Investigation of the coupling of global and local transport processes at contaminated interfaces at Taylor bubbles

Transcription

1 Investigation of the coupling of global and local transport processes at contaminated interfaces at Taylor bubbles A short report about my scientific internship at the research group of Prof. Tomiyama at the Kobe University, Japan. Funded by a scholarship of the ProcessNet Subject Division Reaction Engineering Oct.- Dec Sven Kastens Institute of Multiphase Flows Hamburg University of Technology Eissendorfer Str Hamburg, Germany

2 Introduction In October to December 2013, just a few weeks after I had started my PhD at the Institute of Multiphase Flows, I had the possibility to work together with our first guest PhD student, Mr. Shogo Hosoda, from the research group of Prof. Akio Tomiyama from Kobe University. It was my pleasure to welcome and introduce him into the German work life, culture and habits. Together we have investigated the mass transfer processes at CO 2 Taylor bubbles rising in liquid filled channels. Due to our cooperation we established a Sherwood correlation for Taylor bubbles in various channel sizes (TUHH: D=5.5-8mm, Kobe: D= mm). The results were presented on the second International Symposium on Multiscale Multiphase Process Engineering (MMPE) and published in the CET in 2015 [1]. From October to December 2015, it was my pleasure to welcome the former PhD student Mr. Jiro Aoki in our laboratory. We were investigating the effect of surfactants on the transport processes at single Taylor bubbles. With Laser Induced Fluorescence (LIF) and Particle Image Velocimetry (PIV) the concentration and velocity filed around CO 2 Taylor bubbles were measured and compared with the numerical simulations made at Kobe University. I have presented the comparison of numerical and experimental results at the 9th International Conference on Multiphase Flow 2016 in Florence in Italy. From March 2015 I was also the supervisor of Mr. Motoki Iwakiri, the 1 st visiting Master student from the research group of Prof. Koichi Terasaka from Keio University in Yokohama. We have investigated the transport processes at Fine Bubbles, which is a popular topic in Japan for years, but still mostly unknown in other parts of the world. Due to his effort under my supervision, he presented his experimental results from his Master thesis in Sevilla at the 11 th European Fluid Mechanics Conference 2016 and published a scientific paper in the Japanese Journal of Multiphase Flows in 2017 [2]. During my last years as PhD at the Institute of Multiphase Flows, my research were strongly related to Japan and my wish to make my own experiences about the Japanese work life and habits, but also about the fantastic cuisine, culture and the Japanese people has grown tremendously. Additionally, I was fascinated by the Spatial Filter Velocimetry Method (SFV) and its accuracy, which was established and developed in a contaminated liquid / liquid system at the research group of Prof. Akio Tomiyama at Kobe University [3, 4]. With this method, they measured the effect of a contaminated interface on the local liquid velocity in the vicinity of the accumulated surfactants at the interface. Because I wasn t able to measure these effects on gas / liquid interfaces with our techniques (PIV), I needed to learn SFV to adapt this method to my research field of transport processes at Taylor bubbles. I expected a higher spatial resolution and therefore, more accurate local velocity gradients and the possibility to analyze a difference in the boundary condition (slip vs. no-slip) at a clean or contaminated interface of a Taylor bubble. With the scholarship from the ProcessNet subject division Reaction Engineering I was enabled to travel to our Japanese friends at the Kobe University for two months, where I made faster progress and better experiences than ever expected. The research group of Prof. Akio Tomiyama Prof. Akio Tomiyama is the Dean of the Graduate School of Engineering of the Kobe University and leads a research group in the Department of Mechanical Engineering (Thermo-Fluid Dynamics) with two Associate Professors, Assoc. Prof. Kosuke Hayashi and Assoc. Prof. Shigeo Hosokawa, 5-6 PhD students, 8-9 Master students and 12 undergraduate students. I was amazed by the efficient team work in the student subgroups (3-4 students: 1 PhD, 1-2 Master, 2 Bachelor), which is due to the Japanese system, where the students are working in their free time in the labs to learn from the

