Gas Hold-Up and Volumetric Liquid-Phase Mass Transfer Coefficient In Solid-Suspended Bubble Columns With Draught Tube

The solid suspended-bubble colum ns with draught tube are widely used as a three-phase slurry reactor in industrial chemical process such as absorption, biochemical reactions, coal liquefaction. etc. To design such a column the average gas holdup ( g e ) and the volumetric liquid -phase mass transfer coefficient (K La ) should be known. This study includes the effect of gas velocit y, liquid-phase properties, solid-particles concentration and the static liquid height on both ( g e ) and (K La ). It was found that both ( g e ) and (K La ) increase with increasing gas velocity and decrease with increasing solid particles concentration, static liquid height, viscosity and surface tension o f liquid-phase.


1-Introduction
Bubble column reactors belong to the general class of multiphase reactors which consist of three main categories namely, the trickle bed reactor (fixed or packed bed), fluidized bed reactor, and the bubble column reactor (4) .Slurry columns are similar to fluidized bed columns, in that a gas is passed through a column containing solid catalyst particles suspended in a fluid.In slurries the catalyst is suspended in a fluid, in fluidized beds the suspending fluid is the reacting gas itself (1) .In (sc) an attempt is made to realize intensive and intimate contact between a gas -phase component and a finely dispersed solid.With respect to this purpose (sc) are related to packed bed columns with the different gas-liquid flow regimes that can be realized (such as trickle flow, pulsed flow, dispersed bubble flow etc….).Also there is a lot of similarity with three -phase fluid bed systems.The latter systems share many properties with (sc), but the main difference is the fact that in fluid beds with upward fluid flows the drag force acting on the solids by the gas and liquid flow is on the average balanced by the net weight of the particles, while in (sc) the overall liquid -solid slip velocities are practically zero and particles remain suspended by the action of the turbulence, in the liquid phase (2,3) . .Deckwer and schumpe (5) studied the effect of various design parameters.They found that mixing time decreases initially with increasing gas velocity and then increases in the higher range of velocity.Guerin et.al (6) studied the effect of gas flow rate on mixing time .They found that the mixing time dose not decrease proportionally with the increase of gas flow rate, and this is a main difference between BC and mechanically stirred columns.Solidmixing and solid replacement are important factors in cases where the solids have a short life-time (7) .Fan and chern (8) studied solid-mixing in a gasliquid-solid system .They reported three-states of solid mixing, these states are complete segregation, partial intermixing and complete intermixing.
Particle size in (sc)can be small to very small ,even down to the submicron range.The effect of average particle size (dp) and solid density on the critical superficial gas velocity (Vgc) is given as (9) :- The pressure drop in (sc) is usually more or less independent of the gas flow and close to the hydrostatic pressure.Of course there is also a pressure drop required for the gas distributor.
Slurry columns can be classified according to the phases where the reactant are present.Generally (sc) can be classified according to (10):- a.The chemical system.b.The contacting pattern and mechanical hold -up-fractions devices.
Hydrodynamics of (sc) includes the study of mixing and the volumetric liquid side mass transfer coefficient at the gas-liquid interface (K La ).For the design of (sc), whether agitated only by the flowing gas or assisted by one or more stirrers, the conditions at which the particles are just suspended are very important.
Therefore, generally only a minimum suspension criteria (M.S.C) is considered (2) .Roy et.. al (11) studied by applying the pressure drop technique a large variety of gas solid and liquid systems including nonaqueous systems and particles with different degree of wetting.Kato et.al (12) studied the (M.S.C) for different columns heights.Koide found that the effect of solid particles on reducing g ε value in the transition regims is larger than that in heterogeneous regime.(Fuku et.al (13) studied the (M.S.C) for a column with a draft tube, they showed that with a draft-tube applied in the system, a much smaller volumetric flow rate of gas was required to suspend tha same amount of solid, compared with normal column.Narayanan et.al (14) The average hold-up fractions of gas,solids and liquid should satisfy the eqution: In contrast to three-phase fliudbeds where the relation between the three-phase hold ups can be rather complex, in slurry columns with the much smaller particles and slip velocities, the relation between( ε l ) and (є s ) is often simple as it is fixed by the feed ratio of solids liquid phases, or liquid and solids volumes are constant (in batch systems).
The bubble hold-up is much more difficult to predict, first of all because of the different regimes that might prevail both in stirred vessels and slurry sparger columns.Shah et.al (15) studied the regimes in three-phase flow sparger.They showed that there are three regimes.The regimes are:-1-Uniform bubbling at low gas velocity.
2-Churn-turbulent-flow at higher-gas velocities, with a mixture of large and small bubbles.3-Slugging in small diameter columns, where the largest bubbles are comparable to the column diameter.Koide et.al (16) studied the three phase flow with different solid concentrations, They showed that the effect of solid particles on reducing ε g value in the transition regimes is larger than that in heterogenous regime.Akita et.al (17) studied the gas hold up in BC and proposed a correlation for gas hold up in terms of bond and Galilo numbers.
( ) (18) studied the gas hold up in three phase system and proposed for churn turbulent regime in three phase fluidization to use the Akita , s equation using liquid/solid suspended bulk properties instead of liquid properties.
Koide and Horibe (19) studied the effect of using solids in draught tube in solid suspended bubble column.For an adequate description of mass transfer with chemical reaction in slurry columns, reliable data on the following two types a-parameters which are specific for slurry columns (K G, K L , K La , K Ga ).b-parameters which are not specific for the type of reactor applied (intrinsic reaction kinetics). (2)kita et.al (17) showed that a better empirical equation of (K La ) can be obtained by using (εg) instead of (V g ) in the BC: Koide and Horibe (19) studied the bubble column with draught tube.Koide and Iwamoto (20) studied BC with draught tube and with gas dispersion into annulas, the values of (K La ) and (ε g ) is much larger than those in the bubble column without draught tube, when a liquid with forthing ability is used.

