Granny Smith Gold Mine Tails Loss Improvement
by Peng Zhou, Gold Fields Australia
1. Introduction
Gold Fields’ Granny Smith Gold Mine is an underground gold operation with the processing plant having a capacity of 3.0 million tonnes per annum. Since late 2013, Granny Smith has employed a campaign milling schedule, operating 50% – 60% of the time. This is attributed to the restricted ore supply from underground mine and cease of toll treatment. It was observed, however, that since starting campaign milling, tails grades increased for the first two production days of each campaign start-up.
2. Evaluate the extra tail loss
2.1 Higher Tails Grade
Normal tails solids grade at Granny Smith is between 0.30 – 0.50 g/t. The extra tails loss at start-up of the processing plant is, however, ranging from 0.50 – 2.50 g/t. A typical trend of extra tails loss is shown in Figure 1 below. In February and March 2016, the start-up tails grades are high at 1.15 and 1.01 g/t, which are significantly higher than the normal grade.
Figure 1 - An Example of Shift Tails Solids Grade at Granny Smith in 2016
2.2 Benchmarking
To quantify the value of higher tails grade on campaign start-up days, the following benchmarking method was employed. Tails solids grade of the 3rd production day, after plant start-up, is used as the baseline. Slurry residence time is around 30 hours at Granny Smith processing plant. Hence the 3rd production day performance presents a normal performance without being affected by plant shutdown or start-up. In comparison with the 3rd Day, the 1st Day tails grade is 0.24 ± 0.12 g/t higher and it of the 2nd Day is 0.06 ± 0.07 g/t higher at 95% confidence level.
2.3 Value of the Extra Tails Loss
Basing on the benchmarking results above, revenue impact of the extra tails loss can be evaluated. Assuming gold price of $1,600 AUD/ounce, the extra tails loss is valued at about $4 million AUD from January 2014 to January 2017 as shown in Figure 2 below based on actual mill throughput.
Figure 2 - Revenue Impact of the Extra Tails Loss from Jan 2014 - Jan 2017 at Granny Smith
3. Investigation
3.1 Poor Leaching Performance
Extensive investigations have been conducted to reveal the root cause of the higher tails grade problem. The initial step was to determine the content of cyanide leachable gold in the tails on campaign start-up day. Plant historical bottle roll data has been analysed. Intensive cyanide leachable gold in the final tails is normally 0.05 g/t in the tails during normal production days. It was found that, on the campaign start-up day, it is averaged at 0.29 g/t. That is significantly higher than the normal level.
Furthermore, it was revealed that leach reagent concentrations were not maintained in the Leach Circuit when the grinding circuit was shutting down. At Granny Smith, the shutdown grind-out process takes about 1 – 1.5 hours after the mill feed is turned off. This is to ensure the entire Grinding Circuit is free of slurry and build-up for the upcoming two-three weeks shutdown. The grind-out process, however, introduces excessive amount of water to dilute pH and sodium cyanide concentration in the Leach Circuit.
An example of actual pH and cyanide readings are shown in Figure 3 below. In the chart, pH quickly dropped below 9.3 from setpoint 9.8 one hour after the mill feed was turned off. At a pH of 9.3 – 9.5, cyanide ion, CN-, and hydrogen cyanide, HCN, are in equilibrium, with equal amounts of each present. Once pH is below 9.3, sodium cyanide in the slurry volatilise to hydrogen cyanide gas quickly. As it can be seen in the chart, sodium cyanide concentration decreased from the setpoint, 140 ppm, to 105 ppm after one hour and to 80 ppm after two hours. It should also be noted that the sodium cyanide reading has a 40 minutes delay and pH reading, however, is instantaneous. As the result of lower cyanide concentration, the gold leaching performance for slurry that stayed in the Leach/CIP tanks during the whole shutdown period was poor.
Figure 3 – Example Sodium Cyanide and pH readings before and after Mill Feed was turned off. Note: At 0 hour, the mill feed was turned off. -1 hour means one hour before the mill feed was turned off. 1 hour means one hour after the mill feed was turned off.
3.2 Soluble Gold Transport
Since the start-up tails grade was extremely high on some occasions, it was believed that soluble gold in the slurry was adsorbed by or precipitated onto solids. Gold distribution in the slurry was then analysed by assaying slurry sample from Leach/CIP Circuit in the campaign shutdowns. As noted in Figure 4 below, gold in solids increased significantly from 18-Dec 2016 to 05-Jan 2017, while gold in solution has decreased by a similar amount. It is highly likely that gold was transferred from solution-phase to solids-phase during campaign shutdown at Granny Smith.
The mechanism of this process is currently unclear. There are a few possible causes: 1) Soluble gold could be preg-borrowed on the solids, particularly when clays are present in the ore, 2) soluble gold may precipitate out of solution if the free cyanide concentrations are low.
Figure 4 - Gold Inventory in solids and solution during December 2016 campaign shutdown. Note: 16-Dec is the first day of campaign shutdown.
4. RECTIFY THE PROBLEM
4.1 Solutions
Based on the findings above, two solutions were provided to mitigate the extra tails loss:
- Increased amount of sodium cyanide and lime is to be added 4 hours prior to campaign shutdown. This is to ensure that favourable leaching reagent concentrations are maintained even after mill grind-out water dilution.
- Recirculate slurry in the Leach/CIP Circuit after mill feed is turned off for 48 hours by switching existing valves. This is to reduce soluble gold in the slurry through carbon adsorption.
It should be noted that both solutions are implemented without capital input.
4.2 Results
Both solutions above were implemented since Q1 2017, plant data was analysed and gold inventory surveys were conducted to verify the effectiveness.
Historical plant bottle roll data was analysed again to assess the effectiveness of the above solutions. The intensive cyanide leachable gold grade in the tails on campaign start-up day from 2014 to 2018 was shown in Figure 5 below. The leachable gold grade has decreased significantly since Q1 2017. Overall, the leachable gold in tails on campaign start-up day has decreased from 0.29 g/t to 0.05 g/t since the implementation.
Furthermore, a similar gold inventory test was conducted in February 2017 after the solution was implemented. As it can be seen in Figure 6 below, the soluble gold inventory was decreased from 438 Oz to 15 Oz within two days which is an indication of favourable gold leaching performance. More importantly, no obvious transfer of gold from solution to solids was observed.
Note: 20-Feb is the first day of campaign shutdown
Additionally, a revenue impact analysis has conducted in Figure 7 below. By implementing the above solutions, about $700 Oz of gold or $ 1.1 million AUD are saved per annum.
Figure 5 - Intensive Cyanide Leachable Gold grade in the Tails on Campaign Start-up Day from 2014 - 2018
Figure 6 - Gold Inventory in solids and solution during February 2017 campaign shutdown. Note: 20-Feb is the first day of campaign shutdown
Figure 7 - Revenue Impact of the Extra Tails Loss from 2014 - Jan 2018 at Granny Smith
5. Conclusion and Recommendation
The extra tails loss on campaign start-up day at Granny Smith Gold Mine is successfully rectified by introducing operational solutions. It is an example that some processing problems can be resolved at a low cost once the root causes are identified.
Acknowledgement
The author would like to thank Granny Smith processing department for supporting the investigations and implementation of solutions. Also, Gold Fields Australia regional technical services team has provided opportunity and convenience for the author to write this article.