The solubility of gold in metallurgical slags

The solubility of gold in metallurgical slags has always been assumed to be very low, based on the experimental findings of a single study whose data set was subject to a high degree of scatter. It has also been generally assumed that gold dissolves into slags as uncharged atoms, apparently without any experimental foundation. In this work, the availability of neutron activation analysis, which has a lower detection limit for gold of only 0?005 ppm, was the impetus for a re-examination of gold solubility in slags. It was found that the solubility of gold in iron silicate slag is about a half of that previously reported, and is the same as that in calcium ferrite slag. In both slags, gold solubility is a function of oxygen partial pressure and gold appears to be present as Au+ ions. The solubility of gold in molten PbO slag was found to be very much higher than in iron silicate and calcium ferrite slags and it is proposed that gold dissolves into such highly basic slags as a gold/oxide anion complex. It was shown that these results are consistent with gold being present in this complex in the +1 oxidation state, and that each gold ion appears to be associated with two oxide ions from the slag. The practical implications of these findings are briefly discussed.