Examination of Ultrasonic Treatment of Iron Ore Fines Using Automatic Iron Ore Texture Classification

In order to correctly predict the performance of beneficiation processes, such as hydrocycloning or ultrasonic treatment of iron ore fines, it is important to understand possible changes in size distribution, mineral composition and textural peculiarities. During beneficiation processes, iron ore can be subjected to various physical actions which may result in changes in size distribution. In the case of soft or friable ores, attritioning and/or deagglomeration may have a significant effect on both separation performance and product stream specifications. Optical image analysis tools, which give statistical information on a particle by particle basis, can help in better understanding the response of the ores to these processes. An experimental study has been conducted to investigate the response of a haematitic-goethitic iron ore to a combination of stirring and ultrasonic treatment. The effect of stirring and ultrasonic treatment on the size and mineral distribution has been quantified using image analysis and automatic ore texture classification. This approach has been applied to several size coarse fractions of the initial feed and product. The results show that, following treatment, the mass proportion of ore texture classes which contained higher amounts of ochreous goethite and kaolinite had significantly decreased compared with that in the feed, while the proportion of texture classes where haematite was the major mineral had increased. Liberation analysis also showed a significant increase in the proportion of particles with more than 95 per cent haematite and vitreous goethite.