An automated voltammetric system using solid electrodes was investigated for continuous and on-line remote monitoring of heavy metals in natural waters. Both a polluted river impacted by a copper mine, and costal seawater were monitored automatically every 30 minutes for six months. The huge difference in these two locations demonstrates the wide application range for this system. Results have been automatically monitored every thirty minutes for six months, and can be correlated against a range of parameters, e.g., season, flow of water, temperature, pH etc. The concentration ranges analyzed have been down in the range from 0.2 mg/L in seawater and up to 3 mg/L in polluted river water. Measurements have been performed every 30 minutes in the monitored real samples, with only minor maintenance of the system. Manual sampling with measurements by ICP-MS has been performed frequently for comparison and quality insurance, and the results show good agreement (r 2 avg ¼ 0.98).

Electrochemical properties of silver electrodes with 2, 4, 6, 10 and 15% bismuth have systematically been investigated with cyclic voltammetry (CV), differential pulse voltammetry (DPV), and electrochemical impedance spectroscopy (EIS). Increased overpotential towards hydrogen evolution reaction (HER) was found as a result of increasing amount of added bismuth. This was also demonstrated in acid solution where zinc was successfully detected on mixed electrodes, but failed on pure silver electrodes. Formation and decomposition of oxide products formed on the different electrode surfaces were studied by cyclic voltammetry, and in addition to known species found on sliver, also peaks attributed to bismuth were achieved and examined. Zinc, cadmium, and lead were measured in the low mg/L range on the mixed electrodes, and good linearity (r 2 ¼ 0.998) was found for 2 to 10 mg/L. Lead was measured down to 0.1 mg/L. Further it was found in DPASV that the zinc peak significantly shifted towards a more negative value with increasing amount of bismuth in the silver electrodes. The same was also observed for cadmium and lead, but in a less extent. Finally a silver electrode containing 15% bismuth was used for continuous analyses in a polluted river for 6 weeks.

Some recent developments and results in the field of automatic monitoring of electrolabile concentration of zinc and iron in the low mg/L range in river water, drainage water, and waste water by use of solid dental amalgam electrode (DAM) as a working electrode are reviewed for three different geographical sites representing the mentioned matrixes. At all sites, voltammetric measurements were carried out continuously every 30 or 60 minutes for periods up to 4 months, and compared with total amounts of the metals found by ICP-MS on manually collected samples. In total, the observed concentration ranges analyzed was in the ranges of sub-mg/L to approximately 30 mg/L for zinc, and from approximately 1 mg/L to 150 mg/L. for iron. Results shows good calibration curves for the metals in the different matrixes (r 2 avg ¼ 0.99) with standard deviation within 5%. The voltammetric system showed good stability and gave reliable results which were in a reasonable agreement with ICP-MS measurements for all analyses when comparing the concentration trends. The frequency of maintenance varied from once a week in waste water samples to once a month in river water.

Microelectrodes of silver-copper alloys have been evaluated for use in voltammetric analyses. Increased overpotential towards the hydrogen overvoltage reaction (HER) was found as a function of increased copper content in the silver. A study of oxidizing products by cyclic voltammetry (CV) in NaOH solution showed ten anodic and eight cathodic peaks which are described in the present paper. The behaviour of these alloy electrodes is somewhere between pure silver and pure copper electrodes. Differential pulse anodic stripping voltammetry (DPASV) was used to measure zinc, cadmium and lead in ultrapure water only (18 MOmegacm), and good linearity was found for all metals (r (2)=0.998) in the range of 0.5 to 5 ppb with a 600- to 1,200-s plating time. It was additionally found that cadmium and lead were better separated on the alloy electrodes compared to pure silver electrodes. Measurements of nickel were carried out on alloy electrodes by use of adsorptive differential pulse cathodic stripping voltammetry (Ad-DPCSV), and good linearity (r (2)=1.000) was found in the range from 0.5 to 5 ppb with an adsorption time of 120 s. The alloy electrodes were also found to be sensitive to nitrate, and good linearity (r (2)=0.997) was found in the range from 1 mg L(-1) to 100 mg L(-1) using differential pulse voltammetry (DPV) scanning from -450 mV to -1,500 mV. Addition of nitrate in ultrapure water afforded two different peaks related to the successive reductions of nitrate and nitrite. In ammonium buffer solution (pH 8.6) only one peak resulting from reduction of nitrate was observed. Furthermore, the use of alloy electrodes containing 17% Cu was tested in real samples, by installing it in a voltammetric system for monitoring of zinc and lead in a polluted river, the river Deûle, near the town of Douai in northern France. Results were found to be in agreement with parallel measurements carried out by ICP-MS.

Evaluation of different solid electrode systems for detection of zinc, lead, cobalt, and nickel in process water from metallurgical nickel industry with use of differential pulse stripping voltammetry has been performed. Zinc was detected by differential pulse anodic stripping voltammetry (DPASV) on a dental amalgam electrode as intermetallic Ni-Zn compound after dilution in ammonium buffer solution. The intermetallic compound was observed at -375 mV, and a linear response was found in the range 0.2-1.2 mg L(-1) (r(2)=0.98) for 60 s deposition time. Simultaneous detection of nickel and cobalt in the low microg L(-1) range was successfully performed by use of adsorptive cathodic stripping voltammetry (AdCSV) of dimethylglyoxime complexes on a silver-bismuth alloy electrode, and a good correlation was found with corresponding AAS results (r(2)=0.999 for nickel and 0.965 for cobalt). Analyses of lead in the microg L(-1) range in nickel-plating solution were performed with good sensitivity and stability by DPASV, using a working electrode of silver together with a glassy carbon counter electrode in samples diluted 1:3 with distilled water and acidified with H(2)SO(4) to pH 2. A new commercial automatic at-line system was tested, and the results were found to be in agreement with an older mercury drop system. The stability of the solid electrode systems was found to be from one to several days without any maintenance needed.