- Email: [email protected]
Electroanalysis of selenium in water on an electrodeposited gold-nanoparticle modified glassy carbon electrode
Journal of Electroanalytical Chemistry 758 (2015) 7–11
Contents lists available at ScienceDirect
Journal of Electroanalytical Chemistry journal homepage: www.elsevier.com/locate/jeac
Electroanalysis of selenium in water on an electrodeposited gold-nanoparticle modified glassy carbon electrode A.O. Idris a, N. Mabuba a, O.A. Arotiba a,b,⁎ a b
Department of Applied Chemistry, University of Johannesburg, Johannesburg, South Africa Centre for Nanomaterials Science Research, University of Johannesburg, South Africa
a r t i c l e
i n f o
Article history: Received 9 July 2015 Received in revised form 20 August 2015 Accepted 10 October 2015 Available online xxxx Keywords: Selenium Square wave anodic stripping voltammetry Electrodeposition Gold nanoparticles Water
a b s t r a c t This work presents a simple and cheaper method of detecting selenium in water using an electrochemical sensor based on gold nanoparticle (AuNP) modified glassy carbon electrode. AuNPs were electrochemically deposited on a glassy carbon electrode (GCE) using cyclic voltammetry within the potential range of −0.4 mV to 1.1 mV for 10 cycles. The modification of GCE with AuNPs resulted in an increase in the electroactive surface area of the electrode which led to the enhancement of the redox current peaks of [Fe (CN)6]3−/4− and [Ru (NH3)6]2+/3+ in comparison to the bare GCE. Square wave anodic stripping voltammetry was used to detect Se (IV) in water (in 0.1 M H2SO4 as supporting electrolyte) at the following optimum conditions: pH 1, deposition potential of − 100 mV, pre-concentration time of 60 s. The GCE-AuNP sensor was able to detect Se (IV) to the limit of 0.64 μg L−1 and was not susceptible to many interfering cations except Cu (II) and Cd (II). This method involves a simple one step electrode modification. The sensor was used to detect Se (IV) in a real sample. This method was validated by its good correlation with the result obtained from inductively coupled plasma method. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Selenium is a trace element that is important in our diet and in the health of plants and animals when taken in the correct amount [1]. Exposure to an excess amount of selenium is toxic and can cause health problems such as gastrointestinal upsets, hair loss, white blotchy nails, garlic breathe odour and nerve damage. Selenium deficiency on the other hand, has been linked to cancer, heart diseases, muscular dystrophy, multiple sclerosis, immune system, osteoarthopathy, reproductive disorder in humans and white muscle disease in animals [2]. The World Health Organization has set a limit of 10 μg L−1 selenium concentration in drinking water [3]. Human activities that have facilitated the increase in the concentration of selenium in the environment include the following: the mining and processing of base metal, gold and coal; phosphate deposits; the use of rock phosphate as fertiliser; the application of sewage sludge to land; the manufacture of detergents and shampoo; and so on. The increased use of selenium in the pharmaceutical, glazing, photocopying, ceramics, paint and electronics industries has also increased the amount of selenium entering the environment [4]. Owing to the toxicity and the environmental importance of selenium, its detection and quantification have been studied using various methodologies. The three main methods of selenium detection are ⁎ Corresponding author at: Department of Applied Chemistry, University of Johannesburg, Johannesburg, South Africa. E-mail address: [email protected] (O.A. Arotiba).
http://dx.doi.org/10.1016/j.jelechem.2015.10.009 1572-6657/© 2015 Elsevier B.V. All rights reserved.
inductively coupled plasma (ICP), spectroscopy and electrochemistry. Inductively coupled plasma has been used to detect selenium at low concentrations but the pitfall of this method is that it is expensive, it requires extensive sample preparation, it has a long analysis time and it requires sophisticated operator personnel [5]. The spectroscopic methods predominantly used are graphite furnace atomic absorption spectroscopy [6], hydride generation atomic absorption spectroscopy [7] and atomic fluorescence spectrometry [8]. Electrochemical techniques for selenium detection have an edge over the other techniques based on the grounds such as low cost of instrument, minimum sample preparation for analysis, faster analysis time and amenability to miniaturisation [9]. Bertolino et al. reported the speciation analysis of selenium in natural water using square wave voltammetry after pre-concentration on activated carbon and selenium was detected to a limit of 0.004 μg L−1 within 0.01 μg L−1–20 μg L−1 concentration range [10]. In another report, Se (VI) was reduced to Se (IV) before analysis using flow injection anodic stripping voltammetry at a gold electrode for Se (IV) determination [11]. The Se (IV) concentration was calculated as the difference between the result for total inorganic selenium and Se (IV). The interfering divalent cations were removed by using cation exchange column prior to the injection value. Differential pulse adsorptive stripping voltammetry at a bismuth film electrode with cetyl trimethyl ammonium bromide and p-amino benzene sulphonic acid as a complexing agent, was used to determine trace selenium by Zhang et al. [12]. A detection limit of 0.1 μg L− 1 and accumulation time of 300 s were used. Britta Lange et al. were able to determine Se (IV) by catalytic cathodic
8
A.O. Idris et al. / Journal of Electroanalytical Chemistry 758 (2015) 7–11