2-Purpose:-
The purpose of this study is to clarify experimentally the effects of,gas velocity and properties of the liquid and solid particles on ε g and K La in the solid suspended bubble column with adraught tube in liquid solid batch operation.

3-Experimental Section
A schematic diagram of the experimental set-up used in this work is shown in Figures Aplexiglass column of 0.15 m inside diameter and about 1.60 m total height with draught tube dimensions of 0.075 m inside diameter and 1.40 m total height was used.
The draught tube was fitted with three support legs at the upper and the lower end of the column so as to locate it in central position at any distance above the base.
The column consists of two main sections namely, the gas inlet section and the liquid recycling testing section.The gas inlet section consists of a gas distributor.
At the bottom of this section, two lines are connected together before entering the distributor section each line has a valve to be opened or closed as required.One of these lines is the air inlet flow.
Air compressor supplied the line with the desired amount of air needed, for the experiment, the amount of air was measured using a gas meter, and two calibrated rotameters connected in parallel were used to measure the air flow rate.
The other line is the nitrogen gas inlet flow.It was supplied from a cylinder.
A gate valve was used in the nitrogen flow, which must be shut off when air was sparged to the column, and must be opened during desorption process.
The liquid testing section contains two openings, one for liquid out-flow and the other for liquid in flow.
The circulation of liquid in the column was achieved using a centrifugal pump placed in the recycling line.A ball valve placed in the middle of the recycling line was used to take various samples at various times to measure the concentration of the dissolved oxygen during the operation.The column was filled with water to the desired height above the distributor(0.5, 1,1.4) m.Then the solid particles (plastic-non porous) were added to the liquid in the column.The concentration of solid particles to each level of static liquid were (30,60,100)kg/m3 respectively.
PDF created with pdfFactory Pro trial version www.pdffactory.comThe desired air flow rate was set-up using needle valve and the amount measured with a gas meter.
The dissolved oxygen concentration in the liquid phase was measured using oxygen meter, which consists of a gold metal electrode.The liquid phase (batch) consists of the following systems (only water; water and solids; water, alcohols and solids; water, NaCl and solid; water , aceticacid and solids ).
The gas distributor Fig ( 1.2) was constructed from a ceramic material and the type is multi hole tuyere.
The distributor has equivalent pore diameter of 0.15 mm and free section of 0.61%.

