Electrodeposited Bi-Au Nano Composites Modified Carbon Paste Electrode for the Simultaneous Determination of Copper and Mercury

Composites of bismuth and gold nano particles (Bi-AuNPs) were prepared on carbon paste electrode (CPE) by electrochemical deposition method. The electrochemical parameters were optimized for the effective deposition of Bi-AuNPs composite materials. Microscopic examination revealed that, nanoparticle clusters of gold were embedded well within the bismuth films (Bi films) over the CPE surface. An analytical method has been developed for the simultaneous determination of copper and mercury at ultra trace levels using the Bi-AuNPs modified CPE by anodic stripping voltammetry. It was possible to observe well separated stripping peaks of the two metal ions and the modified electrode was successfully applied for simultaneous determination of Cu and Hg. Limits of detection using the optimized analytical procedure were observed as 0.16 µg L -1 and 0.28 µg L -1 for Cu and Hg respectively. Interference effects of some of the commonly occurring metal ions were investigated and the method was applied for the determination of Cu and Hg in two ground water and two soil samples collected from different places.

Please note that technical editing may introduce minor changes to the text and/or graphics, which may alter content. The journal's standard Terms & Conditions and the Ethical guidelines still apply. In no event shall the Royal Society of Chemistry be held responsible for any errors or omissions in this Accepted Manuscript or any consequences arising from the use of any information it contains.

Introduction
Due to their interesting properties, Cu and Hg have been extensively used in industries and there are enormous concerns of contamination of these metal ions in the environment through the industrial activities. All the heavy metal ions, beyond a certain concentration levels are highly toxic to the environments; Cu and Hg are two important toxic elements in the list. In the body system Cu generally present in the bound form with ceruloplasmin and non toxic [1].
The free Cu in the body system causes toxicity as it generates reactive oxygen species and damages proteins, lipids and DNAs [1]. Stripping voltammetry is one of the most important methods of determination of Cu in wide concentration ranges using mercury drop electrode. Present days, due to the toxicity of mercury, mercury drop electrodes are being avoided in the stripping as among all the heavy metal ions mercury toxicity is more severe than the others. Mercury exhibits a very complex behavior in the food chain and in the human body system [2,3]. There have been some methods of determination of mercury such as, gas chromatography-atomic fluorescence spectrometry and atomic absorption spectrometry [4][5][6][7][8][9].
Other commonly used methods for the determination of total mercury are: cold vapor atomic absorption spectrometry (CVAAS), inductively coupled plasma mass spectrometry (ICP-MS) [10,11]. All the methods mentioned are either less sensitive, lack in selectivity or require complicated instrumentations and also involve high operational cost. Voltammetric techniques represent important method with high sensitivity and low operational cost and have the possibility of portable instrumentations for onsite determination. Bismuth film electrode is an interesting alternative to mercury drop electrode for the determination of toxic heavy metal ions.
Bismuth, known to be an environment friendly material has been used in wide variety of cosmetics and also in medicines [12][13][14][15][16][17][18]. Bi film electrodes were deposited over glassy carbon electrode [19][20][21][22][23][24][25] or over CPE [26][27][28] and applied in stripping voltammetry. However the easy renewability of CPE is a better choice for the present purpose compared to glassy carbon electrode. It has been observed by previous researchers that a strong stripping peak due to Bi was observed before Cu peak, which is detrimental to the appearance of Cu stripping peak on bare Bi film electrodes [13,29]. There was one report based on the decrease in the Bi stripping peak for the determination of Cu using bare Bi film electrodes [13]. We also have reported the determination of Cu from the decrease in the Bi stripping peak on bare Bi film electrode [29] and

