Australia's green metals gambit: Technologies designed to to decarbonize steel using Pilbara iron ores

Switching to 'green' iron production, powered by renewables, could grow Australia's iron ore export earnings from $116 billion in 2025-25 to around $386 billion a year by 2060, according to projections from the Superpower Institute headed by economist Ross Garnaut.

"Making metals is an energy intensive business," said Keith Vining, CSIRO Group Leader for Green Metals Production. He said switching to 'green metals'—steel, aluminum and other metals produced without fossil fuels—could transform Australia's export earnings.

Australia is by far the world's biggest exporter of iron ore, supplying just under 55% of global market share in 2024. Nearly all of it is from the Pilbara region in WA. Australia is also the world's largest alumina exporter.

But with iron and steel production alone responsible for a massive 6% to 8% of global carbon dioxide emissions, and aluminum a further two percent, there's global demand for decarbonization of metal production.

And Australia faces a significant obstacle if it is to transform the world's largest iron deposits into fossil-free metals. With 55 to 62 percent iron, Pilbara ore is not pure enough for current low-carbon and carbon-free technologies.

The Pilbara problem

"Pilbara iron ores are challenging materials in existing low carbon iron-making processes," said Vining.

The problem lies in impurities (mostly silica and alumina) which push the ratio of iron content down and raise the ratio of waste rock and minerals, known as 'gangue." These impurities are tightly bound to iron oxide in the rock and need to be removed via high-temperature processing.

Traditional blast furnaces use coal to remove oxygen and associated impurities from iron ore in the steelmaking process.

The newer direct reduction (DRI) process uses gas or hydrogen as a reducing agent instead of coal. It typically requires ores with an iron content over 67 percent which excludes Australia's lower-grade Pilbara ore.

Instead of waiting decades for new technologies to emerge, CSIRO is currently testing adaptations of existing DRI technology which can work with Pilbara ore.

"We want to use existing technologies because the pathway for brand new technologies is so much longer," said Vining.

CSIRO researchers are part of a cross-industry project testing whether the current narrow specifications around the operating envelope for DRI can be expanded and whether, by accepting some productivity loss, Australian iron ore could be viable for these new processes.

Value-add opportunity

The potential economic benefits are enormous. Australia's steel-making capacity is relatively small and limited with just two major steelworks at Whyalla and Port Kembla using primary ore. The economic opportunity lies in processing iron ores.

The Superpower Institute's $386 billion annual revenue projection is based on ramping up green iron production gradually between now and 2060.

Transforming our metals production away from fossil fuels would shift Australia from ore exporter to value-added producer, embedding renewable energy into what the nation ships overseas.

"Is it doable? Theoretically, yes, absolutely," said Vining.

"Those numbers are not based on everything suddenly getting made into a green iron product. The production ramps up over that period of time, so they're pragmatic sort of numbers; but we need a lot of other things to come together, for us to pull it off."

Infrastructure gap

The biggest barrier to achieving that $386 billion opportunity, Vining said, is that it requires a massive investment in infrastructure, including the development of an industrial-scale renewables grid to service metals processing works.

"We're going to need a renewable energy network at a scale that simply does not exist at the moment," he said.

Beyond the renewable energy network itself, Australia faces the challenge of building processing plants in an area with some of the world's highest capital costs.

The Pilbara region is remote and its ports are configured for outgoing shipments, not incoming materials and equipment.

"Most things have to go into Fremantle, and get trucked to the Pilbara," said Vining.

That's a journey of more than 1,300 kilometers by road. All of the labor will need to fly in and fly out.

Energy costs must also come down to competitive levels. Producing hydrogen from renewable energy is currently too expensive to compete with Chinese blast furnaces, even before factoring in labor and capital costs.

The final piece of the puzzle is market development.

"We need a market that is going to value a green iron product," said Vining.

The new Green Metals Innovation Network (GMIN), launched in 2025 as a $10 million Australian Government initiative, brings CSIRO and the Heavy Industry Low-carbon Transition Cooperative Research Centre (HILT CRC) together to coordinate the national research effort.

The collaboration serves a crucial purpose: de-risking the technology for commercial investors.