stripping voltammetry in the presence of Rhodium. Rhodium was used for this work because of its catalytic ability. U.V. digestion was used to eliminate the interference caused by inorganic material and to convert Se (VI) to Se (IV) and a detection limit of 2.4 pM was obtained [13]. Attempts to use microelectrode in the determination of Se (IV) have also been reported [14]. Tan and Kounaves were able to determine Se (IV) at a micro fabricated Au ultra-micro electrode array using square wave anodic stripping voltammetry; they discovered that Se (IV) redox reaction appears to be kinetically faster and more reversible at the Au Ultra microelectrode (UME) array than at macro electrode or single Au UME. The limit of detection was 0.42 ppb [15]. In this work, gold nanoparticles were electrodeposited on glassy carbon electrode in a one step approach and used to detect Se (IV) by square wave anodic stripping voltammetry. The choice of supporting electrolyte, pH and deposition potential was optimised. The sensor was used to detect Se (IV) in both analytical and real water samples. The concentration of Se (IV) detected in real water sample was correlated with the result obtained from inductively coupled plasma optical emission spectrometry technique. 2. Experimental
emission spectrometry technique was also used to detect the amount of Se (IV) in the tap water sample. 3. Results and discussion 3.1. Electrode characterisation Bare glassy carbon and GCE-AuNP electrodes were electrochemically characterised by using CV in [Fe (CN) 6]3−/4− and [Ru (NH3)6]2+/3+ solutions; this was done by cycling a potential from −400 mV to 600 mV at a scan rate of 50 mV s−1, to compare the peak current signal of the bare GCE and GCE-AuNPs. Fig. 1 shows the electrochemical responses of the bare and modified GCE in two common redox probes. As depicted in Fig. 1, the presence of AuNPs enhanced the interfacial electron transfer of Ferrocyannide and Ruthenium redox probes. This electrochemical enhancement is an indication of the potential application of GCE-AuNP electrodes in electroanalysis. The increase in current can be attributed to an increase in electroactive surface area of the electrode which can enhance the sensitivity of the electrode when used for electroanalysis [17]. 3.2. Optimisation: the effect of supporting electrolyte, pH, deposition potential and deposition time
2.1. Material and methods HAuCl4, KCl, Na2SeO3, HNO3, KNO3, K2HPO4, KH2PO4, NaOH, K3Fe(CN)6, K4Fe(CN)6.3H2O, Ru(NH3)6Cl3, Ru(NH3)6Cl2, H2SO4 and KOH were purchased from sigma Aldrich. All chemicals were of highest analytical grade and deionised water was used for the preparation of solutions. All electrochemical measurements were done on Compactstat electrochemical workstation (Ivium Technologies, Netherlands), using a three-electrode configuration. The working electrode, counter electrode and reference electrode were glassy carbon electrode, platinum wire and Ag/AgCl (3 M KCl) respectively. The electrochemical cells were purged with ultra-pure argon gas for at least 10 min prior to all electrochemical measurements. 2.2. Electrode modification Glassy carbon electrode (GCE) was modified with gold nanoparticles (AuNPs) according to the method reported by Arotiba et al. [16]. Briefly, GCE was modified with 5 mM of HAuCl4 solutions by cycling the potential from −400 mV to 1100 mV for 10 cycles at a scan rate 50 mV s−1, the reference electrode used is Ep vs Ag/AgCl (3 M KCl). The modified electrode was referred to as GCE-AuNPs. The modified electrodes were electrochemically characterised using square wave voltammetry and cyclic voltammetry (CV) in 5 mM mixture of K3Fe(CN)6 and K4Fe(CN)6.3H2O (referred to as [Fe (CN) 6]3−/4−) and 1 mM mixture of Ru(NH3)6Cl3 and Ru(NH3)6Cl2 (referred to as [Ru (NH3)6]2 +/3 +) redox probes.
The effect of supporting electrolyte is presented in Fig. 2a. In this work, 0.1 M H2SO4 was chosen as the supporting electrolyte because it is more suitable for the stripping of Se (IV) due to the highest current signal obtained in comparison to other supporting electrolytes. The use of H2SO4 as a supporting electrolyte in the electrochemical detection of selenium has been reported [18]. The effect of pH on the availability of selenium for stripping was examined at pH of 1, 3, 6 and 10 as shown in (Fig. 2a). Such a study is important because the ionic states of metals can be affected by the proton environment (pH). An electrodeposition potential of −100 mV and electrodeposition time of 60 s were chosen as optimised parameters for the pre-concentration step. Claudete et al. were able to sense selenium by anodic stripping voltammetry using gold electrodes made from recordable CDs using deposition time of 60 s [18]. 3.3. Electrochemical detection of Se At the optimised conditions, GCE-AuNP electrodes gave a mark increased current response during the stripping of 10 ppm selenium (Fig. 3). This marked amplification of selenium signal is due to the fact that the electrodeposited gold has an affinity for selenium and they both interact to form an intermetallic compound Au–Se [19–20]. The
2.3. Selenium detection Square wave anodic stripping voltammetry (SWASV) was used for the detection of selenium (IV) in 0.1 M H2S04 on modified glassy carbon electrode. The pre-concentration potential and pre-concentration time were −100 mV and 100 s respectively. The SWVs were recorded at potentials ranging between 600 mV and 1000 mV to accommodate the stripping of selenium. The peak currents obtained from the SWASV of the various solutions were used to plot a calibration curve from which the regression equation was obtained. The electrode was subjected to a potential of 200 mV for 90 s to remove the excess selenium followed by a SWV sweep in Se (IV) free solution. Different concentrations of selenium were detected with a good detection limit. Furthermore, the GCE-AuNP electrode was used as a sensor for Se (IV) detection in tap water sample. For the purpose of validating the result obtained from this electrochemical sensor, inductively coupled plasma optical
Fig. 1. An overlay of CVs recorded on bare GCE and GCE-AuNPs in 5 mM [Fe (CN) 6]3−/4− (labelled/FC) and in 1 mM [Ru (NH3)6]2+/3+ (labelled/Ru) at 50 mV s−1 scan rate. Ep vs Ag/AgCl (3 M KCl).
A.O. Idris et al. / Journal of Electroanalytical Chemistry 758 (2015) 7–11
9
Fig. 2. (a) Effect of various supporting electrolytes on the Se (IV) peak current evaluated in 20 ppm Se (IV) solution. (b) SWASV of 5 ppm selenium at different pH on GCE-AuNP electrodes. Ep vs Ag/AgCl (3 M KCl).
