Andy Home: US Aluminium smelters compete with Big Tech to get scarce power.

In the United States, it's been 45 years since anyone has built a primary aluminum smelter.

Alumax opened the Mt Holly plant, in South Carolina, in 1980. The country now had 33 smelters with a combined capacity of nearly five million metric tonnes of aluminium per year.

Six is the number today. Two have been completely curtailed. Mt Holly and two other plants are operating below capacity. The annual production has dropped to 700,000 tonnes.

Emirates Global Aluminium is hoping to turn the tide in Oklahoma with a new facility. The new plant joins Century Aluminum which received federal funding from the Joe Biden Administration for a "green" low carbon smelter in the Ohio/Mississippi River Basin.

Both projects are facing the same problem. The high power prices have killed most of the country's metal smelters, and the lack of affordable power has discouraged anyone from building a smelter since the turn of the century.

The fact that tech companies are willing to pay anything for their data centres, which consume a lot of electricity, makes it difficult for any smelter projects to compete with them for power.

No power, no metal

Since ancient times, aluminium compounds have been used as dye fixers by the Egyptians and for pottery by the Persians.

It wasn't until early in the 19th century, however, that someone figured out how to refine it into metal. Even then it was still a costly curiosity. In 1869, the global production of aluminium was only two tons and it was worth more than gold.

Charles Martin Hall, in the United States, and Paul Heroult, in France, independently discovered a solution that involved electrolysing an intermediate product known as alumina.

Hall-Heroult is the dominant process for producing metals that are ubiquitous in vehicles, buildings and consumer packaging. It also requires a large amount of power.

According to the U.S. Aluminum Association, it takes 14,821 Kilowatt-hours to produce a ton aluminium. A smelter of modern size with an annual capacity of 750,000 tonnes requires more electricity than a city of the size of Boston.

It's a huge challenge for primary aluminum producers in the United States, given that the Energy Information Administration has estimated the country to be facing a deficit of energy of 31 million megawatt hours by 2030 and 48 millions by 2035.

ALUMINIUM VERSUS AI

Matt Aboud is Senior Vice President for Strategy & Business Development, Century Aluminum. He says that the power to build a U.S. aluminum smelter is now available.

He explained the problem at last week's CRU Aluminium Conference, held in London. It is that there is no fixed price for a long time, and a smelter would need that to secure its profitability, as well as pay off construction costs which will reach billions of dollars.

According to the Aluminum Association, a new U.S. aluminum smelter needs a minimum of a 20-year contract for power at a cost not exceeding $40 per MWh in order to be financially viable.

Every smelter is competing with Big Tech. They are both on the hunt for energy in order to power their next-generation artificially intelligent data centres.

According to the Aluminum Association report released on rebuilding U.S. Supply Chain Resilience, tech companies are "not limited in what they will pay" for reliable 24/7 electricity.

Microsoft reportedly paid Constellation Energy $115 per megawatt hour in order to restart Three Mile Island Nuclear Plant in Pennsylvania.

It warned that even reactivating idle aluminium lines would be difficult, given that the average price of electricity in 2023 will be $73.42 per megawatt hour (MWh) for the four states where smelters are located.

"WHERE the wind sweeps down the plain"

EGA has not yet signed a deal to provide electricity for its 600,000-ton smelter project in Oklahoma. According to a Memorandum of Understanding, signed by the state governor Kevin Stitt, the final go-ahead depends on an agreement "power solution framework" based on a Special Rate Offer from Public Service Company of Oklahoma.

According to the EIA, Oklahoma produces almost three times as much energy as it consumes.

In 2023, natural gas will account for around half of the electricity generated in Oklahoma. Wind power will make up another 42%. Oklahoma is actually the third-largest wind power state, after Texas and Iowa.

To run an aluminum smelter using intermittent wind power, it would require a large amount of grid storage, so gas would be a part of the energy mix.

It's better than coal, but it isn't ideal for an industry that collectively tries to reduce its carbon footprint in order to produce "greener" aluminium.

DO NOT CHANGE IT!

Even if EGA is able to secure a long-term, viable power deal, it will take until the end of this decade for the project to produce its first hot metal.

According to projections by the Aluminum Association, by then, 14 new remelt facilities would have been established, bringing the U.S. scrap aluminum demand to 6.5 millions tons.

Recycling uses much less energy, usually around 5%, than it does to produce new metal. It also has a lower capital cost.

The shortage of scrap is the main obstacle to growth in secondary production in the United States.

Only 43% of beverage cans are recycled in the country. This equates to 800,000 tonnes of aluminum thrown away every year.

Also, it exports large amounts of scrap aluminium. Exports will increase by 17% annually to 2.4 millions tons in 2024. Most of these are destined for China which is hungry for recyclable materials.

To reduce import dependence of a metal classified by all U.S. government agencies as critical, capturing more recyclable material and sending less abroad is a complementary approach.

This is also more cost-effective and faster than waiting to find out if EGA or Century will win the fight with Big Tech to get enough power for a new primary melter.

These are the opinions of the columnist, who is also an author. Mark Potter edited this article

(source: Reuters)