Due to the high toxicity of heavy metals, it is crucial to detect ultra low levels of the metals, especially in drinking water. The common techniques include spectrometric techniques such as inductively coupled plasma- atomic emission spectroscopy, ICP-AES (Bettinelli et al. 2000; Rahmi et al. 2007; Tuzen et al. 2008) as well as anodic stripping voltammetry (Brainina et al. 2004). Even though ICP techniques have low detection limits (ranges from parts per billion, ppb to parts per trillion, ppt (Berezhetskyy et al. 2008), however, they are unsuitable for in-situ analysis, they are expensive, sophisticated and require skilled operators.For these reasons, the development of alternative techniques such as electrochemical biosensor techniques, offer alternative methods because they are sensitive, low cost and simple to operate (Wang et al. 2009b).
Recent developments have shown the use of electrochemical biosensors as indirect methods for detection of Cd2+, Cu2+, Cr3+, Zn2+, Ni2+ and Pb2+ using urease biosensor (Ilangovan et al. 2006; Tsai et al. 2003); Cd2+, Co2+, Zn2+, Ni2+and Pb2+ using alkaline phosphatase (Berezhetskyy et al. 2008); Cd2+, Cu2+, Zn2+ and Pb2+by glucose oxidase (Ghica and Brett 2008); Hg2+ using glucose oxidase invertase and mutarose (Mohammadi et al. 2005); Cu2+, Cd2+, Mn2+ and Fe3+ using acetylcholinesterase (Stoytcheva 2002); and Cu2+, Cd2+, Zn2+ and Pb2+by nitrate reductase (Wang et al. 2009b).
Horseradish peroxidase (HRP) biosensor has so far only been reported for detection of mercury (Han et al. 2001). This study sought to extend its application for detection of other metals such as lead, cadmium and copper. We have chosen cadmium due to its similarities with mercury with regards to toxicity as both metals, belong to the same group. In addition, we have also chosen copper and lead because of their common occurrence in environmental matrices (Pb from leaded petrol and Cu from wiring activities). Furthermore, copper is reported to show interaction with biological systems (Cecconi et al. 2002; Uriu-Adams and Keen 2005) and therefore interesting to see how it interacts with HRP enzyme.
The main aim of this present work is to investigate the inhibition of HRP enzyme by Cd, Pb and Cu, a phenomenon that can be employed for their indirect determination. Kinetic studies were done to determine the nature of enzyme inhibition (whether it is reversible or irreversible and if reversible whether it is competitive or noncompetitive). The apparent Michealis-Menten constant () as well as maximum current () values in the absence and the presence of metal inhibitor were investigated. The developed biosensor was applied for determination of the Cd, Pb and Cu in tap water and landfill leachate sample.
2. Methodologies, results and discussion
2.1. Experimental reagents
All the chemicals used in this work were of analytical grade unless otherwise stated. Horseradish peroxidase (E.C. 22.214.171.124, 169 Units mg-1 powder, Sigma) aniline (99%), hydrochloric acid (37%), N,N-dimethylformamide (DMF), disodium hydrogen phosphate (dehydrated) and sodium dihydrogen phosphate (dehydrated) were all obtained from Sigma-Aldrich (South Africa). Cadmium and copper stock solutions (1000 ppm) were obtained from KIMIX Chemicals & Lab Supplies; and lead stock solution (1000 ppm) was obtained from Saarchem-Holpro Analytic (PTY) Ltd. Working solutions of hydrogen peroxide were prepared from 30% v/v stock solution obtained from Merck Chemical (PTY) Ltd.. Phosphate buffer (PBS, 0.1 M, pH 7.0) was used as a supporting electrolyte as per Songa et al. 2009.
All electrochemical experiments were performed using BAS100W Electrochemical Analyzer (Bioanalytical Systems, West Lafayette, IN, USA). A 15 mL electrochemical cell consisting of Pt working electrode (A = 0.018 cm2), Pt wire auxiliary electrode and Ag/AgCl (saturated 3 M NaCl) reference electrode. Supporting electrolyte solutions was degassed with argon gas before measurements performed at room temperature (20-25 °C). The PANI film was characterized using both Perkin Elmer Spectra 100 FT-IR Spectrometer (attenuated total reflectance, ATR) and UV-Vis Perkin Elmer Spectra spectrophotometer (PANI in DMF solution in quartz cuvette). UV photolysis of the leachate water sample was carried out by UV digester 705 equipped with a 500 W Hg lamp from Metrohm (Herisau, Switzerland). Inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis of Cd, Cu, and Pb was performed using an Optima 5300 ICP-OES system (Perkin Elmer LLC, 761 Main Avenue, Norwalk, USA) equipped with AS 93plus autosampler.
