New aluminium production process avoiding CO₂-emissions

The project’s goal is to develop an alternative process to produce aluminium from aluminium oxide that may result in lower CO₂ emissions than the current well-established electrolysis process (Hall-Heroult process). In Norway, more than two million tonnes of CO₂ per year are emitted from the electrolysis plants. Globally, more than hundred million tonnes are emitted. CO₂ stems from the reaction between aluminium oxide and carbon anodes. During the electrolysis process, aluminum oxide is split into liquid aluminium metal on the cathode and CO₂ is on the anode. In the alternative process, aluminium oxide is first converted to anhydrous aluminum chloride, AlCl3, which is then electrolysed. In this electrolysis process, aluminium chloride is split into liquid metal on the cathode and chlorine gas on the anode.

The electrolysis process itself takes place without CO₂ emissions. Anhydrous aluminum chloride does not exist naturally and must be made as an integral part of the process. On a large scale, the only proven method is so-called carbochlorination of aluminium oxide at around 700°C: Al2O3(s) + 1.5C(s) + 3Cl2(g) = 2AlCl3(g) + 1.5CO₂(g). Chlorine for this reaction will come from the electrolysis of AlCl3. Even if CO₂ is not formed in the electrolysis process itself, CO₂ will form during carbochlorination. But this CO₂ can be captured far more easily than the one from the current electrolysis process since it is almost concentrated. It is also possible to recycle it in this process.

It is known that CO gas can be used instead of carbon to produce AlCl3. CO₂ from the carbochlorination can therefore be converted to CO and thus recycled. For example, conversion of CO₂ to CO can occur using hydrogen or electrolysis. Alternatively, CO can be made from natural gas with hydrogen as a by-product. Then CO₂ from production of AlCl3 must be sent to storage (CCS). An additional option is to use biomass to make CO. Then produced CO₂ could be released to the atmosphere climate neutral or, if stored, give net capture of CO₂. All these options will be evaluated in the project. In addition, the carbochlorination process will be studied in depth and a techno-economic model will be developed for the whole process.

The first phase of the project has now been completed. The results are positive enough to merit further development of the process. It has been found that a mixture of CO and chlorine gas effectively converts aluminum oxide into anhydrous aluminum chloride. A detailed description of a factory based on the new process has also been developed and is used as a basis for the techno-economic study. The study shows that the new process has a cost level comparable to the current Hall-Héroult process. In the next phase, the various process steps will be further developed experimentally to provide a technical basis for significant upscaling. The goal is that towards the end of this decade the process has been demonstrated at full industrial scale and is ready for commercial expansion.