Marilyn Scales is a field editor for the Canadian Mining Journal, Canada’s first mining publication. She is one of Canada’s most senior mining commentators.This article was originally published – April/2006
New Caledonia, a French island territory 1,600 km off the northeast coast of Australia, is home to an estimated 25% of the world’s known nickel reserves. With rich laterite and saprolite deposits, it is no wonder this island nation is the scene of increased mining activity. A subsidiary of Paris-based Eramet currently owns five mines and a smelter scattered across the island. The other producer is Société Minière du Sud Pacifique S.A. It, too, has several mines supplying an Australian smelter.
The Goro Nickel Deposit, tucked away on the southern tip of New Caledonia, is one of the world’s largest undeveloped laterite deposits. But not for long. Construction of the mine, mineral processing plant, and extensive infrastructure is moving ahead quickly toward a start-up date of late 2007.
As of the end of February 2006, engineering is over 70% done, with about 1,600 workers on the site. Earthworks for the process plant were completed in March 2006, and will continue at the residue storage facility and on road realignment. The test mine extends to the saprolite horizon and exposed bedrock. The first of almost 2,000 skilled Filipino workers will soon arrive to start on construction.
The first berth of the port will be completed in time to receive the first module of the processing plant in May. The next milestone will be completion of the first half of the coal-fired power plant in September. The second berth of the port and the raw water pipeline will be finished in time for that event.
Goro Nickel, owned by Toronto’s Inco Ltd. (69%), as well as a joint venture of Japan’s Sumitomo Metal Mining and Mitsui Co. (21%), and three local provinces (10%), has committed a huge sum — US$1.878 billion including a 15% overage — to the development.
That amount of commitment came only after careful planning. Inco first became involved at Goro in 1992. By the end of the decade, the company commissioned a pilot plant at the site to test its high-pressure acid leaching (HPAL) for nickel and cobalt recovery. In 2001, Inco decided to go ahead with the Goro project at a cost of US$1.4 billion, and anticipated production in late 2004.
A year later, Inco put the brakes on the project. In 2002 it began a comprehensive, two-part review of the Goro development in light of skyrocketing costs. Every aspect of the project was re-examined. On Oct. 19, 2004, Inco chairman and CEO Scott Hand announced that Goro would finally move forward toward production of 60,000 tonnes of nickel oxide and between 4,300 and 5,000 tonnes of cobalt in the form of carbonate annually. Cash costs per pound of nickel are estimated at US$1.10-$1.15 after byproduct credits.
The Goro deposit will support at least 20 years of mining, and probably many more. Proven and probable reserves are 120 million tonnes grading 1.48% Ni and 0.12% Co. There are 75 million tonnes of measured and indicated resources grading 1.49% Ni and 0.12% Co. And for an even brighter future, the inferred resources are estimated at 128 million tonnes grading 1.7% Ni and 0.1% Co.
How to recover Goro ore
The Goro deposit is a classic nickel laterite deposit — flat-lying, near surface and measuring 45- to 50-m thick. It is covered by an iron oxide crust at surface and a high-hematite zone below, which measure 10 to 20 m thick. The first ore zone is a 10-m-thick limonite layer, which grades about 1.5% Ni. The transition zone between the limonite and underlying saprolite is enriched with cobalt and magnesium. The bottom layer (the saprolite zone) is approximately 5 m thick with higher nickel grades, averaging 2-3%.
The orebody, like the topography, slopes gently from 330 m above sea level in the north to 200 m in the south. Mining will begin at the lowest elevation and proceed uphill. In this way mine water and runoff can be collected and will interfere as little as possible with the mining sequence. The mine will expand during the initial 20 years of production from the south in a more-or-less counter-clockwise direction to the northeast and then northwest. The plan anticipates removing a “slice” of ore roughly 100 m wide each year.
