 |
 |
 |
 |
 |
|
|
address: |
DelftChemTech |
|
room: |
Proeffabriek (0.020) |
|
|
telephone: |
015-2784753 |
|
|
fax: |
015-2785006 |
|
|
e-mail: |
||
|
http: |
www.dct.tudelft.nl/pcmt/html/christensen.html |
Controlling fluidised bed dynamics with fractal gas injection systems
supervisor: prof.dr.ir. M.-O. Coppens
sponsor: NWO (the Netherlands Organization for Scientific Research)
When a fixed bed of solid (catalytic) particles is suspended in a sufficiently large gas flow, a process called fluidisation sets in; the bed initially expands homogeneously, but, as the gas flow is further increased, bubbles form at the bottom of the bed and grow as the gas moves upward.
Whereas the strong mixing in fluidised
beds leads to the high mass and heat transfer required for fast chemical reactions, the formation of bubbles is unfavourable. This is because, inside the bubbles, there is no adequate contacting between the (catalytic) solids and the (reactant) gas. Another major disadvantage of fluidised beds is that they are difficult to scale from small laboratory set-ups to full-size industrial installations.
All these difficulties arise from the chaotic movements of the solid bed particles, which makes the behaviour of the bed irregular and difficult to predict. To substantially improve on this situation, we propose to control the bed dynamics with an extra gas injection system having fractal geometry.
We often find branched or tree-like 
structures in nature, for example in lungs, trees, or riverbeds. Mathematically, such structures can be described with fractals, which repeat a basic pattern (like a Y-shaped split) on increasingly smaller scales. This can lead to complex structures that efficiently distribute themselves over a given volume. In addition, such structures are similar on each scale and are therefore by definition easy to scale-up or -down. Hence, by controlling the bed dynamics via a fractal injection system, the bed should operate more uniformly and should be easier to scale.
Preliminary results on small 2-dimensional fluidised sand beds indicate that a secondary injection system with a simple T-shaped fractal geometry is indeed already capable of (1) inducing a more regular bubble pattern and (2) reducing the average bubble size. Future work includes more rigorous experiments on these systems using video image analysis, and measurements on pressure fluctuations and retention time distribution at different positions in the bed. Also, a set-up has been built for measuring the effects on the conversion of a fast chemical gas reaction in a bed filled with catalytically active particles.
Physical Chemistry and Molecular Thermodynamics
Julianalaan 136, 2628 BL Delft, The Netherlands, Phone: +31 15 2784748, Fax: +31 15 2788047,
Last update: 12 July, 2006 webmaster