Academia.eduAcademia.edu

Figure 5 – uploaded by Prasert Pavasant

See full PDFdownloadDownload figure
* [28]. Estimates of disjoining pressures (/7) at different salinity levels Table 5

Table 5 * [28]. Estimates of disjoining pressures (/7) at different salinity levels Table 5

Related Figures (15)

Fig. 1. Schematic diagram of concentric internal loop airlift contactor employed in this work
Fig. 1. Schematic diagram of concentric internal loop airlift contactor employed in this work
Liquid properties Table 3 Table 2 2.2. Bubble size distribution measurement
Liquid properties Table 3 Table 2 2.2. Bubble size distribution measurement
Operating conditions for each ALC system Table 4
Operating conditions for each ALC system Table 4
Locations of digital video camera for bubble size measurement axes of bubble images were measured. The equivalent size of the bubble (dg), representing the diameter of a sphere whose vol- ume was equal to that of the bubble, is calculated using Eq. (1) l16.171.
Locations of digital video camera for bubble size measurement axes of bubble images were measured. The equivalent size of the bubble (dg), representing the diameter of a sphere whose vol- ume was equal to that of the bubble, is calculated using Eq. (1) l16.171.
Fig. 2. Frequency distribution of bubble sizes at various superficial gas velocities in ALC with Aq/A,; = 0.661 (salinity = 30 ppt). sections, the distribution changed from uni-modal to multi- modal curve at usy © 0.019 m/s whereas the bottom section saw this change at usp * 0.029 m/s. The breakage of the bubbles at high gas throughput was caused by higher amount of energy dis- sipation and turbulent which promoted more interaction between bubbles. The results suggested, therefore, that there was a higher level of turbulence in the top and middle sections than that in the bottom. Fig. 3 illustrates the mean values of the bubble sizes with the highest occurrence frequencies (Fig. 3(a)) compared with the average, Sauter mean diameter of bubbles (Fig. 3(b)) in the three sections in the airlift system with Ag/A,; of 0.661. This revealed that bubble size in the bottom section was slightly larger
Fig. 2. Frequency distribution of bubble sizes at various superficial gas velocities in ALC with Aq/A,; = 0.661 (salinity = 30 ppt). sections, the distribution changed from uni-modal to multi- modal curve at usy © 0.019 m/s whereas the bottom section saw this change at usp * 0.029 m/s. The breakage of the bubbles at high gas throughput was caused by higher amount of energy dis- sipation and turbulent which promoted more interaction between bubbles. The results suggested, therefore, that there was a higher level of turbulence in the top and middle sections than that in the bottom. Fig. 3 illustrates the mean values of the bubble sizes with the highest occurrence frequencies (Fig. 3(a)) compared with the average, Sauter mean diameter of bubbles (Fig. 3(b)) in the three sections in the airlift system with Ag/A,; of 0.661. This revealed that bubble size in the bottom section was slightly larger
Fig. 3. Axial variation in bubble size in ALC with Ag/A;=0.661 (salin- ity = 30 ppt): (a) the average high bubble size (dp nigh) and average low bubble size (dplow), and (b) Sauter mean diameter.
Fig. 3. Axial variation in bubble size in ALC with Ag/A;=0.661 (salin- ity = 30 ppt): (a) the average high bubble size (dp nigh) and average low bubble size (dplow), and (b) Sauter mean diameter.
Fig. 4. Bubble sizes in ALC with different downcomer to riser cross-sectional area ratios (salinity = 30 ppt).
Fig. 4. Bubble sizes in ALC with different downcomer to riser cross-sectional area ratios (salinity = 30 ppt).
Fig. 6. Relationship between bubble size and AP. In the airlift with Ag/A; of 0.067 running with 45 ppt salin- ity, AP was about 43-121 N/m? at Usg > 0.02 m/s and therefore
Fig. 6. Relationship between bubble size and AP. In the airlift with Ag/A; of 0.067 running with 45 ppt salin- ity, AP was about 43-121 N/m? at Usg > 0.02 m/s and therefore
Estimates of pressure driving forces for average bubble size in different salinity levels in ALC with Ag/A; =0.067
Estimates of pressure driving forces for average bubble size in different salinity levels in ALC with Ag/A; =0.067
Fig. 7. Overall volumetric mass transfer coefficient (ALC with Aq/A,; =0.661). in the airlift system as described earlier. At low range of gas flow rate (<0.02 m/s) as shown in Fig. 5(a), the pressure difference was in the range of 15-20 N/m’, and this enhanced bubble size. Ata higher range of superficial gas velocity (>0.02 m/s), AP was higher than 20 N/m* which promoted the breakup of the bub- bles. In addition, at this high gas throughput conditions, the airlift contained a relatively high gas holdup which also enhanced the chance of bubbles collision and breaking up.
Fig. 7. Overall volumetric mass transfer coefficient (ALC with Aq/A,; =0.661). in the airlift system as described earlier. At low range of gas flow rate (<0.02 m/s) as shown in Fig. 5(a), the pressure difference was in the range of 15-20 N/m’, and this enhanced bubble size. Ata higher range of superficial gas velocity (>0.02 m/s), AP was higher than 20 N/m* which promoted the breakup of the bub- bles. In addition, at this high gas throughput conditions, the airlift contained a relatively high gas holdup which also enhanced the chance of bubbles collision and breaking up.
