Talanta 131 (2015) 121–126
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Electrochemical OFF–ON ratiometric chemodosimeters for the selective and rapid detection of fluoride Veerappan Mani a, Wen-Yung Li a, Jiun-An Gu a, Chun-Mao Lin b, Sheng-Tung Huang a,n a b
Department of Chemical Enginnering and Biotechnology, National Taipei University of Technology, Taipei 10608, Taiwan (ROC) Department of Biochemistry, College of Medicine, Taipei Medical University, No. 250, Wu-xing Street, Taipei 110, Taiwan (ROC)
art ic l e i nf o
a b s t r a c t
Article history: Received 27 May 2014 Received in revised form 23 July 2014 Accepted 24 July 2014 Available online 4 August 2014
We have described two “OFF–ON electrochemical latent ratiometric redox chemodosimeters”, 1,4-Bis (tert-butyldimethylsiloxy)benzene (H2Q0 ) and 1,4-Bis (tert-butyldimet hylsiloxy)-2-methoxybenzene (MH2Q0 ) for the selective detection of inorganic fluoride. The electrochemical signals of hydroquinone (H2Q) and o-methoxy hydroquinone (MH2Q) within this latent redox probes (H2Q0 and MH2Q0 ) were completely masked by protecting their hydroxyl group as silylether (OFF state). The externally added fluoride ions triggered the deprotection of H2Q0 and MH2Q0 and unmasked the electrochemical properties of H2Q and MH2Q respectively. The electrochemical reporters (H2Q and MH2Q) presented a pair of redox peaks at the electrode surface (ON state) and the peak currents are linearly dependent with the concentration of fluoride which leading to the ratiometric detection of fluoride. The limit of detection (signal-to-noise ratio¼ 3) observed for the probes are 23.8 mM and 2.38 mM for H2Q0 and MH2Q0 respectively. The deprotection is highly selective for fluoride over other anions investigated. The probes are highly stable and the proposed approach offers rapid response time and promising practical applicability. The proposed strategy holds great promise for the commencement of new H2Q based electrochemical probes by tuning the electrochemical behavior of H2Q. & 2014 Elsevier B.V. All rights reserved.
Keywords: Hydroquinone Latent redox ratiometric probe Chemodosimeters Sensor Fluoride Selectivity
1. Introduction Fluoride is widely used as the active ingredient in toothpaste formulations for the prevention of dental caries and other health benefits. In addition, fluorides are one of the essential trace elements required to form bones and teeth in human [1]. However, fluorides are potentially toxic, if their concentrations exceed levels that cause severe health risks such as dental fluorosis and bone diseases [2,3]. Therefore, the permissible overall quantity of fluoride present in toothpaste is regulated within the concentration range of 0.50–1.50 mg g 1, while US Environmental Protection Agency's (US EPA) has recommended the upper allowed limit of 2 ppm (105 mM) fluorides in drinking water to prevent dental fluorosis [4]. Therefore, determination of fluorides is of great significance [5–9]. In the past years, several analytical methods have been developed for the determination of fluoride such as chromatography [10], ion selective electrodes [11–13], anion recognition [14], fluorimetry [15], spectrophotometry [16], and colorimetry methods [17]. But, most of these traditional ana-
n
Corresponding author. Tel.: +886 2271 2171 2525; fax: +886 02 2731 7117. E-mail address: [email protected] (S.-T. Huang).
http://dx.doi.org/10.1016/j.talanta.2014.07.074 0039-9140/& 2014 Elsevier B.V. All rights reserved.
lytical methods involve time-consuming and tedious procedures. Recently, several research groups have published optical chemodosimeters for the determination of fluoride; in this case, the detection approach involves typical fluoride triggered cleavage of Si–O bond in the designed probes with concomitant release of the optical reporters [8,17–21]. These optical chemodosimeters are able to provide excellent selectivity for the determination of fluoride attributed to the strong chemical affinity between fluoride and silicon. Compared with optical techniques, electrochemical techniques have relatively better performances in terms of short reaction time, high sensitivity, simplicity of the voltammetric instrumentation, simple operating protocols, direct use in pointof-care assays, potential for miniaturization and portability [22]. In the present work, we incorporated the similar concept of optical chemodosimeters into electrochemical redox probes for the simple and selective detection of fluoride at electrode surface. The schematic representation for the mechanistic pathway of the designed “OFF-ON electrochemical latent ratiometric redox chemodosimeters” is given as Scheme 1: (1) Masking the electrochemical signal of hydroquinone (H2Q) through silylation to obtain the probe, 1,4-Bis(tert-butyldimethylsiloxy)benzene (H2Q0 , OFF state), (2) fluoride triggered cleavage of Si–O bond, and subsequent removal of silyl protecting group through quinone type
122
V. Mani et al. / Talanta 131 (2015) 121–126
Scheme 1. Schematic representation for the electrochemical determination of fluoride via fluoride induced electrochemical “ON–OFF” ratiometric latent redox switch.
rearrangement with concomitant release of electrochemical reporter (H2Q), and (3) H2Q readily undergoes electrochemical redox reaction and presented its redox peaks at the electrode surface. The deprotection is highly specific for sodium fluoride and the amount of released H2Q is quantitative to the concentration of fluoride. The same “latent redox probe approach” of H2Q0 was extented to 1,4-Bis (tert-butyldimet hylsiloxy)-2-methoxybenzene (MH2Q0 ), which is a methoxy derivative of H2Q0 . Preparation of the probes (H2Q0 and MH2Q0 ) involves very simple silylation reaction and the assay pathway involves rapid and quantitative determination of fluoride. The aim of the present work is to develop a new electrochemical platform for the selective determination of fluoride. To the best of our knowledge, there is no report available in the literature employing H2Q as the “Electrochemical OFF–ON Ratiometric Chemodosimeters” for the determination of fluoride. Considering the advantages of electrochemical methods and the need for the development of novel electrochemical approaches to selectively detect fluoride, herein we have designed two “electrochemical ratiometric redox probes”, by incorporating HQ and o-methoxy hydroquinone (MH2Q) as the electrochemical reporters. The limit of detection (LOD) observed for these probes (23.8 mM for H2Q0 and 2.38 mM for MH2Q0 ) are within the upper allowed limit mandated by the US EPA.
