A prevalent technique for the recovery of silver from wastes generated in silver-tungsten electrical contact fabrication is electrowinning. Such scrap is placed in a permeable plastic basket which constitutes the anode compartment and this basket is suspended in a larger cell in which the cathode is mounted.
The solution to this problem lies in the selective removal of the W from the AgNO3 electrolyte either by liquid-liquid or liquid-solid extraction. Although W extraction in the liquid-liquid mode with a quaternary amine or other reagents as extractants appears feasible, particularly in view of the low pH (-1.5) of the system, it was eliminated from consideration because it is essentially a batch type process and it is a more elaborate industrial operation than extraction by ion exchange resins.
A useful material is a gel-type, weakly basic anion exchange resin with a polyamine functionality and in the free base form, the commercial Amberlite IRA-68 manufactured by Rohm and Haas Company. Initial slurry tests with W-contaminated AgNO3 electrolyte showed that the resin extracts W quantitatively but it also coextracts substantial amounts of Ag.
The flow diagram of the process for the separation of W from AgNO3 electrolyte
- Resin conditioning
- Column preparation
- W extraction
- Column elution and regeneration
The rapid and accurate determination of the Ag and W content of the feed and effluent streams is essential to the assessment of the extraction and separation process. A dc plasma atomic emission spectrometer (Spectraspan IV, manufactured by Spectrametries Inc. , Andover, MA) was used for this purpose. The instrument is capable of measuring metal concentrations of less than 1 ppm. Two different feed streams were used in this study. On contained 91.8 g/l of Ag and 210 ppm W while the other had 49.3 g/l of Ag, 10.0 g/l of Cu and 590 ppm of W. The emission wavelength used for monitoring the W concentration was 400.88 nm. An examination of a possible interference effect of the large excess of Ag on the W emission indicated a maximum signal enhancement of about 4% which is negligible in the context of this work.
The measured concentration of W in the column effluent stream is given from which is calculated the difference in percent between the W concentration of the input feed and that of the effluent stream. This value is, of course, the measure of the W sorbed by the column. In a similar vein, the difference between the measured Ag concentration of the two streams is given; a value of 0.0 indicates that there was no measurable difference between the two streams within the experimental error of the determination.
At the same time the extraction of Ag is minimal, if it takes place at all. The large difference in Ag concentration in the first fraction is not due to extraction but is caused by the dilution of the feed stream with water present in the column at the start of the process.
The initial fraction of the water wash contains about half the Ag and W concentration of the feed stream but this represents a dilution and replacement of the stock present in the column at the start of the elution step rather than a recovery of the sorbed metal. The concentration of these metals decreases substantially with continued washing. T
The resin column, on which the aforementioned data were obtained, was evaluated in four complete extraction-elution cycles and the results were comparable.