- Title
- A novel sink-hole fluidization method for dry separation of 1-10 mm particles at cut points above 2500 kg/m³
- Creator
- Kumar, Deepak
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2020
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Increasing water restrictions make it desirable to increase the use of dry processes in mineral beneficiation. Air dense medium fluidized beds (ADMFBs) are one of the most effective methods of dry separation, especially for fine coal and minerals applications. However, dry processes usually struggle to separate particles as efficiently as wet processes, especially for particles finer than about 10 mm. This research was concerned with the separation of 1-10 mm particles of density between 2.4 and 4.8 RD. Experiments were conducted in two phases. The Phase 1 device consisted of a 1 mm mesh screen above a 200 mm diameter dense medium fluidized bed, mounted on top of a Kason vibratory device with the help of struts. The screen mesh above the fluidized bed contained a centrally located opening, which was large compared to its 1 mm aperture, referred to as a “Sink-Hole”. The experiments in this phase were performed using 2.0-2.8 and 2.8-4.0 mm tracer particles using nominal 200 μm volume median diameter sand or a 85:15 mix of sand with hematite as the fluidized medium with a single centrally located 13 mm diameter Sink-Hole. The performance of the system was monitored to separate the particles based on their density. However, there were reproducibility issues with the Phase 1 system which lead to changes being made. The Phase 2 device consisted of a pair of 1 mm aperture screens above a 200 mm diameter dense medium bed. These were directly mounted on top of the same Kason vibratory device as in Phase 1, but this time it was set in a different vibration mode that better assisted the transport of the particles across the Sink-Hole. The screen on top contained a centrally located Sink-Hole. The secondary screen mounted 15 mm below the top screen improved the bed homogeneity by suppressing the bubbles within the fluidized bed, and also enabled the underflow particles to be more easily captured and counted. Again, the experiments in Phase 2 started with a 13 mm diameter Sink-Hole. However, during Phase 2 experiments, the diameter of the Sink-Hole was also varied from 13 to 60 mm. Most of the experiments were again performed using the sand of nominal median size 200 μm, skeletal density 2643 kg/m³ and Umf of 2.5 cm/s. Tracer particles ranging from 2.0 to 8.0 mm in size and 2400 to 4800 kg/m³ in density were used to measure the partition performance. Different operating conditions were explored. The optimum (lowest Ep) operating conditions were found to be at superficial velocities of 2.5-3.0 times the minimum fluidization velocity, with the fluidised bed just reaching the level of the Sink-Hole. Sand flowed upwards through the Sink-Hole and then spread out and percolated back through the surrounding mesh. The system was found to reach a dynamic equilibrium separation state, in which sometimes particles that had previously passed through the Sink-Hole would re-emerge onto the upper screen. Robust performance with relatively sharp separations (Ep values below 0.2 RD) was obtained over a wide range of conditions, and results were reproducible within the expected 95 % confidence intervals. Increasing the Sink-Hole size from 13 to 60 mm in diameter improved the separation kinetics, causing the time to reach equilibrium to decrease from about 300 s down to less than 20 s. Either increasing the Sink-Hole diameter or the superficial air velocity under the same operating condition lowered the separation density. Remarkably, the separation densities covering the range 2.5 to 4.0 RD, were higher than the medium density and, invariably higher even than the skeletal density of the medium sand particles. Furthermore, over the range of particle sizes from 2.8 to 8.0 mm, the separation density was fairly insensitive to the particle size, a useful feature for a potential beneficiation device. These features indicate that the separation mechanism at work was very different to the buoyancy mechanisms that determine the separation performance of conventional dense-medium fluidised beds. To further improve the understanding of this intriguing behaviour, a transparent two-dimensional rig was constructed and used to observe the granular flow. A column of upwards flowing material was observed to form beneath the Sink-Hole, extending upwards from a considerable distance below the bed. This observation might explain how particles that had previously sunk through the Sink-Hole could re-emerge at later times. It is speculated that the material in this column was in a critical jammed state part way between that of a packed bed (able to support large loads) and a fluidised state (unable to support loads from particles denser than the bed bulk density). Theoretical modelling using Discrete Element Modelling (DEM) would likely illuminate further this proposed mechanism based on a jamming condition between the particles. Further work is also needed to explore the potential to achieve scaled-up to a commercial-scale continuous device.
- Subject
- mineral beneficiation; dry processes; sink hole; fluidization
- Identifier
- http://hdl.handle.net/1959.13/1424255
- Identifier
- uon:38042
- Rights
- Copyright 2020 Deepak Kumar
- Language
- eng
- Full Text
- Hits: 648
- Visitors: 1103
- Downloads: 489
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Thesis | 32 MB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Abstract | 5 MB | Adobe Acrobat PDF | View Details Download |