Between six and 24 months before mining begins, the vegetation will be removed and the topsoil stored for reclamation. The overburden will be excavated and sorted. About 70% of it will be used to build access and haulage roads, the tailings dam and other such earthworks. The remainder will be placed in the waste rock dumps.
A truck-and-shovel fleet will remove the ore. The main haulage level will be established immediately above the limonite layer. Besides accommodating the haul trucks, the level will be home to hydraulic excavators with 11-m3 buckets that will either dig down into the ore or up into the overburden. Trucks will be loaded selectively with either limonite or overburden.
To remove the saprolite portion of the orebody, smaller excavators with 4.4-m3 buckets will travel along the transition layer, digging downward into the ore. They will dump their buckets up on the haulage level where the saprolite can be loaded into the trucks by the larger excavators.
The bi-level excavation and single haul road arrangement is important for at least two reasons. First, only one haul road must be built and maintained in a rainy, tropical climate. That saves considerable costs. Second, only the smaller excavators will traverse the saprolite, which can have a very rough profile due to its rocky nature. That saves wear and tear on expensive equipment.
The ore preparation plant will handle ore sizing, blending, grinding and conditioning. Trucks will discharge their loads of limonite or saprolite into separate screening plants. The soft, clay-like limonite will be pulped. The hard saprolite will pass over a grizzly to remove +760-mm rocks and then be ground. The two slurries will be blended and cycloned, and the portion with a nominal particle size of -100 microns will be stored in tanks at the prep plant.
Inco engineers are confident that their HPAL method of recovering nickel and cobalt is efficient and cost-effective. They have the experience of a two-and-a-half-year run with the pilot plant at Goro. Recoveries of both metals are expected to top 90% from either type of ore. And HPAL is energy efficient because there is no drying, calcining and melting as is found in conventional pyrometallurgical smelters.
Building the processing plant is unique in that it is being constructed in more than 400 separate modules, each weighing approximately 200 or 300 tonnes, at two yards in the Philippines. The first module, the boiler plant, is expected to arrive at Goro in May.
The Goro processing plant is not the first to be built in modules or at a location other than the mine site. There are several advantages. The modules are built where there is infrastructure and skilled labour. They are pre-assembled and pre-tested. They are sized for ease of transport. Installation at the mine site will be faster than if the plant were built there from scratch. And — this is the big one — the plant can be expanded in the future with the addition of more modules.
Briefly, the HPAL plans will work along these lines:
The pulp at the ore preparation plant will be piped to the process plant and thickened to 33% solids ahead of three identical HPAL trains. Each 27.9-m-long autoclave has six agitated chambers. The thickened pulp is introduced to be preheated and pressurized. Sulphuric acid is injected, and the metals are leached for a total of 30 minutes at a temperature of 270°C and a pressure of 60 bars.
Nickel- and cobalt-laden solution leaving the autoclaves will pass into flash tanks (fleeters) where the pressure is reduced ahead of two counter-current decantation (CCD) circuits. This is followed by partial neutralization with lime and limestone to precipitate iron, aluminum, chrome, silica and part of the copper in solution. The remainder of the copper is removed in pulsed ion exchange columns.
The pregnant solution will next pass through the first solvent extraction (SX) circuit. Between this circuit and the second SX circuit, zinc will be removed from solution in another ion exchange circuit. The second SX circuit produces two pure chloride solutions of nickel and cobalt.
The nickel chloride solution will pass through pyrohydrolysis at 800°C to create the final nickel oxide product. This step also regenerates the hydrochloric acid used in the SX circuit.
The cobalt chloride solution will be treated with the addition of sodium carbonate in a series of three agitated tanks. This will precipitate the saleable cobalt carbonate product.
For every tonne of ore that is mined, the plant will produce approximately 10 kg of nickel and 1 kg of cobalt.
The Goro project will stand as one of Inco’s most important producers when it starts up in late 2007. There have been difficulties getting it off the ground, but Inco and its team of engineers and contractors have solved them creatively. They can now proceed with confidence, building a project with environmental, employment and economic stability.