RSC Advances Accepted Manuscript
a direct determination of Cu using Bi-Graphene composite electrode [30]. In the present investigation we have incorporated gold into the Bi film so that the stripping peak due to Cu could be observed. Gold has the strong affinity towards mercury thus the Bi-AuNPs composite material would be the ideal choice for the simultaneous determination of Cu and Hg.
It has been reported that the nanoparticle arrays have increased the effective mass transport and also increased the signal to noise ratio of analytical measurements [31][32][33][34][35][36]. Merged diffusion layers of nanoparticle modified electrodes also reduced the requirement of active materials to obtain similar or enhanced response to the whole electrode. There are a few reports on the gold nanoparticles (AuNPs) modified electrode for the determination of mercury either on the carbon electrode or on screen printed electrode [37][38][39][40][41]. One paper reported the modified electrode using EDTA-like films for the simultaneous determination of Cu and Hg [42]. In a recent publication heavy metal ions were determined using bismuth nano tube modified glassy carbon electrode [43]. Other than analytical application Au and Bi 2 O 3 nano composite modified electrode was also applied for photocatalytic decomposition of dyes [44].

Instrumentation and procedure
Electrochemical studies were carried out using the cell assembly with a three-electrode

RSC Advances Accepted Manuscript
Energy dispersive X-ray fluorescence (EDXRF) was carried out using EX-3600 M spectrometer from Xenemetrix. Electrodeposited samples were dried under vacuum before the measurements with SEM and EDXRF.

Reagents and Solutions
All the chemicals used in the present study were AR grade from Merck India limited or SD fine-chem limited. A stock solution of Au (0.01 M) was prepared by dissolving an accurately weighed amount of high purity Au metal in aqua regia. Appropriate quantities of all the reagents were weighed, dissolved in ultrapure water and made up to desired volume. High purity nitrogen gas (from Indian Oxygen Ltd.) was used for removing the dissolved oxygen from the voltammetric solution.

Deposition of Bi-AuNPs composite materials on carbon paste electrode surface
Preparation of carbon paste has been reported previously [45]; we have adapted the similar method of preparation [29. 30]. Bi and Au films were simultaneously deposited over the CPE surface using electrodeposition techniques from 0.1 M H 2 SO 4 solution containing 0.1 mM Bi 3+ and 0.2 mM Au 3+ . Electrodeposition was carried out under mild stirring condition (at 500 RPM) at -0.3 V. Deposited films were washed with ultrapure water and then dried under air flow. SEM images were recorded using these films, which are shown in Figure 1A, 1B and 1C.
In Figure 1 A, Au, Bi and the carbon paste substrate are shown in different colours. It was observed that Au and Bi composite were embedded well within the carbon paste matrix. The sizes of one assembly of the gold nanoparticles were around 100 nm. Individual particle size of these assemblies was in the range of 10-20 nm. (cf. Figure S1 of supporting information). Bi and Au component of the composite film were individually mapped and show in Figure 1B and 1C respectively. Both Bi and Au were dispersed well over the electrode surface. Energy dispersive X-ray fluorescence (EDXRF) measurements were carried out on CPE/Bi-AuNPs substrate and the results are shown in Figure 2(A) and 2B. Presence of Bi and Au on the CPE/Bi-AuNPs substrate was observed from the measurements at the elemental ratio of Au/Bi as 2:1. electrochemical impedance measurements. Impedance measurements were recorded at a single frequency of 100 Hz with varying potentials as shown in Figure 3. Impedance performance of CPE/Bi-AuNPs electrode was compared with CPE modified with gold nanoparticles (CPE/AuNP) and also with the CPE modified with Bi film (CPE/Bi). It was observed that the capacitance of the Bi film modified CPE was lowest. Capacitance of the CPE/Bi-AuNPs composite was also less, which is useful for the application of the composite electrode in analytical measurements as modified electrode. The CPE/AuNP composite electrode has shown very high capacitance, much higher than the other two electrodes. The lower capacitance of the CPE/Bi-AuNPs composite electrode is beneficial to have lower background in stripping analysis.