2. Experimental 2.1. Materials and methods H2Q, sodium fluoride, and all the other chemicals were purchased from Sigma-Aldrich and used without further purification. Electrochemical measurements were recorded in a conventional three electrode cell employing BAS glassy carbon electrode (GCE) as the working electrode (area 0.071 cm2), saturated Ag/Ag þ (AgNO3) filled with acetonitrile containing tetra-n-butylammonium bromide (TBAP) as the reference electrode and Pt wire as the counter electrode. 1 mM of TBAP dissolved in DMF was used as the supporting electrolyte. The determination of fluoride was carried out in DMF containing TBAP (1 mM), HEPES buffer (0.05 M, pH 7.8) (DMF–TBAP–HEPES solution), and probe (1 mM). The solutions were deoxygenated with pre-purified nitrogen for 5 min before performing each electrochemical experiment. 2.2. Preparation of H2Q0 and MH2Q0 H2Q0 and MH2Q0 was prepared in accordance to the procedure reported in the literature [23,24]. H2Q (1.67 g, 15.2 mmol) was treated with TBDMS chloride (11.4 g, 75.3 mmol) and imidazole (10.2 g, 150 mmol) in DMF (5 ml) for overnight to yield H2Q0 as a colorless solid (4.64 g, 90%). 2-methoxybenzene-1,4-diol (2.13 g,
15.2 mmol) was treated with TBDMS chloride (11.4 g, 75.3 mmol) and imidazole (10.2 g, 150 mmol) in DMF (7 ml) for overnight to yield the MH2Q0 as a colorless solid (4.784 g, 85%). 1H NMR data of the H2Q0 and MH2Q0 were in good agreement with those reported in the literature.
3. Results and discussion 3.1. Determination of fluoride using H2Q0 The cyclic voltammogram (CV) acquired at GCE in DMF–TBAP– HEPES solution containing H2Q0 (1 mM) (curve b, Fig. 1A) did not exhibit any characteristic redox peaks in the absence of fluoride, indicating the complete masking of H2Q redox sites. However, the CV obtained in the presence of 0.24 mM NaF (curve c, Fig. 1A) exhibited well defined quasi reversible redox peaks related to the characteristic redox reaction of H2Q. The redox reaction involves two electrons and two protons and occurred via semiquinone formation which was established fabulously in the literature [25]. The oxidation peak corresponding to the oxidation of H2Q to pbenzoquinone (Q) was observed at the potential of þ0.10 V, whereas the reduction peak corresponding to the reduction of Q to H2Q was observed at the potential of 0.71 V. Thus, the CV results clearly indicating that the addition of fluoride triggered the silyl deprotection and uncloaks the redox active center of H2Q0 . Moreover, we observed that short time (5 min) is sufficient to complete the deprotection; thus our assay platform involves short reaction time allowing rapid detection of fluoride. CV obtained at GCE in DMF–TBAP–HEPES solution containing 1 mM of pristine H2Q has been carried out to correlate the electrochemical behavior of pristine H2Q (inset to Fig. 1A) and regenerated H2Q from H2Q0 (curve c, Fig. 1A). CV obtained for the pristine H2Q is quite consistent with that of released H2Q from H2Q0 in terms of peak potentials and peak shapes supports the proposed protection/ deprotection mechanistic pathway. The effect of scan rate (ν) on the redox behavior of H2Q was examined in DMF–TBAP–HEPES solution containing 1 mM H2Q0 and 0.24 mM of NaF at the scan rates from 0.1 to 1.2 V s 1 (Fig. 1B). Both the anodic (Ipa) and cathodic peak current (Ipa) increased linearly with increase in scan rates from 0.1 to 1.2 V s 1. A plot of square root of scan rates (ν1/2) and anodic peak current (Ip) exhibited linear relationship indicating that the redox reaction of released H2Q occured at the GCE is a diffusion controlled electron transfer process (inset to Fig. 1B). Fig. 2A shows the CVs obtained at GCE in DMF–TBAP–HEPES solution containing H2Q0 (1 mM) and different concentrations of NaF (a ¼0.024, b ¼0.12, c¼0.24, d ¼0.48, e¼ 0.72, f ¼0.96, and g¼ 1.19 mM). As evident from the figure, a pair of well-defined redox peaks of H2Q was present for the 0.024 mM addition of fluoride. Moreover, both Ipa and Ipc were increased linearly
V. Mani et al. / Talanta 131 (2015) 121–126
123
Fig. 2. (A) CVs obtained at GCE in DMF–TBAP–HEPES solution containing H2Q0 (1 mM) with different concentrations of fluoride; 0.024 (a), 0.12 (b), 0.24 (c), 0.48 (d), 0.72 (e), 0.96 (f), and 1.19 mM (g). (B) Plot of [fluoride] vs. Ipa and Ipc.
Fig. 1. (A) CVs obtained at GCE in DMF–TBAP–HEPES solution (a) containing H2Q0 (1 mM) in the absence (b) and presence of 0.24 mM NaF (c) and 1 mM pristine H2Q (inset). (B) CVs obtained at GCE in DMF–TBAP–HEPES solution containing H2Q0 (1 mM) in the presence of 0.24 mM NaF at different scan rates from 0.1 to 1.2 V s 1 (curves a–i). Inset: ν1/2 vs. Ipa and Ipc.
upon increase in the concentration of fluoride. A plot of Ipa and Ipc vs. concentration of fluoride exhibited a linear relationship (Fig. 2B) with the slope of 7.153 mA mM 1 and 8.586 mA mM 1 respectively. The linear range was observed beween 23.8 mM and 1.19 mM, while limit of detection (LOD) was found to be 23.8 mM (S/N ¼ 3) which is sufficiently enough to detect the US EPA mandated upper limit of fluoride concentration in drinking water [4]. Monitoring Ipc of the H2Q redox peaks offers more sensitivity than Ipa and therefore we choose Ipc for the determination of fluoride. 3.2. Determination of fluoride using MH2Q0 MH2Q0 also follows similar electrochemical protection/deprotection pathway as H2Q0 (Scheme 1). No redox peaks were observed in the
absence of fluoride (protected, OFF) (curve a, Fig. 3A), whereas well defined redox peaks were observed in the presence of 0.24 mM NaF (deprotected, ON) (curve b, Fig. 3A). Here, the redox peaks are ascribed to the redox conversion between MH2Q and MQ (o-methoxy benzoquinone). The CV obtained at the pristine MH2Q (inset to Fig. 3A) is quite consistent with that obtained at the regenerated MH2Q from MH2Q0 . MH2Q0 follows concentration-dependent increase in Ipc with different concentrations of fluoride (Fig. 3B) and exhibited linear concentration range from 2.38 mM to 0.96 mM (inset to Fig. 3B). LOD (S/N ¼3) of MH2Q0 was calculated to be 2.38 mM which is 10 fold lower than that observed at H2Q0 . The probable reason for the lower LOD of MH2Q0 might be the chelating effect caused by the formation of stable five membered chelating ring when sodium fluoride approaches MH2Q0 which brings the fluoride anion more closer to silicon (Scheme 2). As a result, the removal of silyl ether become more easy even at low concentration of fluoride, whereas formation of this kind of chelating effect is not possible in H2Q0 . Perhaps, this result indicating that any substituent on the ring of H2Q0 may influence the analytical performance of the fluoride detection. Therefore the OFF/ON switching ability of the H2Q0 can be tuned by different substituents to boost the performance of the resulting sensor. The sensor performance of the probes were listed in Table 1. It is evident from the table that, LOD and linear range of these probes are within the US EPA mandated upper limit of fluoride concentration in drinking water.
