Cracking the chrome conundrum
Source: Modern Mining
Author: Andrew Lanham
Northam Platinum has been in the news recently, with the intended acquisition of the company by Impala Platinum. However, Northam has also been the source of a technological development that could have an even greater influence on South Africa’s platinum industry.
Northam, South Africa’s fifth largest platinum producer, is also the country’s deepest. From a technological perspective, it has been a pioneer. As the quest for platinum inevitably takes mining ever deeper, Northam will be the template for those who follow.
In the past six years, with little fanfare, the mine’s metallurgical team has also been pioneering new technologies, which are having a notable effect on the future of the company.
To recap briefly, platinum mines on South Africa’s Bushveld Complex exploit two platinum-bearing bodies, the Merensky and UG2 reefs. The challenge for platinum mine metallurgists lies in the UG2 orebody, which contains high levels of chromite (Cr2O3). Merensky Reef does not have this problem, making it a more attractive target. However, to establish a major platinum mine requires very substantial investment. As Merensky and UG2 reefs occur in similar quantities in a given mining area, if a mine can exploit both resources optimally, the advantages in terms of operating flexibility and consequent life of mine are enormous.
The Merensky and UG2 ores are brought to surface separately. The ores are concentrated in two dedicated plants, each designed to handle the respective mineralogical compositions. Northam’s UG2 concentration circuit was designed as an industry-standard MF2 circuit, with a capacity of 75 000 tons a month. Flotation hardware consists of Outokumpu roughers, cleaners and re-cleaner cells. Initial pilot plant test work on the UG2 indicated an 85% PGM recovery and 3% Cr2O3 in final concentrate, with a plant availability of 85%.
The two concentrate streams are blended in the first stage of smelting in the furnace. The furnace is a massive rectangular refractory brick ‘box’, 22 metres long, nine metres wide and five metres high. Six huge carbon electrodes are inserted through its roof into the molten mix below. It is at this first stage of smelting, that excess chrome poses a problem.
After milling and flotation, the blended concentrates, which now contain 0.012% PGMs (or 120g/t), are sent to Northam’s furnace, where the bulk of the barren material is removed as slag. The drawback arises in that, if the levels of chrome in the concentrate are too high, the chrome accumulates in the furnace. Excess chrome reduces the capacity of the furnace to build up metal-bearing matte. A point is reached when the efficiency of the furnace is impaired to a point where it has to be demolished, and the chrome mechanically removed. To carry out this exercise involves a very large capital expense for a mine such as Northam.
A way to prevent this is to increase the solubility of the chrome in the slag by operating the slag temperature of the furnace at a temperature well above the current temperature of 1 485˚C. However, temperatures of this magnitude would soon cause damage to the refractory linings inside the furnace.
The presence of UG2-derived chromite affects three critical areas of metallurgical processing:
- the overall recovery of platinum group metals (PGMs)
- the 3PGM+Au upgrade ratio (final concentrate grade to head grade ratio)
- the amount of chrome in the UG2 concentrate
These three factors are interdependent, and thus a fine balance between the three has to be achieved to maximize the ultimate value recovered from the UG2 ore. The key to achieving the balance lies in the concentration process.
In addition, the successful treatment of UG2 ore is having a marked effect on the life of Northam mine. Normal practice has been to send a blend of Merensky and UG2 concentrate to the furnace to minimise the amount of chrome-in-concentrate emanating from the UG2 ore. Traditionally, Northam has blended its concentrates in a 4:1 Merensky/UG2 ratio by mass.
Underground, the Merensky Reef is geologically more complex that the UG2. This makes the former reef more difficult to mine. With Merensky Reef being a more favoured mining target, it has, over time, been mined more extensively. Today, Northam is in a situation where its reserves of this reef are becoming limited.
The metallurgical team at Northam, under the leadership of metallurgical manager Danie Minnaar launched a multi-pronged attack on the problem of chrome in concentrate.
The process of flotation is based on making certain mineral particles hydrophobic (repelled by water) and others hydrophilic (attracted to water). Incoming ore is milled fine and the resulting slurry is then pumped to flotation tanks. In these tanks, with agitation and chemicals, a froth is created. Simplistically, hydrophobic PGM particles cling to the froth, which can be skimmed off. The hydrophilic particles, which include most of the chrome, remain in the non-frothing suspension, which is discarded.
However, there is no absolute dividing line. If one attempts to recover 100% of the PGMs, an unacceptably high level of chrome will be entrained in the froth phase. Conversely, if one attempts to eliminate all chrome in concentrate, then the recovery of PGMs in the final concentrate will be unacceptably low. In the original design of Northam’s UG2 concentrator, it was estimated that 85% of the PGMs could be recovered, while limiting the level of chrome-in-concentrate to an acceptable two percent. To manipulate the balance, much research was put into the blend of frothers, collectors and depressants, the esoteric chemicals that are used in flotation.
Minnaar points out that, until four years ago, the UG2 concentrator achieved a balance of a 79% PGM recovery with chrome in concentrate at about 4.5%. At this level, the Merensky Reef concentrate was essential to reduce the levels of chrome being fed to the smelter. To contain the rising accumulation of chrome in the furnace, the levels of this metal in the concentrate needed to be kept below 1.5%.