4.Preparation of polyaniline (PANI) film modified electrode
Aniline was distilled before use. The platinum working electrode was first polished thoroughly with successive alumina slurries particle size of 1.0, 0.3 and 0.05 µm, and then rinsed with distilled water after each polishing step followed by 10 min sonication with ethanol and then water. The polyaniline (0.2 M aniline in 1.0 M HCl degassed in argon for 10 min) was electrochemically deposited on the platinum electrode (-200 mV to +1100 mV at 50 mVs-1 for 20 cycles). The PANI- modified electrode was rinsed with water before use. The modified electrode was used in subsequent biosensor fabrication.
The PANI film was reduced in PBS at a constant potential of –500 mV until the current signal reached a steady state value. This was followed by the oxidation at +0.65 V for 20 min in the presence of HRP solution (50 µl of 2.0 mg ml-1 in 1.0 ml fresh PBS). During the oxidation process, the heme protein of HRP became electrostatically attached onto the PANI film (Songa et.al. 2009; Mathebe et al. 2004).The biosensor was stored in PBS at 4 °C when not in use.
6. HRP Biosensor response to hydrogen peroxide
The response of the biosensor (Pt//PANI/HRP) to H2O2 was studied at pH 7.0 in PBS. Cyclic voltammetric (CV), differential pulse voltammetry (DPV) and amperometric responses of the biosensor were recorded by adding small aliquots of 0.01-0.05 M H2O2.
7. Determination of Cd2+, Cu2+ and Pb2+ in model solutions
Amperometric measurements of HRP inhibition by cadmium, copper and lead were carried out in a cell containing 2.0 ml of 0.1 M PBS (pH 7.02) and constant concentration H2O2 (0.5 mM) with continuous stirring. The experiments were carried out at -0.20 V versus Ag/AgCl (3 M NaCl) and allowing the steady-state current to be attained. An appropriate volume (µl) of the inhibitor stock solution (10 ppm of each Cd2+, Cu2+ and Pb2+) was then added using a micropipette. After each experiment the enzyme electrode activity was regenerated by rinsing the electrode with distilled water.
8. Analysis of heavy metals in tap water and landfill leachate samples
Water samples were collected as follows: Tap water was collected from the Laboratory Tap at University of Western Cape, Bellville, Cape Town. Landfill leachate sample was collected from the Marrianhill landfill (Ethekwini municipal solid waste deposit). The leachate water sample was collected in polyethylene container and stored in the fridge at 4 °C.
Determination of heavy metals in tap water was achieved using standard addition method. The pH of the tap water samples was first adjusted from 8.90 to 7.04 before the analysis was carried out. The tap water sample (10 ml) was spiked with 0.1 ppm of each metal solution (Cd2+, Cu2+ and Pb2+) followed by amperometric analysis. For ICP-OES, the tap water sample was analysed without the addition of metal standards.
Leachate water sample is rich with organics; therefore prior electrochemical analysis, the organics were removed by passing the water sample through C-18 SPE column. The cartridges were first conditionedwith 5 mL methanol followed by 5 mL water. The C-18 column retained the organics and the water sample containing inorganics was collected. The collected leachate water sample was spiked with 0.1 ppm of each metal solution (Cd2+, Cu2+ and Pb2+) followed by Pt/PANI/HRP biosensor analysis.
For ICP-OES analysis, the leachate samples were filtered with 0.45 µm pore size filter before they were subjected to UV digester. This procedure was done in order to destroy all dissolved organic matter in the landfill leachate sample. A UV digester 705 equipped with a 500 W Hg lamp from Metrohm was used. The quartz vessels were arranged concentrically around the Hg lamp with a distance of 2.5 cm. Ten mL of leachate samples were placed in quartz vessels and 100 μL H2O2 was added to each sample. The solution was irradiated with UV light for about 2 hour. The leachate water sample was then analyzed by ICP-OES.
