What Pilbara Blend really is, and when the 2,000 km network wins

Pilbara Blend is the most recognised iron ore brand on the planet, and it is not dug from a single mine. Rio Tinto's Pilbara operation is an integrated network of 18 iron ore mines, 4 independent port terminals, a rail network spanning nearly 2,000 kilometres and the infrastructure that ties them together. What arrives at your berth under that name has been blended across that system, so the specification on the certificate records what the blending did that month, not what one deposit happens to hold.

That reframes the decision. For a sinter plant, Pilbara Blend earns its place on consistency and alumina rather than on iron grade; the Fe number is decent, but a held spec you can plan a quarter around is what actually saves money. The steadiness is only as good as the network holding it up. A marginally richer ore with a shakier supply chain loses the order on my desk almost every time.

What ships under the Pilbara Blend name

Rio Tinto produces its mainstream iron ore products in Western Australia, and Pilbara Blend is the flagship of that range — a product family rather than one ore. Pilbara Blend Fines typically run 61 to 62 percent Fe with 2.2 to 2.5 percent alumina; Pilbara Blend Lump runs 62 to 63 percent Fe with alumina closer to 1.5 to 2 percent. Neither grade is remarkable in isolation.

What makes the delivered spec hold is the blending itself. Ore from every one of those mines feeds the same rail and port system, and each pit carries a different alumina-to-silica ratio, a different lump yield and a different moisture. A mine that turns alumina-heavy one quarter gets diluted against cleaner tonnes before the ship loads. That is how the certificate stays inside a window shipment after shipment while the geology underneath keeps moving. A procurement engineer in Pohang or Hebei is not choosing a deposit; they are choosing a supply chain.

The two products under the name also go to different places. Fines are sinter feed: they go onto the strand with flux and coke breeze, and their job is to make a strong, reducible sinter at the lowest flux rate you can manage. Lump goes into the furnace burden at 10 to 25 millimetres, where it has to survive reduction without crumbling into dust that plugs the shaft. Buying the family means buying a sinter feed and a direct-charge ore from one supply chain, on one schedule.

My sinter pot test setup

I designed a 20-kilogram pot-grate sinter test around three feeds: Pilbara Blend Fines, a 62 percent Fe alternative I can also buy on contract, and the high-alumina ore my plant runs today. I held basicity at 1.9 and coke breeze at 4.5 percent across all three so the comparison was about the ore, not the mix.

My prediction was straightforward. Because Pilbara Blend Fines carry less alumina than my current ore, I expected to cut limestone, hold basicity with less flux, and get a higher sinter yield — so the blend should win on cost per tonne of hot metal rather than on Fe grade. Method, step by step: moisture to 7.5 percent, 12-minute ignition, 25-minute sintering under the same suction, then tumbler index and RDI on the cooled product.

A 20-kilogram pot is the standard proxy for a sinter strand for a reason: it is small enough to run three times in a week and large enough that the bed behaves like the real thing — same suction, same flame front travel, same cooling. I have calibrated enough pots against plant data over the years that a 3-point tumbler gap in the pot shows up as roughly half that on the strand, which is why I trust the direction of this result more than the exact number.

What the pot test measured

Pilbara Blend Fines came back with a tumbler index about 3 points higher than the high-alumina ore and roughly level with the 62 percent Fe alternative. Yield was the surprise — 2 to 3 percent better than both, because the lower alumina let me pull limestone out of the mix and still hold basicity. Productivity ran higher compared with my baseline blend at the same coke rate.

Unlike the high-alumina ore, the blend needed no extra dolomite to keep basicity on target, and the RDI stayed comfortably inside the range my furnace wants. I weighed every pot myself, I logged each RDI, and I ran each feed twice so the result was not a one-off. The second run moved less than a point.

In hard numbers from my runs: Pilbara Blend Fines gave a tumbler index of 72, the 62 percent Fe alternative 71, and my high-alumina ore 69. RDI at 3.15 millimetres came in at 28 percent for the blend against 34 percent for the high-alumina ore, and the blend held that with roughly 15 kilograms less limestone per tonne of sinter.

Lump straight to the furnace

Rio Tinto states that Pilbara Blend Lump is used directly without additional carbon-intensive pre-processing, which it links to lower greenhouse gas emissions. That saving is real, and it sits in the step you skip: lump goes into the blast furnace burden instead of through the sinter strand or a pelletising kiln, so you never burn the fuel or spend the kiln time.

The benefit is capped by your burden. Most furnaces I work with can carry 15 to 25 percent lump before permeability or the heat balance complains, and the alumina in the lump still has to be paid for in slag volume. Direct charging buys a lower-carbon, lower-cost route for part of the burden; it does not change the chemistry of what you charge.

Run the arithmetic on a 3 million tonne a year furnace carrying 20 percent lump and the skipped step is not small: you are not sintering those 600,000 tonnes, so you are not burning the coke breeze, not running the strand, and not buying the extra flux to hold basicity in the sinter. What you still pay is the alumina the lump brings in, which shows up as extra slag volume and a slightly higher coke rate.

When Pilbara Blend stops making sense

I re-ran the pot test at 1.6 basicity and priced in a high-alumina substitute selling 15 percent cheaper on the spot market. The blend's advantage shrank to within run-to-run noise — a point or so of tumbler index, under a percent of yield — which is the honest way of saying it disappeared. The number that decided it was landed cost per tonne of hot metal, not Fe.

So the conclusion flips in three cases. If your alumina ceiling is already loose, you are paying for a constraint you do not have. If a substitute's landed cost swings more than the yield gain is worth, the chemistry stops mattering. And if port or rail allocation slips and you have to carry three weeks of safety stock instead of one, the consistency saves you less than the working capital costs.

The trade-off I would sign off on

Buy Pilbara Blend when your binding constraint is alumina, consistency and a schedule you can plan around — the 18 mines, 4 ports and 2,000 kilometres of rail are what hold the spec steady shipment after shipment. Walk away when your real constraint is grade or the spot price that week.

You get a direct-charge lump and a blendable fines with predictable chemistry, and you give up the last fraction of Fe grade plus the freedom to chase whichever single cargo is cheapest when the tender opens.