RPT-COLUMN-A quiet revolution is unfolding in the mining sector: Andy Home

By Andy Home

LONDON, Aug 29 (Reuters) - The world is going to need a lot of copper and other critical metals if it is going to pivot away from fossil fuels. But can the mining industry deliver?

The challenges are huge. Ore grades at existing copper mines are steadily falling, big new discoveries are becoming rarer and development times can stretch up to a decade.

Part of the solution is to increase the efficiency of the mining process, which has historically been both highly polluting and wasteful.

BACK TO THE FUTURE

The world dug up 650 million metric tons of copper between 1910 and 2010 but 100 million tons never made it to market, according to a 2020 research paper by Germany's Fraunhofer Institute.

All that metal is still there lying in tailings ponds, a potentially massive resource awaiting the right technology to unlock it.

Rio Tinto RIO.L has already successfully separated critical metals such as scandium and tellurium from waste streams at existing operations.

Others are now looking at ways to extract value from the vast legacy of past mining activity.

Hudbay Minerals HBM.TO, for example, is evaluating the potential for re-mining tailings at the Flin Flon site in Canada's Manitoba. The mine closed in 2022, leaving nearly a century's worth of minerals-rich waste.

Australia's Cobalt Blue Holdings COB.AX, which has been collaborating on the Flin Flon project, has also signed an agreement with the Mount Isa city council in Queensland to explore re-working pyrite tailings as a potential alternative source of sulphur once the town's copper smelter closes.

These and many similar projects are still only at conceptual or pilot stage but India's Hindustan Zinc HZNC.NS is scaling up with a $438 million commitment to process 10 million tons per year of tailings at its Rampura Agucha mine, the world's largest zinc mine.

LESS WASTE

While miners are collectively reassessing the value of legacy waste, they are also working out how to produce less waste in the first place.

This comes with both economic and environmental upside. The mining industry currently generates over seven billion tons of tailings per year and the amount is rising as ore grades fall.

Much of the work in this area is incremental in nature. Glencore Technology, for example, has been steadily improving its ISAMill grinder to handle increasingly coarser particle sizes. The aim is to reduce the amount of ore grinding to save water and reduce tailings waste.

The company's Albion Process for leaching can lift copper recovery rates to over 99% and reduce operating costs by up to a third, allowing development of complex ore-bodies that wouldn't be viable with traditional technologies.

Others such as Allonnia, which describes itself as a bio-ingenuity company, are pioneering more revolutionary approaches.

The company's D-Solve technology uses microbes to selectively extract impurities such as magnesium from concentrates.

Allonnia has just partnered with the Eagle nickel mine in the United States for an onsite unit to pilot technology that in laboratory tests can improve nickel grades by 18% and cut magnesium impurities by 40%.

BIG TECH MEETS OLD TECH

The new overarching technology that can bind all these innovations together is artificial intelligence (AI).

Majors such as Rio Tinto and BHP are already using AI in autonomous haulage systems and to predict maintenance downtime rather than reacting to equipment failures.

Generative AI is the next big leap forward. BHP BHP.AX uses it in combination with "digital twin" technology, a real-time virtual replica of the mining process, at its South Australian copper mine and the giant Escondida mine in Chile.

GenAI models at Escondida "inform ore blasting and blending strategies, identify mine areas with challenging ore characteristics, and support the implementation of SAG mill model predictive control," according to BHP.

U.S. copper producer Freeport-McMoRan FCX.N has partnered with consultancy group McKinsey to use AI to boost production at its North American operations, which were facing declining output due to mature mines and aging process technology.

Integrating traditional mining with data engineering allows for real-time adjustments to processing rates to handle variable ores. When AI was trialled at the Baghdad mine in Arizona, it led to a 5%-10% increase in copper production.

Rolling it out across the company's other American operations is projected to lift output by 90,000 tons each year.

That's equivalent to a new processing plant, which would come at a cost of over $1.5 billion and a timeline of eight to ten years for planning, construction and commissioning.

FUTURE MINING

Mining, it is often said, is a dirty business.

The proof lies in the billions of tons of sludge sitting in tailings ponds around the world. The consequence is public antipathy to new mine projects, which is one of the reasons it takes so long to build and commission a new one.

Mining has also been a highly inefficient business in the past. Too much mineral value has been either discarded as waste or simply left in the ground because the technology didn't exist to treat such low-grade ore.

That's changing as one of the world's oldest industries rapidly modernizes, combining innovations in traditional processing with new technologies such as bio-engineering and AI.

This is a quiet revolution playing out in multiple laboratories, pilot plants and data centres around the world.

But the promise is one of a much cleaner and more efficient sector, which may just mean the world isn't going to run out of copper after all.

The opinions expressed here are those of the author, a columnist for Reuters.