Gas Holdup in a Slurry Bubble Column: Influence of Electrolyte and

It is suggested that a layer of carbon particles around the gas bubbles results in a lower average bubble rise velocity. Both the addition of carbon p...
1 downloads 0 Views 179KB Size
Download PDF
5326

Ind. Eng. Chem. Res. 2001, 40, 5326-5333

Gas Holdup in a Slurry Bubble Column: Influence of Electrolyte and Carbon Particles Jeroen H. J. Kluytmans,* Berend G. M. van Wachem, Ben F. M. Kuster, and Jaap C. Schouten Laboratory of Chemical Reactor Engineering, Eindhoven University of Technology, P. O. Box 513, 5600 MB Eindhoven, The Netherlands

This study deals with the effects of electrolyte and particle concentrations on the gas holdup in both the homogeneous and the heterogeneous flow regimes in a slurry bubble column. Gas holdup measurements and video recordings of the bubble behavior were carried out in a 2D slurry column (0.015 × 0.30 × 2.00 m) under ambient conditions. The additions of electrolyte (sodium gluconate, 0.05-0.2 M) and of solid carbon particles (diameter 30 µm, 0.1-1.0 g L-1) both lead to a considerable increase in gas holdup. In both cases, critical concentrations exist above which no further increase in gas holdup is observed. The transition from the homogeneous to the heterogeneous regime is not significantly affected by electrolyte but is increased by the presence of particles. Three mechanisms are proposed that might account for the gas holdup increase resulting from particle and electrolyte addition. It is suggested that a layer of carbon particles around the gas bubbles results in a lower average bubble rise velocity. Both the addition of carbon particles and the addition of electrolyte lead to bubble stabilization, a decreased rate of coalescence, and thus a higher gas holdup. It is further suggested that the presence of electrolyte changes the surface tension, leading to smaller bubbles, a lower average bubble rise velocity, and thus a higher gas holdup. The combined addition of electrolyte and carbon particles confirms these hypotheses. Introduction Increasingly, bubble columns find their application in chemical industries, for example, in Fischer-Tropsch processes or in biological wastewater treatment. Bubble columns incorporate many advantages such as easy construction, easy operation, and strong mixing of the phases by gas aeration only. The hydrodynamic behavior, governed by bubble-bubble, bubble-particle and bubble-liquid interactions, is still a major research topic, judging from the recent review of Joshi et al.,1 which covers 253 references on bubble column reactors. In the literature, attention to gas holdup prevails because gas holdup affects mixing and mass transfer and, therefore, the performance of the system. Gas holdup is influenced by many parameters, including particle concentration,2-4 electrolyte concentration,5 liquid viscosity, and surface tension. Increase in gas holdup was reported to occur upon addition of electrolyte5 and addition of wettable particles.2-4 Although many mechanisms have been proposed for this gas holdup increase, still there is no agreement as to which mechanism prevails. Objective In the literature, many gas holdup studies have been performed with model systems using distilled water as the liquid phase, air as the gas phase, and silica or glass beads as the solid phase. Because the present study is part of a research project that aims to investigate an actual reaction system,6 the effect of carbon particles * Author to whom correspondence should be addressed. Phone: +31 40 2472850. Fax: +31 40 2446653. E-mail: [email protected].

and electrolyte concentration on the hydrodynamics was studied. Both particle and electrolyte concentrations were kept low to study the effect of particles and electrolyte on bubble-bubble and bubble-particle interactions, without significantly changing the bulk properties of the liquid. With a combination of experimental techniques, including local and overall gas holdup measurement and high-speed video imaging, the mechanisms accounting for the increase in gas holdup resulting from electrolyte and particle additions were investigated. Experimental Setup and procedures Gas holdup experiments were carried out with different carbon particle and electrolyte concentrations. Carbon particles with a mean diameter of 30 µm were used, and sodium gluconate was used to study electrolyte concentration effects. 2D Slurry Bubble Column. All experiments were carried out in a 2D slurry bubble column as shown in Figure 1. Gas Holdup Measurement. The pressure difference between two pressure sensors represents the local gas holdup according to eq 1 local i,j )

(p jj - p j i)0 - (p jj - p j i)aerated (p j j - pi)0

with i, j ) 1-4 (1)

