Complete process plants (5): ALUMINIUM PRODUCTION

We start first with curve 1 in Figure 5, which gives the capital requirements for petroleum coking facilities. Such coke provides the electrodes used in large quantities for converting alumina into the metal. Note that differences in technology and location- for example, the configuration of the refinery where the coking plant is based - may give rise to a wide spread of costs.

Clearly the data in Table 5 need to be used with great caution because of the spread which they represent.

Moving on to curve 2 and correlation 5b, we have budget costs for plants for the manufacture of alumina from bauxite ore. The cost of open pit mining is excluded since this operation may be very remote from the alumina plant and involve substantial transport costs. The product of such a plant is high grade Al2O3, which is of course the feedstock for the aluminium smelter.

Area 3 covers the capital costs of plants for the production of aluminium from alumina using electrodes from the coking process. Again, as with curves 1 and 2, the spread represented by the correlations (5c - 5e) and area 3 indicates that the data are very approximate and are particularly location-dependent. The other requirement, of course, for aluminium production is the availability of large quantities of cheap (or subsidised) electrical power. This accounts for the establishment of the Gulf aluminium industry and similar developments in regions of plentiful hydroelectricity.

An interesting feature of the data for smelters is the exponent of unity on capacity rather than the usual rough correspondence with two-thirds power law. It seems that this is due to scale-up being achieved by simply increasing the number of cells, rather than the size of individual cells. This is the first time that Costimator has identified a scale-up exponent which is unity throughout.

* Although this series on complete plant will continue, should any readers be looking for a quick budget cost of a complete process plant not covered so far, they are welcome to send their enquiry into the Editor who will pass it to Costimator, who may then be able to give a very fast response, depending on the information in his database.

{{Table 5: Key and correlations to Figure 5

Curve/area Product and main Correlation Equationnumber feedstocks no

Curve 1 Petroleum coke from Cost = 1.05(Capacity)0.61 5a petroleum residuesCurve 2 Alumina from bauxite Cost = 1.96(Capacity)0.54 5b

Area 3 Aluminium from alumina Upper approx limit using carbon Cost = 1.08(Capacity)1.0 5c electrodes Average approx Cost = 0.836(Capacity)1.0 5d Lower approx limit Cost = 0.57(Capacity)1.0 5e