124
V. Mani et al. / Talanta 131 (2015) 121–126
Scheme 2. Schematic representation for the chelating effect observed in MH2Q0 .
Table 1 Electroanalytical parameters and sensor performance of the probes H2Q0 and MH2Q0 .
Fig. 3. (A) CVs obtained at GCE in DMF–TBAP–HEPES solution containing MH2Q0 (1 mM) in the absence (a) and presence of 0.24 mM NaF (b) and 1 mM pristine MH2Q (inset). (B) CVs obtained at GCE in DMF–TBAP–HEPES solution containing MH2Q0 with different concentrations of fluoride; 0 (a), 2.38 10 3 (b), 11.9 10 3 (c), 23.8 10 3 (d), 0.12 (e), 0.24 (f), 0.48 (g), 0.72 (h), and 0.96 mM (i). Inset: [fluoride] vs. Ipc.
3.3. Selectivity In order to study the selectivity of H2Q0 towards fluoride detection, CV studies were carried out in the presence of variety of likely anionic interferences such as NaI, NaCl, NaBr, KH2PO4, Na2HPO4, LiClO4, NaOAc, NaNO3, KCN and NaN3. The CV obtained at GCE in DMF–TBAP–HEPES solution containing H2Q0 (1 mM) exhibited well defined Ipc response towards addition of 2 mM NaF (a), whereas no noteworthy responses were observed for the addition of 2 mM of NaI (b), NaCl (c), NaBr (d), KH2PO4 (e), Na2HPO4 (f), LiClO4 (g), NaOAc (h), NaNO3 (i), KCN (j) and NaN3 (k) (Fig. 4A). The selectivity studies clearly revealed that only fluoride was able to unmask the redox active sites of H2Q from H2Q0 among all the anions tested (Fig. 4B). This must be ascribed to the special chemical affinity between fluoride and silicon [5]. Thus, H2Q0 is a highly selective redox probe for the determination of fluoride even in the presence of other common interferrants. Similar to H2Q0 , MH2Q0 has also found to be insensitive to other
Probes
H2Q0
MH2Q0
Limit of detection Linear range Repeatability Reproducibility
23.8 mM 23.8 mM–1.19 mM 2.53% 3.22%
2.38 mM 2.38 mM–0.96 mM 2.72% 3.44%
anions tested and selective for the determination of fluoride (Figure not shown). 3.4. Stability, repeatability and reproducibility To investigate the stability of the probes, DMF–TBAP–HEPES solutions containing 1 mM of each probe (H2Q0 and MH2Q0 ) were treated under thermal (60 1C) and ultrasonic conditions for 1 h and their respective CVs were recorded. Under these conditions, no characteristic redox peaks of their corresponding redox reporters (H2Q and MH2Q) were observed in the absence of fluoride. On the other hand, the redox reporters were liberated from the probes within few seconds in the presence of fluoride (Figs. 1B and 3B). Therefore, we concluded that both the probes are highly stable under thermal and ultrasonication conditions, while fluoride ions only able to trigger the deprotection. Both the two probes exhibited appreciable repeatability and reproducibility with acceptable recoveries for five experiments. Repeatability experiments were carried out using single electrode, while reproducibility experiments were carried out with five different electrodes (Table 1). 3.5. Real sample analysis We demonstrated the practical feasibility of the proposed approach towards determination of fluoride present in water
V. Mani et al. / Talanta 131 (2015) 121–126
125
Table 2 Real sample analysis in water samples. Probes Tap water
River water
Added (mM)
Found (mM)
Recovery (%)
Added (mM)
Found (mM)
Recovery (%)
H2Q0
25 250
24.2 256
96.8 102.4
25 250
26 258
104 103.2
MH2Q0
25 250
25.83 245.1
103.3 98
25 250
24.13 244
96.52 97.6
ferrocenyl compounds as receptors are facing difficulties in the detection of aqueous fluoride ions [28–33]. On the other hand, electrochemical techniques based on the modified electrodes often lack selectivity. Most recently, a potentiometric sensor based on poly(3-aminophenylboronic acid) modified graphite rod electrode was reported for the determination of fluoride [34], however its detection range was beyond the upper limit imposed by US EPA. In the present work, we established a highly selective electrochemical inorganic fluoride sensing platform which allowed the direct transferring of sensing signal electronically and applicable for sensing fluoride in the biological sample. The LOD observed for the probes (23.8 mM for H2Q0 and 2.38 mM for MH2Q0 ) are within the relevant window of US EPA's mandated upper limit [4]. H2Q is a potential electrochemical reporter attributed to its stable, predictable, and tailorable electrochemical properties, and its electrochemical signal could be easily measured by simple electrochemical methods such as cyclic voltammetry [25,35]. To the best of our knowledge, this is the first report demonstrating “latent redox ratiometric probe” approach through simple selective protection/deprotection strategy for the detection of fluoride.
