Saturday, November 19, 2016

Molten metal metamorphosis

The Australian Aluminium smelting industry is having a rough time. Built to utilize electricity from Australian coal from the 1960’s through the 1980’s, our smelters are ill equipped to deal with the migration of the Aluminium industry to a rapidly industrializing China or cheap low-carbon energy areas such as Iceland or New Zealand. As a result, the Kurri Kurri smelter closed in 2012, the Point Henry smelter closed in 2014, and the future for the Portland smelter is currently uncertain, with the contract for electricity due to be renegotiated this month.

At the same time, Australia is lagging the rest of the developed world in the transition to low emissions electricity. Although certain jurisdictions, like South Australia, are making progress, the fragile nature of the grid connections and the intermittent nature on renewable energy is slowing its uptake, and potentially contributing to supply instability, as was seen during this winter’s South Australian storm.

The production of aluminium metal requires a huge amount of electricity. An aluminum smelter basically consists of a huge tub of molten salt, from which the enormous electrical currents basically force the electrons onto aluminum ions, depositing them on the cathode atom by atom at a rate that allows several tons of production per day.
As a result, aluminium smelters are typically located in areas where there is a large, cheap supply of electricity. Traditionally these have been areas of hydroelectric power, or in Australia’s case, cheap open cut thermal coal. With coal getting more expensive, and with concerns over the impact of CO2 production on the climate, these coal-powered smelters are finding it harder to compete in high wage countries. So Australia has facilities which are designed to take a substantial proportion of the energy grid’s electricity, which are getting closed down just as the requirement for storage of large amounts of variable renewable energy is appearing.

One proposed solution of the “storage problem” is the use of a new technology known as the liquid metal battery. Like the aluminium smelting process, the liquid metal battery consists of a molten salt, which can have ions driven out of it to the anode and the cathode when power is applied. Unlike aluminium, the anode is a base metal instead of graphite, so instead of oxidizing the anode and making CO2, the metal is deposited. This allows the battery to discharge by dissolving the anode and cathode back into the molten salt. So if aluminum smelters are going obsolete in areas which are in desperate need of battery storage, it seems like modifying the smelter to store energy is a option worth at least considering.

There are technical issues, of course. An industrial Hall-Héroult cell is the size of a city bus, and a smelter contains lots of them. The liquid metal technology is being developed by a small company, Ambri, which seems to be starting small (like bottlecap scale), and scaling up. So there is a bit of a gap between the emerging battery technology and the aging smelter technology. But it is in everybody’s interest to bridge it.

Ambri is trying to raise cash and start production. South Australia is still investigating their state-wide blackout.  Alcoa and Hydro have two shuttered smelters which they need to remediate or repurpose, and Portland has 11% of its population working at the smelter. In addition, Boyne Island and Tomago are supposedly facing similar market pressures.

Portland would be a particularly useful place for a pilot project, since the smelter is still operating, even though the pain of closing a big industrial center in a small isolated town looms. It is also located in prime wind power country, on the Victoria / South Australia border, close to the interconnector. So it would be nice if the union, the council, the state and federal governments, and the industry groups could work together to see if there is a solution that benefits everybody.

As for Kurri Kurri and Port Henry, the Kurri Kurri remediation plan comment period closed in August, but Port Henry is still open, even though the last public hearing was last week.Thus the rushed, not completely researched blog post.


Chris Phoenix said...

Things the size of a city bus don't cool down very fast. Could it work to leave the smelters running as smelters, providing grid-smoothing by only using power when there's extra power? So you'd build out lots of, say, wind, enough to power the grid without the smelters 80% of the time; then whenever there was more wind than the grid needed, which is usually, you electrolyze some more aluminum.

C W Magee said...

I have no idea if the quality of the aluminum would be affected by variable electrolysis rates.

lyle said...

The interesting problem Australia has is a lot of its land is desert and therefore good for solar power but no one lives in that part of the country. In particular a lot of Northern Western Australia should be good for solar being semi arid at best. Plus this is were the big natural gas reserves are in Australia. Either consider several hvdc lines down into South Australia or do like Alcan did back in the 50s with Kitamat,and build a town for the smelter on the Wa Coast, to avoid shipping electricity the 2000 mi needed to say Portland Vic. (HVDC is low loss compared to AC and is typically used for very long distance electric links particularly with few intermediate points that need the energy.

lyle said...

It turns out that a number of organizations are working on the ability to turn the electric demand for Al smelters up and down. Current cells won't work it appears because if you turn the current off the cell freezes over as the heat from the smelting keeps everything molten. It is not clear how close to commercialization this technology is.

BTW at least one plant in the US that has an attached coal mine is shutting down, the Warrick works of Alcoa as well.