Introduction
The analysis of platinum group metals’ (PGMs) ores is complicated and time-consuming due to their low grade, inhomogeneity and complex mineralogy. Platinum group metals are usually produced as by-products of copper-nickel ore refining. The classical lead fire assay is still an efficient method to extract and concentrate gold and platinum group metals like platinum, palladium and rhodium, together called the 4E elements as it is relatively inexpensive, accurate and have low detection limits. Even though the nickel sulfide fire assay is the only internationally recognized method for determination of all the six elements of platinum group metals, it is quite expensive, time consuming and hazardous compared to lead fire assay. Moreover, it is found that the lead fire assay gives a good recovery for platinum, palladium and gold compared to the nickel sulfide fire assay. Traditional lead fire assay techniques for the PGMs use a pyrometallurgical fusion using litharge, various fluxes and silver collector at 1000-1100℃ to obtain the lead button followed by an oxidative fusion of the button called cupellation at 950℃ to isolate a silver containing PGM bead called a silver prill. The silver prill is further heated at 1300℃ to remove the silver in order to isolate the PGM and gold bead which is subsequently weighed as 4E (the sum of gold, platinum, palladium and rhodium). Even though gravimetric 4E determination is simple and has fast turn around time, it leads to some losses of the elements especially rhodium and therefore it is not suitable for highly accurate analysis. For accurate determination of gold, platinum and palladium in rock samples, lead fire assay is a superior and efficient method compared to nickel sulfide fire assay. In the present study we took one pyroxenite and one chromite ore based CRMs as quality control samples and a commercial sample of high grade chromite ore in order to determine platinum, palladium and rhodium using lead fire assay followed by ICP-OES finish.
Results and Discussion
All the three samples were analyzed in triplicate using lead fire assay. Normal silicate rocks like pyroxenite was fused with a flux mixture whose composition was different compared to refractory ores like chromite ores. Chromite ores are highly refractory in nature and hence are very difficult to fuse compared to silicate rocks. We have optimised a flux mixture (consisting of litharge, flour, anhydrous sodium carbonate, borax and silica) which is suitable for difficult to fuse samples like chromite ores by increasing the amount of acidic flux in the mixture and increasing the temperature of the furnace to 1180 ℃. For silicate rocks like pyroxenite, acidic fluxes such as the amount of silica and borax were reduced to completely fuse the sample and obtain a lead button of about 40 gm. The buttons were cupelled at 950℃ and the PGMs containing silver beads were dissolved in aqua regia and the PGM concentrations were determined using ICP-OES. A triplicate analysis was performed for all the three samples and the results obtained for quality control samples show a good recovery of 98-101 % for platinum and 101-104% for palladium (as shown in table 1). Rhodium recovery was very poor owing to its tendency to form volatile rhodium oxide and is lost during the cupellation stage. The amount of loss is variable and hence no fixed correction factor can be applied. Therefore, rhodium is best determined using nickel sulfide fire assay as it does not involve the cupellation step.
Table 1: Platinum, palladium and rhodium values obtained for the three samples
| Sample | Platinum (ppm) | Palladium (ppm) | Rhodium (ppm) | |||
| Declared | Obtained | Declared | Obtained | Declared | Obtained | |
| AMIS 0504 | 2.88 | 2.90 | 1.54 | 1.60 | 0.614 | 0.17* (0.6) |
| OREAS 684 | 3.87 | 3.79 | 1.72 | 1.74 | 0.28 | 0.06*(0.20) |
| Chromite ore (commercial sample) | 0.37 | 0.23 | 0.057*(0.20) | |||
*Considerable loss of rhodium is observed as rhodium oxide during the cupellation stage. Considering a loss 70-72% of rhodium during cupellation, the concentration of rhodium after correction is given in parenthesis. From the above data it can be concluded that no fixed correction factor can be applied to the samples and hence rhodium is best determined by nickel sulfide fire assay.
Conclusion
In this work, we studied the extraction of platinum, palladium and rhodium using lead fire assay. The excellent recovery obtained for the above metals except rhodium is especially important because it proves the complete collection of palladium and platinum in the lead button using the new optimised flux composition. Two different quality control samples, one based on pyroxene and the other based on chrome were studied alongwith a high grade chromite ore sample. The method can be applied for the extraction of palladium and platinum from even difficult to fuse samples like chromite ore. It gives an advantage over nickel sulfide fire assay as it is fast, cheap and involves less hazardous materials. However, a considerable loss of rhodium was observed and hence the method is suitable for determination of only gold, palladium and platinum in rocks.
Contributed by Dr. Satirtha Sengupta under the guidance of Prof. Barun K. Gupta