-
strated that sensor LS was highly sensitive toward Hg in a All chemicals and solvents were obtained from commercial
relatively wide pH range (6.15–9.96). Considering that sources and used without further purification, except when
major biological imaging experiments are performed in neu- specified. Solvents used in fluorescent experiments were of
Experimental
2+
1
13
tral environments, sensor LS could be suitable for measur- spectroscopic grade. The H NMR and C NMR spectra of
2+
ing Hg in biological samples.
the ligand were recorded on an Agilent 400-MR DDR2
2
+
Based on the above results, a proposed Hg -promoted spectrometer. The chemical shifts (δ) are recorded in ppm,
hydrolysis desulfurization mechanism for the fluorescence relative to tetramethylsilane (SiMe ). The Fourier trans-
4
2
+
response of LS toward Hg is proposed. As shown in form infrared spectrum was recorded on a Thermo-Nocilet
Scheme 2, sensor LS emits strong light-blue fluorescent IR200 spectrophotometer. Mass spectra were obtained
emission. Upon addition of Hg to the solution of LS, the using a Shimadzu LCMS-2020 spectrometer. A.Mapada
coordination between Hg(NO ) and the sulfur atom of LS UV-6100 UV–Vis spectrophotometer was used for record-
2
+
3
2
caused activation of the carbon atom on thioacetal. The ing the UV–Vis spectra. Fluorescence emission spectra
resulting activated carbon atom is then attacked by a water were recorded on a Dual-FL fluorescence spectrophotom-
molecule to afford the corresponding aldehyde.
eter. Elemental analysis (C, H, and N) was performed with

Ke et al.
5
1
2+
Figure 5. H NMR spectra of sensor LS and LS-Hg in CDCl3.
Table 2. Determination of Hg2+ in real water samples with
sensor LS.
room temperature (25°C) for 12h. After completion of that
reaction, the resulting mixture was made alkaline with satu-
rated sodium bicarbonate to pH 8–9 and then extracted with
dichloromethane (50mL). The organic layer was separated,
dried over anhydrous sodium sulfate, and then concentrated
in vacuo. The crude product was purified by flash chroma-
tography (petroleum/ethyl acetate=2:1, v/v) to afford the
desired product (179mg) as white solid. Yield 61%. M.p.
b
Sample
pHa
Added
Found
Recovery RSD
2
+
2+
Hg (nM) Hg (nM) (%)
(%)
Tap water 6.75
2
4
6
2
4
6
2
4
6
1.91
3.93
6.06
1.96
3.99
6.02
2.05
3.97
6.04
95.50
98.25
101.00
98.00
2.75
Domestic 7.91
sewage
1.21
1.63
−1
2
03–204°C. IR (KBr, cm ): 3310, 2919, 2891, 1712, 1607,
99.75
1
1572, 1392, 1362, 1317, 1064, 855, 809, 675. H NMR
100.33
102.50
99.25
(
400MHz, CDCl ): δ=7.56 (s, 1H), 7.47 (d, J=8.8 Hz, 1H),
3
Industrial
sewage
8.22
6.88 (d, J=8.8 Hz, 1H), 6.26 (s, 1H), 6.14 (s, 1H), 3.13-3.20
(
1
m, 2H), 2.89-2.94 (m, 2H), 2.39 (s, 3H), 2.20–2.26 (m,
H), 1.92–1.99 (m, 1H). C NMR (100 MHz, CDCl3):
100.67
13
δ=160.74, 159.86, 153.10, 151.07, 125.88, 114.45, 113.21,
111.56, 110.96, 39.23, 31.20, 24.58, 18.85. MS (m/z):
RSD: relative standard deviation.
a
Determined with a pH meter before treating with Hg2+
.
b
+
+
Results are based on three measurements.
3
16.95 [M 23] . Anal. calcd for C H O S : C, 57.12; H,
14 14 3 2
4.79; S, 21.78; found: C, 57.15; H, 4.79; S, 21.79%.
an Elementar Vario MICRO cube and the results are within
0.4% of the calculated values. All the solvents were puri-
fied before use. Compounds 1 and 2 (Scheme 1) were pre- The author(s) declared no potential conflicts of interest with
Declaration of conflicting interests
±
pared starting from resorcinol according to a literature respect to the research, authorship, and/or publication of this
4
0
article.
procedure.
Funding
Preparation of 8-(1,3-dithian-2-yl)-7-hydroxy-
-methylcoumarin (sensor LS)
4
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article: This
work was supported by the National Key Research and
Development Program of China (No. 2018YFC0807400) and the
To a solution of compound 2 (204mg, 1mmol) in dichlo-
romethane (10mL) was added 1,3-dimercaptopropane
(108mg, 1mmol), followed by five drops of boron trif- Analysis and Measurement Foundation of Zhejiang Province (No.
luoride-diethyl etherate. The mixture was then stirred at LGC19B050003).

6
Journal of Chemical Research 00(0)
ORCID iD
19. Kim JH, Kim HJ, Kim SH, et al. Tetrahedron Lett 2009; 50:
5
958.
0. Lee DN, Kim GJ and Kim HJ. Tetrahedron Lett 2009; 50:
766.
1. Wang F, Nam SW, Guo Z, et al. Sens Actuators B 2012; 161:
48.
2. Lee HY, Swamy KHK, Jung JY, et al. Sens Actuators B
013; 182: 532.
Yilei Fan
https://orcid.org/0000-0003-0737-9757
2
2
2
4
Supplementary material
Supplementary material is available online for this article.
9
References
2
2
2
3. Kaur M and Choi DH. Sens Actuators B 2014; 190: 542.
4. Prabhu J, Velmurugan K and Nandhakumar R. J Lumin
1
2
. Selin NE. J Environ Monit 2011; 13: 2389.
. Henriques MC, Loureiro S, Fardilha M, et al. Reprod Toxicol
2
014; 145: 733.
5. Kumari N, Dey N and Bhattacharya S. Analyst 2014; 139:
370.
6. Wang G, Xu G, Zhu Y, et al. Chem Commun 2014; 50:
47.
2
019; 85: 93.
2
2
3
4
. Faghiri F and Ghorbani F. J Hazard Mater 2019; 374: 329.
. WHO. The international programme on chemical safety.
Environmental health criteria I. Mercury. Geneva: World
Health Organization, 1976.
2
7
2
2
7. Chen KY, Guo Y, Lu ZH, et al. Chin J Chem 2010; 28: 55.
8. Shinohara Y, Tsukamoto K and Maeda H. J Photoch
Photobiol A 2019; 371: 407.
5
6
7
. Park H, Naveen M, An KL, et al. J Nanosci Nanotechnol
2
019; 19: 6893.
. Sharma BM, Sanka O, Kalina J, et al. Environ Int 2019; 125:
00.
. Kim KH, Kabir E and Jahan SA. J Hazard Mater 2016; 306:
76.
2
9. Cheng XH, Li QQ, Qin JG, et al. ACS Appl Mater Interfaces
3
2
010; 2: 1066.
3
3
3
0. Gao Y, Yi N, Ou Z, et al. Sens Actuators B 2018; 267: 136.
1. Gao Y, Ma T, Ou Z, et al. Talanta 2018; 178: 663.
2. Zhou Y, He X, Chen H, et al. Sens Actuators B 2017; 247:
3
8
9
. Fong B, Mei W, Siu TS, et al. J Anal Toxicol 2007; 31: 281.
. Mandil H, Sakur AA and Alulu S. Asian J Chem 2012; 24:
6
26.
3. Prabhu J, Velmurugan K and Nandhakumar R. J Lumin
014; 145: 733.
4. Tang L, Ding S, Zhang X, et al. J Photochem Photobiol A
017; 340: 15.
5. Xu QW, Wang C, Sun ZB, et al. Org Biomol Chem 2015; 13:
032.
6. Song C, Yang W, Zhou N, et al. Chem Commun 2015; 51:
443.
7. Chang IJ, Hwang KS and Chang SK. Dyes Pigments 2017;
37: 69.
2
980.
3
3
3
3
3
1
1
1
0. Wang L, Zhou JB, Wang X, Wang ZH and Zhao RS. Anal
Bioanal Chem 2016; 408: 4445.
1. Zohar M, Auslender M, Faraone L, et al. Opt Quantum
Electron 2012; 44: 95.
2. Chandrasoma A, Hamid AAA, Bruce AE, et al. Anal Chim
Acta 2012; 728: 57.
3. Guo Y, An J, Tang HY, et al. Mater Res Bull 2015; 63: 155.
4. Wang H and Chan WH. Tetrahedron 2007; 63: 8825.
5. Fan JL, Guo KX, Peng XJ, et al. Sens Actuators B 2009; 142:
2
2
3
1
1
1
4
1
1
91.
3
3
4
8. Qu SZ, Zheng CH, Liao GM, et al. RSC Adv 2017; 7: 9833.
9. Wu X, Chen J and Zhao JX. Analyst 2013; 138: 5281.
0. Huang DD, Chen YH and Zhao JZ. Dyes Pigments 2012; 95:
1
1
1
6. Ho IT, Lai TL, Wu RT, et al. Analyst 2012; 137: 5770.
7. Qi Z and Tian H. Sens Actuators B 2010; 149: 20.
8. Cheng XH, Li S, Zhong AS, et al. Sens Actuators B 2011;
7
32.
1
57: 57.