With increasing steel production, an exponential increase in demand for steel is rapidly depleting the high-grade iron ore deposits. Hence, in search of an alternative source of iron ores, the steel industries are forced to use low-grade ores. The large amount of iron ore slimes accumulated in slime ponds at mine sites also poses a challenge for the environment. Iron ore slimes typically consist of iron, alumina, and silica at a wide range of sizes and densities. The marginally varying density components (iron oxide and alumina particles) of iron ore slimes can be separated by hydrocyclones based on size and density. Therefore, this work is aimed to experimentally study the iron ore slime beneficiation using a novel 100 mm hydrocyclone (with small cone angle, tapered vortex finder and optimum cylindrical length) design developed by computation fluid dynamics (CFD) simulation studies. As the design is developed using numerical studies, the initial experiments are conducted with mono density silica particles to test and validate the performance. Further, the experiments are continued with iron ore slimes. The novel hydrocyclone is able to provide desired cut size and sharpness of separation required to separate iron from alumina and silica. It also enhances the classification by reducing coarse and fine particle misplacement compared to the conventional hydrocyclone design. The product obtained from the underflow of the optimised hydrocyclone operation is suitable for direct pellet grade feed with iron and alumina as ∼61% and 2.77%. © 2022 Elsevier Ltd