Square-wave anodic stripping voltammetry of cadmium and lead in waters using a glassy carbon electrode modified with multi-walled carbon nanotubes and bismuth

Multi-walled carbon nanotubes (MWCNTs), attached with the help of Nafion film, were used to modify the surface of a glassy carbon electrode (GCE). Bismuth (which has been studied in recent years as replacement for mercury) was then coated on the MWCNT-modified GCE to enhance its sensitivity for metal detection. The modified electrode was used in the simultaneous analysis of cadmium and lead by square-wave anodic stripping voltammetry (SWASV). Studies of solution composition, instrument parameters, Nafion film composition and Bi-coating conditions resultied in the following optimum analytical conditions: supporting electrolyte 0.1M KNO3 in 0.01M HNO3, containing 0.33 mg/L Bi, deposition potential of -0.900V (vs Ag/AgCl), deposition time of 240 s and stripping potential scan rate of 75 mV/s. A comparison of the three types electrode surfaces (see table) revealed that the MWCNT/Nafion-Bi-GCE had the best calibration sensitivities and limits of detection (LOD, based on 3sd) for Cd and Pb.Working electrodeCdPbSensitivity,A/gL-1Linearity (R2)LOD, g/LSensitivity,A/gL-1Linearity (R2)LOD, g/LMWCNT/Nafion-GCE0.3530.87331.030.98220Bi-GCE0.3710.95610.6110.9532MWCNT/Nafion-Bi-GCE0.6860.9820.50.8900.9962Recovery tests of Cd and Pb using the MWCNT/Nafion-Bi-GCE electrode in the presence of other possible interfering metals (eg Cu) indicated major interfering effects. At ratios of 1:1 (analyte:Cu interferent), low recoveries for both Cd and Pb were obtained (ranging between 27-40% recovery); on the other hand at 1:10 (analyte:Cu interferent) Cd and Pb recoveries were depressed in the presence of Cu, but were extremely high (ranging between 200-350% recovery of analyte). A stripping peak associated with copper (occurring between 0-100 mV) was observed to overlap with the Bi stripping peak (occurring between -20 to 20 mV).The MWCNT/Nafion-Bi-GCE electrode was applied to the analyses of Cd and Pb in samples of a natural water (pH 6.94) and highly acidic (pH 2.88) mine pit water (both taken from Mount Morgan, Queensland, Australia). The samples were filtered and diluted (25% v/v and 1% v/v respectively) prior to analyses. The natural water sample showed Cd and Pb levels were below their respective SWASV LODs. The pit water (from a decomissioned copper and gold mine) showed Cd at 212 gL-1 while Pb was below LOD. These results compared well with those obtained by ICP-mass spectrometry: natural water contained < 0.02 gL-1 Cd while pit water had 205 gL-1 Cd; both samples had <0.5 gL-1 Pb. It is noted that the ICP-MS analyses showed expectedly that the pit water contained high levels of contaminants, notably: 58.4 mgL-1 Cu (vs less than 10 gL-1 Cu for the natural water sample), 947 mgL-1 Al, 193 mgL-1 Fe and 3449 mgL-1 S (while the corresponding levels in the natural water were <0.02-1.5 mgL-1). Despite this complex sample matrix, the MWCNT/Nafion-Bi-GCE electrode performed very well, unlike what had been expected based on the study of interferences. The sensor and the analytical procedure therefore proved to be simple, inexpensive yet reliable method for the determination of the trace metal analytes in water samples. Acknowledgments: The study was supported by the CQU Centre for Environmental Management. Water samples were provided by the Queensland Department of Mines & Energy.