Optimization of the analytical parameters of the measurements.
After electrochemical deposition of Bi-AuNPs over CPE surface, the electrode was washed with ultrapure water and dipped into the stripping solution containing the analytes. The stripping solution was 0.1 M acetate buffer; pH of the buffer solution was optimized from the stripping peak of Cu and Hg and is shown in Figure. 4. At pH lower than 4 the stripping peak of Cu was merged with that of the Bi peak (cf. Plot for pH 3 of Figure.4). At pH values higher than 4, the Cu stripping peak was well separated from the Bi stripping peak. However at pH more than 6 the stripping response of Cu was decreased. Thus a solution pH of 5 was optimized for the analysis for well separated peaks between Bi and Cu and also for good stripping current of Cu.
Stripping peak of Hg has also shown strong pH dependency. The highest peak current of Hg was observed in between pH 4 and 5. Stripping peak was reduced at a higher pH values than 5.5.
Considering the stripping response of both Cu and Hg, pH 5 was optimum for their simultaneous determination in a stripping solution of 0.1 M acetate buffer. Furthermore it was observed that the stripping peak of Cu was positively shifted compared to their positions on bare Bi film electrodes [29]. Cu might have formed the inter-metallic compound with AuNPs, which made the positive shifting of the stripping peak. Effect of deposition potential of the analytes on the modified electrodes was investigated by varying the deposition potentials of the measurements from -0.3 V to -1.2 V and the results are shown in Figure S2 of the supporting information.
Initially the peak currents for both Cu and Hg were increased from -0.3 V to -0.5 V, beyond -0.5 6 V, though the base line of the plots were modified the stripping peak of Cu and Hg remained almost unaffected. Therefore a deposition potential of -0.5 V was required for the deposition of Hg and Cu. A more negative deposition potential was avoided as the quality of deposits will be affected due to the hydrogen evolution process during deposition. In view of this the deposition potential of -0.5 V was optimized in the present case. In order to optimize the deposition time, it was varied from 50 s to 800s by keeping the deposition potential fixed at -0.5 V, the results were shown in Figure 5. Stripping peak current due to Cu 2+ was increased with increase in the deposition time up to 800s. Unlike Cu stripping peak, the stripping peak of Hg was not increased Bi produces strong stripping signal during anodic scanning process and it is difficult to separate the Cu stripping peak from that of Bi, due to which determination of Cu is difficult on Bi film electrode. Proportional decrease of Bi-stripping peak with the addition of Cu was successfully attempted as an alternate way to determine Cu [13]. It would always be good idea to observe the direct electrochemical stripping response of Cu with its proportional increment with

RSC Advances Accepted Manuscript
concentration. Interestingly, in the present case on simultaneous deposition of AuNPs and the Bifilms, the stripping response of Cu was shifted to positive potential and it was clearly observed.
Cu has the stronger affinity to form inter-metallic with Au which has shifted its stripping peak towards more positive directions. Due to that, its stripping peak was well separated from the Bistripping peak. Stripping response of Cu and Hg was compared using three different electrodes; CPE/Bi, CPE/AuNPs and CPE/Bi-AuNPs, the results are shown in Figure 7. The surface area from the cyclic voltammetry measurements were obtained as It was observed that on CPE/Bi electrode, no stripping response due to Cu and Hg was observed. Though the stripping peak heights in the case of CPE/Bi-AuNPs modified electrode was only marginally higher than that of

RSC Advances Accepted Manuscript
peak was also unaffected with the addition of Hg 2+ . This observation concluded that both Cu and Hg stripping responses remained unaffected due to the presence of other at high concentration.
Reproducibility of the stripping responses was evaluated by recording voltammetric scans repeatedly and the results are shown in Figure S6 of supporting information. Percent standard deviation for Cu 2+ was obtained as 3.6 % and for Hg 4.5% for 6 repetitive measurements