Also, the overall gas holdup in the column was measured, both visually and with a float. The float accurately follows the movement of the liquid surface. The movement of the float is measured with a Honeywell K180E ultrasonic distance sensor. Signals from the float position and the four pressure sensors were measured

10.1021/ie001078r CCC: $20.00 © 2001 American Chemical Society Published on Web 08/21/2001

Ind. Eng. Chem. Res., Vol. 40, No. 23, 2001 5327 Table 1. Surface Tension of Distilled Water and Different Electrolyte Solutions surface tension [mN m-1] distilled water 0.05 M electrolyte 0.1 M electrolyte 0.2 M electrolyte 0.5 M electrolyte

Figure 1. 2D Perspex bubble column (d × w × h ) 0.015 × 0.30 × 2.00 m) with 20 sensor connections located 2.5, 48.5, 83.5, and 118.5 cm above the gas sparger. When the sensor connections were not used, they were closed flush with the wall. Four Dru¨ck PTX 400 pressure sensors were connected in the middle position of each sensor row, from bottom to top numbered from P1 to P4.

for 2 min with a sample frequency of 50 Hz. The time series were averaged to obtain average pressures and average liquid heights. The total gas holdup was calculated with eq 2

) total g

H1 - H0 H1

(2)

where H0 is the nonaerated liquid height and H1 is the average aerated liquid height. Transition Point Measurement. The homogeneous regime is characterized by a low superficial gas velocity, a narrow bubble size distribution, small bubbles, and a low rate of bubble coalescence. When the superficial gas velocity is increased, the homogeneous regime changes into the heterogeneous regime. The heterogeneous regime is characterized by a higher rate of bubble coalescence and a wider bubble size distribution, in which both large and small bubbles exist. The superficial gas velocity at which the homogeneous regime changes into the heterogeneous regime is called the transition velocity. Changes in the transition point between the homogeneous and heterogeneous flow regimes can provide insight into the mechanisms of the increase in gas holdup. Several techniques have been proposed to determine the transition point. Vial et al.7 compared different techniques for estimating the transition point from pressure signals. Part of their findings are in agreement with the work of Letzel et al.8 Letzel et al. employed Kolmogorov entropy to determine the transition point, whereas Vial et al. chose the relative standard deviation of the pressure signal. In the present study, the relative standard deviation and the average cycle frequency are used. The average cycle frequency is directly related to the Kolmogorov entropy used by Letzel et al. The average cycle frequency is however, much quicker and easier to calculate, which is the reason that we used it in this work. Using the average cycle frequency, the transition point between the homogeneous and the heterogeneous flow regimes is estimated from a graph of the average

74 ( 2 68 ( 2 61 ( 2 55 ( 2 48 ( 2

cycle frequency fc of the measured pressure signals as a function of the superficial gas velocity. The average cycle frequency is defined as fc ) nt/(2tm), where tm is the total measuring time of the pressure signal and nt is the number of times the pressure signal crosses its mean value. The average cycle frequency provides a characteristic measure of the bubble dynamics and therefore serves as an indication of regime transition. Experimental Conditions. From several experiments it was observed that within a small range of liquid viscosity (1.0-2.0 kg m-1 s-1), gas density (0.171.3 kg m-3) and type of gas (nitrogen, oxygen, and air) do not influence the gas holdup significantly. Therefore all experiments in this paper were carried out with nitrogen. The initial liquid height (H0) did not influence the gas holdup if it was above 1 m. Therefore, the initial liquid height in all experiments was between 1.0 and 1.5 m. All experiments were carried out at ambient pressure and temperature. Distilled water is preferred over tap water because the properties of tap water are poorly defined. Sodium gluconate was used as the lectrolyte in concentrations of 0.05-0.5 M. Sodium gluconate is one of the products formed during glucose oxidation. This oxidation reaction is used as a model reaction for further research. The surface tension and viscosity of the electrolyte solutions were measured. The viscosity varied between 1.0 × 10-3 and 1.2 × 10-3 kg m-1 s-1 for all solutions. The surface tension was measured with a digital tensiometer. From Table 1 the surface tension is shown to decrease with increasing electrolyte concentration. At first the decrease in surface tension is high, but it becomes smaller at higher electrolyte concentrations. Carbon particles (Engelhard Q500-130) with a mean particle diameter of 30 µm were used. The carbon particles are similar to the carrier material of the Pt/ carbon catalyst that will be employed in the actual reaction system.6 Prior to each experiment, the carbon particles were washed with distilled water and dried at 378 K, to remove any organic contamination. Because the carbon particles tend to be hygroscopic, they were stored at 378 K. The wettability of the carbon particles is one of the main parameters influencing the gas holdup in a bubble column. To ensure that all particles arwee completely wetted at the start of each measurement, the particles were mixed with distilled water for 1 h in the column. Spargers. Two spargers were used. The first sparger was a porous plate with a mean pore size of 30 µm. The second sparger was a perforated plate with 49 holes of 0.5-mm diameter. The holes were positioned as shown in Figure 2. Both spargers were 0.2 × 0.01 m and were located in the bubble column as shown in Figure 1. If not mentioned differently, the 0.5-mm sparger was used in all experiments. Bubble Size Imaging. Video images were recorded with an image size of 0.15 × 0.15 m, using a Dalsa