SWVs of different Se (IV) concentrations between 0.001 mg L−1 and 50 mg L−1 were recorded using the optimised conditions. The peak currents increased linearly with selenium concentration as shown in Fig. 3b. The linear regression equation for Se (IV) detection was y = 2.6841 × 10−6 + 7.09669 × 10−6 with R = 0.9898. The limit of detection defined as CL = 3SB/m [21] (where CL, SB and m are the limit of detection, standard deviation of the blank and the slope of the calibration graph respectively) was calculated to be 0.64 μg L−1. The experiments were carried out in triplicates with good reproducibility as depicted by a RSD of 4.3%. The detection limit obtained in this work was compared to other reports (Table 1). The sensor developed in this work ranks among those with the lowest detection limits and has a wide linear range (Table 1). Furthermore, the detection limit is far below the maximum allowable limit of 10 ppb [3].
Fig. 3. (a) SWASV overlay of 10 ppm solution of Se (IV) on bare GCE and GCE-AuNP electrodes. (b) SWASV responses of different concentrations of selenium on GCE-AuNPs with the calibration graph (inset) where n = 3. (c) SWASV of 10 ppm solutions of Se (IV), Cu (II) and Cd (II). All Ep vs Ag/AgCl (3 M KCl).
3.4. Interference studies While the GCE-AuNP sensor was not susceptible to interference from common ions such as Ca2+, Na+, K+ and Mg2+; it was found to be susceptible to interferences from Cd2 + and Cu2 + as depicted in Fig. 3c. In the presence of Cd2 +, two peaks were observed in Fig. 3c. The selenium peak was reduced to shoulder while a more pronounced peak at 950 mV may contain Se4+, Cd2+ and their alloy. Cd had been reported to be an interference ion in selenium sensing [11–12,22]. This interference could be the possibility of the formation of intermetallic compounds or alloys such as Cd–Se during the electrodeposition step, which may amplify the selenium signal [23–24].
Table 1 Detection limits of some reports on the electrochemical detection of Se (IV). Electrode type
Mode
Linear range
Detection limit
References
Pt BiFe HMDE MGUA GCE-AuNPs
OSWV SWASV CCSV SWASV SWASV
0.01–20 ppb 0–100 ppb 50 pM–0.5 nM 0–100 ppb 0.001–50 ppm
0.004 ppb 0.1 ppb 2.4 pM 0.42 ppb 0.34 ppb
[10] [11] [12] [15] This study
OSWV — Osteryoung Square Wave Voltammetry, BiFe — Bismuth Film Electrode, HMDE — Hanging Mercury Drop Electrode, CCSV — Catalytic Cathodic Stripping Voltammetry.
10
A.O. Idris et al. / Journal of Electroanalytical Chemistry 758 (2015) 7–11
Fig. 4. a) SWASV of selenium in tap water sample; b) corresponding calibration curve. Ep vs Ag/AgCl (3 M KCl).
Copper and selenium are well known to interact during electrochemical reductions as they form a bi-metallic compound Cu2Se [22, 25–26]. The mechanism for the formation of Cu2Se bimetallic compounds at gold electrodes has been proposed [26]. The addition of Cu in the detection of Se (IV) resulted in a 38% lower current than in the absence of copper (Fig. 3c). A reduction wave at ca +150 mV when 5 μM of copper was added to 100 μM selenium solutions has also been reported [22]. This interference may be as a result of the proximity between the stripping potential of copper and selenium. 3.5. Selenium detection in real water sample The GCE-AuNP sensor was used in the detection of Se (IV) in tap water sample (Fig. 4). Briefly, a measured volume of tap water sample was placed in the vial and 0.1 M H2SO4 was added to the tap water sample as supporting electrolyte. Even without the need to spike. GCE-AuNP sensor electrode was sensitive enough to detect Se (IV) in water denoting the applicability and sensitivity of this electrode. When spiking was carried out, the peak increased in the same potential range and a recovery of 101% average was obtained. A concentration of 8.86 (± 0.34) μg L− 1 selenium was calculated for the tap water. In comparison, ICP-MS analysis at n = 3 gave a selenium concentration of 9.42 (±0.25) μg L−1. Using t test at 95% confidence level, the value of tobserved (2.30) was less than tcritical (2.78). Thus the methods are in good agreement and the difference is only due to random error. The concentration of Se (IV) in the tap water is however within the limit of 10 μg L−1 guideline of the World Health Organization [3]. Piech and Kubiak [25] detected a 0.19 ppb concentration Se (IV) in tap water with a recovery of 95% using hanging copper amalgam drop electrode. It should be noted however that real water sample (such as tap water) correlation may be difficult owing to different compositions or matrices which can result from factors such as geological location, water treatment efficiencies and so on. 4. Conclusion The work has shown that Se (IV) can be detected by square wave anodic stripping voltammetry on a gold nanoparticle modified glassy carbon electrode. The presence of gold as a modifier enhanced the detection of selenium. The electrochemical sensor has a detection limit of 0.64 μg L−1 (which is lower than the permitted limit) and also possesses a wide linear range. The simple method was used for real water sample and the method was validated by its good correlation with inductively coupled plasma optical emission spectrometry technique. Thus the electrochemical sensor reported in this work has analytical significance.
Acknowledgements Financial support from the following are gratefully acknowledged: The National Research Foundation of South Africa Thuthuka Research Grant (grant number TTK13070620648); the Centre for Nanomaterials Science Research, University of Johannesburg (UJ); Faculty of Science, University of Johannesburg; and Eskom Tertiary Education Support Programme (TESP) grant (awarded in 2014 and 2015). The authors wish to thank Ms Esther Tsotetsi and Ms Siposetu Madyibi for the contributions made during their Work Integrated Learning Programme which was supported by ESKOM TESP 2014/2015 grants.
