Valuable titanium is the incentive to process red mud, but I like increasingly the production of pig iron to remove Fe from Al, Ti and Si oxides.
Red mud (freed of Na and Ca oxides) is a very poor iron "ore" (50% Fe2O3) but it's free, or even, we want to get rid of it.
Aluminium plants are not installed near coal mines nor gas wells. Though, iron ore travels 800km by train in South Africa, so 2.5 times less coke can travel 2,000km if needed. Petrol coke (freed of sulphur!) could even travel from Alberta's oil sands deposits to Quebec's aluminium plants.
The blast furnace must be bigger to accommodate the iron-poorer red mud. At 0.25$/kg/yr for a complete plant it looks affordable.
More heat is necessary to pre-heat the iron-poorer red mud (but as much carbon to reduce only the iron oxide). I suggest to recycle this heat by injecting as well the slag in a modified cooper, in addition to the flue gas.
The amount of CO2 is the same as if producing pig iron from good ore.
Alternately, perhaps the iron oxide can be melted away from aluminium, titanium and silicon oxides, prior to reduction in the blast furnace if pig iron is desired. Around 1560°C, Fe2O3 decomposes to Fe3O4 which melts at 1597°C while SiO2, TiO2 and Al2O3 keep solid up to 1713°C, 1843°C and 2072°C. It depends much on how the oxides combine in red mud and when heated - maybe, but maybe not. If sunlight is available all the year, 1km2 of solar concentrators provide the heat for the red mud output of a typical aluminium plant - recycling the heat would improve. At averaged 3mm steel thickness, the concentrators' material costs 50M€.
Marc Schaefer, aka Enthalpy