5328

Ind. Eng. Chem. Res., Vol. 40, No. 23, 2001

Figure 2. Illustration of 0.5-mm sparger holes located equidistant from each other at a distance of 7 mm.

CA-D6 high-speed camera at a rate of 955 frames per second. Because of the high shutter speed, strong illumination of the imaged surface was required. For this purpose, 10 halogen lamps of 500 W each were pointed at a white screen behind the 2D column depicted in Figure 1. All of the lamps could be dimmed between 0 and 100% of their capacity. Ten light sensors, located at the back of the 2D bubble column, measured the amount of light falling on the imaged surface, ensuring that the same amount of light was used for each video image. Error Analysis of Gas Holdup Data. Reproducibility and measurement error were estimated for each average pressure value obtained from a time series of 2 min that was sampled with a sample frequency of 50 Hz. Each of these average points suffers from an error that is given by

error ) (measured pressure × accuracy of the sensor) + read out error + reproducibility error The accuracy of the sensor is expressed as 0.25% percent of the measured value. For P1 and P2, a read-out error of 0.25 mbar and a reproducibility error of 0.353 mbar were measured (95% confidence interval). For P3 and P4 a read-out error of 0.08 mbar and a reproducibility error of 0.615 mbar were measured. Hence, the maximum error in each gas holdup point is 15%. For the data interpretation, this maximum error was taken into account. Experimental Results Distilled Water. Figure 3a shows the pressure holdups measured at positions 3 and 4 and with the float. The parity plot of these gas holdup data is shown in Figure 3b. The local gas holdup calculated from the pressure signals is in agreement with the overall gas holdup obtained from the float measurements. The deviation from the overall “float” gas holdup compared to the local “pressure” gas holdup is due to a somewhat lower gas holdup below pressure sensor 3 and a slightly higher gas holdup above pressure sensor 4 that results from gas expansion. Holdup data obtained from pressure sensors 1 and 2 show a deviation from the overall gas holdup because of acceleration effects resulting from the air entering the bottom of the column. Electrolyte Concentration. Figure 4 shows the increase in gas holdup with increasing electrolyte concentration. No changes in gas holdup take place above a critical concentration. The critical concentration is between 0.05 and 0.1 M. These results are in ac-

Figure 3. (a) Gas holdup in distilled water measured with float and pressure sensors P3 and P4. (b) Parity plot of the gas holdup calculated from pressure signals at sensors P3 and P4, compared to gas holdup calculated from float signals. Sparger: 0.5-mm perforated plate.