Fig. 8. Effects of superficial gas velocity, usg, on gas holdups. bubble size (Fig. 5(b)) and gas holdup. Fig. 8 illustrates that the effect of salinity on gas holdups in the system was only marginal and the specific area should only vary with bubble size. As discussed earlier, the bubble size in sea water was smaller than that in fresh water and became smaller with an increase in superficial gas velocity. Therefore, the specific interfacial areas obtained in the systems at all salinity levels were higher than that in the fresh water system.
Fig. 8. Effects of superficial gas velocity, usg, on gas holdups. bubble size (Fig. 5(b)) and gas holdup. Fig. 8 illustrates that the effect of salinity on gas holdups in the system was only marginal and the specific area should only vary with bubble size. As discussed earlier, the bubble size in sea water was smaller than that in fresh water and became smaller with an increase in superficial gas velocity. Therefore, the specific interfacial areas obtained in the systems at all salinity levels were higher than that in the fresh water system.
Fig. 10. Effects of superficial gas velocity, usg, on overall specific mass transfer coefficient, k,, in ALC (Aqg/A; = 0.661). Fig. 9. Specific interfacial area in ALC (Aq/A; = 0.661). k_. Among the saline solutions, ky, was highest in the system with salinity of 30 ppt, followed by that with 45 ppt and 15 ppt, respectively. The reason fore this variation still could not be retrieved from this experiment. However, it was observed that the trend in ka followed that of ky quite closely, i.e. the highest was in pure water system, followed by those in saline concentrations of 30, 45 and 15 ppt, respectively.
Fig. 10. Effects of superficial gas velocity, usg, on overall specific mass transfer coefficient, k,, in ALC (Aqg/A; = 0.661). Fig. 9. Specific interfacial area in ALC (Aq/A; = 0.661). k_. Among the saline solutions, ky, was highest in the system with salinity of 30 ppt, followed by that with 45 ppt and 15 ppt, respectively. The reason fore this variation still could not be retrieved from this experiment. However, it was observed that the trend in ka followed that of ky quite closely, i.e. the highest was in pure water system, followed by those in saline concentrations of 30, 45 and 15 ppt, respectively.
Parameter estimates for correlation in Eq. (14) Sh=a+bGr* Sct + eRe Sc! Table 7 natural convection whereas the force convection as represented by the Reynolds term also was significant in the system operated with saline water. Fig. 11 illustrates the comparison between the calculated and experimental kya of the airlift contactor operated with various saline solutions. downcomer liquid velocity, u_, by assuming that the liquid must have velocity equaled to the terminal velocity to be able to drag the bubble down into the downcomer, or
Parameter estimates for correlation in Eq. (14) Sh=a+bGr* Sct + eRe Sc! Table 7 natural convection whereas the force convection as represented by the Reynolds term also was significant in the system operated with saline water. Fig. 11 illustrates the comparison between the calculated and experimental kya of the airlift contactor operated with various saline solutions. downcomer liquid velocity, u_, by assuming that the liquid must have velocity equaled to the terminal velocity to be able to drag the bubble down into the downcomer, or
Fig. 11. Comparison of ka from experiment and kya estimated by Eq. (14) in ALC (Agq/A; = 0.067-0.661). The parameters a—h in Eq. (14) were evaluated using non- linear parameter fittings using all the results available in this work, and the results are given in Table 7 (noted that these param- eters were obtained from the solver function in the MS Excel 97 where the objective was a minimal error between experimental and simulation data). For the case of tap water, the results from parameter fitting were reasonably close to those proposed from Wongsuchoto et al. [4] (as shown in the last row of Table 7). The fittings for the saline water gave somewhat different results from that for pure water in that the terms Reynolds number was not involved in the pure water system, but it was, to certain extent, for the saline water systems. This meant that the mechanism con- trolling the mass transfer coefficient in pure water was only the
Fig. 11. Comparison of ka from experiment and kya estimated by Eq. (14) in ALC (Agq/A; = 0.067-0.661). The parameters a—h in Eq. (14) were evaluated using non- linear parameter fittings using all the results available in this work, and the results are given in Table 7 (noted that these param- eters were obtained from the solver function in the MS Excel 97 where the objective was a minimal error between experimental and simulation data). For the case of tap water, the results from parameter fitting were reasonably close to those proposed from Wongsuchoto et al. [4] (as shown in the last row of Table 7). The fittings for the saline water gave somewhat different results from that for pure water in that the terms Reynolds number was not involved in the pure water system, but it was, to certain extent, for the saline water systems. This meant that the mechanism con- trolling the mass transfer coefficient in pure water was only the
Related topics:
Chemical EngineeringDraft TubeMass Transfer CoefficientSauter Mean Diameter (SMD)
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved