Feature Articles
Aluminium Smelting
June 2021
Aluminium is one of the most abundant metallic elements on Earth but is rarely found in its elemental form. Prior to 1886, the cost of producing aluminium metal was greater than that of gold and platinum, making it highly valuable. However, since 1886 when an American, Charles Hall and a Frenchman, Paul Heroult simultaneously discovered you could dissolve alumina in cryolite, apply a bit of electricity, and Bob’s your uncle— there's aluminium. The Hall-Heroult process has seen the cost, and therefore value of, aluminium fall off a comparative cliff. This process, combined with cheaper electric power transitioned aluminium from a precious metal to a comparatively inexpensive commodity.

Since its first use in industrial production, the Hall-Heroult process has, in essence, not changed. Improved efficiency and increasing current amperage have achieved greater production per unit area and been the predominate focus of technological develop. Striving to increase production and efficiency, operational amperage levels have increased from sub 50kA to pushing 600kA—though are starting to reach the limit of how high they can be pushed in a controlled manner.

However, in an increasingly carbon emission conscious world, the major technological development being pursued is replacement of the carbon anode, responsible for significant direct CO2 emissions from the production process. While inert anode technology is not new, with a number of producers having invested in the research. The current push to a green economy is seeing new impetus on utilising it at an, as yet to be seen, commercial scale.


The Hall-Heroult Process

Smelter grade alumina (Al2O3) is treated by electrolytic reduction at temperatures of 960–1,000°C in cells (‘pots’) containing a bath of molten cryolite (sodium aluminium fluoride - Na3AlF6). The electrolysis dissociates the aluminium and oxygen. Oxygen bonds with carbon in the anode forming CO2 and liquid aluminium, heavier than the cryolite, solution collects at the bottom and is periodically tapped.

The pots are shallow rectangular basins encased in a steel shell and lined with fireclay brick for heat insulation, which in turn is lined with carbon bricks to hold the electrolyte. Steel bars carry the electric current through the insulating bricks into the carbon cathode in the floor of the cell. The cell lining may be either thick carbon blocks or a rammed mixture of carbon and pitch.

The anodes dip into the electrolyte, suspended on steel rods. They are either pre-baked carbon blocks (a ‘pre-bake’ anode) or a mixture of unbaked petroleum coke and coal tar pitch, which is baked in-situ with the heat of the pot (a ‘Söderberg’ anode).

During electrolysis, an electric current flows through the electrolyte, breaking down the dissolved alumina into its constituent elements: aluminium and oxygen ions. Oxygen migrates to and combines with the carbon anode to form carbon monoxide or carbon dioxide, liberating the aluminium at the cathode as pure molten metal. The liquid aluminium settles to the bottom of the cell and is vacuum siphoned off (‘tapped’) periodically into crucibles.

Incoming alternating current is rectified to direct current and supplied to a line of reduction cells. Modern solid-state systems have vastly increased the efficiencies of rectification, so that the usage of direct current is now the prime measure of energy efficiency. In a modern plant, the ‘specific energy’ factor is typically in the range of 13.0–13.5 kilowatt hours per kilogram (kWh/kg), though newer, high amperage units are pushing towards 12kWh/kg. Older plants operate at up to 17kWh/kg.



The current passing through the cell is the primary determinant of the rate at which it produces metal and affects the physical dimensions of the cell. Outside China, due to their age the size of each cell in newly constructed plants is typically in the range of 250–350 kiloamperes (kA) with more recent capacity likely closer to 400+kA. More recently installed capacity in China is of the order of 500+kA.

A series of cells is connected electrically to form a potline through which high amperage direct current is passed through the cryolite bath to create the reaction:

2Al2O3 + 3C → 4Al + 3CO2

The electrolyte solution is principally synthetic molten cryolite—sodium aluminium fluoride produced by treating sodium aluminate from the Bayer process with hydrofluoric acid. Dissolving alumina in cryolite lowers its melting point for easier electrolyisis. Small amounts of aluminium fluoride, calcium fluoride, magnesium fluoride, lithium cryolite or sodium chloride may be added to lower the bath temperature in order to increase current efficiency.

Molten aluminium metal is deposited on the cathode floor of the cell and is removed by vacuum siphoning into a crucible, in which it is carried to a holding furnace. The liquid metal may be directly transported from there to a semi-fabrication plant or cast into solid form, often with some alloying, as ingots or billets.



The Anodic Concern

Beyond the usually indirect emissions from generating the large amounts of electricity to undertake the electrolysis process, it is the direct carbon emissions associated with the production and consumption of the carbon anode creating consternation about primary aluminium’s place in a carbon-constrained ‘green economy’.

Approximately 0.45 units of carbon are consumed per unit of aluminium produced in modern cells. The carbon is liberated as CO2 – making the current reduction process a significant contributor to direct (Scope 1) carbon emissions. They are present regardless of the power generation source. Inert anodes, replace the carbon block with a ceramic and produce oxygen as opposed to carbon dioxide.


The Green Future

The transition to a ‘green economy’ presents significant opportunity for aluminium. Aluminium’s light weight is seeing significant research into its application in vehicles to increase battery range. Advancement in aluminium alloy’s is revealing opportunities for its application in areas previously requiring strength better met by steel.

While the benefits of lighter materials are a positive for the transition, the carbon emissions to produce them can become a sticking point. Given its energy intensity, this is particularly the case for primary aluminium. While the carbon emissions associated with the process’s electricity requirements can—to a degree—be overcome through the use of hydropower; the direct emissions generated by the use of a carbon anode in the reduction process remains an often-overlooked issue and potential area of concern.

Primary aluminium produced without direct CO2 production at the anode is the holy grail of low carbon ‘green’ aluminium. While power can be sourced from renewable or low emission sources, lowering scope 2 emissions, there is currently no commercially viable replacement for the carbon anode. Though this appears to be changing.

Elysis, a JV between Rio Tinto and Alcoa are developing carbon-free reduction technology which has the potential to eliminate direct carbon dioxide emissions from the reduction process. It recently announced a commercial scale demonstration at the Alma smelter in Canada. Rusal which has also invested heavily in research into inert anode, has also started at scale production at its Krasnoyarsk smelter, albeit at very small production volumes.

Current aluminium reduction technology is also relatively inefficient with the electricity it does consume—losing large amounts of energy as radiant heat—with estimates of up to 50% of incoming energy lost. Improved refractories in cells, capable of withstanding the corrosive conditions, could also help insulate the process, keeping more heat in. Alternate anodes reducing resistance losses and improved process control are also being investigated with the aim of reducing electricity requirements and therefore Scope 2 emissions per tonne of metal production.