A design of experiments investigation into the processing of fine low specific gravity minerals using a laboratory Knelson Concentrator. (May 2019)

• Record Type: Journal Article
• Authors: Marion, C.; Langlois, R.; Kökkılıç, O.; Zhou, M.; Williams, H.; Awais, M.; Rowson, N.A.; Waters, K.E.
• Is Part Of: Minerals engineering. Volume 135(2019)
• Page Start: 139
• Page End: 155
• Publication Date: 2019-05
• Subjects: Gravity separation -- Knelson Concentrator -- Response surface method -- Central composite design; Mines and mineral resources -- Periodicals; Ressources minérales -- Périodiques; Mines and mineral resources; Periodicals; Electronic journals; 622
• DOI: 10.1016/j.mineng.2018.08.023
• Languages: English
• ISSNs: 0892-6875

Abstract

Highlights: Fine, synthetic ore processed using a laboratory Knelson Concentrator. DOE used to determine optimum operating conditions. Feeds with three different grades (5%, 10% and 15%) were investigated. Optimal operating conditions are dependent on feed grade. Grade and recovery can be optimized simultaneously for high-grade feeds. Abstract: Fine grinding, required to liberate valuable minerals in finely disseminated deposits, creates significant challenges for beneficiation. For these deposits, traditional gravity separation techniques are often ineffective, and centrifugal separators, such as the Knelson Concentrator, are required. The Knelson Concentrator is well established for treating gold ores, and due to its relatively low cost and small environmental impact when compared to other separation techniques, it has become an active area of research for the processing of lower specific gravity (SG) minerals. This work investigates the optimum operating conditions when processing fine (−53 μm) low SG material using a laboratory Knelson Concentrator. A synthetic feed comprised of magnetite (SG 5.2) and quartz (SG 2.65), with grades of 5%, 10% and 15% magnetite, was used to mimic a low-density ore. Central composite design was used to design the experiments and response surface method was used for optimization, with the experimental variables being bowl speed (G), fluidizing water rate (L/min) and solids feed rate (g/min). The results indicate, for 5% and 10% magnetite feeds, that bowl speed impacts concentrate grade negatively and heavy mineral recovery positively, while the fluidizing water rate has an opposite effect on separation. A trade off between grade and recovery must therefore be made when processing this material. When processing the 15% feed, maximum concentrate grade and magnetite recovery were achieved at high bowl speeds and low fluidizing water rates.