9. Results and discussion
9.1. Electrosynthesis of PANI film
Multiscan voltammetry of Pt/PANI electrode was performed (result not shown). The redox peak currents increased with increasing scan rate while the peak potentials showed slight increase in positive potential. These observations shows that the polymer is electroactive and the peak currents are diffusion controlled (Mathebe et al. 2004). In order to calculate surface concentration of the PANI film, Г*PANI, Brown-Anson equation (1) (Bard &Faulkner 2000) was used.
where n is the number of electrons (n = 2) transferred, F is the Faraday constant (96584 C mol−1), Γ* PANI is the surface concentration of the PANI film (mol cm−2), A is the surface area of the electrode (0.0177 cm2), υ is the scan rate (V s−1), R is the gas constant (8.314 J mol K−1), and T is the absolute temperature of the system (298 K). A graph of peak current versus scan rate was obtained and the slope of the curve was used to calculate the surface concentration of the PANI film. The surface concentration was found to be 7.8×10-7 mol cm-2. The surface concentration obtained in our study was comparable to that reported by Mathebe et al. 2004 (1.85×10-7 mol cm-2).
where is the peak current (A), n is the number of electrons appearing in half-reaction for the redox couple, A is the area of the electrode (cm2), D is the diffusion coefficient (cm2/s), C is the concentration (mol/cm3) and v is scan rate (V/s). Equation 2 was used to plot peak current versus the square root of the scan rate and the slope of the linear regression was used to estimate the diffusion coefficient of the electrons within the polymer () as 4.07×10-8cm2 s-1.
10. Spectroscopic characterization of polyaniline
The absorption spectrum of PANI (dissolved in DMF) shows two characteristic absorption peaks at 340 nm and 660 nm. The first absorption peak was assigned to π–π * transition of the benzenoid rings and the second was attributed to the transition of benzenoid rings into quinoid rings (Laska & Widlarz 2005; Kanet al. 2006). The results (UV-Vis characterization of PANI) obtained in this study are in close agreement with the literature values (Laska &Widlarz 2005; Kan et al. 2006; Singh et al. 2008; Mazeikiene et al. 2007; Kang, Neoh & Tan 1998).
The FTIR absorption band at 3325 cm−1 was assigned to N-H stretching of the amine group of polyaniline (spectrum not shown). The peaks at 1596 and 1493 cm−1 which are characteristics of, polyaniline, were most likely due to the C=C stretching of quinoid and benzenoid groups, respectively (Lakshmi et al. 2009; Kim et al. 2001).
11. Cyclic Voltammetric (CV) and Differential Pulse Voltammetric (DPV) response of Pt/PANI/HRP biosensor to hydrogen peroxide
The electrochemical behaviour of Pt/PANI/HRP electrode in the absence and presence of H2O2 in PBS (0.1 M, pH 7.02) was studied using CV and DPV. Figure 1 shows CV (A) and DPV(B) of Pt/PANI/HRP electrode in different concentrations of H2O2 (0-6.9 mM) at scan rate of 10 mV s-1 and 20 mV s-1 for CV and DPV, respectively. The value used in this study for the optimum concentration of HRP was as per Ndangili et al. (2009) and Songa et al. (2009). The effect of pH on HRP electrode response was investigated by CV in the pH ranges from 5.5 to 8.5 in the presence of 1.0 mM H2O2. The HRP electrode response current achieved a maximum value at pH 7.0. Therefore, in order to obtain maximum sensitivity, 0.1 M PBS solution of pH 7.0 was used throughout this study.
As expected, in the absence of H2O2, no significant current was observed. However increasing the amount of H2O2 showed increased cathodic peak current intensity due to the reduction of H2O2. In order to confirm whether the change in current intensity observed was due to the enzymatic catalytic reduction of H2O2, control experiments in the absence of HRP were carried out. At both the bare electrode (Pt) and polymer modified surface (Pt//PANI) no H2O2 reduction current was observed at -200 mV. This is because, the reduction reaction of H2O2 at both electrodes in the absence of HRP, is very slow and usually occurs at higher potentials. The difference in the observations made for the control experiments (Pt and Pt/PANI) compared to that for Pt/PANI/HRP confirms that the increase in the cathodic current was due to the direct electron transfer between the HRP molecules and the electrode (Wang & Wang 2004).Moreover, PANI provides a suitable platform for the immobilization of HRP on the platinum electrode surface and it also mediates in electron transfer between HRP and the electrode (Gerard et al. 2002). Thus the reduction peak is an indication of the electrocatalytic activity of the enzyme on H2O2 (Sun et al. 2004).
