Sensor-based sorting of spent refractory bricks

Refractory materials are essential for all high-temperature applications in industry. Depending on the position in the furnace and specific process requirements, refractories are produced from several raw materials, such as dolomite, magnesite, bauxite, graphite or chromite. Most of these raw materials are not rare but strict quality requirements limit the producers of refractories to only a few deposits in the world. Thus, raw material supply is a critical factor for the production of refractories and accounts for 40-50 % of the total costs of refractories. A possible way to lower dependency on raw material supply is recycling of spent refractories. Main challenge in recycling is the separation of different brick types from each other and the removal of impurities, such as steel or slag. The European FP7 project REFRASORT aims to apply sensor-based sorting for separation of spent refractories. Sensor-based sorting is known from food industry, recycling and mineral processing and has proven its reliability. Nevertheless, some adjustments are necessary compared to applications in mineral processing because of other sensor-types and a higher number of products in one sorting step. This paper presents the results of feasibility tests for sensor-based sorting of refractories with near infrared spectroscopy and laser-induced breakdown spectroscopy (LIBS). Both sensors showed the potential to identify different brick types but the most promising results were obtained using LIBS technology.