Mass transfer dispersed systems, hydrodynamics

Elperin, I., Enyakin, Yu, P. and Meltzer, V. (1968). Experimental investigation of the hydrodynamics of impinging gas-solid particles streams. Heat and Mass Transfer in Dispersed Systems, (5) 454-496. 

The elimination or estimation of the axial dispersion contribution presents a more difficult problem. Established correlations for the axial dispersion coefficient are notoriously unreliable for small particles at low Reynolds number(17,18) and it has recently been shown that dispersion in a column packed with porous particles may be much greater than for inert non-porous particles under similar hydrodynamic conditions(19>20). one method which has proved useful is to make measurements over a range of velocities and plot (cj2/2y ) (L/v) vs l/v2. It follows from eqn. 6 that in the low Reynolds number region where Dj. is essentially constant, such a plot should be linear with slope Dj, and intercept equal to the mass transfer resistance term. Representative data for several systems are shown plotted in this way in figure 2(21). CF4 and iC io molecules are too large to penetrate the 4A zeolite and the intercepts correspond only to the external film and macropore diffusion resistance which varies little with temperature. 

Effective rates of sorption, especially in subsurface systems, are frequently controlled by rates of solute transport rather than by intrinsic sorption reactions perse. In general, mass transport and transfer processes operative in subsurface environments may be categorized as either macroscopic or microscopic. Macroscopic transport refers to movement of solute controlled by movement of bulk solvent, either by advection or hydrodynamic (mechanical) dispersion. For distinction, microscopic mass transfer refers to movement of solute under the influence of its own molecular or mass distribution (Weber et al., 1991). 

Advantages of three-phase fluidized beds over trickle beds and other fixed bed systems are temperature uniformity, high heat transfer, ability to add and remove catalyst particles continuously, and limited mass transfer resistances (both external to the particles and bubbles, because of turbulence and limited bubble size, and inside the particles owing to relatively small particle diameters). Disadvantages include substantial axial dispersion (of gas, liquid, and particles), causing substantial deviations from plug flow, and lack of predictability because of the complex hydrodynamics. There are two major applications of gas-liquid-solid-fluidized beds biochemical processes and hydrocarbon processing. 

In Section 2.9, various aspects were considered of the hydrodynamics of a constrained flow past a system of particles based on the cell model. Here we briefly describe mass and heat transfer in such systems at high Peclet numbers. We investigate either sufficiently rarefied systems of particles or systems with an irregular structure, where the diffusion interaction of isolated particles can be neglected. (Regular disperse systems, where the interaction between diffusion wakes and boundary layers must be taken into account, were investigated in [172, 365].)... 

The authors hope that the book will be useful for researchers and engineers, as well as postgraduate and graduate students, in chemical engineering science, hydrodynamics, heat and mass transfer, mechanics of disperse systems, physicochemical hydrodynamics, power engineering, meteorology, and biomechanics. 

Nikitin, V.S. and Puchkov, G.F. 1983. Hydrodynamics of vibro-rotational bed. In Heat and Mass Transfer in Dispersed Systems, Minsk, AN BSSR, pp. 116-123 (in Russian). 

The models of van Baten and Krishna (2004) and Vandu et al. (2005), for gas-Uquid bubble flows, showed little or no agreement with the experimental results. Van Baten and Krishna (2004) developed their model (Eq. 7.1.1) over a wide range of parametric values (ID = 1.5-3 mm, Luc = 0.015-0.05 m). Their model underestimated the current mass transfer coefficients for all the channels. It is worth noting that in this work the length of the unit cells (Luc) and the velocity of the dispersed phase (Up) were one order of magnitude lower than those used by Van Baten and Krishna (2004). In the model by Vandu et al. (2005) (Eq. 7.1.2), which was evaluated for channel sizes from 1 to 3 mm ID and unit cell lengths from 5 to 60 mm, the only contribution on the mass transfer coefficient is by the film. The kuu obtained for 0.5 and 1 mm ID channel seem to fall within the predictions of their model (for C = 8.5), whilst mass transfer is underestimated in all cases for the 2 mm ID channel with a relative error from 40 to 60 %. The discrepancies between the experimental results and the gas-liquid models may be attributed to the more complex hydrodynamics in the liquid-liquid systems. In addition, there is less resistance to mass transfer by diflusion within a gas plug compared to a liquid one. 

Hydrodynamics and Heat or Mass Transfer in Finely Dispersed Systems... 

Indeed, as for hydrodynamics, mass transfer depends strongly on the physico-chemical properties of the gas-liquid system and many correlations have been proposed to predict the interfacial areas a and liquid mass transfer coefficient kLa, reported to the unit volume of dispersion. They have been recently reviewed by Botton et al. (97) and Hikita et al. (111). It seems that for the scale-up prevision in bubble flow regime (u <0.3 m/s), small scale experiments with the system of interest will allow scale-up on the basis of equal superficial velocity of the gas. So the data in Fig. 17, or those found in the many literature references, or of specific experiments can be used noting that a, k a, k a and a vary approximatively as For other flow regimes and for... 

---