cordance with the literature.7,9,10 Above an electrolyte concentration of 0.1 M, no changes in gas holdup with increasing electrolyte concentration were observed. One thousand video images (approximately 1 s) were captured at a superficial gas velocity of 0.07 m s-1. The pictures shown in Figure 5 are representative for each of the series of 1000 images. In each image of Figure 5, a large bubble of approximately 5-cm diameter can be seen. In the electrolyte solutions many more smaller bubbles with diameters of less than 0.5 cm are present compared to the image of distilled water. The number density of small bubbles increases with increasing electrolyte concentration. Carbon Particle Concentration. The gas holdup in different carbon particle slurries was measured as a function of the superficial gas velocity. Figure 6 shows that a critical concentration exists above which no changes in gas holdup occur. This critical carbon particle concentration is found between 0.2 and 0.5 g L-1. Video images of different slurries were recorded at a superficial gas velocity of approximately 0.07 m s-1. Characteristic pictures are shown in Figure 7. The images in Figure 7 are clearly different from the images in Figure 5. For both the distilled water and for the suspension containing 0.1 g L-1 carbon particles, a large bubble with a diameter of approximately 5 cm is observed. This typical-sized bubble is not present in the picture of the suspension containing the 0.5 g L-1 slurry. For the 0.5 g L-1 suspension, many more smaller bubbles exist compared to the 0.1 g L-1 carbon slurry and distilled water cases. However, these small bubbles are larger than the small bubbles observed in the 0.2 M electrolyte picture from Figure 5.

Ind. Eng. Chem. Res., Vol. 40, No. 23, 2001 5329

Figure 4. Gas holdup measured with pressure sensors 3 and 4 with different electrolyte solutions and distilled water. Sparger: 0.5-mm perforated plate.

Joint Effect of Carbon Particle and Electrolyte Concentrations. Experiments were carried out with addition of both particles and electrolyte well above the critical concentrations. This led to a very high gas holdup that exceeded the holdup of the two separate cases (see Figure 8). Because of the limited height of the column, it was not possible to measure gas holdups at superficial gas velocities higher than 0.07 m s-1. At low gas velocities, in the homogeneous flow regime, the addition of carbon particles and electrolyte results in the same gas holdup as for the separate cases. At high gas velocities, in the heterogeneous flow regime, the gas holdup clearly exceeds the holdup of the two separate cases. Initial Bubble Size. The initial bubble size immediately above the sparger was changed by using a 30-µm porous plate instead of the 0.5-mm perforated plate. It was observed from the video image recordings that the initial bubble size was considerably smaller (approximately 0.2-0.5 mm) than in case of the perforated plate (approximately 1-2 mm). Figure 9a shows that the initial bubble size affects the gas holdup in the case of the electrolyte solution. Apparently, in the homogeneous regime, no change in gas holdup occurs as a function of the initial bubble size. However, in the heterogeneous regime, a smaller initial bubble size leads to a higher gas holdup. At a gas velocity of about 0.5 m s-1, coalescence starts resulting in a steep gas holdup decrease. At higher gas velocities, the initial gas holdup (Figure 9b) indicates no effect of the initial bubble size on gas holdup in the case of the carbon particle slurry. Transition Point. Carbon particles and electrolyte might have an effect on the transition between the

homogeneous and heterogeneous flow regimes. Figure 10a and b shows the relative standard deviation (σ/µ) of the pressure signals in 1 g L-1 carbon particle suspension and in distilled water, respectively. Vial et al.7 suggested that a measured value of the relative standard deviation of 1.5 (dotted lines in Figure 10a and b) indicates that the end of the homogeneous regime is reached. In the case of distilled water, this would be around a superficial gas velocity of 0.03 m s-1, which is in agreement with the average cycle frequency measurement of pressure sensor 3 (Figure 10d). For the carbon particle slurry, however, the transition point proposed by Vial et al. would predict an incorrect transition point at a superficial gas velocity of 0.06 m s-1 (Figure 10a). The average cycle frequency of the pressure signals in Figure 10c shows a clear transition point around a superficial gas velocity of 0.035 m s-1, which is significantly different from the transition point obtained from Figure 10a. In general, Figure 10c and d shows that it is possible to estimate a regime transition point by means of pressure signal analysis using the average cycle frequency. A comparison of the two plots shows that bubble dynamics influences the average cycle frequency. In the case of Figure 10c, many small bubbles were present, resulting in a sharp transition point. This effect is caused by the passage of the small bubbles along the pressure sensor, which results in a higher frequency in the pressure time series than the passage of a large bubble. When more small bubbles are present, the frequency changes more drastically upon appearance of large bubbles. This results in a sharp transition point. With the average cycle frequency, transition points were determined. This resulted in a transition point for distilled water at a superficial gas velocity of 0.03 ( 0.005 m s-1. For all carbon slurries, a slight shift in transition velocity of about 0.005 m s-1 was measured, resulting in a transition velocity of 0.035 ( 0.005 m s-1. For all electrolyte solutions above 0.1 M, a transition velocity of 0.04 ( 0.005 m s-1 was measured. The transition point for the experiment with both carbon particles and electrolyte could not be found within the range of superficial gas velocities studied because of severe foam formation. Mechanisms for Gas Holdup Increase From Figures 4 and 6, it was concluded that the addition of carbon particles and electrolyte leads to a significant increase in gas holdup in a bubble column.