Figure 2 shows the possible mechanism of electric transduction between the platinum electrode, PANI and HRP enzyme active site combined with electrocatalytic reduction process of H2O2 by HRP. It can be seen from figure 2 that H2O2 is reduced by HRP to form water and in turn HRP gets oxidized to form Compound I. The latter is converted to HRP through the formation of intermediate (Compound II) via a two-electron reduction step (Iwuoha et al. 1997). The reduction of Compound I is due to the direct electron transfer that takes place between the PANI modified electrode and the enzyme (Liu & Ju 2002).
The relatively low potential value for H2O2 reduction (-200 mV) in the presence of HRP, ensures minimal risk of interfering reactions of other electroactive species in solution as well as low background current and noise levels (Tong et al. 2007; Wang &Wang 2004).
12. Amperometric responses of Pt/PAN/HRP to H2O2
Amperometric responses of Pt/PANI/HRP biosensor were investigated by consecutively increasing the concentration of H2O2 at a working potential of -200 mV. Figure 3 presents a typical steady state current-time plots obtained with the fabricated biosensor upon successive additions of 10 μL of 0.001 M H2O2 into 2.0 mL PBS with the calibration plot as an inset. It was observed that, upon the addition of H2O2 into the PBS, the reduction current rises sharply to reach the steady state value. In addition, the biosensor attained 95% steady state current within 5 seconds after each addition of 10 µL 0.010 M H2O2. This observation
implied a fast response of the fabricated biosensor.The response currents at the biosensor were linear to H2O2 in the range from 0.05 to 3.17 mM with a correlation coefficient of 0.9991 (n = 18); sensitivity of 1.75 µA mM-1; and a detection limit of 36.8 nM (0.0368 µM) (estimated at a signal-to-noise ratio of 3).
13. Kinetics of Pt/PANI/HRP electrode
A plot (not included) of the reciprocal of current versus the reciprocal of H2O2 concentration (Lineweaver–Burk plot) showeda linear relationship implying a kinetic behavior (Liu and Ju 2002; Wang, et al. 2009a). Using equation 3, Michaelis–Menten constant of the enzyme could be calculated as per previous studies (Liu, and Ju 2002; Mathebe, Morrin, and Iwuoha2004):
where is the observed response current, is the maximum steady state current that can be attained for the system (Pt//PAN/HRP biosensor), is Michaelis–Menten constantand [H2O2] is the concentration of H2O2 in the bulk solution. The slope (1.18 µA-1mM) and the y-intercept (0.958 µA-1) of the Lineweaver–Burk plot are equal to and, respectively. The values of andwere estimated as 1.04 µA and 1.23 mM, respectively. The value obtained in our study was relatively better (lower) than previous studies (Wang et al. 2009a; Liu & Ju 2002; Ndangili et al. 2009. Since is inversely proportional to the affinity of the enzyme for the substrate (Wang et al. 2009; Ansari et al. 2009), based on the value 1.23 mM it can be concluded that Pt/PANI/HRP electrode exhibited high affinity towards H2O2. The differences in the values obtained by different workers indicate that the affinity between H2O2 and HRP is plainly dependent on the immobilizing materials and procedure used (Wang et al. 2009).
14. Reproducibility, repeatability and stability of Pt/PAN/HRP biosensor
To evaluate the reproducibility of the HRP based biosensor, six biosensors were prepared under similar conditions separately. The amperometric responses of each biosensor to successive additions of H2O2 (0.05 to 0.30 mM) were recorded and gave a relative standard deviation (RSD) of 3.8% (n = 6). The precision (repeatability) of a selected biosensor was investigated by recording the response current using 0.05 mM H2O2 for replicate experiments (n = 10). The biosensor showed a high precision with a %RSD of 2.1%. The stability of the Pt/PANI/HRP electrode (stored in 0.1 M PBS at 4 °C in between measurements) was tested every two days by amperometric measurements in the presence of 0.05 mM H2O2. After two weeks, the biosensor retained > 95% of its initial response, indicating good stability. The latter implies that the HRP molecules were firmly immobilized in the PANI films which provided a biocompatible microenvironment.