The breakthrough in the chrome conundrum came when Minnaar and his colleagues revisited an earlier technology, the sparger column, which had been tried elsewhere in the platinum industry and had been subsequently discarded. Whereas conventional flotation uses tanks with internal agitators and air blown into the pulp to make bubbles, the sparger column relies solely on a jet of high-pressure air being fed through a nozzle into the bottom of a tall cylinder or ‘column’. The slurry from the final stages of flotation is fed in to the column. While sparger columns can’t produce the volumes that flotation tanks can, they do give superior separation of the PGMs from the gangue.
However, Northam made a fundamental design change from the above internal sparger to an external sparger column, which concerned the way air and slurry were mixed by an inline mixer, external to the column. This design has been patented by Northam, which was the first mine to install such a column to treat UG2.
The sparger column circumvents the problem that is experienced with conventional flotation tanks. Traditionally, to get more performance out of the latter, the level of agitation is increased and more air is pumped through the pulp. However, at a certain stage, this results in such turbulent conditions in the cell that no further advantage is gained.
After extensive test work, it was shown that with the addition of the first sparger column after the re-cleaner circuit, it was possible to increase PGM recovery by more than 2% while the chrome-in-concentrate decreased by 44.4%, a drop from 4.5% to 2.5%. The PGM upgrade ratio improved by more than 9%.
The first column was installed in February 2004. Building on the success of the first column, Northam put in a second column a year ago running this in series mode with the first. The new column is a 2.2 metre diameter column, which is much larger than the original 0.9 metre column. Now the 2.2 metre unit is used as the primary sparging circuit and the 0.9 metre column as a secondary. The 3PGM+Au recovery improved to more than 85% and the chrome grade in final concentrate was reduced to less than 2%.
Apart from the new sparger column, the Northam metallurgical team introduced a number of other measures to improve the concentrator performance, while containing the level of chrome in the furnace concentrate. One comparatively inexpensive measure was the installation of high-energy flotation tanks, which were placed at the end of the flotation circuit to improve the upgrade ratio.
With both the sparger additions and the high-energy cells, the comparatively modest capital expenditure was recovered in a matter of months.
The characteristics of the ore feeding a concentrator plant change from hour to hour. Responding to these fluctuations taxes the skills of the plant operators. Part of the multi-pronged attack on chrome-in-concentrate involved a substantial investment in further training of concentrator staff.
A further challenge that faces Northam is that the mine is getting to the point where the amount of Merensky Reef available to the mining team is decreasing. To maintain production, the mine will have to produce more UG2.
Here, the UG2 rod mill posed a constraint. In contrast to a conventional ball mill, the particles produced by the rod mill were in a narrow size range. Importantly, if the chrome fraction of the UG2 ore is milled too fine, it is more readily entrained with the PGMs in the flotation froth. The rod mill was thus the first line of defence against having high chrome levels in the output from the UG2 concentrator.
The drawback of the rod mill was its capacity. At regular intervals, the mill had to be shut down to replace rods. Even though the Northam team had this down to a slick procedure, the rod mill had already been stretched to the limit of its capacity at 75 000 tons a month.
To increase the milling throughput, it was decided to convert the rod mill into a ball mill. The catch here was that balls mills give a wider size range of particles, coarser at one end of the scale and finer at the other. The answer here required new technology in the form of a high-pressure rolls crusher (HPRC). The HPRC, which feeds directly to the converted ball mill, introduces micro-cracks in the ore, allowing for more effective liberation of the PGMs. This process also ensures that the chrome particles are of sufficient coarseness not to be carried with the froth into the concentrate.
In addition, to further prevent the creation of fines, the ball mill is being run with a lower ball load, thereby limiting the grinding during the first milling phase.
Whereas the UG2 milling rate was 100 tons an hour, now with the HPRC and the ball mill, this figure has climbed to 160 tons an hour. Contrary to expectations, the PGM recovery improved at the same time that the milling rate was increased. An added bonus is that the HPRC/ball mill combination has decreased the power usage of the plant.
“While our chrome in final concentrate used to run at 4.5 to 5.0%, with the combination of new technology and improved expertise we have dropped it down to less than 2%. At the same time, the PGM recoveries improved to more than85%,” explains Minnaar.
This means that the concentrates of the two reefs can now be blended into the furnace at a 1:1 ratio. With further work, Minnaar and his colleagues are working towards a point where UG2 concentrate can be the sole feed to the furnace, without affecting its long-term performance.
With the groundbreaking work by the Northam team, its metallurgical performance is among the best in the industry on the western Limb. The work that has been done will stand the company in good stead when it comes time to process the production from Northam’s future prospect at Booysendal.
“At present the UG2 concentrator chrome-in-concentrate runs between 1.8 and 2% and the PGM recovery at 85%, which is the best in our western limb industry,” says Minnaar.
“However, our prediction is that we will reduce UG2 chrome in concentrate to 1.5%,” he concludes.
