Motion of objects immersed in a bubbling fluidized bed
- Autores: A. Soria Verdugo
- Directores de la Tesis: Ulpiano Ruiz-Rivas Hernando (dir. tes.)
- Lectura: En la Universidad Carlos III de Madrid ( España ) en 2010
- Idioma: inglés
- Tribunal Calificador de la Tesis: Bo Leckner (presid.), Domingo Santana Santana (secret.), Jesús Guardiola Soler (voc.), José Antonio Almendros Ibáñez (voc.), Jesús Arauzo Pérez (voc.)
- Materias:
- Enlaces
- Tesis en acceso abierto en: e-Archivo
- Resumen
Fluidized beds are employed in industry because of their excellent properties involving heat and mass transfer, and their capability to establish and promote chemical reactions inside them. A variety of processes can occur inside a fluidized bed, including drying, heat exchange, thermal conversion of solid fuels, and coating of particles. Most of the applications of fluidized beds involve the motion of objects inside the bed. Fuel particles, catalysts, and agglomerates are examples of typical objects found inside a fluidized bed. It is necessary to characterize the motion of these objects within the bed to establish the region for proper performance and to prevent operational problems such as the existence of hot or cold spots in a reactor or the appearance of de-fluidized zones due to the existence of agglomerates. In this work, the motion of large objects immersed in a bubbling fluidized bed was experimentally studied using digital image analysis. The experiments were performed in two facilities designed for such purpose, a 2-D bed and a lab-scale 3-D bed. Different objects were tested, varying density and size. The main characteristics of the object motion were studied on the 2-D bed. By direct visualizations of the object trajectory, the preferential paths and the homogeneity of its spatial distribution were characterized. Then, the cycles described by the object in his way from and towards the surface of the bed were studied in detail. Every cycle consists of processes of descent and processes of ascent. In most cases, a series of movements of ascent and decrease interleaved along the path are observed. From this experimental evidence and considering every cycle independent from the previous history, a simple model was developed to characterize the cycles, based on four fundamental parameters: the average rising and sinking velocities of the object, the maximum depth attained along the cycle and the number of independent rising movements (number of jumps) that take place in each cycle. Concerning the rising and sinking velocities, a methodology was established for the averaging calculations, as the existence of sudden changes of trend and vibratory movements complicates the separation of the processes. The object sinking motion is linked to the dense phase sinking motion and the object rising motion is linked to the evolution of bubbles. The probability of reaching the surface by the action of a single bubble or jump was quantified, along with the existence and relative incidence of cycles with multiple jumps. Finally a simple semi-empirical model was defined to characterize the cyclical motion of the object, using the four fundamental parameters, the number of jumps during the cycle, the maximum attained depth and the average rising and sinking velocities. These latter two parameters were associated to well-known correlations for the average sinking velocity of the dense phase and the average bubble velocity, while the former ones were characterized in relation with the time of circulation of the object and also in his respective distributions of probability. The procedure is tested for a neutrally buoyant object and the results are presented. Then, the procedure is applied to objects with different sizes and densities to study the incidence of buoyant forces. The effect of the gas velocity is also studied. The results show that the semi-empirical model possesses general validity within the range of our experiments. This includes variations of the object density from lightly higher to that of the bed to values lower than half of it, changes of object size around one order of magnitude and even changes of the height of the bed and of the distribution of sizes of the material that conforms the bed. The distribution of probabilities for the number of jumps follows a geometric decay. As a consequence, a value of 45 % is obtained for the average probability by which an object that starts rising by the action of a bubble finishes in the surface of the bed without detaching from it or its trail. This also implies that there is a 55 % probability that the cycle will have a new jump. This average value is kept constant for all the experimental conditions tested. Also a parabolic profile is obtained for the distribution of depths, which can be explained considering the preferential paths of both objects and bubbles. Throughout the study, a negligible effect of buoyant forces is observed during the object rising motion, while it is relatively important and coherent with the above mentioned forces in the sinking path. Finally, in a third part a practical application of object motion in a 3-D bed is presented. The time of circulation of the objects is measured acquiring images of the surface of the bed. A comparison of the distributions of circulation times for a standard bed and for a bed in which an actuator is used to modify and improve the dynamics of the bed. The actuator consists on the low-frequency rotation of the bed distributor. The results show an improvement of the circulation of objects in the whole range being measured. The fundamental aspect consists of the fact that in a standard bed with a perforated plate distributor, the object usually disappears after several cycles, and remains in the dead zones between the holes of the distribution plate. On the other hand the incidence of long periods increases as a consequence of objects passing though the surroundings and being affected proportionally. These effects do not exist (disappearance of the object) or are minimized (long periods) when using the rotating distributor. This result suggests the possibility of increasing the range of bed parameters that assure a proper object circulation, or that to recover objects, by means of the application of the actuator. This work also includes a study of the evolution of the probability distributions of the circulation times depending on the size of the bed, on the speed of the gas and on the density and size of the object.