3 older students, without any clocking or payments. This assures that the knowledge about their scientific topic is taught from generation to generation and the scientific group progress is extremely fast. The weekly discussion with the Professors was very well organized during the night before by my group (Fig.1), where the new results were fortunately presented in English. But in Japan a discussion with the Professors does not mean that someone wants to hear your opinion. It is more about the students present and the Professor gives advices, which I hardly recognized after I started to discuss as usual about an advice.i think I have dropped a clanger, because no one else ever did this, but the Japanese are so polite, friendly and welcoming, that this was no problem at all. Actually, I still think that Prof. Tomiyama liked to hear the honest opinion of a PhD student, but it seems to be very unusual. Research Figure 1: Subgroup Taylor bubble: Yohei Hori, Jiro Aoki, me, Masaaki Hashida, Yohei Hirota The subgroup Taylor bubbles (Fig.1) was headed by former PhD student Aoki, who stayed in Hamburg in We arranged a reconstruction of their experimental setup, to enable experiments in similar condition than I have in TUHH. Taylor bubbles were observed in a vertical arranged and liquid filled channel with a diameter of D=7mm (Fig.2, a). The injection of the gas phase, were inside the lower tank, where the gas was injected with a precise gas tight syringe into the rotatable cup. The bubble was released after the rotation of the cup and formed a Taylor bubble when it reached the vertical channel. Those Taylor bubbles can rise in stagnant liquid or being kept in the observation area in a counter current flow. From a continuous 3W green laser a laser sheet was formed by two lens and actuators for precise arrangement. This optical system creates a thin laser light sheet inside the channel center. A Photron Fastcam S1 (up to fps) with two combined Nikkor lenses were used to enable a macro lens system with a high spatial resolution (Fig.2 b & c). Tracer particles were added to the liquid to enable the velocimetry due to local particle deposition (Fig.2 d). Assoc. Prof. Hosokawa taught me how to use the Japanese in-house-code for the post processing of High-speed images, which I have translated to English. The detailed description about the method can be read in their publication [3, 4]. SFV is analyzing the fluctuation frequency of fluorescence intensity of the particle, while moving through a digital stripe filter, which is multiplied with the image. Therefore, this method is somehow like a digital 2D LDV. I have learned a lot about the required equipment, the method, the post processing and the improvement of the experiment, that it was possible to visualize the partly contaminated interface due to the reduced tangential liquid velocity at the interface of the bubble rear (Fig. 2, e). The bubble rear is typical for the accumulation of surface active agents, which are adsorbed at the interface and sheared down by the liquid, where the boundary condition could change due to layer formation of surface active agents. This result was much more than I have expected before I have travelled to Japan, because I didn t even know, if it would be ever possible to proof a contamination of a gas/liquid interface just by measuring the reduced interfacial mobility by the tangential velocity of the continuous phase.

4 Figure 2: a) Experimental Setup; b) Taylor bubble with Region of Interest for SFV (ROI: red dotted line); c) backlight image of ROI; d) HS image of particles in ROI (red line: interface) ; e) velocity field at partly contaminated interface by SFV. Summary The whole group was so welcoming and helped me a lot during the scientific work, but also organizing my stay with an accommodation and a private shuttle service from airport and trips on the weekends, that I think that my earlier effort for hosting our Japanese guest at our institute was paid back a hundred times. I am so thankful for the opportunity to do a scientific internship at the research group of Prof. Akio Tomiyama at Kobe University in Japan. Without the scholarship from the ProcessNet subject division Reaction Engineering, I would not have been able to learn so much about how science work in Japan, how kind and helpful Japanese are, how delicious and versatile the Japanese cuisine is, and many things more. I had the chance to learn a very rare and quite new method, which will help me to do further progress during my upcoming projects about contaminated gas/liquid interfaces for my further scientific career. I had unforgettable experiences in Japan during my scientific exchange and made Japanese friends for life (Fig.3). Figure 3: Left: Group photo with former guest students at TUHH: Dr. Jiro Aoki, Dr. Shogo Hosoda, Mr. Motoki Iwakiri and me in Tokyo. Right: Soccer tournament with the students of the research group of Prof. Tomiyama in Kobe.

5 Reference [1] Kastens, S.; Hosoda, S.; Schlüter, M.; Tomiyama, A.: Mass Transfer from Single Taylor Bubbles in Mini Channels, Chemical Engineering & Technology, 2015, 38(11), pp , DOI: /ceat [2] Iwakiri, M., Koichi T., Fujioka, S., Schlüter, M., Kastens, S., Tanaka, S.: Mass Transfer from a Shrinking Single Microbubble Rising in Water. Japanese Journal of Multiphase Flow, 2017, 30(5), pp , DOI: /jjmf [3] Hosokawa, S., Tomiyama, A.: Spatial filter velocimetry based on time-series particle images. Exp. Fluids, 2012, (52), pp DOI: /s z [4] Kurimoto, R., Hayashi, K., Tomiyama, A.: Terminal velocities of clean and fully-contaminated drops in vertical pipes. Int. J. Multiphase Flows, 2013, (49), pp 8-23, DOI: /j.ijmultiphaseflow