"To get the investment that we need in one of the highest capital cost jurisdictions in the world, plus the energy infrastructure, an investor would need to have very significant confidence it's technically possible," Vining said.

Bridge technology

Near-term projects are already moving forward using reformed natural gas (methane) as a bridge fuel.

At 700–1000 degrees Celsius, gas reforming uses steam to convert methane into hydrogen and carbon monoxide.

"Using that process is actually 50 percent less CO₂ intensive than using solid carbon in a blast furnace," Vining said.

Critically, new furnaces designed for this gas-based direct reduction are 'hydrogen ready." They can start with natural gas and switch to renewable hydrogen when costs become competitive.

BHP, in a partnership with POSCO, the Korean steelmaker has announced plans for a hydrogen-ready DRI demonstration plant adjacent to POSCO's steelworks in Korea established to process Pilbara ores without these first going through a pelletizing process.

Another joint venture, NEOSMELT—led by BlueScope, in partnership with Rio Tinto, BHP, Mitsui and Woodside Energy, and supported by the Australian Renewable Energy Agency—will build an electric smelting demonstration facility in Kwinana, south of Fremantle in Western Australia.

The project involves feeding direct reduced iron into an electric smelting furnace, with plans to operate for three years to test the technology at scale.

The timeline for moving to fully hydrogen-based steelmaking depends on multiple factors.

"The infrastructure that needs to be built and the policy mechanisms that need to be in place. This is what will probably take longer than resolving the technical issues," said Vining.

Hydrogen is not the only option for metals processing. Another potential pathway could use a technology called Molten Oxide Electrolysis (MOE), which is an electrical process originally developed by NASA for iron manufacturing on the moon.

"We know MOE works, at small scale, but scaling that up to get to 10 million tonnes? That's a big challenge," said Vining.

The projects making Australian ore work

A key focus for CSIRO is adapting Pilbara ore for these new processes, with research spanning the full production chain, from grinding to pelletizing to electric smelting.

CSIRO's three-year India-Australia Green Steel Partnership supports five different projects to reduce emissions and address the challenge of processing low-grade iron ores without fossil fuels.

Under this program, researchers have developed what Vining describes as "a very promising method for making the pellets that will go into the shaft furnace."

The challenge lies in Pilbara ore's mineral composition. When it is heated, there's a structural change in the material, which becomes riddled with microscopic holes (porosity), potentially weakening it.

"We think that porosity leads to a drop in physical strength," Vining explained. In a shaft furnace, under the weight of material above, "they can break apart."

Successfully demonstrating the pelletizing method will be a significant step in the journey.

On the pre-processing front, CSIRO is working both sides of the energy-intensive grinding circuit that breaks ore down into fine particles suitable for processing.

In the first stage of the process, researchers are finding ways to selectively remove more of the unwanted material like silica and alumina, so less ore goes through the energy-intensive grinding step. After grinding, they're improving classification systems to avoid recycling fine material back through the circuit.

The key technology in that classification stage is the hydrocyclone, which uses water and centrifugal force to sort particles at an industrial scale.

"The challenge is trying to get the accuracy of a sieve with the throughput of a hydro-cyclone," Vining said.

Currently, the hydrocyclone has lower accuracy than simpler methods. The technology has been demonstrated at scale with industry partners. A commercialization partner is now deploying it in operating environments.

CSIRO is also building end-to-end pilot-scale capability, from pelletizing to gas-based DRI to electric smelting, to simulate the complete process with Australian ores. The pilot is in the early stages with industry and university partners being established and will couple an electric smelting furnace with existing pelletizing equipment.

Competitive pressure

First-mover advantage matters in heavy industry. Once infrastructure investments are made they're extremely difficult to shift which makes the competitive threat real.

The Middle East already does DRI production with extremely cheap gas, very high solar potential and empty gas fields suitable for carbon dioxide storage during the transition from natural gas to renewable hydrogen.

Vining warned Australia could lose the value-add opportunity entirely if infrastructure investments flow to more competitive jurisdictions first.

"We're here to resolve this problem," said Vining. "The benefit is a $360 billion a year industry for Australia."

Provided by CSIRO