15. Detection of heavy metals (Cd2+, Cu2+ and Pb2+) in model solutions
Heavy metals such as Cd2+, Co2+, Cu2+, Fe3+, Mn2+, Ni2+ and Pb2+ as well as some organic and inorganic compounds such as sodium azide, cyanide, L–cystine, dichromate, ethylenethiourea, hydroxylamine, sulfide, vanadate, p–aminobenzoic acid are well-known inhibitors of horseradish peroxidase (Zollner 1993). Figure 4 (I) shows amperometric responses of Pt/PANI/HRP to successive additions of 0.05 mM H2O2 (denoted by r) followed by addition of Cd2+ (denoted by i). It can be observed that the current for the electrocatalytic reduction of H2O2 decreases upon the addition of Cd2+. This observation clearly showed that the catalytic activity of HRP at Pt/PANI electrode is inhibited by Cd2+ by binding to the enzymes active sites. The inhibition of HRP activity was analyzed by monitoring the current decrease at -0.20 V using the following procedure: H2O2 was added to PBS (electrolyte solution) and the current response was recorded. After the stabilization of the steady-state current response, Cd2+ was added, leading to an immediate decrease in the biosensor response.
The other two heavy metal ions (Pb2+ and Cu2+) showed the same behaviour. Calibration plots for the determination of the metals in the range 0.05 to 60.8 ppb were obtained. The linear ranges were established as 4.76-55.3 ppb for Cd2+ and Pb2+ while that for Cu2+ was 2.38-52.8 ppb. Limits of detection (LOD=) and limits of quantification (LOQ=)
were calculated and their values are presented in Table 1. Where is standard deviation of the blank signal (n = 8) obtained in PBS (= 2.4×10-4 µA) and m is the slope of the calibration curve. The low LOD and LOQ values confirmed good sensitivity of the proposed biosensor method for the determination of heavy metals. Table 6 shows a summary of analytical response and linear regression characteristics of calibration curves for heavy metals. The inhibition of heavy metals to HRP enzyme is reversible, and the biosensor can be reused (beyond 10 runs) after rinsing with double distilled water.
16. Inhibition studies
The values of the steady state current in the absence () and in the presence () of an inhibitor were determined from the recorded amperograms in Figure 4.The inhibition percentage () was calculated from the following expression (Nwosu et al. 1992):
The values obtained were used to compare the inhibitory effects of the different metal ions on the activity of the immobilized HRP. Inhibition plots showing the effects of heavy metals on the activity of the immobilized HRP are presented in Figure 5. The order of inhibition was found to increase from Pb2+ (32.8 %), Cu2+ (43.4 %) to Cd2+ (51.1 %) in the presence of 0.95 mM H2O2. It should be noted that the higher the value the higher the degree of enzyme inhibition by the heavy metal.
There are two approaches that can be used for inhibition studies; incubation method and direct method (without incubation) and in this study the latter was employed. It was observed that when using the direct method, it is difficult to attain the concentration that causes 50% inhibition () (Songa et al. 2009); this worked reasonably well for Cd2+ (51.1%) with the value for calculated as 3.13 ppm but was not successful for Pb2+ and Cu2+. Therefore unlike Cd2+, in order to attain the 50% inhibition (value) for Pb2+ and Cu2+, either a higher concentration of the metal can be used or the incubation method can be employed.
After successfully obtaining the value for Cd2+ using direct approach, the results from the latter approach were used to investigate the inhibition kinetics and the type of inhibition (competitive or non-competitive).
17. Investigation of inhibition kinetics and mechanism
The amperometric response of the biosensor to various H2O2 concentrations (in 0.1 M PBS, pH 7.02) in the absence (0 ppm Cd2+) and presence (3.13 ppm Cd2+, ) of an inhibitor was recorded at -200 mV (Figure 6A). The response of the biosensor to various H2O2 concentrations in the presence of an inhibitor was done as follows; HRP electrode was first incubated in PBS containing 3.13 ppm Cd2+ for 20 min followed by the successive addition of different concentrations of H2O2. A fast response of the biosensor to the additions of different concentrations of H2O2 was observed in the absence of Cd2+ whereas the presence of 3.13 ppm Cd2+ gave a slow response. Studies suggest that during incubation, Cd2+ slowly binds to the enzyme and leads to gradual conformational changes (Keyhani et al. 2003, Tayefi-Nasrabadi et al. 2006 ). Therefore, it can be suggested that the difference in background current could be due to the conformational changes of HRP. Figure 6B shows the Lineweaver–Burk plots for enzymatic reactions. The slope and the y-intercept values of the linear plots were used to calculate apparent Michealis-Menten constantsand maximum current () in the absence and the presence of Cd2+ (Table 2).
