- Title
- Influence of narrow inclined channels on fne particle separations
- Creator
- Walton, Kelly Joel
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2012
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- The Reflux Classifier is an industrial technology that separates particles on the basis of size and / or density. It consists of a conventional vertical fluidised bed with a section of parallel inclined channels positioned above. The research described in this thesis is an examination of the effects of using “narrow” inclined channels (with widths as fine as 1.77 mm) on the performance of a Reflux Classifier. The finest channels studied prior to this work were nominally 7 mm wide. The understanding of particle elutriation from these relatively wide inclined channels was developed by Laskovski et al. (2006). They noted that there was an increased preferential re-suspension of low density particles as the channel spacing narrowed. However, this re-suspension phenomenon was not fully investigated, thus, the mechanisms for this re-suspension were not fully understood or accounted for in their generalised correlation. Experiments were performed in laboratory-scale vessels, with 1 m long channels, inclined at 70° to the horizontal. Channel widths from 18 mm down to 1.77 mm were investigated. Batch experiments were performed using well-defined feeds with particles of a single density (PVC, glass ballotini and ilmenite) and also using an industrial coal sample. It was found that: 1) relatively narrow channels produced a previously unobserved size suppression effect; and 2) the maximum particle size to channel width ratio for optimum performance was determined to be 0.33. A theoretical model, with no adjustable parameters, was developed which successfully predicted and explained the observed results. The main cause of the suppression of size effects is that when laminar flow occurs in narrow channels, the effective fluid velocity experienced by the particles lying on the channel wall is proportional to their size. Further experiments were performed in continuous mode with simultaneous feed addition, and underflow and overflow removal. A number of different industrial coal feeds were used. Comparing these results against those from the more traditional wider channelled Reflux Classifiers, it was demonstrated that the size suppression effect observed in batch narrow channel operations also occurred in continuous operations. The new narrow 5.5 mm channel Reflux Classifier was capable of processing an extended size range of - 2.0 mm + 0.25 mm while achieving a separation density of 1560 kg/m³ and Ep of 0.062. Where Ep is a measure of how accurately the separation device performed, with an Ep of “0” being a perfect splitting of a stream over a point and an Ep of “1” showing that there was no preferential segregation in the splitting of the stream. The traditional wide 30 mm or 120 mm channel Reflux Classifier could only achieve for a similar separation density and feed size range of - 2.0 mm + 0.25 mm could only typically produce an Ep of 0.15 (55). While the narrow channels operate best at a capacity of 20 t/(m²h), allowing significant separation down to 0.075 mm, increasing the processing rate of the classifier is possible, though at a diminished capacity. At 30 t/(m²h) the narrow channel Reflux Classifier will competently process particles as fine as 0.125 mm and at throughputs of ~ 40 t/(m²h) the channel Reflux Classifier would only be able to process a feed range of - 2 mm + 0.25 mm which is comparable to the maximum capacity of the more traditional wide channel Reflux Classifier.
- Subject
- reflux classifier; inclined channels; particles; fluidization
- Identifier
- http://hdl.handle.net/1959.13/932170
- Identifier
- uon:11274
- Rights
- Copyright 2012 Kelly Joel Walton
- Language
- eng
- Full Text
- Hits: 1077
- Visitors: 1256
- Downloads: 341
Thumbnail | File | Description | Size | Format | |||
---|---|---|---|---|---|---|---|
View Details Download | ATTACHMENT01 | Abstract | 250 KB | Adobe Acrobat PDF | View Details Download | ||
View Details Download | ATTACHMENT02 | Thesis | 2 MB | Adobe Acrobat PDF | View Details Download |