Using Laboratory Scale Flotation Testing to Predict Pilot Scale Flotation Performance

"While methods exist to allow the simulation and optimization of existing flotation operations, there is a deficit in the ability to predict the performance of a new ore type pre-processing. A laboratory scale batch flotation testing methodology has been developed to assess the flotation potential of ores (known as the JK Floatability Index or JKFI). Previous work conducted by Alexander and Collins (2008) showed that the test could be used to predict full scale results in base metal ores. In this work, JKFI tests were applied to three different types of ore and one blend from a greenfields gold deposit to show the test applicability to flotation of precious metals. The results from the JKFI test on one ore type were then used to predict the response of all ore types (including the blend). To validate the predicted kinetic parameters, a comprehensive survey was performed on a pilot plant where each of ore types was processed in the circuit. An acceptable agreement was found between the experimental recoveries and predicted recoveries which indicate the potential of this method for use in predicting the flotation response of an ore.INTRODUCTIONOver 100 years, flotation has been considered an efficient method to concentrate different types of sulphide and oxide minerals and coal. Thus, numerous research studies have been undertaken to find an appropriate model to predict the behaviour of particles in flotation circuits (Barnwal et. al., 2006; Bascur and Herbst, 1983; Cecile and Villeneuve, 1991; Deng et. al., 1996; Zheng et. al., 2006). Most methods are based on the ore flotation response in an industrial plant which is expensive and time consuming to characterise. It is difficult to reliably predict flotation response for new types of ores in industrial scale plants.The floatability component modelling (FCM) approach, developed within the Australian Minerals Industry Research Association (AMIRA) P9 project, has been used to derive models of operating flotation plants for the purpose of simulation and optimisation (Alexander and Morrison, 1998; Alexander et. al., 2000; Varadi, 2006). This approach has been applied at a large number of operating sites treating a wide variety of ores. The approach requires metallurgical surveys to be undertaken in conjunction with batch flotation tests and cell characterisation. This allows ore and machine parameters to be derived, which are then used in the P9 Flotation Model. The development of a FCM generally requires a large commitment of resources and time. While there has been significant support of this approach in industry, the FCM approach is limited to ores already being processed (Runge, 2004; Wightman, 2006)."