in the absence and the presence of Cd2+ (3.125 ppm). The values for Cd2+ in the absence (1.16 mM) and the presence (1.13 mM) of the inhibitor were not significantly different (based on statistical t-test, n =3, p = 0.05), whereas the values decreased. The decrease in indicates that the type of inhibition in this study is reversible and also non-competitive inhibition and the inhibition constant was found to be 27.8 µM (Amine et al. 2006; Stoytcheva 2002). Non-competitive inhibition occurs when the inhibitor binds to both the enzyme and enzyme-substrate complex. The possible mechanism of non-competitive inhibition can be seen in Scheme 1 (Amine et al. 2006); where E represents HRP enzyme, S represents H2O2, ES represents Compound I containing an oxyferryl centre with the iron in the ferryl state (), and a porphyrin π cation radical; EIS presents HRP-heavy metal-H2O2 complex; I represents heavy metal cation and P represents H2O.
18. Selectivity of Pt/PAN/HRP biosensor
In order to show the selectivity of Pt/PANI/HRP biosensor, the response of possible interferences in tap water (drinking water) components on the determination of heavy metals were investigated. A range of cations were measured to examine if they interfered with the determination of heavy metal cations. The effect of possible interferences in drinking water such as Mg2+, Zn2+, Ca2+, K+, Fe3+ and Na+ was investigated by Pt/PANI/HRP biosensor under the same working conditions. Table 3 presents the possible interferences tested with the biosensor. It can be observed that most cations had minimal effect (<5%) on the determination of heavy metals apart from Fe3+ which also inhibit HRP activity. However, at the same concentration (1.0 ppm), the degree of inhibition by interferentis still less than that of the analyte cations
Apart from the degree of inhibition of the interfering species, the selectivity of the biosensor was also evaluated by two methods; mixed and separate solutions (Stefan et al. 2001; Macca & Wang 1995) with respect to Cd2+, Cu2+ and Pb2+. For the mixed solution method, the concentration of the interfering cation was 1.0 ppm and the concentration of the analyte was 0.5 ppm. In the case of separate solution method, the concentration of the analyte and that of interfering cation were equal (1.0 ppm). Amperometric selectivity coefficients and response ratio were calculated using equations (5) and (6) for mixed solutions and separate solutions (Macca & Wang 1995), respectively.
where, and are current values recorded for mixed solution, interfering cation and analyte of interest, respectively; and are concentrations of the interfering cation and the analyte, respectively. The and values in Table 4 suggest that Fe3+ is a strong interferent while the other cations species are relatively not interfering with the determination of cadmium, copper and lead. Therefore, the results suggest that Pt/PANI/HRP electrode can be used for the determination of cadmium, copper and lead in the presence of other cations except Fe3+.
Table 4. Interference studies showing the effects of cations on the response of Cd2+, Pb2+ and Cu2+ using mixed (amperometric selectivity coefficients,) and separate (response ratio,) solution methods.
19. Application of Pt/PANI/HRP for analysis heavy metals in tap water and landfill leachate samples
The performance of the biosensor (Pt/PANI/HRP) was tested using both tap water and landfill leachate water samples. The latter were collected from Sensor Lab in the University of Western Cape (Cape Town, South Africa) and Marrianhill landfill in Durban (South Africa). The quantification of Cd2+, Pb2+ and Cu2+ in water samples was achieved by employing the standard addition method. For tap water sample, the sample preparation was not required. This is because the possible interferences that are normally present drinking water samples did not have much effect on the catalytic activity of immobilized HRP. The procedure involved initially measuring the response of the biosensor after subsequent additions of 0.05 mM H2O2 followed by metal-spiked tap water (0.1 ppm metal ion solution), into the PBS solution containing H2O2 (0.5 mM). On addition of the water sample, current intensity was observed to decrease and the magnitude of the decrease was proportional to the amount of metal spiked. The decrease in the response current was most likely due to the inhibition of the enzyme by the metal ions. A possible dilution effect due to increased volume of added tap water was checked by setting a control experiment where deionized water was used instead of metal spiked tap water. The inhibition effect was not observed with the deionized water. In order to double-check the effect of metal inhibition, H2O2 was added into the mixture and resulted in an increase in current intensity. Similar observations have been made by Han et al. (2001); Stoytcheva (2002) and Ghica & Brett (2008).
