Reinventing Gyro Gearloose machine, this time, to purify scrap aluminum

One of SINTEF Materials and Chemistry's laboratory halls on Gløshaugen in Trondheim is dominated by a four metre-long, silver-coloured apparatus. Both pipework and cables have yet to be installed, and its "lid" still lies on the floor, awaiting installation some time this spring.

Nevertheless, this is a machine into which molten impure aluminium will be poured at the top, undergo a refining process and emerge through two outlets, as pure aluminium and contaminated metal, respectively.

A number of projects all over the world have tried to recycle and refine aluminium in this way, but so far with limited success. The SINTEF scientists, at least, are not aware of any positive results in the literature, so now they themselves are now making the attempt. For more than ten years they have been working on their machine to recycle and refine contaminated aluminium. This May, they will know whether they have succeeded.

The equipment

The left-hand part of the giant machine contains the purification system: molten aluminium at a temperature of around 660 degrees Celsius is poured in through a vertical insulated pipe and spreads through a horizontal channel immediately below. The whole system is then tilted slightly and a large rotating screw made of graphite forces (delete the) solidified crystals in equilibrium with pure aluminium upwards to the hottest point, while the impurities settle down to the cooler section. Here the temperature will be around 650 degrees. Today, the screw has been taken out of the machine and is lying on the skid that forms the right-hand side of the machine.

Some 40 heating elements are installed in various parts of the machine, giving the scientists complete control of the temperature.

Aims

Today, the world recycles about 9.9 million tonnes of scrap metal a year. In the near future, more such scrap is expected to become available also from buildings that have come to the end of their useful life after fifty or sixty years.

"Hydro aims to raise the level of recycling, and recycle a million tonnes of aluminium a year by 2020. The company has provided financial support for this project for more than ten years," says SINTEF's Anne Kvithyld. "Today, Hydro buys scrap metal from scrap-merchants, in addition to the company's own resources. But a pallet of scrap metal is a mixture of many things; for example, contaminated with iron. Some of this can be magnetically separated, but a good deal remains. Impure metal can be used in products that do not need to be pure, such as certain vehicle components, but not in aluminium kitchen foil, which must be extremely pure aluminium. It is important for Hydro to be able to extract such pure products from scrap metal."

The problem is that every time metal is recycled, the proportion of unwanted elements in the melt rises. This is now giving industry such serious problems that it is occasionally forced to add pure aluminium (delete to the smelt) in order not to end up with "rubbish alloys."

"If all goes well, our refining furnace will be able to reverse this trend. We can remove undesirable impurities such as iron at the dirty end and get pure aluminium at the clean end, where it can be just as pure as primary metal. We can even make it purer than primary, but then we have to start to take the economics of the process into account," says Arne Nordmark a scientific colleague of Anne Kvithyld.

The story behind the process

It was Christian Simensen and his colleagues at SINTEF Materials and Chemistry in Oslo who started the project more than ten years ago. Simensen was working for the Australian company Comalco Ltd at the end of the 80s, and was involved in experimental purification in pilot-scale aluminium smelters in a project that involved Comalco and CSIRO, the Australian research council.

"The trials were successful," says Simonsen, "and they produced high-purity aluminium. But Comalco went on to build a full-scale system, which failed because the materials of the furnace were not good enough. The furnace sagged; (delete and) its internal surfaces corroded, and contaminated the melt. Then the company decided to stop the trials.

SINTEF's apparatus differs from Comalco's, but the principle involved is identical; the Pattinson process which was invented in 1833.

The same stuff

Anne Kvithyld's speciality is aluminium recycling. She knows that we can never get rid of impurities in a metal; whatever goes into a purification process can only be separated out in different ways.

"It is just the same with the process that we are working on just now; if we put in aluminium that contains 2% iron, there will still be 2 percent in the smelt when it emerges, but we can choose for equal amounts to take out 3.9% iron on the "dirty" side and 0.1% on the clean side, or 3.5% on the dirt side and 0.5% on the clean side. It all depends on how we run the process. We can even set it to produce completely pure recycled aluminium. But for the time being, the most important thing is to upgrade the aluminium to a quality that can be reused, and the idea is to separate out just enough of the impurities it contains."

Testing in the spring

The project group is now in its last year of self-financed research, with additional support from the Research Council of Norway and Hydro. In the big test in May, 50 kilos of contaminated aluminium will be poured into the apparatus, taken out at the clean/dirty end and subsequently analysed.

"The next steps will depend on the results of the test," says Arne Nordmark. "One thing we will have to do is to find the correct balance between the degree of purity and the time need to achieve it. The longer we run the process, the purer will be the end-product."

"Yes, we are excited," adds Anne Kvithyld, and it would certainly be nice if, after the test is finished,, we could say like Gyro Gearloose: "As you can see- it is dead easy!"