4. Conclusions
Fig. 4. (A) CVs obtained at DMF–TBAP–HEPES solution containing H2Q0 (1 mM) with the presence of 2 mM of NaF (a) and NaI (b), NaCl (c), NaBr (d), KH2PO4 (e), Na2HPO4 (f), LiClO4 (g), NaOAc (h), NaNO3 (i), KCN (j), and NaN3 (k). (B) Plot between various anions and their response towards NaF.
samples. The spiked fluoride concentrations are 25 mM and 250 mM. The found and recovery values are given in Table 2. As evident from the table, both the probes presented acceptable recoveries in tap and river water samples revealed the practical feasiblity of the proposed approach. 3.6. Comparison of the proposed “latent ratiometric redox probe approach” with other methods In the past decades, several optical probes have been reported which are relying on hydrogen bonds or Lewis acid coordination of fluoride with the probes, and most of these probes could only be operated in organic solvents to detect tetrabutylammonium (TBA þ ) fluoride rather than inorganic fluoride salts which limited the application of sensing fluoride in the biological sample [26,27]. Furthermore, the LOD of this probe was in the upper mg/L range and transferring the sensing results electronically required additional optical sensing equipment. Nevertheless, the detection of low concentrations of fluoride in polar and aqueous solutions remains challenging without expensive analytical equipment. The electrochemical fluoride recognition strategies based on the Lewis acid coordination with arylboronic acids/esters incorporating
A facile electrochemical “latent ratiometric redox probe” approach has been developed for the detection of fluoride employing H2Q as the electrochemical reporter. The selectivity studies revealed that the deprotection is highly specific for fluoride. Both H2Q0 and MH2Q0 are highly stable and exhibited rapid response time. Real sample analysis revealed the practical applicability of the sensor. LOD of both the probes (23.8 mM for H2Q0 and 2.38 mM for MH2Q0 ) are within the upper limit prescribed by US EPA. Undoubtedly, the developed H2Q based latent ratiometric approach holds great promise for the commencement of potential electrochemical probes for the detection of fluoride.
Acknowledgments This work was supported by the National Science Council of Taiwan (NSC 103-2811-M-027-002 and 102-2113-M-027-002MY3). References [1] I.C. Guimaraes, C.C. Rezende, J.A.F. da Silva, D.P. de Jesus, Talanta 78 (2009) 1436–1439. [2] F.L. Cerklewski, Nutrition, Burbank, Los Angeles County, California (1998) 475. [3] M. Kaykhaii, M.H. Ghalehno, Anal. Methods 5 (2013) 5622–5626. [4] U. EPA, Total Coliforms (Including Fecal Coliforms and E. Coli), 54, 2009. [5] J. Du, M. Hu, J. Fan, X. Peng, Chem. Soc. Rev. 41 (2012) 4511–4535. [6] R. Perry-Feigenbaum, E. Sella, D. Shabat, Chem. – Eur. J. 17 (2011) 12123–12128. [7] M.S. Baker, S.T. Phillips, Org. Biomol. Chem. 10 (2012) 3595–3599. [8] S.Y. Kim, J.I. Hong, Org. Lett. 9 (2007) 3109–3112.
126
V. Mani et al. / Talanta 131 (2015) 121–126
[9] Y. Michigami, Y. Kuroda, K. Ueda, Y. Yamamoto, Anal. Chim. Acta 274 (1993) 299–302. [10] J.J. Potter, A.E. Hilliker, G.J. Breen, J. Chromatogr. A 367 (1986) 423–427. [11] R.A. Durst, Ion-selective electrodes: proceedings, US National Bureau of Standards; for sale by the Supt. of Docs, US Govt. Print. Off, Washington, 1969. [12] E. Kissa, Clin. Chem. 33 (1987) 253–255. [13] D.A. Skoog, Fundamentals of Analytical Chemistry, Grupo Editorial Norma, USA, 2004. [14] M. Cametti, K. Rissanen, Chem. Commun. (2009) 2809–2829. [15] H.C. Huang, K.L. Wang, S.T. Huang, H.Y. Lin, C.M. Lin, Biosens. Bioelectron. 26 (2011) 3511–3516. [16] T.L. Marques, N.M. Coelho, Talanta 105 (2013) 69–74. [17] J.A. Gu, Y.J. Lin, Y.M. Chia, H.Y. Lin, S.T. Huang, Microchim. Acta 180 (2013) 801–806. [18] K. Ghosh, D. Kar, R. Fröhlich, A.P. Chattopadhyay, A. Samadder, A.R. KhudaBukhsh, Analyst 138 (2013) 3038–3045. [19] L. Fu, F.L. Jiang, D. Fortin, P.D. Harvey, Y. Liu, Chem. Commun. 47 (2011) 5503–5505. [20] H. Lenormand, J.P. Goddard, L. Fensterbank, Org. Lett. 15 (2013) 748–751. [21] J. Wang, L. Yang, C. Hou, H. Cao, Org. Biomol. Chem. 10 (2012) 6271–6274. [22] L. Hu, S. Han, Z. Liu, S. Parveen, Y. Yuan, G. Xu, Electrochem. Commun. 13 (2011) 1536–1538.
