257154 Interaction Between Fluid Mixing and Chemical Reaction -Effect of Non-Uniform Chaotic Mixing On Concentration Distribution for Periodical Reaction System-

Tuesday, October 30, 2012: 8:55 AM
Frick (Omni )
Shunsuke Hashimoto, Yusuke Chikamochi and Akitoshi Nishimura, Division of Chemical Engineering, Department of Materials Engineering Science, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka, Japan

Interaction between Fluid Mixing and Chemical Reaction -Effect of Non-Uniform Chaotic Mixing on Concentration Distribution for Periodical Reaction System-

S. Hashimoto*, Y. Chikamochi, A. Nishimura

Division of Chemical Engineering, Graduate School of Engineering Science, Osaka University, 1-3, Machikaneyama, Toyonaka, Osaka 560-8531, Japan.

*corresponding author and presenter: shunsuke@cheng.es.osaka-u.ac.jp, Phone & Fax: +81-6-6850-6294 (S. Hashimoto)

Mixing in stirred tank reactors in a wide variety of tank sizes and impeller shapes has been often utilized to homogenize different substances and to conduct chemical reactions in industrial chemical processes. Recently in various industrial processes, a wide range of operation for stirred tank is required depending on purposes and conditions. "Mixing time" is essential as one of index to evaluate the mixing process in stirred tank experimentally. Various techniques for measuring mixing time have been used such as coloration, decolorization, measurements of electric conductivity and temperature, and so on. Coloration is most simple technique to observe the mixing state in a whole tank, while it has a few demerits: it is difficult to detect the mixing state in a central part of stirred tank and the required time for complete mixing. Measurement of electric conductivity [1] enables us to detect delicate time variation of conductivity, but only local information for mixing near the sensor of conductivity can be obtained. Measurement of temperature has also demerits that only local information for mixing near thermocouples can be obtained, which is similar to measurement of electric conductivity. There are additional problems such as the variation of fluid property due to temperature change and thermal insulation for the measurement of temperature. It is expected that tomography [2] can compensate these demerits for other techniques efficiently in future, while its low resolution and high cost remain the key issue for the evaluation of mixing state. Decolorization enables us to observe easily the difference of mixing state at every position in stirred tank. In the case of decolorization, it is easy to determine the completion time of mixing because the termination of decolorization is directly equivalent to the completion time of mixing. Consequently, decolorization is generally used as a simple method for the measurement of mixing time. In the case of the visualization using chemical reactions such as decolorization, however, the relation between the rate of chemical reaction and that of convective mixing is quite important.

Based on general mixing theory, the completion time for mixing should be defined as the time that is required to homogenize the concentration of each fluid component all over the place in the system. In above general techniques for the measurement of mixing time, fluid is judged to mixed completely at the time that its concentration, conductivity, and temperature are homogenized in stirred tank. The properties such as concentration, conductivity, and temperature are considered to have one-to-one relationship with substance quantity of fluid component (that is, concentration). Hence, it is reasonable to use these properties as an index to evaluate mixing state. On the other hand, there are unique cases containing fluid properties that are not one-to-one relationship with concentration. For example, in the periodical vibrational system by Beloousov-Zhabotinskii (BZ) reaction, fluid has not only the property of concentration but also the properties of period and/or phase of concentration oscillations, which depend on the dynamical mode of reaction and do not have simple relationships with the concentration. In this case, it is inadequate to judge the mixing completion based on only the homogeneity of concentration of each component in stirred tank. In addition, final pattern by convective mixing would potentially depend on only the structure of flow field. For example, under low agitating Reynolds number (Re) conditions, the presence of segregated mixing regions from the chaotic mixing regions (CMR) in the form of toroidal vortices above and below an impeller, which is called as "isolated mixing region (IMR)" [3, 4], is well known. Material exchange by diffusion is dominant at the interface between CMR and IMR. In this case, even if the decolorization (homogenization of concentration) in the whole region of stirred tank, IMR must remain there essentially. Hence, it is necessary to evaluate mixing degree based on not only the homogeneity of concentration but also various index.

The present study focused attention on reconsidering the traditional concept of mixing that was based on the homogeneity of concentration. In the present study, required time for complete mixing and mixing patterns containing "partially mixing regions" such as IMR in CMR were investigated by use of two chemical reactions: conventional decolorizing reaction (with iodine and sodium thiosulfate) and periodical oscillating reaction (BZ reaction). Based on experimental results, the body of matter that was mixed by impeller agitation and the correspondence between chemical reaction and convection were discussed with the view of "fluid-informatic" engineering. In addition, the availability of periodical reaction for the visualization of partially mixing regions was briefly discussed.

Figure 1 shows the simple summary of the present study. The periodical steady color-variation in BZ reaction system remained after the sufficient time required for the complete decolorization. The sequential spatial color-patterns obtained in steady periodical variation process were similar to the transitional color patterns observed in the decolorization process. The color-patterns obtained in BZ periodical reaction were consistent with the outline of partially mixing regions where the exchange of substance is relatively slow in the vessel and they depended on Re. The phase of periodical concentration oscillation in each partially mixing region was shifted with one another in spite of the same period of oscillation. Whether the period and/or phase synchronize or not in each partially mixing region would depend on the relative speed between the exchange of substance and the synchronization of periodical oscillation of concentration there.

Figure 1 The experimental images obtained from the decolorizing experiment and the BZ reaction experiment; (a)Re = 117 and (b)Re = 780. Panel (c) shows the color-pattern diagrams corresponding to these experimental images.

Keywords: Mixing; Mixing Time; Visualisation; Fluid Mechanics; Nonlinear Dynamics; Periodical Reaction

References

[1] Kramers H., Baars G.M. Knoll, W.H. (1935). Chem. Eng. Sci, 2, 35-42.

[2] Kaminoyama M., Taguchi S., Misumi R., Nishi K. (2005). Chem. Eng. Sci., 60, 5513-5518.

[3] Bresler L., Shinbrot T., Metcalfe G., Ottino M.J. (1997). Chem. Eng. Sci, 52, 1623-1636.

[4] Hashimoto S., Ito H. Inoue Y. (2009). Chem. Eng. Sci., 64, 5173-5181.


Extended Abstract: File Uploaded
See more of this Session: Mixing In Single Phase Systems
See more of this Group/Topical: North American Mixing Forum