Figure 5. Video images taken at a superficial gas velocity of approximately 0.07 m s-1 at a distance of 70 cm above the sparger. Image size is 15 × 15 cm. Sparger: 0.5-mm perforated plate.

5330

Ind. Eng. Chem. Res., Vol. 40, No. 23, 2001

Figure 6. Gas holdup measured with pressure sensors 3 and 4 with different carbon particle solutions and distilled water. Sparger: 0.5-mm perforated plate.

This conclusion is underlined by the images shown in Figures 5 and 7, where the additions of electrolyte and carbon particles lead to a large increase in the volume occupied by bubbles, which is caused by a change in the bubble size distribution. Three possible mechanisms might account for this gas holdup increase. A holdup increase might be attributed to one or a combination of these mechanisms. First, these possible mechanisms are briefly outlined. Subsequently, these mechanisms are used to explain the experimental results. Mechanism 1: Effect of Surface Tension on Gas Holdup. It is possible that the presence of electrolyte or particles changes the surface tension of the slurry, as well as the surface properties of the gas-liquid interface. This change leads to a change in bubble size distribution: as the surface tension is a measure for the stability of the gas-liquid interface, a smaller surface tension leads to a less stable gas-liquid interface and thus to a smaller average bubble size. The residence time of a small bubble is longer than that of a large bubble, because the rise velocity of a bubble increases with the square root of its size. Hence, a smaller average bubble size leads to an increase in gas holdup. The average bubble size is primarily dictated by the surface tension and, consequently, is not affected by the initial bubble size at the sparger. Mechanism 2: Effect of Wettability of Particles and Ionic Forces on Gas Holdup. Another possible effect of the addition of electrolyte and active carbon particles is the stabilization or destabilization of bubbles. This (de)stabilization is a result of the formation of a layer of particles or electrolyte around the gas bubble, which hinders or promotes bubble coalescence.

Jamialahmadi and Muller-Steinhagen2 described this effect for wettable and nonwettable particles. As shown in Figure 11a, wettable particles tend to repel the gas interface, thus acting as a buffer between two adjacent gas bubbles and resulting in a decreased rate of coalescence. Nonwettable particles have the opposite effect (Figure 11b). Marrucci10 performed small-scale experiments with two approaching bubbles in electrolyte solutions. It was proposed that, a result of ionic forces, the film drainage speed between two approaching bubbles is decreased, resulting in a lower rate of coalescence and thus a higher gas holdup. According to this theory, electrolyte decreases the liquid film drainage speed between two approaching bubbles, thus decreasing the rate of coalescence. In this mechanism, the initial bubble size affects the average bubble size and thus the gas holdup. A different initial bubble size results in a different bubble size distribution, leading to a higher or lower gas holdup. This mechanism causes the transition between the homogeneous and heterogeneous regimes to occur at a larger or smaller superficial gas velocity. For example, wettable particles stabilize small bubbles and, therefore, delay the formation of large bubbles by coalescence. Mechanism 3: Effect of Viscosity and Density on Gas Holdup. Electrolyte and carbon particles influence the rise velocity of the bubbles. The rise velocity of a single bubble is decreased, resulting in a higher gas holdup, when the density or viscosity of the liquid-slurry layer around the bubble is significantly increased because of the presence of electrolyte or carbon particles. The initial bubble size does not affect the gas holdup when only mechanism 3 prevails. The bubble size distribution is not affected by local changes in viscosity or density around the bubbles. Discussion of Results Electrolyte. Figure 4 shows a significant increase in gas holdup in electrolyte solutions. The images in Figure 5 clearly show that this increase is caused by an increase in the number of small bubbles compared to distilled water. Changing the initial bubble size by applying a different sparger shows a significant increase in the gas holdup in the presence of electrolyte. Moreover, the transition from the homogeneous regime to the heterogeneous regime in systems containing electrolyte is delayed compared to that occurring in systems with distilled water. The difference in gas holdup with different initial bubble size occurs because small bubbles

Figure 7. Video images taken at a superficial gas velocity of approximately 0.07 m s-1 at a distance of 70 cm above the sparger. Image size is 15 × 15 cm. Sparger: 0.5-mm perforated plate.

Ind. Eng. Chem. Res., Vol. 40, No. 23, 2001 5331

Figure 8. Gas holdup measured with pressure sensors P3 and P4 in 0.1 M electrolyte solution, 1.0 g L-1 carbon particle slurry and in a combined experiment with 0.1 M electrolyte and 1.0 g L-1 carbon particles. Sparger: 0.5-mm perforated plate.

Figure 9. (a) Local gas holdup measured by P3 and P4 as a function of the superficial gas velocity in 0.1 M electrolyte solution with two different spargers. (b) Local gas holdup measured by P3 and P4 as a function of the superficial gas velocity in a 1 g L-1 carbon slurry with two different spargers.

are more stabilized than large bubbles. When the coalescence starts, at a superficial gas velocity of about 0.05 m s-1, the collision probability is greater in the experiment with a smaller initial bubble size, thus resulting in a higher rate of coalescence than in the case of the perforated plate, where larger bubbles are already present. This causes a steep decrease in the gas holdup in the experiment with the porous plate. Also, electrolyte changes the bubble size distribution because of the decrease in surface tension. This is supported by the images in Figure 5. At low superficial gas velocities, many more small bubbles are formed, an effect that can only be attributed to the change in surface tension, because coalescence and separation of bubbles at these low superficial gas velocities is negligible. These observations indicate that mechanisms 1 and 2 explain the increase in gas holdup with addition

of electrolyte. A reduced rise velocity as explained in mechanism 3 is not likely, because the viscosity of the electrolyte solutions hardly changes with increasing electrolyte concentration. Prince et al.9 calculated the critical electrolyte concentration in electrolyte solutions for different electrolytes. Their calculations were based on the model of Marrucci.10 This critical concentration is of the same order of magnitude as that found in our study. Marrucci10 and Prince et al.9 concluded that a decreased rate of coalescence, due to a reduced film drainage speed between two approaching bubbles, is responsible for the increase in gas holdup. Our study shows that, in addition to this effect, also a smaller bubble size distribution, due to a change in surface tension of the liquid, accounts for a gas holdup increase upon electrolyte addition. Carbon Particles. Figure 9b shows that the gas holdup is not affected when the initial bubble size is changed for a slurry of carbon particles. However, the gas holdup shown in Figure 6 increases upon addition of particles, and the transition from the homogeneous to the heterogeneous regime occurs at a higher gas velocity in systems with carbon particles compared to distilled water. These observations favor of mechanism 2. If only bubble stabilization accounted for the higher gas holdup due to particle addition, one would not expect a higher gas holdup compared to distilled water, already in the homogeneous regime, where bubble coalescence is almost absent. However, Figure 6 shows the opposite. Therefore, it is expected that the density of the layer around a bubble is significantly increased by the addition of carbon particles, lowering the rise velocity of a single bubble. This suggests mechanism 3. The surface tension is barely influenced by the presence of carbon particles, and thus, mechanism 1 is less likely to occur. Hence, upon addition of carbon particles, a combination of mechanisms 2 and 3 appears to adequately describe the increase in gas holdup. Electrolyte and Carbon Particles. The experiment with both carbon particles and electrolyte in Figure 8 shows the joint effect of the suggested mechanisms. Smaller bubbles are formed because of the addition of electrolyte and are stabilized by the carbon particles, resulting in a considerable increase of the gas holdup. Conclusions In this study, we have shown the following: (1) The addition of electrolyte changes the bubble size distribution and leads to an increase of the gas holdup in a bubble column reactor. Experiments show that addition of electrolyte changes the surface tension of the solution, thus leading to a smaller average bubble size. Adding electrolyte also stabilizes bubbles, decreasing the rate of coalescence and, therefore, increasing the gas holdup in the bubble column. The postponing of the transition point going from the homogeneous to the heterogeneous regime supports this conclusion. It was found that addition of electrolyte does not significantly change the density or viscosity of the liquid layer around a bubble. (2) The addition of carbon particles changes the bubble size distribution through bubble stabilization. Experiments show that adding carbon particles leads to a delay in bubble coalescence and thus to an increase in the superficial gas velocity at which the transition from the homogeneous to the heterogeneous regime occurs. Also, as a result of particle addition, the density

