In this project a detailed investigation of electrochemical sensing principles for the detection of certain elements in molten aluminium is proposed
A prototype sensor, based on sodium conducting beta-alumina, was constructed and tested in semi-industrial (50 kg) as well as industrial circumstances (250 kg of aluminium-silicon alloy). The obtained emf values with oxidic reference materials have a maximum scatter of 5 ppm in the range 30 to 150 ppm. Since this accuracy is sufficient for the application involved, further development work must lead to a practical sensor for measuring the sodium content of casting alloys. Moreover, the thermal shock stable beta-alumina ceramic may also be used for other sensor applications.
A potentiometric hydrogen measuring cell for molten aluminium, using gaseous reference, has been constructed and shown to work on laboratory scale. In the 50 kg volumes however, a severe and unwanted offset was observed and hampered the measurements although the sensors clearly responded to hydrogen changes induced by bubbling hydrogen around the sensor. Additionally, a production route of proton conducting ceramics was established and can be used for other applications. Finally, a commercial measuring system was purchased and evaluated. It was shown that, in order to be used under semi-industrial circumstances, a specific methodology was needed and developed.
In order to develop a magnesium sensor a probe based on a molten salt immobilized in a porous MgO membrane was successfully tested under inert atmosphere, as well as in air. The obtained values correspond well with the off-line analyses obtained by classical spectroscopic techniques, the response is instantaneously once the probe has stabilised in the melt and the life-time is at least several hours. A first test of three sensors at Hoogovens Aluminium N.V. yielded reliable and very promising results for two of the three sensors. One of these indicated the expected value, within 0.1 wt% Mg, during the whole casting operation, proving the excellent opportunities for developing sensors based on immobilized salts.
As a general conclusion, it can be stated that the project reported has demonstrated that for all three sensors particular concepts have been designed, constructed and tested, Each of them has sufficient potential to be developed into a commercial product or prototype. Due to the integration of the material development phase into this 'sensors development project', the production routes of the key ceramic elements are available now and yield the strategic basis for further research.