The concentrations of Cd2+, Pb2+ and Cu2+ (0.4507 ppb Cd2+, 0.2201 ppb Pb2+, 41.77 ppb Cu2+) in the tap water sample (analysed in triplicate for each metal) were calculated from the calibration curves and the results are presented in Table 5.
The values obtained were compared against the allowed MCLs by USEPA in drinking water. The MCLs are given as 5 ppb, 15 ppb and 1300 ppb for cadmium, lead and copper, respectively. The World Health Organization (WHO 2004) on the other hand has given the guideline values for cadmium, lead and copper in drinking water as 3.0 ppb, 10 ppb and 2000 ppb, respectively. It can be seen that the concentrations of Cd2+, Pb2+ and Cu2+ obtained in this study are much lower than the USEPA and WHO guideline values, implying that the drinking water was not contaminated and therefore safe for human consumption.
In the case of landfill leachate water sample, sample preparation was performed. Solid phase extraction (amperometric biosensor) and UV digestion (ICP-OES) were used to minimize organic interferences. After removal of organics the same procedure that was applied for tap water sample was followed. Summary of the concentrations found by the fabricated biosensor are presented in Table 5.
Results of heavy metals obtained by amperometric biosensor and ICP-OES were compared statistically by Student t-test (two-tailed). At 95% confidence interval, the results by the two analytical techniques for the determination of Cu2+ were not significantly different (p = 0.035 < 0.05). However, the two methods differed significantly for the determination of cadmium, at 95% confidence interval (P = 0.116 > 0.05). The reason for this discrepancy could be the difference in limits of detection (LOD) capabilities by the two techniques. The detection limit for cadmium was 2.70 ppb with the ICP-OES and 0.091 ppb (91 ppt) with the biosensor. Thus the detection limit for Cd2+ obtained with the biosensor is lower by a magnitude of 30 to that by ICP-OES. It should be noted that the concentration of Cd2+ in the tap water sample was less than the LOD of ICP-OES (LOD = 2.70 ppb cited from ICP-OES operation manual). This explains the poor precision when the tap water sample was analyzed by ICP-OES in triplicate (3 repeated runs gave 0.41, 0.25 and 0.35 ppb). Lead on the other hand, was not detected probably because its level was lower than 90 ppb (conc. of Pb2+ was 0.22 ppb by the biosensor). It can be concluded that the Pt/PANI/HRP biosensor technique is suitable for determination of ultra trace levels of heavy metals in drinking water.
The amount of copper in landfill leachate samples (Table 5) showed good correlation between the results obtained with the biosensor and the standard technique. Applying the Student t test, it was possible to verify that the averages obtained by the both methods are not significantly different at a confidence level of 95% (P = 0.009 < 0.05). With both methods, cadmium and lead were not detected.
An inhibition amperometric biosensor for the determination of selected heavy metals was fabricated on the base of the inhibition to horseradish peroxidase, which was immobilized on platinum-polyaniline electrode. Inhibition of HRP activity by heavy metals followed noncompetitive reversible mechanism. The HRP biosensor exhibited fast response, high sensitivity towards the determination of heavy metals (LOD of 0.091, 0.033 and 0.10 ppb for Cd2+, Pb2+ and Cu2+, respectively). The fabricated biosensor was applied for the determination of heavy metals in real samples (landfill leachate and tap water). The evaluation of the amperometric biosensor measurements against the standard technique (ICP-OES technique) verified the suitability of biosensor for rapid analysis of heavy metals. Moreover, the amperometric biosensor requires minimal sample preparation as compared to the tiresome sample pretreatment procedures required prior to metal determination by the conventional ICP-OES method.
The authors are grateful for NRF funding through grant-holder Dr Anna Soares. They also acknowledges the Sensor Lab at University of Western Cape where part of the work for this study was carried outas well as NRF sponsorships through grant-holder Prof E I Iwuoha