5332

Ind. Eng. Chem. Res., Vol. 40, No. 23, 2001

Figure 10. (a) Relative standard deviation of pressure signals measured in a 1 g L-1 carbon slurry. (b) Relative standard deviation of pressure signals measured in distilled water. (c) Average cycle frequency of pressure signals measured in a 1 g L-1 carbon slurry. (d) Average cycle frequency of pressure signals measured in distilled water.

addition of electrolyte, and these bubbles are stabilized and slowed because of the presence of carbon particles. (4) The average cycle frequency is a very powerful tool for determining the transition point between the homogeneous and the heterogeneous regimes. The standard deviation of the pressure signal does not provide this information. List of Symbols p ) pressure measured by P1-P4, N m-2 g ) gas holdup fc ) average cycle frequency, s-1 nt ) number of times the pressure signal crosses its mean value tm ) measuring time, s H ) liquid height, m

Literature Cited

Figure 11. (a) Gas-liquid adsorption on a wettable particle. (b) Gas-liquid adsorption on a nonwettable particle.

of the liquid layer around a bubble significantly increases, decreasing its rise velocity andthus increasing the gas holdup. The surface tension is barely influenced by the presence of carbon particles. (3) The addition of both carbon particles and electrolyte increases the gas holdup significantly through a joint effect. Smaller bubbles are formed because of tha

(1) Joshi, J. B.; Parasu Veera, U.; Prasad, Ch. V.; Phanikumar, D. V.; Deshphande, N. S.; Thakre, S. S.; Thorat, B. N. Gas holdup structure in bubble column reactors. Proc. Indian Natl. Sci. Acad. 1998, 64A, 441-567. (2) Jamialahmadi, M.; Muller-Steinhagen, H. Effect of solid particles on gas hold-up in bubble columns. Can. J. Chem. Eng. 1991, 69, 390-393. (3) Koide, K. Design parameters of bubble column reactors with and without solid suspensions. J. Chem. Eng. Jpn. 1996, 29 (5), 745-759. (4) Krishna, R.; De Swart, J. W. A.; Ellenberger, J.; Martina, G. B.; Maretto, C. Gas Holdup in Slurry Bubble ColumnssEffect of Column Diameter and Slurry Concentrations. AIChE J. 1997, 43 (2), 311-316.

Ind. Eng. Chem. Res., Vol. 40, No. 23, 2001 5333 (5) Jamialahmadi, M.; Muller-Steinhagen, H. Effect of electrolyte concentration on bubble size and gas hold-up in bubble columns. Trans Inst. Chem. Eng. 1990, 68 (A), 202-204. (6) Kluytmans, J. H. J.; Markusse, A. P.; Kuster, B. F. M.; Marin, G. B.; Schouten, J. C. Engineering aspects of the aqueous noble metal catalysed alcohol oxidation. Catal. Today 2000, 57, 143-155. (7) Vial, C.; Camarasa, E.; Poncin, S.; Wild, G.; Midoux, N.; Bouillard, J. Study of hydrodynamic behaviour in bubble columns and external loop airlift reactors through analysis of pressure fluctuations. Chem. Eng. Sci. 2000, 55, 2957-2973. (8) Letzel, H. M.; Schouten, J. C.; Krishna, R.; van den Bleek, C. M. Characterisation of regimes and regime transitions in bubble

columns by chaos analysis of pressure signals. Chem. Eng. Sci. 1997, 52, 4447-4459. (9) Prince, M. J.; Blanch, H. W. Transition electrolyte concentrations for bubble coalescence. AIChE J. 1990, 36 (9), 14251429. (10) Marrucci, G. A theory of coalescence. Chem. Eng. Sci. 1969, 24, 975-985.

Received for review December 13, 2000 Revised manuscript received June 6, 2001 Accepted June 6, 2001 IE001078R