4-Resalts on gas hold up ( ε g ) and solid hold up(ε s )
The averge gas hold up ε g was calculated from eqution (1) using the data of the static slurry height(H l ) and the height of the aerated slurry(H f ) which were determined by visual observation .....
The solid hold up was calculated from equation(2) using the data of static liquid height(H l ) and the height of slurry after adding solid particles The experimental gas hold up was found by measuring the difference between initial liquid height and final liquid height.Since it was rather difficult to read directly the level of the aerated liquid the values of gas hold up thus obtained probably involves an error of about 5%, established via repeated measurements.Fig (2) shows the effect of gas velocity for water system with and without solid particles.The gas hold up was found to increase with increasing gas velocity ,because the large bubble holdup increases with increasing gas velocity leading to the increase of the overall holdup.But the effect without solid particles is larger than that with solid particles.
When the gas velocities above 0.01m/sec and the liquid phase without solid particles, bubble size is independent of gas flow rate.The buoyant force of the bubble is opposed by viscous drag of the fluid, and if bubble size is independent of gas flow rate, the velocity of bubble rise will show similar independence.
Hence increased gas flow rate will increase the gas hold up ,and ultimately the liquid will seen to be filled with bubbles.Fig (3) show,in slurries the presence of solid particles in the liquid might enhance bubble coalescence into larger bubbles and therefore reduces the value of gas hold up.
The solid particles retard the bubble rise velocity and prevent increase in bubble size.Fig( 4) show, the effect of the static liquid height on gas hold up.It is clear that increasing the static liquid height will decrease the gas hold up.As the static liquid height is increased , the bubbles have time to coal further and ultimately reduce gas hold up.
PDF created with pdfFactory Pro trial version www.pdffactory.comIn our experimental studies the maximum static liquid height used was 1.4 m.In slurries the the presence of solid particles in the liquid might enhance bubble coalescence into larger bubbles and therefore reduce gas hold up .
Fig (5) shows the effect of using different liquid phase (alcohols and electrolytes) respectively,on gas hold up.Low electrolyte concentrations have no noticeable effect on the surface tension of the solution.However the ionic forces in the liquid bulk reduce the bubble rise velocity and the bubble coalescence.As a result, the gas hold up increase.For high electrolyte concentration,the interfacial tension increases,resulting in increased bubble size and reduce gasholdup.

5-Mass transfer coefficient results.
The physical absorption of oxygen in the air by the liquid was employed to determine the mass transfer coefficient.Amaterial balance of oxygen in the liquid gives (23) :- ) Plotting the left hand side of equation ( 4) with(t),the avarge slope of the plot will give the term ( ) The values of (ε g ) and (ε s ) were determine as mentioned in ( 1) and ( 2) respectively, then (k La ) can be calculated .Fig (6) shows that the mass transfer coefficient increase with increasing gas velocity.The axial dispersion coefficient (D L ) increases with increasing gas velocity and therefore increase (k La ).
Fig (7) shows the effect of solid particle concentration on (K La ).The presence of solid particle in the liquid will decrease the axial dispersion coefficient and it enhance bubble coalescence.The bubble size will be larger and occupying larger space in the column and therefore reduces (K La ).At higher gas velocities (0.1-0.2) m/sec, the effect of solid particles on (K La ) will be less than in low gas velocities (0.03m/sec).Fig (8) show the effect of static liquid height on mass transfer coefficient.
As the static liquid height is increased, however the bubbles have time to coalesce further and ultimately decreases the axial dispersion coefficient and the mass transfer coefficient.Fig (9) shows the effect of liquid phase properties on (K La ).As mentioned before, the ionic forces in the liquid bulk reduce the bubble rise velocity and the bubbles coalescence, so that the mass transfer coefficient is increased.6-Conclusions 1.The gas hold up and mass transfer coefficient increase with increasing gas velocity, for Vg≤ 0.2m/sec.2-The addition of solid particles reduces both gas hold up and mass transfer coefficient.3. Increasing the static liquid height reduces both gas hold up and mass transfer coefficient.Testing-section

Table ( 1) physical properties for pure liquids at T=25 O C Table (2) Physical properties for mixtures used with concentration 20% at T=25 O
CPDF created with pdfFactory Pro trial version www.pdffactory.comEng.& Tech.

Gas hold up for C s =0 Kg/m 3 Gas hold up for Cs=60 Kg/m 3 Figure (2) Gas hold up vs gas velocity for water system
PDF created with pdfFactory Pro trial version www.pdffactory.comEng.

Figure (3) Gas hold up vs solid concentration for water system for various gas velocities
PDF created with pdfFactory Pro trial version www.pdffactory.comEng.

Figure (4) Gas hold up vs static liquid height for various gas velocities
PDF created with pdfFactory Pro trial version www.pdffactory.comEng.

3 Figure (5) Gas hold up vs gas velocity for different liquid phase systems
PDF created with pdfFactory Pro trial version www.pdffactory.comEng.