References [1] S. Khanna, C. Udas, G.K. Kumar, S. Suvarna, F.R. Karjodkar, Trace elements (copper, zinc, selenium and molybdenum) as markers in oral sub mucous fibrosis and oral squamous cell carcinoma, J. Trace Elem. Med. Biol. 27 (2013) 307–311. [2] T. Himoto, H. Yoneyama, K. Kurokohchi, M. Inukai, H. Masugata, F. Goda, R. Haba, S. Watababe, S. Kubota, S. Senda, T. Masaki, Selenium deficiency is associated with insulin resistance in patients with hepatitis C virus-related chronic liver disease, Nutr. Res. 31 (2011) 829–835. [3] M.R. Awual, T. Yaita, S. Suzuki, H. Shiwaku, Ultimate selenium (IV) monitoring and removal from water using a new class of organic ligand based composite adsorbent, J. Hazard. Mater. 291 (2015) 111–119. [4] S. Santos, G. Ungureanu, R. Boaventura, C. Botelho, Selenium contaminated waters: an overview of analytical methods, treatment options and recent advances in sorption methods, Sci. Total Environ. 521–522C (2015) 246–260. [5] G. Nogay, E. Ozkol, S. Ilday, R. Turan, Structural peculiarities and aging effect in hydrogenated a-Si prepared by inductively coupled plasma assisted chemical vapour deposition technique, Vacuum 110 (2014) 114–120. [6] M. Ghaedi, K. Niknam, A. Shokrollahi, E. Niknam, H.R. Rajabi, M. Soylak, Flame atomic absorption spectrometric determination of trace amounts of heavy metal ions after solid phase extraction using modified sodium dodecyl sulfate coated on alumina, J. Hazard. Mater. 155 (2008) 121–127. [7] N. Maleki, A. Safavi, M.M. Doroodmand, Determination of selenium in water and soil by hydride generation atomic absorption spectrometry using solid reagents, Talanta 66 (2005) 858–862. [8] V. Stibilj, P. Smrkolj, A. Krbavcic, Investigation of the declared value of selenium in food supplements by HG-AFS, Microchim. Acta 150 (2005) 323–327. [9] M.B. Gumpu, S. Sethuraman, U.M. Krishnan, J.B.B. Rayappan, A review on detection of heavy metal ions in water — an electrochemical approach, Sensors Actuators B Chem. 213 (2015) 515–533. [10] F.A. Bertolino, A.J. Torriero, E. Salinas, R. Olsina, L.D. Martinez, J. Raba, Speciation analysis of selenium in natural water using square-wave voltammetry after preconcentration on activated carbon, Anal. Chim. Acta 572 (2006) 32–38. [11] D.W. Bryce, A. Izquierdo, M.D.L. De Castro, Flow-injection anodic stripping voltammetry at a gold electrode for selenium (IV) determination, Anal. Chim. Acta 308 (1995) 96–101. [12] Q. Zhang, X. Li, H. Shi, Z. Yuan, Determination of trace selenium by differential pulse adsorptive stripping voltammetry at a bismuth film electrode, Electrochim. Acta 55 (2010) 4717-472. [13] B. Lange, C.M.G. Van den Berg, Determination of selenium by catalytic cathodic stripping voltammetry, Anal. Chim. Acta 418 (2000) 33–42. [14] A. Kolliopoulos, J. Metters, C. Banks, Electroanalytical sensing printed graphite electrodes, Anal. Methods 5 (2013) 851–856.
A.O. Idris et al. / Journal of Electroanalytical Chemistry 758 (2015) 7–11 [15] S.H. Tan, S.P. Kounaves, Determination of selenium (IV) at a micro fabricated gold ultramicroelectrode array using square wave anodic stripping voltammetry, J. Electroanal. Chem. 10 (1998) 364–368. [16] O.A. Arotiba, A. Ignaszak, R. Malgas, A. Al-Ahmed, P.G.L. Baker, S.F. Mapolie, E.I. Iwuoha, An electrochemical DNA biosensor developed on novel multinuclear nickel (II) salicylaldimine metallodendrimer platform, Electrochim. Acta 53 (2007) 1689–1696. [17] T. Ndlovu, B.B. Mamba, S. Sampath, R.W. Krause, O.A. Arotiba, Voltammetry detection of arsenic on a bismuth modified exfoliated graphite, Electrochim. Acta 128 (2014) 48–53. [18] C.F. Pereira, F. Gonzaga, A. Guaritasantos, J. Souzade, Determination of Se (IV) by anodic stripping voltammetry using gold electrodes made from recordable CDs, Talanta 69 (2006) 877–881. [19] B.J. Foss, A. Ion, V. Partali, H.R. Sliwka, F.G. Banica, Electrochemical and EQCM investigation of selenium derivatized carotenoid in the self-assembled state at a gold electrode, J. Electroanal. Chem. 593 (2006) 15–28. [20] G. Riveros, R. Henríquez, R. Cordova, R. Schrebler, E.A. Dalchiele, H. Gómez, Electrochemical study concerning the deposition of copper on selenium covered gold electrodes, J. Electroanal. Chem. 504 (2001) 160–165.
11
[21] H. Chen, Y. Teng, S. Lu, Y. Wang, J. Wang, Contamination features and health risk of soil heavy metals in China, Sci. Total Environ. 512–513 (2015) 143–153. [22] V. Beni, G. Collins, D.W.M. Arrigan, Investigation into the voltammetric behaviour and detection of selenium (IV) at metal electrodes in diverse electrolyte media, Anal Chim. Acta 699 (2011) 127–133. [23] G.H. Hwang, W.K. Han, S.J. Hong, J.S. Park, S.G. Kang, Determination of trace amounts of lead and cadmium using a bismuth/glassy carbon composite electrode, Talanta 77 (2009) 1432–1436. [24] W. Yantasee, Y. Lin, G.E. Fryxell, B.J. Busche, Simultaneous detection of cadmium, copper, and lead using a carbon paste electrode modified with carbamoylphosphonic acid self-assembled monolayer on mesoporous silica (SAMMS), Anal. Chim. Acta 502 (2004) 207–212. [25] R. Piech, W.W. Kubiak, Determination of trace selenium on hanging copper amalgam drop electrode, Electrochim. Acta 53 (2007) 584–589. [26] A. Marlot, J. Vedel, Electrodeposition of copper–selenium compounds onto gold using a rotating electrochemical quartz crystal microbalance, J. Electrochem. Soc. 146 (1999) 177–183.