[23] T. Greene, P. Wuts, Protecting groups in organic synthesis, Wiley Interscience, New York, 1999. [24] J.P. Willis, K.A. Gogins, L.L. Miller, J. Org. Chem. 46 (1981) 3215–3218. [25] B. Eggins, J. Chambers, J. Electrochem. Soc. 117 (1970) 186–192. [26] S. Rivadehi, E.F. Reid, C.F. Hogan, S.V. Bhosale, S.J. Langford, Org. Biomol. Chem. 10 (2012) 705–709. [27] Y. Qu, J. Hua, H. Tian, Org. Lett. 12 (2010) 3320–3323. [28] C.R. Wade, A.E. Broomsgrove, S. Aldridge, F.P. Gabbai, Chem. Rev. 110 (2010) 3958–3984. [29] Q. Tan, L. Wang, H. Yu, L. Deng, J. Phys. Chem. B. 111 (2007) 3904–3909. [30] C. Dusemund, K.S. Sandanayake, S. Shinkai, J. Chem. Soc. Chem. Commun. (1995) 333–334. [31] H. Yamamoto, A. Ori, K. Ueda, C. Dusemund, S. Shinkai, Chem. Commun. (1996) 407–408. [32] F. Otón, A. Tárraga, M.D. Velasco, A. Espinosa, P. Molina, Chem. Commun. (2004) 1658–1659. [33] J. Xiong, L. Cui, W. Liu, J.E. Beves, Y.Y. Li, J.L. Zuo, Tetrahedron Lett. 54 (2013) 1998–2000. [34] H. Çiftçi, Y. Oztekin, U. Tamer, A. Ramanavicine, A. Ramanavicius, Talanta 126 (2014) 202–207. [35] M. Rafiee, D. Nematollahi, Electroanalysis 19 (2007) 1382–1386.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Electrochemical OFF–ON ratiometric chemodosimeters for the selective and rapid detection of fluoride Veerappan Mani a, Wen-Yung Li a, Jiun-An Gu a, Chun-Mao Lin b, Sheng-Tung Huang a,n a b
Department of Chemical Enginnering and Biotechnology, National Taipei University of Technology, Taipei 10608, Taiwan (ROC) Department of Biochemistry, College of Medicine, Taipei Medical University, No. 250, Wu-xing Street, Taipei 110, Taiwan (ROC)
art ic l e i nf o
a b s t r a c t
Article history: Received 27 May 2014 Received in revised form 23 July 2014 Accepted 24 July 2014 Available online 4 August 2014
We have described two “OFF–ON electrochemical latent ratiometric redox chemodosimeters”, 1,4-Bis (tert-butyldimethylsiloxy)benzene (H2Q0 ) and 1,4-Bis (tert-butyldimet hylsiloxy)-2-methoxybenzene (MH2Q0 ) for the selective detection of inorganic fluoride. The electrochemical signals of hydroquinone (H2Q) and o-methoxy hydroquinone (MH2Q) within this latent redox probes (H2Q0 and MH2Q0 ) were completely masked by protecting their hydroxyl group as silylether (OFF state). The externally added fluoride ions triggered the deprotection of H2Q0 and MH2Q0 and unmasked the electrochemical properties of H2Q and MH2Q respectively. The electrochemical reporters (H2Q and MH2Q) presented a pair of redox peaks at the electrode surface (ON state) and the peak currents are linearly dependent with the concentration of fluoride which leading to the ratiometric detection of fluoride. The limit of detection (signal-to-noise ratio¼ 3) observed for the probes are 23.8 mM and 2.38 mM for H2Q0 and MH2Q0 respectively. The deprotection is highly selective for fluoride over other anions investigated. The probes are highly stable and the proposed approach offers rapid response time and promising practical applicability. The proposed strategy holds great promise for the commencement of new H2Q based electrochemical probes by tuning the electrochemical behavior of H2Q. & 2014 Elsevier B.V. All rights reserved.
Keywords: Hydroquinone Latent redox ratiometric probe Chemodosimeters Sensor Fluoride Selectivity
1. Introduction Fluoride is widely used as the active ingredient in toothpaste formulations for the prevention of dental caries and other health benefits. In addition, fluorides are one of the essential trace elements required to form bones and teeth in human [1]. However, fluorides are potentially toxic, if their concentrations exceed levels that cause severe health risks such as dental fluorosis and bone diseases [2,3]. Therefore, the permissible overall quantity of fluoride present in toothpaste is regulated within the concentration range of 0.50–1.50 mg g 1, while US Environmental Protection Agency's (US EPA) has recommended the upper allowed limit of 2 ppm (105 mM) fluorides in drinking water to prevent dental fluorosis [4]. Therefore, determination of fluorides is of great significance [5–9]. In the past years, several analytical methods have been developed for the determination of fluoride such as chromatography [10], ion selective electrodes [11–13], anion recognition [14], fluorimetry [15], spectrophotometry [16], and colorimetry methods [17]. But, most of these traditional ana-
n
Corresponding author. Tel.: +886 2271 2171 2525; fax: +886 02 2731 7117. E-mail address: [email protected] (S.-T. Huang).
http://dx.doi.org/10.1016/j.talanta.2014.07.074 0039-9140/& 2014 Elsevier B.V. All rights reserved.
lytical methods involve time-consuming and tedious procedures. Recently, several research groups have published optical chemodosimeters for the determination of fluoride; in this case, the detection approach involves typical fluoride triggered cleavage of Si–O bond in the designed probes with concomitant release of the optical reporters [8,17–21]. These optical chemodosimeters are able to provide excellent selectivity for the determination of fluoride attributed to the strong chemical affinity between fluoride and silicon. Compared with optical techniques, electrochemical techniques have relatively better performances in terms of short reaction time, high sensitivity, simplicity of the voltammetric instrumentation, simple operating protocols, direct use in pointof-care assays, potential for miniaturization and portability [22]. In the present work, we incorporated the similar concept of optical chemodosimeters into electrochemical redox probes for the simple and selective detection of fluoride at electrode surface. The schematic representation for the mechanistic pathway of the designed “OFF-ON electrochemical latent ratiometric redox chemodosimeters” is given as Scheme 1: (1) Masking the electrochemical signal of hydroquinone (H2Q) through silylation to obtain the probe, 1,4-Bis(tert-butyldimethylsiloxy)benzene (H2Q0 , OFF state), (2) fluoride triggered cleavage of Si–O bond, and subsequent removal of silyl protecting group through quinone type
122
V. Mani et al. / Talanta 131 (2015) 121–126
Scheme 1. Schematic representation for the electrochemical determination of fluoride via fluoride induced electrochemical “ON–OFF” ratiometric latent redox switch.
rearrangement with concomitant release of electrochemical reporter (H2Q), and (3) H2Q readily undergoes electrochemical redox reaction and presented its redox peaks at the electrode surface. The deprotection is highly specific for sodium fluoride and the amount of released H2Q is quantitative to the concentration of fluoride. The same “latent redox probe approach” of H2Q0 was extented to 1,4-Bis (tert-butyldimet hylsiloxy)-2-methoxybenzene (MH2Q0 ), which is a methoxy derivative of H2Q0 . Preparation of the probes (H2Q0 and MH2Q0 ) involves very simple silylation reaction and the assay pathway involves rapid and quantitative determination of fluoride. The aim of the present work is to develop a new electrochemical platform for the selective determination of fluoride. To the best of our knowledge, there is no report available in the literature employing H2Q as the “Electrochemical OFF–ON Ratiometric Chemodosimeters” for the determination of fluoride. Considering the advantages of electrochemical methods and the need for the development of novel electrochemical approaches to selectively detect fluoride, herein we have designed two “electrochemical ratiometric redox probes”, by incorporating HQ and o-methoxy hydroquinone (MH2Q) as the electrochemical reporters. The limit of detection (LOD) observed for these probes (23.8 mM for H2Q0 and 2.38 mM for MH2Q0 ) are within the upper allowed limit mandated by the US EPA.
