Gas—solid cyclones

Cyclone separators are also frequently used for gas-liquid separation. They can be designed using the same methods for gas-solids cyclones. The inlet velocity should be kept below 30 m/s to avoid pick-up of liquid form the cyclone surfaces. 

Hydroclones were introduced in 1891 to remove sand from water. They function like a gas-solid cyclone, have no moving parts, and rely on centrifugal force for separation, clarification, and dewatering processes. Hydroclones find use in concentrating slurries, in classifying solids in liquid suspensions, and in washing solids. They may be used alone or in conjunction with clarifiers, thickeners, or filters (Besendorfer, 1996). 

Fig. 10.5. A schematic representation of a circulating fluidized bed. The CFB loop consists of a riser, gas-solid cyclone separators, standpipe type of downcomer, and a non-mechanical solids flow control device. Reprinted from [82] with permission from...

The liquid-solid hydrocyclone, shown schematically in Fig. 3.4-3, functions like a gas-solid cyclone. The hydrocyclone is also known as a hydroclone. The primary independent parameters that influence the ability of a hydrocyclone to make a separation are size and geometry of the hydrocyclone, particle size and geometry, solids loading, inlet velocity, split between overflow and underflow, density differential, and liquid viscosity. A reasonable estimate of Ae particle cut diameter (50% in underflow and overflow) d o) is given by the following dimensionless relationship, developed initially by Bradley ... 

Figure 14.4-6. Gas-solid cyclone separator (a) side view, (b) top view.

Gas-solid Cyclones Depends on density diflercnces. Solid particles thrown to wall Recovery of fine solids from gas streams... 

The method we wish to present here for modeling the performance of vapor-liquid ( demisting ) cyclones follows closely the method presented by Muschelk-nautz and Dahl (1994) and that presented for gas-solids cyclones previously reported in Chap. 6. All the same, to avoid repetition of the formulism, here we shall focus on pointing out differences in the two methods. The reader is encouraged to refer back to Chap. 6 while reading the discussion below. 

The calculation method reported below is rather rudimentary in comparison to that which we have reported earlier for gas-solids cyclones. There is still substantial room for further refinement in the modeling of vapor-liquid cyclones. 

As was the case for gas-solids cyclones, we begin by computing the entrance constriction coefficient, a, from Eq. (6.1.1). If, as is often the case for vapor-liquid cyclones, the entrance duct is a circular pipe, the width variable, b,... 

In analogy to our gas-solids cyclones, the amount of liquid that the gas phase can hold in turbulent suspension upon its entrance into a cyclone depends on the mass average drop size of the feed, (x), the cut-point of the inlet wall region , x oin, md, to a lesser extent, the inlet loading itself, Cq. For gas-liquid cyclones, Muschelknautz and Dahl (1994) report for the limit-loading concentration ... 

As we reported for gas-solids cyclones, the overall or total efficiency for gas-liquid cyclones under mass loading or saltation conditions also includes a... 

Very few correlations are available for predicting the inlet-to-underflow pressure loss. This is also the case for gas-solids cyclones but is especially true of gas-liquid cyclones. The Muschelknautz method described in Chap. 6 for gas-solids cyclones may be used for rough estimation purposes if one substitutes the friction factor for gas-liquid cyclones (see Chap. 13) in place of the gas-solids friction factor. However, as a word of caution, the writers have observed that this correlation tends to overpredict actual pressure losses. Clearly, neglecting the inlet-to-underflow pressure loss altogether in Eq. (14.4.5) will lead to a conservative estimate of the minimum submergence, H,... 

Gas-solid cyclones exhibit not only strongly unsteady flow behavior (Derksen et al, 2008) but also another type ofselGorganization the particles collect on the inner walls of the cyclone in downward-driven ropes, as illustrated by the computational large-eddy simulations (LESs) results of Derksen which are confirmed by visual observations (Fig. 8). 

Figure 8 The phenomenon of roping in a gas-solid cyclone. Left external view of a cyclone with two types of particles exhibiting roping as found in a two-way coupled large-eddy simulation by Derksen et al (2008)—figure not published before right picture of three gas-solid cyclones operated in parallel.

Derksen (2003), Shalahy et al (2008), and De Souza et al (2012) tracked point particles in a gas flow field obtained by means of LESs with the view of studying separation performance of gas—solid cyclones, while Derksen et al... 

---