Mixing time in shaking bioreactors
Tan, Rung-Kai; Büchs, Jochen (Thesis advisor)
Aachen : Publikationsserver der RWTH Aachen University (2012)
Mixing plays an important and essential role in various chemical processes. In addition, shake flasks have been widely used in cultivation of small-scale and academic researches. However, the mixing process in shake flasks has not been well understood. Therefore, the purpose of this study is to measure the mixing time by colorimetric method and to characterize the mixing process in combination with a rotating camera. By using a rotating camera, the mixing process in the shake flask could be successfully observed. Due to the centrifugal force, the motion of fluid in shake flasks was presented as sickle form. The rough position where an acid drop was impacted into the bulk solution was observed. It was obviously seen that the mixing process always occurred from the rear to front of the fluid sickle. The last trace of blue color was identified by the software, Virtualdub and an Image Analysis Software written by Labview. Thus, the mixing time in the shake flask was defined as the exact time between the impacting of the acid drop in the bulk solution and the exact time when the last trace of blue color disappeared. In this study, different operational parameters were varied to investigate the effect on the mixing time. From the experimental results, the mixing time increased with increasing the shaking frequencies in both deionied water and viscous aqueous polyvinylpyrrolidone (PVP) solution. The mixing time is inverse proportional to the shaking frequency. The motion of fluid showed similar behavior in both solutions. Moreover, different flask diameters, i.e. 25 mm and 50 mm, did not have significant influence on the mixing time. In addition, the mixing time was a little longer in the flask of large volume than in the flask of small volume. The mixing time stayed constant until 0.8 kW/m3 of power input, but decreased smoothly when the power input increased furthermore. The mixing time in shake flasks were also compared with that in conventional reactors. The mixing time in shake flasks stayed constant in the region of the Reynolds number of 104 to 105. In other words, the mixing time in the region of high Reynolds number was independent of the geometry of flasks. Furthermore, the mixing number increased with increasing of Reynolds number. The influence of the Phase number was also investigated. It was found that the mixing number decreased clearly when the motion of fluid changed from in-phase condition to out-of-phase. This agreed that the out-of-phase is unwanted in shake flasks, because the mixing process became worse in the out-of-phase condition.