2. Experimental 2.1. Materials and methods H2Q, sodium fluoride, and all the other chemicals were purchased from Sigma-Aldrich and used without further purification. Electrochemical measurements were recorded in a conventional three electrode cell employing BAS glassy carbon electrode (GCE) as the working electrode (area 0.071 cm2), saturated Ag/Ag þ (AgNO3) filled with acetonitrile containing tetra-n-butylammonium bromide (TBAP) as the reference electrode and Pt wire as the counter electrode. 1 mM of TBAP dissolved in DMF was used as the supporting electrolyte. The determination of fluoride was carried out in DMF containing TBAP (1 mM), HEPES buffer (0.05 M, pH 7.8) (DMF–TBAP–HEPES solution), and probe (1 mM). The solutions were deoxygenated with pre-purified nitrogen for 5 min before performing each electrochemical experiment. 2.2. Preparation of H2Q0 and MH2Q0 H2Q0 and MH2Q0 was prepared in accordance to the procedure reported in the literature [23,24]. H2Q (1.67 g, 15.2 mmol) was treated with TBDMS chloride (11.4 g, 75.3 mmol) and imidazole (10.2 g, 150 mmol) in DMF (5 ml) for overnight to yield H2Q0 as a colorless solid (4.64 g, 90%). 2-methoxybenzene-1,4-diol (2.13 g,
15.2 mmol) was treated with TBDMS chloride (11.4 g, 75.3 mmol) and imidazole (10.2 g, 150 mmol) in DMF (7 ml) for overnight to yield the MH2Q0 as a colorless solid (4.784 g, 85%). 1H NMR data of the H2Q0 and MH2Q0 were in good agreement with those reported in the literature.
3. Results and discussion 3.1. Determination of fluoride using H2Q0 The cyclic voltammogram (CV) acquired at GCE in DMF–TBAP– HEPES solution containing H2Q0 (1 mM) (curve b, Fig. 1A) did not exhibit any characteristic redox peaks in the absence of fluoride, indicating the complete masking of H2Q redox sites. However, the CV obtained in the presence of 0.24 mM NaF (curve c, Fig. 1A) exhibited well defined quasi reversible redox peaks related to the characteristic redox reaction of H2Q. The redox reaction involves two electrons and two protons and occurred via semiquinone formation which was established fabulously in the literature [25]. The oxidation peak corresponding to the oxidation of H2Q to pbenzoquinone (Q) was observed at the potential of þ0.10 V, whereas the reduction peak corresponding to the reduction of Q to H2Q was observed at the potential of 0.71 V. Thus, the CV results clearly indicating that the addition of fluoride triggered the silyl deprotection and uncloaks the redox active center of H2Q0 . Moreover, we observed that short time (5 min) is sufficient to complete the deprotection; thus our assay platform involves short reaction time allowing rapid detection of fluoride. CV obtained at GCE in DMF–TBAP–HEPES solution containing 1 mM of pristine H2Q has been carried out to correlate the electrochemical behavior of pristine H2Q (inset to Fig. 1A) and regenerated H2Q from H2Q0 (curve c, Fig. 1A). CV obtained for the pristine H2Q is quite consistent with that of released H2Q from H2Q0 in terms of peak potentials and peak shapes supports the proposed protection/ deprotection mechanistic pathway. The effect of scan rate (ν) on the redox behavior of H2Q was examined in DMF–TBAP–HEPES solution containing 1 mM H2Q0 and 0.24 mM of NaF at the scan rates from 0.1 to 1.2 V s 1 (Fig. 1B). Both the anodic (Ipa) and cathodic peak current (Ipa) increased linearly with increase in scan rates from 0.1 to 1.2 V s 1. A plot of square root of scan rates (ν1/2) and anodic peak current (Ip) exhibited linear relationship indicating that the redox reaction of released H2Q occured at the GCE is a diffusion controlled electron transfer process (inset to Fig. 1B). Fig. 2A shows the CVs obtained at GCE in DMF–TBAP–HEPES solution containing H2Q0 (1 mM) and different concentrations of NaF (a ¼0.024, b ¼0.12, c¼0.24, d ¼0.48, e¼ 0.72, f ¼0.96, and g¼ 1.19 mM). As evident from the figure, a pair of well-defined redox peaks of H2Q was present for the 0.024 mM addition of fluoride. Moreover, both Ipa and Ipc were increased linearly
V. Mani et al. / Talanta 131 (2015) 121–126
123
Fig. 2. (A) CVs obtained at GCE in DMF–TBAP–HEPES solution containing H2Q0 (1 mM) with different concentrations of fluoride; 0.024 (a), 0.12 (b), 0.24 (c), 0.48 (d), 0.72 (e), 0.96 (f), and 1.19 mM (g). (B) Plot of [fluoride] vs. Ipa and Ipc.
Fig. 1. (A) CVs obtained at GCE in DMF–TBAP–HEPES solution (a) containing H2Q0 (1 mM) in the absence (b) and presence of 0.24 mM NaF (c) and 1 mM pristine H2Q (inset). (B) CVs obtained at GCE in DMF–TBAP–HEPES solution containing H2Q0 (1 mM) in the presence of 0.24 mM NaF at different scan rates from 0.1 to 1.2 V s 1 (curves a–i). Inset: ν1/2 vs. Ipa and Ipc.
upon increase in the concentration of fluoride. A plot of Ipa and Ipc vs. concentration of fluoride exhibited a linear relationship (Fig. 2B) with the slope of 7.153 mA mM 1 and 8.586 mA mM 1 respectively. The linear range was observed beween 23.8 mM and 1.19 mM, while limit of detection (LOD) was found to be 23.8 mM (S/N ¼ 3) which is sufficiently enough to detect the US EPA mandated upper limit of fluoride concentration in drinking water [4]. Monitoring Ipc of the H2Q redox peaks offers more sensitivity than Ipa and therefore we choose Ipc for the determination of fluoride. 3.2. Determination of fluoride using MH2Q0 MH2Q0 also follows similar electrochemical protection/deprotection pathway as H2Q0 (Scheme 1). No redox peaks were observed in the
absence of fluoride (protected, OFF) (curve a, Fig. 3A), whereas well defined redox peaks were observed in the presence of 0.24 mM NaF (deprotected, ON) (curve b, Fig. 3A). Here, the redox peaks are ascribed to the redox conversion between MH2Q and MQ (o-methoxy benzoquinone). The CV obtained at the pristine MH2Q (inset to Fig. 3A) is quite consistent with that obtained at the regenerated MH2Q from MH2Q0 . MH2Q0 follows concentration-dependent increase in Ipc with different concentrations of fluoride (Fig. 3B) and exhibited linear concentration range from 2.38 mM to 0.96 mM (inset to Fig. 3B). LOD (S/N ¼3) of MH2Q0 was calculated to be 2.38 mM which is 10 fold lower than that observed at H2Q0 . The probable reason for the lower LOD of MH2Q0 might be the chelating effect caused by the formation of stable five membered chelating ring when sodium fluoride approaches MH2Q0 which brings the fluoride anion more closer to silicon (Scheme 2). As a result, the removal of silyl ether become more easy even at low concentration of fluoride, whereas formation of this kind of chelating effect is not possible in H2Q0 . Perhaps, this result indicating that any substituent on the ring of H2Q0 may influence the analytical performance of the fluoride detection. Therefore the OFF/ON switching ability of the H2Q0 can be tuned by different substituents to boost the performance of the resulting sensor. The sensor performance of the probes were listed in Table 1. It is evident from the table that, LOD and linear range of these probes are within the US EPA mandated upper limit of fluoride concentration in drinking water.