Contents lists available at ScienceDirect
Journal of Electroanalytical Chemistry journal homepage: www.elsevier.com/locate/jeac
Electroanalysis of selenium in water on an electrodeposited gold-nanoparticle modified glassy carbon electrode A.O. Idris a, N. Mabuba a, O.A. Arotiba a,b,⁎ a b
Department of Applied Chemistry, University of Johannesburg, Johannesburg, South Africa Centre for Nanomaterials Science Research, University of Johannesburg, South Africa
a r t i c l e
i n f o
Article history: Received 9 July 2015 Received in revised form 20 August 2015 Accepted 10 October 2015 Available online xxxx Keywords: Selenium Square wave anodic stripping voltammetry Electrodeposition Gold nanoparticles Water
a b s t r a c t This work presents a simple and cheaper method of detecting selenium in water using an electrochemical sensor based on gold nanoparticle (AuNP) modified glassy carbon electrode. AuNPs were electrochemically deposited on a glassy carbon electrode (GCE) using cyclic voltammetry within the potential range of −0.4 mV to 1.1 mV for 10 cycles. The modification of GCE with AuNPs resulted in an increase in the electroactive surface area of the electrode which led to the enhancement of the redox current peaks of [Fe (CN)6]3−/4− and [Ru (NH3)6]2+/3+ in comparison to the bare GCE. Square wave anodic stripping voltammetry was used to detect Se (IV) in water (in 0.1 M H2SO4 as supporting electrolyte) at the following optimum conditions: pH 1, deposition potential of − 100 mV, pre-concentration time of 60 s. The GCE-AuNP sensor was able to detect Se (IV) to the limit of 0.64 μg L−1 and was not susceptible to many interfering cations except Cu (II) and Cd (II). This method involves a simple one step electrode modification. The sensor was used to detect Se (IV) in a real sample. This method was validated by its good correlation with the result obtained from inductively coupled plasma method. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Selenium is a trace element that is important in our diet and in the health of plants and animals when taken in the correct amount [1]. Exposure to an excess amount of selenium is toxic and can cause health problems such as gastrointestinal upsets, hair loss, white blotchy nails, garlic breathe odour and nerve damage. Selenium deficiency on the other hand, has been linked to cancer, heart diseases, muscular dystrophy, multiple sclerosis, immune system, osteoarthopathy, reproductive disorder in humans and white muscle disease in animals [2]. The World Health Organization has set a limit of 10 μg L−1 selenium concentration in drinking water [3]. Human activities that have facilitated the increase in the concentration of selenium in the environment include the following: the mining and processing of base metal, gold and coal; phosphate deposits; the use of rock phosphate as fertiliser; the application of sewage sludge to land; the manufacture of detergents and shampoo; and so on. The increased use of selenium in the pharmaceutical, glazing, photocopying, ceramics, paint and electronics industries has also increased the amount of selenium entering the environment [4]. Owing to the toxicity and the environmental importance of selenium, its detection and quantification have been studied using various methodologies. The three main methods of selenium detection are ⁎ Corresponding author at: Department of Applied Chemistry, University of Johannesburg, Johannesburg, South Africa. E-mail address: [email protected] (O.A. Arotiba).
http://dx.doi.org/10.1016/j.jelechem.2015.10.009 1572-6657/© 2015 Elsevier B.V. All rights reserved.
inductively coupled plasma (ICP), spectroscopy and electrochemistry. Inductively coupled plasma has been used to detect selenium at low concentrations but the pitfall of this method is that it is expensive, it requires extensive sample preparation, it has a long analysis time and it requires sophisticated operator personnel [5]. The spectroscopic methods predominantly used are graphite furnace atomic absorption spectroscopy [6], hydride generation atomic absorption spectroscopy [7] and atomic fluorescence spectrometry [8]. Electrochemical techniques for selenium detection have an edge over the other techniques based on the grounds such as low cost of instrument, minimum sample preparation for analysis, faster analysis time and amenability to miniaturisation [9]. Bertolino et al. reported the speciation analysis of selenium in natural water using square wave voltammetry after pre-concentration on activated carbon and selenium was detected to a limit of 0.004 μg L−1 within 0.01 μg L−1–20 μg L−1 concentration range [10]. In another report, Se (VI) was reduced to Se (IV) before analysis using flow injection anodic stripping voltammetry at a gold electrode for Se (IV) determination [11]. The Se (IV) concentration was calculated as the difference between the result for total inorganic selenium and Se (IV). The interfering divalent cations were removed by using cation exchange column prior to the injection value. Differential pulse adsorptive stripping voltammetry at a bismuth film electrode with cetyl trimethyl ammonium bromide and p-amino benzene sulphonic acid as a complexing agent, was used to determine trace selenium by Zhang et al. [12]. A detection limit of 0.1 μg L− 1 and accumulation time of 300 s were used. Britta Lange et al. were able to determine Se (IV) by catalytic cathodic
8
A.O. Idris et al. / Journal of Electroanalytical Chemistry 758 (2015) 7–11
