UNIVERSITA’ DEGLI STUDI DI PISA
Facoltà di Ingegneria
Corso di Laurea in Ingegneria Chimica
Tesi di Laurea
ANALISI SPERIMENTALE SUL COMPORTAMENTO
FLUIDODINAMICO DI REATTORI OPERANTI
CON MISCELE BIFASE
Candidato:
Andrea Ghionzoli
Relatori:
Dott. Ing. Elisabetta Brunazzi
Prof.
Ing.
Alessandro
Paglianti
Prof.
Ing.
Roberto
Mauri
In collaboration with
THE UNIVERSITY OF BIRMINGHAM
School of Engineering
Chemical Engineering
Project title:
SUSPENSION OF PARTICLES
IN STIRRED TANKS:
THE INFLUENCE OF BOTTOM ROUGHNESS
Supervisors:
Dr. Waldemar Bujalski
Abstract
An investigation of the effect of the vessel bottom roughness on the suspension of solid particles in a stirred tank reactor was carried out.
The experiments were performed in a baffled vessel of diameter 0.20 m, which was mechanically stirred with a 45°-pitch angle blade turbine pumping down. The impeller was set at an off-bottom clearance of 1/4 the tank diameter; the impeller to tank diameter ratio was 1/3. The vessel was filled with distilled water to a height equal to its diameter. Four different vessel bottoms were used; one was completely made of Perspex with a smooth surface exposed to the liquid, while the remaining three had different degrees of surface roughness, as the side of the bottom exposed to the liquid was covered with an abrasive sheet. Eight different sets of spherical particles were used as the solid phase. Their density was between 2500 and 8743 kg/m3 and they were characterized by narrow size distributions and a mean diameter between 128 and 1850 µm.
For single phase systems the power drawn by the impeller was measured for each of the four vessel configurations. The constant values of the power number in the turbulent regime were calculated. The work has shown that the power number is not dependent on roughness, as it has the same value for all the bottoms used.
For solid-liquid systems the minimum impeller speed for “just complete suspension” NJS was measured. It was determined using a
technique similar to Zwietering’s, which is based on visual observation. The Zwietering’s criterion was applied in order to determine the rotational speed NJS at which the just-suspended state was achieved by
the particles. Solid concentrations up to 2% by weight were investigated. The results of the just-suspended speed are given in dimensionless form using the parameter S of Zwietering’s correlation, which allows a better comparison between the different bottoms. The analysis of the
parameter S has proved that suspension of the smallest particles is helped by rough bottoms, i.e. complete suspension is achieved at lower rotational speeds than with the smooth bottom. With increasing particle size (dP), rough bottoms show a negative influence and suspension is
hindered further. It was found that, depending on the roughness, there was a critical size of particles whose pick up is hindered most. This critical value increases as the bottom roughness increases. However, dividing the critical value by the roughness parameter (e) of the bottom, the same value (dP/e=2) was obtained for all the rough bottoms.
Eventually, with further increasing particle size, the influence of the rough surfaces became weaker and weaker. It was also found that with decreasing bottom roughness the effect on solid suspension becomes less strong.
Interaction between the roughness elements, the particles and the turbulent field is proposed as a means to explain this behaviour. Some considerations on the effect of roughness on turbulence structure in pipe flow allow an understanding of the results obtained in this study. Moreover, studies of the settling velocity of particles in stirred vessels are useful to give a physical interpretation of the particles pick-up mechanism from a rough bottom. In fact, the relevance of the Kolmogoroff length scale in explaining the effect of the roughness is shown.
Finally, a model for predicting the speed of just suspension for the smooth bottom has been developed and compared to the Zwietering’s correlation.
Acknowledgements
I gratefully acknowledge the extraordinary support of my English supervisors Dr. W. Bujalski and Prof. A. W. Nienow. They have always inspired my research and helped me continuously during my stay at the University of Birmingham. It was a pleasure to work and interact with such formidable people.
I would like to thank my Italian supervisors Dr. Ing. E. Brunazzi and Prof. A. Paglianti. Their brilliant ideas and suggestions were essential for developing my research and achieving such important results.
I wish to thank the Department of Chemical Engineering for allowing the use of the facilities and laboratories, which enabled this experimental work to be carried out. Special thanks go to Mr. Robert Sharpe and the many technicians who were always ready to help.
I am really grateful to my family; my parents and Vanessa, my sister, for their moral and material support throughout all these years; my grandmother and Alessio, my cousin, for their continuous encouragement; Riccardo, for his IT support.
I would like to thank all the people I met in Birmingham: Fabio (for his help and friendship), the Mixing group and the people I met at the MIH residence, especially Pamila, a very good friend, for all the time we spent together.
I would also like to say a big thank you to Bridget for correcting my English.
Finally, very important thanks go to Federico, my university colleague throughout these years and house mate in Birmingham, and to Luca, Calca and Mirko, who visited me during my time in England and made me feel less far from my friends in Italy.