124
V. Mani et al. / Talanta 131 (2015) 121–126
Scheme 2. Schematic representation for the chelating effect observed in MH2Q0 .
Table 1 Electroanalytical parameters and sensor performance of the probes H2Q0 and MH2Q0 .
Fig. 3. (A) CVs obtained at GCE in DMF–TBAP–HEPES solution containing MH2Q0 (1 mM) in the absence (a) and presence of 0.24 mM NaF (b) and 1 mM pristine MH2Q (inset). (B) CVs obtained at GCE in DMF–TBAP–HEPES solution containing MH2Q0 with different concentrations of fluoride; 0 (a), 2.38 10 3 (b), 11.9 10 3 (c), 23.8 10 3 (d), 0.12 (e), 0.24 (f), 0.48 (g), 0.72 (h), and 0.96 mM (i). Inset: [fluoride] vs. Ipc.
3.3. Selectivity In order to study the selectivity of H2Q0 towards fluoride detection, CV studies were carried out in the presence of variety of likely anionic interferences such as NaI, NaCl, NaBr, KH2PO4, Na2HPO4, LiClO4, NaOAc, NaNO3, KCN and NaN3. The CV obtained at GCE in DMF–TBAP–HEPES solution containing H2Q0 (1 mM) exhibited well defined Ipc response towards addition of 2 mM NaF (a), whereas no noteworthy responses were observed for the addition of 2 mM of NaI (b), NaCl (c), NaBr (d), KH2PO4 (e), Na2HPO4 (f), LiClO4 (g), NaOAc (h), NaNO3 (i), KCN (j) and NaN3 (k) (Fig. 4A). The selectivity studies clearly revealed that only fluoride was able to unmask the redox active sites of H2Q from H2Q0 among all the anions tested (Fig. 4B). This must be ascribed to the special chemical affinity between fluoride and silicon [5]. Thus, H2Q0 is a highly selective redox probe for the determination of fluoride even in the presence of other common interferrants. Similar to H2Q0 , MH2Q0 has also found to be insensitive to other
Probes
H2Q0
MH2Q0
Limit of detection Linear range Repeatability Reproducibility
23.8 mM 23.8 mM–1.19 mM 2.53% 3.22%
2.38 mM 2.38 mM–0.96 mM 2.72% 3.44%
anions tested and selective for the determination of fluoride (Figure not shown). 3.4. Stability, repeatability and reproducibility To investigate the stability of the probes, DMF–TBAP–HEPES solutions containing 1 mM of each probe (H2Q0 and MH2Q0 ) were treated under thermal (60 1C) and ultrasonic conditions for 1 h and their respective CVs were recorded. Under these conditions, no characteristic redox peaks of their corresponding redox reporters (H2Q and MH2Q) were observed in the absence of fluoride. On the other hand, the redox reporters were liberated from the probes within few seconds in the presence of fluoride (Figs. 1B and 3B). Therefore, we concluded that both the probes are highly stable under thermal and ultrasonication conditions, while fluoride ions only able to trigger the deprotection. Both the two probes exhibited appreciable repeatability and reproducibility with acceptable recoveries for five experiments. Repeatability experiments were carried out using single electrode, while reproducibility experiments were carried out with five different electrodes (Table 1). 3.5. Real sample analysis We demonstrated the practical feasibility of the proposed approach towards determination of fluoride present in water
V. Mani et al. / Talanta 131 (2015) 121–126
125
Table 2 Real sample analysis in water samples. Probes Tap water
River water
Added (mM)
Found (mM)
Recovery (%)
Added (mM)
Found (mM)
Recovery (%)
H2Q0
25 250
24.2 256
96.8 102.4
25 250
26 258
104 103.2
MH2Q0
25 250
25.83 245.1
103.3 98
25 250
24.13 244
96.52 97.6
ferrocenyl compounds as receptors are facing difficulties in the detection of aqueous fluoride ions [28–33]. On the other hand, electrochemical techniques based on the modified electrodes often lack selectivity. Most recently, a potentiometric sensor based on poly(3-aminophenylboronic acid) modified graphite rod electrode was reported for the determination of fluoride [34], however its detection range was beyond the upper limit imposed by US EPA. In the present work, we established a highly selective electrochemical inorganic fluoride sensing platform which allowed the direct transferring of sensing signal electronically and applicable for sensing fluoride in the biological sample. The LOD observed for the probes (23.8 mM for H2Q0 and 2.38 mM for MH2Q0 ) are within the relevant window of US EPA's mandated upper limit [4]. H2Q is a potential electrochemical reporter attributed to its stable, predictable, and tailorable electrochemical properties, and its electrochemical signal could be easily measured by simple electrochemical methods such as cyclic voltammetry [25,35]. To the best of our knowledge, this is the first report demonstrating “latent redox ratiometric probe” approach through simple selective protection/deprotection strategy for the detection of fluoride.