stripping voltammetry in the presence of Rhodium. Rhodium was used for this work because of its catalytic ability. U.V. digestion was used to eliminate the interference caused by inorganic material and to convert Se (VI) to Se (IV) and a detection limit of 2.4 pM was obtained [13]. Attempts to use microelectrode in the determination of Se (IV) have also been reported [14]. Tan and Kounaves were able to determine Se (IV) at a micro fabricated Au ultra-micro electrode array using square wave anodic stripping voltammetry; they discovered that Se (IV) redox reaction appears to be kinetically faster and more reversible at the Au Ultra microelectrode (UME) array than at macro electrode or single Au UME. The limit of detection was 0.42 ppb [15]. In this work, gold nanoparticles were electrodeposited on glassy carbon electrode in a one step approach and used to detect Se (IV) by square wave anodic stripping voltammetry. The choice of supporting electrolyte, pH and deposition potential was optimised. The sensor was used to detect Se (IV) in both analytical and real water samples. The concentration of Se (IV) detected in real water sample was correlated with the result obtained from inductively coupled plasma optical emission spectrometry technique. 2. Experimental
emission spectrometry technique was also used to detect the amount of Se (IV) in the tap water sample. 3. Results and discussion 3.1. Electrode characterisation Bare glassy carbon and GCE-AuNP electrodes were electrochemically characterised by using CV in [Fe (CN) 6]3−/4− and [Ru (NH3)6]2+/3+ solutions; this was done by cycling a potential from −400 mV to 600 mV at a scan rate of 50 mV s−1, to compare the peak current signal of the bare GCE and GCE-AuNPs. Fig. 1 shows the electrochemical responses of the bare and modified GCE in two common redox probes. As depicted in Fig. 1, the presence of AuNPs enhanced the interfacial electron transfer of Ferrocyannide and Ruthenium redox probes. This electrochemical enhancement is an indication of the potential application of GCE-AuNP electrodes in electroanalysis. The increase in current can be attributed to an increase in electroactive surface area of the electrode which can enhance the sensitivity of the electrode when used for electroanalysis [17]. 3.2. Optimisation: the effect of supporting electrolyte, pH, deposition potential and deposition time
2.1. Material and methods HAuCl4, KCl, Na2SeO3, HNO3, KNO3, K2HPO4, KH2PO4, NaOH, K3Fe(CN)6, K4Fe(CN)6.3H2O, Ru(NH3)6Cl3, Ru(NH3)6Cl2, H2SO4 and KOH were purchased from sigma Aldrich. All chemicals were of highest analytical grade and deionised water was used for the preparation of solutions. All electrochemical measurements were done on Compactstat electrochemical workstation (Ivium Technologies, Netherlands), using a three-electrode configuration. The working electrode, counter electrode and reference electrode were glassy carbon electrode, platinum wire and Ag/AgCl (3 M KCl) respectively. The electrochemical cells were purged with ultra-pure argon gas for at least 10 min prior to all electrochemical measurements. 2.2. Electrode modification Glassy carbon electrode (GCE) was modified with gold nanoparticles (AuNPs) according to the method reported by Arotiba et al. [16]. Briefly, GCE was modified with 5 mM of HAuCl4 solutions by cycling the potential from −400 mV to 1100 mV for 10 cycles at a scan rate 50 mV s−1, the reference electrode used is Ep vs Ag/AgCl (3 M KCl). The modified electrode was referred to as GCE-AuNPs. The modified electrodes were electrochemically characterised using square wave voltammetry and cyclic voltammetry (CV) in 5 mM mixture of K3Fe(CN)6 and K4Fe(CN)6.3H2O (referred to as [Fe (CN) 6]3−/4−) and 1 mM mixture of Ru(NH3)6Cl3 and Ru(NH3)6Cl2 (referred to as [Ru (NH3)6]2 +/3 +) redox probes.
The effect of supporting electrolyte is presented in Fig. 2a. In this work, 0.1 M H2SO4 was chosen as the supporting electrolyte because it is more suitable for the stripping of Se (IV) due to the highest current signal obtained in comparison to other supporting electrolytes. The use of H2SO4 as a supporting electrolyte in the electrochemical detection of selenium has been reported [18]. The effect of pH on the availability of selenium for stripping was examined at pH of 1, 3, 6 and 10 as shown in (Fig. 2a). Such a study is important because the ionic states of metals can be affected by the proton environment (pH). An electrodeposition potential of −100 mV and electrodeposition time of 60 s were chosen as optimised parameters for the pre-concentration step. Claudete et al. were able to sense selenium by anodic stripping voltammetry using gold electrodes made from recordable CDs using deposition time of 60 s [18]. 3.3. Electrochemical detection of Se At the optimised conditions, GCE-AuNP electrodes gave a mark increased current response during the stripping of 10 ppm selenium (Fig. 3). This marked amplification of selenium signal is due to the fact that the electrodeposited gold has an affinity for selenium and they both interact to form an intermetallic compound Au–Se [19–20]. The
2.3. Selenium detection Square wave anodic stripping voltammetry (SWASV) was used for the detection of selenium (IV) in 0.1 M H2S04 on modified glassy carbon electrode. The pre-concentration potential and pre-concentration time were −100 mV and 100 s respectively. The SWVs were recorded at potentials ranging between 600 mV and 1000 mV to accommodate the stripping of selenium. The peak currents obtained from the SWASV of the various solutions were used to plot a calibration curve from which the regression equation was obtained. The electrode was subjected to a potential of 200 mV for 90 s to remove the excess selenium followed by a SWV sweep in Se (IV) free solution. Different concentrations of selenium were detected with a good detection limit. Furthermore, the GCE-AuNP electrode was used as a sensor for Se (IV) detection in tap water sample. For the purpose of validating the result obtained from this electrochemical sensor, inductively coupled plasma optical
Fig. 1. An overlay of CVs recorded on bare GCE and GCE-AuNPs in 5 mM [Fe (CN) 6]3−/4− (labelled/FC) and in 1 mM [Ru (NH3)6]2+/3+ (labelled/Ru) at 50 mV s−1 scan rate. Ep vs Ag/AgCl (3 M KCl).
A.O. Idris et al. / Journal of Electroanalytical Chemistry 758 (2015) 7–11
9
Fig. 2. (a) Effect of various supporting electrolytes on the Se (IV) peak current evaluated in 20 ppm Se (IV) solution. (b) SWASV of 5 ppm selenium at different pH on GCE-AuNP electrodes. Ep vs Ag/AgCl (3 M KCl).
SWVs of different Se (IV) concentrations between 0.001 mg L−1 and 50 mg L−1 were recorded using the optimised conditions. The peak currents increased linearly with selenium concentration as shown in Fig. 3b. The linear regression equation for Se (IV) detection was y = 2.6841 × 10−6 + 7.09669 × 10−6 with R = 0.9898. The limit of detection defined as CL = 3SB/m [21] (where CL, SB and m are the limit of detection, standard deviation of the blank and the slope of the calibration graph respectively) was calculated to be 0.64 μg L−1. The experiments were carried out in triplicates with good reproducibility as depicted by a RSD of 4.3%. The detection limit obtained in this work was compared to other reports (Table 1). The sensor developed in this work ranks among those with the lowest detection limits and has a wide linear range (Table 1). Furthermore, the detection limit is far below the maximum allowable limit of 10 ppb [3].
Fig. 3. (a) SWASV overlay of 10 ppm solution of Se (IV) on bare GCE and GCE-AuNP electrodes. (b) SWASV responses of different concentrations of selenium on GCE-AuNPs with the calibration graph (inset) where n = 3. (c) SWASV of 10 ppm solutions of Se (IV), Cu (II) and Cd (II). All Ep vs Ag/AgCl (3 M KCl).
3.4. Interference studies While the GCE-AuNP sensor was not susceptible to interference from common ions such as Ca2+, Na+, K+ and Mg2+; it was found to be susceptible to interferences from Cd2 + and Cu2 + as depicted in Fig. 3c. In the presence of Cd2 +, two peaks were observed in Fig. 3c. The selenium peak was reduced to shoulder while a more pronounced peak at 950 mV may contain Se4+, Cd2+ and their alloy. Cd had been reported to be an interference ion in selenium sensing [11–12,22]. This interference could be the possibility of the formation of intermetallic compounds or alloys such as Cd–Se during the electrodeposition step, which may amplify the selenium signal [23–24].
Table 1 Detection limits of some reports on the electrochemical detection of Se (IV). Electrode type
Mode
Linear range
Detection limit
References
Pt BiFe HMDE MGUA GCE-AuNPs
OSWV SWASV CCSV SWASV SWASV
0.01–20 ppb 0–100 ppb 50 pM–0.5 nM 0–100 ppb 0.001–50 ppm
0.004 ppb 0.1 ppb 2.4 pM 0.42 ppb 0.34 ppb
[10] [11] [12] [15] This study
OSWV — Osteryoung Square Wave Voltammetry, BiFe — Bismuth Film Electrode, HMDE — Hanging Mercury Drop Electrode, CCSV — Catalytic Cathodic Stripping Voltammetry.
10
A.O. Idris et al. / Journal of Electroanalytical Chemistry 758 (2015) 7–11
Fig. 4. a) SWASV of selenium in tap water sample; b) corresponding calibration curve. Ep vs Ag/AgCl (3 M KCl).
Copper and selenium are well known to interact during electrochemical reductions as they form a bi-metallic compound Cu2Se [22, 25–26]. The mechanism for the formation of Cu2Se bimetallic compounds at gold electrodes has been proposed [26]. The addition of Cu in the detection of Se (IV) resulted in a 38% lower current than in the absence of copper (Fig. 3c). A reduction wave at ca +150 mV when 5 μM of copper was added to 100 μM selenium solutions has also been reported [22]. This interference may be as a result of the proximity between the stripping potential of copper and selenium. 3.5. Selenium detection in real water sample The GCE-AuNP sensor was used in the detection of Se (IV) in tap water sample (Fig. 4). Briefly, a measured volume of tap water sample was placed in the vial and 0.1 M H2SO4 was added to the tap water sample as supporting electrolyte. Even without the need to spike. GCE-AuNP sensor electrode was sensitive enough to detect Se (IV) in water denoting the applicability and sensitivity of this electrode. When spiking was carried out, the peak increased in the same potential range and a recovery of 101% average was obtained. A concentration of 8.86 (± 0.34) μg L− 1 selenium was calculated for the tap water. In comparison, ICP-MS analysis at n = 3 gave a selenium concentration of 9.42 (±0.25) μg L−1. Using t test at 95% confidence level, the value of tobserved (2.30) was less than tcritical (2.78). Thus the methods are in good agreement and the difference is only due to random error. The concentration of Se (IV) in the tap water is however within the limit of 10 μg L−1 guideline of the World Health Organization [3]. Piech and Kubiak [25] detected a 0.19 ppb concentration Se (IV) in tap water with a recovery of 95% using hanging copper amalgam drop electrode. It should be noted however that real water sample (such as tap water) correlation may be difficult owing to different compositions or matrices which can result from factors such as geological location, water treatment efficiencies and so on. 4. Conclusion The work has shown that Se (IV) can be detected by square wave anodic stripping voltammetry on a gold nanoparticle modified glassy carbon electrode. The presence of gold as a modifier enhanced the detection of selenium. The electrochemical sensor has a detection limit of 0.64 μg L−1 (which is lower than the permitted limit) and also possesses a wide linear range. The simple method was used for real water sample and the method was validated by its good correlation with inductively coupled plasma optical emission spectrometry technique. Thus the electrochemical sensor reported in this work has analytical significance.
Acknowledgements Financial support from the following are gratefully acknowledged: The National Research Foundation of South Africa Thuthuka Research Grant (grant number TTK13070620648); the Centre for Nanomaterials Science Research, University of Johannesburg (UJ); Faculty of Science, University of Johannesburg; and Eskom Tertiary Education Support Programme (TESP) grant (awarded in 2014 and 2015). The authors wish to thank Ms Esther Tsotetsi and Ms Siposetu Madyibi for the contributions made during their Work Integrated Learning Programme which was supported by ESKOM TESP 2014/2015 grants.