Interference effect
Interference effects due to some of the commonly occurring metal ions Fe(III), Cr(VI), As(III), Ag(I), Pb(II), Cd(II), Zn(II), U(VI) were investigated on the stripping peak of 20 µg L -1 Cu and 8 µg L -1 Hg. Except due to Ag(I) and As(III) no interference was observed from all the other metal ions up to 500 time higher concentration than Cu(II) and Hg (II). Ag(I) interfered with both the Cu and Hg peaks. Effect of interference due to Ag(I) on the Hg stripping peak was shown in Figure 9. It was observed that Hg stripping peak was reduced with the addition of Ag(I) in the test solution from 50 µg L -1 Ag(I) concentration. Observable Hg stripping peak was observed with the addition of Ag(I) up to 1000 µg L -1 . Hg stripping peak was recoverable with the increase in the gold content in the modified electrode. During preparation of the modified electrode a gold concentration of 100 times higher compared to the Ag(I) concentration is recommended to minimize the interference due to Ag(I) on the Hg stripping peak. Tolerance of Ag(I) on the stripping peak has increased due to the incorporation of Bi along with the AuNPs.
As only AuNPs modified electrodes were heavily interfered by Ag(I) for Hg 2+ determination.
As(III) has interference with the Cu stripping peak beyond 200 times higher concentration than Cu by broadening the Cu stripping peak.

Recover test and analysis of water and soil sample
Recovery test was carried out in tap water samples by spiking both 5 µg L -1 of Cu 2+ and Hg 2+ standards. Test results are shown in Table 1. Both the elements were recovered within 10% deviation from the added concentration. Two ground water samples were collected from two different regions of India, and analysed using the modified electrode, results are shown in Table   2. OSCOM (Odissa sands company) sample was collected from the industrial region, important for the mining activities. Other sample was collected from bore well from the Pujab state of India. Lowering of ground water is a concern in Punjab due to the excessive use of ground water for irrigation purposes. Both the samples were analyzed at undigested conditions using the presently developed CPE/Bi-AuNPs electrode. Two soil samples were collected; one from Jadugoda, a mining region of and the other from Vizac a coastal region. Digestion of the soil samples was carried out using Anton Paar Microwave Digester, Multiwave-3000. An aliquot of 0.5 g of the solid was taken with 5 ml of concentrated nitric acid and digested at 150° C for 20 min. After complete digestion, the solutions were diluted with ultrapure water and made up to the volume of 50 mL. Aliquot of this sample solution was taken in to the voltammetric cell for electrochemical measurements. The analysis results are shown in Table 2. Values were verified using the results from inductively coupled plasma optical emission spectrometry (ICPOES) technique for Cu and mercury analyser based on thermal decomposition atomic absorption technique for Hg. Results were comparable thus, with CPE/Bi-AuNPs electrode digestion was not required to analyse Cu and Hg in ground water samples. The Cu content in OSCOM ground water sample was higher due to the mining activity in the region other values were within the prescribed limit by World Health Organization (WHO) in potable water. Once fabricated, this sensor can be used for 15 scans. We have tested the stability of the sensor; it remains stable up to 7 days when stored inside water. Analytical performance of the presently developed electrode was compared with the methods reported previously in the literature for Cu and Hg and the results are reported in Table 3. As reflected from the table, present method stands well in terms of sensitivity and easy of application compared to the reported literature.

Conclusion
Bismuth gold composite material was electrodeposited on CPE surface. Microscopic examination revealed that gold nanoparticle clusters were well spread over the Bi-films.
Conductivity of the modified surface was improved due to the Bi-films. Well spread gold nanoparticles on Bi film have enhanced the sensitivity of mercury determination. Three sigma detection limits of Cu and Hg were 0.16 µg L -1 and 0.28 µg L -1 respectively. Direct determination of Cu and Hg was reported in ground water samples collected from two different regions. High affinity of Cu towards Au has positive shifted the Cu stripping peak; useful for the direct determination of Cu using Bi containing film electrode.            Table 2 Analysis results of Cu + and Hg in two ground water samples and two soil samples collected from different important regions of India, using CPE/Bi-AuNPs electrode.