4. Conclusions
Fig. 4. (A) CVs obtained at DMF–TBAP–HEPES solution containing H2Q0 (1 mM) with the presence of 2 mM of NaF (a) and NaI (b), NaCl (c), NaBr (d), KH2PO4 (e), Na2HPO4 (f), LiClO4 (g), NaOAc (h), NaNO3 (i), KCN (j), and NaN3 (k). (B) Plot between various anions and their response towards NaF.
samples. The spiked fluoride concentrations are 25 mM and 250 mM. The found and recovery values are given in Table 2. As evident from the table, both the probes presented acceptable recoveries in tap and river water samples revealed the practical feasiblity of the proposed approach. 3.6. Comparison of the proposed “latent ratiometric redox probe approach” with other methods In the past decades, several optical probes have been reported which are relying on hydrogen bonds or Lewis acid coordination of fluoride with the probes, and most of these probes could only be operated in organic solvents to detect tetrabutylammonium (TBA þ ) fluoride rather than inorganic fluoride salts which limited the application of sensing fluoride in the biological sample [26,27]. Furthermore, the LOD of this probe was in the upper mg/L range and transferring the sensing results electronically required additional optical sensing equipment. Nevertheless, the detection of low concentrations of fluoride in polar and aqueous solutions remains challenging without expensive analytical equipment. The electrochemical fluoride recognition strategies based on the Lewis acid coordination with arylboronic acids/esters incorporating
A facile electrochemical “latent ratiometric redox probe” approach has been developed for the detection of fluoride employing H2Q as the electrochemical reporter. The selectivity studies revealed that the deprotection is highly specific for fluoride. Both H2Q0 and MH2Q0 are highly stable and exhibited rapid response time. Real sample analysis revealed the practical applicability of the sensor. LOD of both the probes (23.8 mM for H2Q0 and 2.38 mM for MH2Q0 ) are within the upper limit prescribed by US EPA. Undoubtedly, the developed H2Q based latent ratiometric approach holds great promise for the commencement of potential electrochemical probes for the detection of fluoride.
Acknowledgments This work was supported by the National Science Council of Taiwan (NSC 103-2811-M-027-002 and 102-2113-M-027-002MY3). References [1] I.C. Guimaraes, C.C. Rezende, J.A.F. da Silva, D.P. de Jesus, Talanta 78 (2009) 1436–1439. [2] F.L. Cerklewski, Nutrition, Burbank, Los Angeles County, California (1998) 475. [3] M. Kaykhaii, M.H. Ghalehno, Anal. Methods 5 (2013) 5622–5626. [4] U. EPA, Total Coliforms (Including Fecal Coliforms and E. Coli), 54, 2009. [5] J. Du, M. Hu, J. Fan, X. Peng, Chem. Soc. Rev. 41 (2012) 4511–4535. [6] R. Perry-Feigenbaum, E. Sella, D. Shabat, Chem. – Eur. J. 17 (2011) 12123–12128. [7] M.S. Baker, S.T. Phillips, Org. Biomol. Chem. 10 (2012) 3595–3599. [8] S.Y. Kim, J.I. Hong, Org. Lett. 9 (2007) 3109–3112.
126
V. Mani et al. / Talanta 131 (2015) 121–126
[9] Y. Michigami, Y. Kuroda, K. Ueda, Y. Yamamoto, Anal. Chim. Acta 274 (1993) 299–302. [10] J.J. Potter, A.E. Hilliker, G.J. Breen, J. Chromatogr. A 367 (1986) 423–427. [11] R.A. Durst, Ion-selective electrodes: proceedings, US National Bureau of Standards; for sale by the Supt. of Docs, US Govt. Print. Off, Washington, 1969. [12] E. Kissa, Clin. Chem. 33 (1987) 253–255. [13] D.A. Skoog, Fundamentals of Analytical Chemistry, Grupo Editorial Norma, USA, 2004. [14] M. Cametti, K. Rissanen, Chem. Commun. (2009) 2809–2829. [15] H.C. Huang, K.L. Wang, S.T. Huang, H.Y. Lin, C.M. Lin, Biosens. Bioelectron. 26 (2011) 3511–3516. [16] T.L. Marques, N.M. Coelho, Talanta 105 (2013) 69–74. [17] J.A. Gu, Y.J. Lin, Y.M. Chia, H.Y. Lin, S.T. Huang, Microchim. Acta 180 (2013) 801–806. [18] K. Ghosh, D. Kar, R. Fröhlich, A.P. Chattopadhyay, A. Samadder, A.R. KhudaBukhsh, Analyst 138 (2013) 3038–3045. [19] L. Fu, F.L. Jiang, D. Fortin, P.D. Harvey, Y. Liu, Chem. Commun. 47 (2011) 5503–5505. [20] H. Lenormand, J.P. Goddard, L. Fensterbank, Org. Lett. 15 (2013) 748–751. [21] J. Wang, L. Yang, C. Hou, H. Cao, Org. Biomol. Chem. 10 (2012) 6271–6274. [22] L. Hu, S. Han, Z. Liu, S. Parveen, Y. Yuan, G. Xu, Electrochem. Commun. 13 (2011) 1536–1538.
[23] T. Greene, P. Wuts, Protecting groups in organic synthesis, Wiley Interscience, New York, 1999. [24] J.P. Willis, K.A. Gogins, L.L. Miller, J. Org. Chem. 46 (1981) 3215–3218. [25] B. Eggins, J. Chambers, J. Electrochem. Soc. 117 (1970) 186–192. [26] S. Rivadehi, E.F. Reid, C.F. Hogan, S.V. Bhosale, S.J. Langford, Org. Biomol. Chem. 10 (2012) 705–709. [27] Y. Qu, J. Hua, H. Tian, Org. Lett. 12 (2010) 3320–3323. [28] C.R. Wade, A.E. Broomsgrove, S. Aldridge, F.P. Gabbai, Chem. Rev. 110 (2010) 3958–3984. [29] Q. Tan, L. Wang, H. Yu, L. Deng, J. Phys. Chem. B. 111 (2007) 3904–3909. [30] C. Dusemund, K.S. Sandanayake, S. Shinkai, J. Chem. Soc. Chem. Commun. (1995) 333–334. [31] H. Yamamoto, A. Ori, K. Ueda, C. Dusemund, S. Shinkai, Chem. Commun. (1996) 407–408. [32] F. Otón, A. Tárraga, M.D. Velasco, A. Espinosa, P. Molina, Chem. Commun. (2004) 1658–1659. [33] J. Xiong, L. Cui, W. Liu, J.E. Beves, Y.Y. Li, J.L. Zuo, Tetrahedron Lett. 54 (2013) 1998–2000. [34] H. Çiftçi, Y. Oztekin, U. Tamer, A. Ramanavicine, A. Ramanavicius, Talanta 126 (2014) 202–207. [35] M. Rafiee, D. Nematollahi, Electroanalysis 19 (2007) 1382–1386.