References [1] S. Khanna, C. Udas, G.K. Kumar, S. Suvarna, F.R. Karjodkar, Trace elements (copper, zinc, selenium and molybdenum) as markers in oral sub mucous fibrosis and oral squamous cell carcinoma, J. Trace Elem. Med. Biol. 27 (2013) 307–311. [2] T. Himoto, H. Yoneyama, K. Kurokohchi, M. Inukai, H. Masugata, F. Goda, R. Haba, S. Watababe, S. Kubota, S. Senda, T. Masaki, Selenium deficiency is associated with insulin resistance in patients with hepatitis C virus-related chronic liver disease, Nutr. Res. 31 (2011) 829–835. [3] M.R. Awual, T. Yaita, S. Suzuki, H. Shiwaku, Ultimate selenium (IV) monitoring and removal from water using a new class of organic ligand based composite adsorbent, J. Hazard. Mater. 291 (2015) 111–119. [4] S. Santos, G. Ungureanu, R. Boaventura, C. Botelho, Selenium contaminated waters: an overview of analytical methods, treatment options and recent advances in sorption methods, Sci. Total Environ. 521–522C (2015) 246–260. [5] G. Nogay, E. Ozkol, S. Ilday, R. Turan, Structural peculiarities and aging effect in hydrogenated a-Si prepared by inductively coupled plasma assisted chemical vapour deposition technique, Vacuum 110 (2014) 114–120. [6] M. Ghaedi, K. Niknam, A. Shokrollahi, E. Niknam, H.R. Rajabi, M. Soylak, Flame atomic absorption spectrometric determination of trace amounts of heavy metal ions after solid phase extraction using modified sodium dodecyl sulfate coated on alumina, J. Hazard. Mater. 155 (2008) 121–127. [7] N. Maleki, A. Safavi, M.M. Doroodmand, Determination of selenium in water and soil by hydride generation atomic absorption spectrometry using solid reagents, Talanta 66 (2005) 858–862. [8] V. Stibilj, P. Smrkolj, A. Krbavcic, Investigation of the declared value of selenium in food supplements by HG-AFS, Microchim. Acta 150 (2005) 323–327. [9] M.B. Gumpu, S. Sethuraman, U.M. Krishnan, J.B.B. Rayappan, A review on detection of heavy metal ions in water — an electrochemical approach, Sensors Actuators B Chem. 213 (2015) 515–533. [10] F.A. Bertolino, A.J. Torriero, E. Salinas, R. Olsina, L.D. Martinez, J. Raba, Speciation analysis of selenium in natural water using square-wave voltammetry after preconcentration on activated carbon, Anal. Chim. Acta 572 (2006) 32–38. [11] D.W. Bryce, A. Izquierdo, M.D.L. De Castro, Flow-injection anodic stripping voltammetry at a gold electrode for selenium (IV) determination, Anal. Chim. Acta 308 (1995) 96–101. [12] Q. Zhang, X. Li, H. Shi, Z. Yuan, Determination of trace selenium by differential pulse adsorptive stripping voltammetry at a bismuth film electrode, Electrochim. Acta 55 (2010) 4717-472. [13] B. Lange, C.M.G. Van den Berg, Determination of selenium by catalytic cathodic stripping voltammetry, Anal. Chim. Acta 418 (2000) 33–42. [14] A. Kolliopoulos, J. Metters, C. Banks, Electroanalytical sensing printed graphite electrodes, Anal. Methods 5 (2013) 851–856.
A.O. Idris et al. / Journal of Electroanalytical Chemistry 758 (2015) 7–11 [15] S.H. Tan, S.P. Kounaves, Determination of selenium (IV) at a micro fabricated gold ultramicroelectrode array using square wave anodic stripping voltammetry, J. Electroanal. Chem. 10 (1998) 364–368. [16] O.A. Arotiba, A. Ignaszak, R. Malgas, A. Al-Ahmed, P.G.L. Baker, S.F. Mapolie, E.I. Iwuoha, An electrochemical DNA biosensor developed on novel multinuclear nickel (II) salicylaldimine metallodendrimer platform, Electrochim. Acta 53 (2007) 1689–1696. [17] T. Ndlovu, B.B. Mamba, S. Sampath, R.W. Krause, O.A. Arotiba, Voltammetry detection of arsenic on a bismuth modified exfoliated graphite, Electrochim. Acta 128 (2014) 48–53. [18] C.F. Pereira, F. Gonzaga, A. Guaritasantos, J. Souzade, Determination of Se (IV) by anodic stripping voltammetry using gold electrodes made from recordable CDs, Talanta 69 (2006) 877–881. [19] B.J. Foss, A. Ion, V. Partali, H.R. Sliwka, F.G. Banica, Electrochemical and EQCM investigation of selenium derivatized carotenoid in the self-assembled state at a gold electrode, J. Electroanal. Chem. 593 (2006) 15–28. [20] G. Riveros, R. Henríquez, R. Cordova, R. Schrebler, E.A. Dalchiele, H. Gómez, Electrochemical study concerning the deposition of copper on selenium covered gold electrodes, J. Electroanal. Chem. 504 (2001) 160–165.
11
[21] H. Chen, Y. Teng, S. Lu, Y. Wang, J. Wang, Contamination features and health risk of soil heavy metals in China, Sci. Total Environ. 512–513 (2015) 143–153. [22] V. Beni, G. Collins, D.W.M. Arrigan, Investigation into the voltammetric behaviour and detection of selenium (IV) at metal electrodes in diverse electrolyte media, Anal Chim. Acta 699 (2011) 127–133. [23] G.H. Hwang, W.K. Han, S.J. Hong, J.S. Park, S.G. Kang, Determination of trace amounts of lead and cadmium using a bismuth/glassy carbon composite electrode, Talanta 77 (2009) 1432–1436. [24] W. Yantasee, Y. Lin, G.E. Fryxell, B.J. Busche, Simultaneous detection of cadmium, copper, and lead using a carbon paste electrode modified with carbamoylphosphonic acid self-assembled monolayer on mesoporous silica (SAMMS), Anal. Chim. Acta 502 (2004) 207–212. [25] R. Piech, W.W. Kubiak, Determination of trace selenium on hanging copper amalgam drop electrode, Electrochim. Acta 53 (2007) 584–589. [26] A. Marlot, J. Vedel, Electrodeposition of copper–selenium compounds onto gold using a rotating electrochemical quartz crystal microbalance, J. Electrochem. Soc. 146 (1999) 177–183.