Copper: flotation circuits sized on closed circuit test data

Head grade is the number every enquiry leads with. It's also the number that tells you least about which plant you need. Two deposits can carry the same copper grade and still demand completely different plants, because the copper sits in different minerals.

The assay that decides the flowsheet

Ask for a sequential copper analysis, not just total copper. It splits the copper three ways: acid-soluble, cyanide-soluble, residual. Acid-soluble copper sits in oxide minerals. The USGS porphyry copper deposit model names the usual ones — brochantite, malachite, azurite, chrysocolla, atacamite — as relatively soluble carbonate, silicate and sulphate minerals. They don't answer to the collector suite a sulphide circuit is built around. Cyanide-soluble copper points at secondary sulphides such as chalcocite and covellite. The residual fraction is your chalcopyrite and bornite, and that is what the rougher, scavenger and cleaner arrangement described above is designed to recover.

The same USGS model makes a second point worth carrying into your drilling plan: the split changes with depth. Supergene enrichment leaves a leached cap over a chalcocite blanket, with oxidised copper minerals sitting above the sulphides. The ore you drill near surface is not necessarily the ore the mill sees in year six. Composite by domain and by bench. One drum of oxide-rich outcrop will design you a plant that later runs out of feed it can treat.

What the alternatives ask of you

Three routes open up when the sulphide answer comes back wrong.

• Sulphidisation flotation. A sulphidising reagent ahead of the collector brings some oxide and mixed ores back into a flotation circuit. Reagent consumption and pH control decide whether it pays, and bench work settles that quickly.
• Leaching with solvent extraction and electrowinning. USGS records that low-grade and oxide ores may be crushed less intensively and treated with acidic solutions on leach pads. Your decision data here isn't a flotation test at all. It's acid consumption of the gangue, column leach kinetics, and permeability after crushing and agglomeration. Carbonate-rich host rock can consume enough acid to end the conversation before the metallurgy gets interesting.
• Two plants. Mixed deposits sometimes justify a flotation circuit and a leach pad side by side, producing two products for two different buyers.

That last option has a commercial tail. Concentrate goes to a smelter. Cathode from electrowinning goes to the metal market directly, with no treatment or refining charge in between. In the United States alone, the USGS 2026 summary counts 2 primary smelters against 14 electrowon refineries, so the leach route is nobody's fringe option. If your sample comes back oxide-dominant, we'll tell you, and the equipment list on this page becomes the wrong shopping list. Which route holds is settled by test work and then carried into plant design.

That mass pull is the honest summary of a copper concentrator: one small valuable stream, one very large one. The USGS porphyry model states it directly — because ore minerals make up such a small percentage of the rock, flotation tailings tonnage is essentially identical to the tonnage mined, and on average about 1.5 tons of waste rock and overburden are moved for every ton of ore. Your tailings facility, water system and waste dumps are the larger civil job. They are usually the longer permitting job too.

Acid rock drainage gets designed for, or inherited

Pyrite is the reason. USGS describes acid mine drainage as a significant challenge associated with porphyry copper deposits, driven by the abundance of pyrite and partly offset by the acid-neutralising capacity of feldspars and trace carbonate minerals in the same rock. Which way your material sits is answered by acid-base accounting on tailings and waste rock. That testing is cheap, and it should run alongside the metallurgy rather than after it. Leave it to a closure study and you'll be retrofitting liners and cover systems around a plant that's already standing.

The water you tested on is not the water you'll run

Bench flotation is usually done on clean laboratory water. The plant drinks reclaim from the tailings pond, and after months of recirculation that water carries residual reagents, dissolved ions and, on some sites, enough salinity to move the pulp chemistry. Ask for the reagent scheme to be confirmed on site water, or on synthetic water matched to it. A few water samples cost far less than a reagent regime that quietly stops working in month four.

Where the contract line falls

One boundary question deserves settling before anyone prices anything. Do the tailings facility, the power connection and the water supply sit inside the scope, or outside it? Under an EPC contract those interfaces belong to the contractor at a defined price. Under EPCM they stay yours to manage while we direct the work and hold the engineering. Neither model is better in the abstract. On a copper project, where most of the tonnage never reaches the concentrate, that line covers more of the job than owners expect.