Översikt
- Datum:Startar 8 januari 2026, 10:00Slutar 8 januari 2026, 13:00
- Plats:Hörsalsvägen 4, Lecture Hall - HA1 (Room 110); Zoom password: BFB
- Opponent:Prof. Ruud van Ommen, Full Professor of Chemical Engineering at Delft University of Technology (TU Delft), The Netherlands
- AvhandlingLäs avhandlingen (Öppnas i ny flik)
Bubbling fluidized beds with horizontal solids crossflow can be broadly grouped according to function: 1) those designed for solids looping, as in indirect pyrolysis or gasification, and chemical or calcium looping; and 2) those employed in high-throughput solids processing, such as drying, iron ore reduction, and pharmaceutical manufacturing. Despite their widespread industrial application and the growing demand for their use in new process designs, there remains a lack of a detailed understanding of the flow characteristics in fluidized beds with horizontal solids throughput. This knowledge gap represents a major challenge for optimizing reactor design and advancing industrial implementation.
This thesis aims to develop a mechanistic understanding of how forced horizontal convection of solids in a bubbling fluidized bed influences the solids flow characteristics. The main research objectives are to: (i) assess experimental methods for quantifying the solids circulation rate; (ii) evaluate the efficiencies of different mechanisms to induce horizontal convection of solids; (iii) characterize bed solids transport—specifically, the interrelated effects of solids convection and dispersion—and the resulting fluidization quality; (iv) investigate how the solids crossflow influences overall flow structures; (v) examine the influences of frictional losses on bed solids flow, including the rheological properties of the dense suspension; and (vi) explore how the solids crossflow affects the mixing of a secondary solids phase consisting of large, light particles.
Experiments were conducted in a cold-flow model that comprised a closed-loop system in which solids were circulated horizontally via a solids-conveying module. The apparatus was designed and operated according to Glicksman’s simplified scaling laws. In this cold model, fine bronze particles are fluidized with ambient air to fluid-dynamically resemble conditions representative of industrial-scale thermochemical fuel conversion applications. The industrial unit being modeled features a bed channel with a cross-sectional width of 0.92 m and a transport loop length of 10.35 m, in which coarse, sand-like (Geldart B-type) particles are fluidized with flue gas at approximately 800°C.
Four measurement methods for quantifying solids circulation were evaluated—namely, integral mass accumulation, differential mass accumulation, thermal tracing, and magnetic solids tracing—with the latter proving to be the most precise and robust. Using this method, five solids-conveying configurations based on different fluid-dynamical mechanisms were tested: (a) free solids splashing, which relies on bubble bursts to eject particles; (b) confined solids splashing, whereby turbulent fluidization creates particle transport that is dominated by eddies and bubble buoyancy; (c) slugging, whereby gas slugs drive particle movement in vertical ducts; (d) solids entrainment, which is achieved by elutriation at high gas velocities; and (e) directed gas injection, which imparts lateral momentum through angled nozzles. Conveying solids under a controlled bubbling fluidization regime was found to be the most efficient configuration for promoting horizontal transport of solids in the system.
The horizontal solids flow established in the channel was evaluated under various operational conditions using a combination of experimental and modeling approaches, ranging from reduced-order descriptions to Eulerian–Eulerian computational fluid dynamics (CFD) simulations. A positive linear relationship was observed between the solids dispersion coefficient and the mean solids velocity, both evaluated in the streamwise direction. This was further explained by the CFD simulations, which indicated that at low crossflow rates, the solids flow organizes into coherent, counter-rotating vortices along the bubble paths. In contrast, at high crossflow rates, these structures are disrupted, resulting in less streamlined (stronger mixing) and more elongated (longer characteristic lengths) flow patterns. In addition, microscale dispersion was found to be dominated by bubble- and eddy-induced mixing, rather than by random particle motion or collisions. Rheological analysis revealed that the bed exhibits shear-thinning behavior and that single-phase models for non-circular open-flow channels underestimate the influence of geometry on gas-solids flows.
Analysis of the transport behavior of large, light particles added to the bed as a lean phase showed that, similar to bulk solids, a positive correlation exists between dispersion and convection, with sensitivity strongly dependent on the degree of fluidization. The dispersion of the lean particles was greater than that of the bulk solids; it remained at a similar level for both low and high fluidization velocities. However, compared to the bulk solids, the lean particles were transported at lower convection velocities under high fluidization levels. In contrast, at low fluidization levels, the particles formed a layer above the dense bed and were conveyed in a plug-like manner, exhibiting much higher horizontal velocities than the bulk solids.
This thesis aims to develop a mechanistic understanding of how forced horizontal convection of solids in a bubbling fluidized bed influences the solids flow characteristics. The main research objectives are to: (i) assess experimental methods for quantifying the solids circulation rate; (ii) evaluate the efficiencies of different mechanisms to induce horizontal convection of solids; (iii) characterize bed solids transport—specifically, the interrelated effects of solids convection and dispersion—and the resulting fluidization quality; (iv) investigate how the solids crossflow influences overall flow structures; (v) examine the influences of frictional losses on bed solids flow, including the rheological properties of the dense suspension; and (vi) explore how the solids crossflow affects the mixing of a secondary solids phase consisting of large, light particles.
Experiments were conducted in a cold-flow model that comprised a closed-loop system in which solids were circulated horizontally via a solids-conveying module. The apparatus was designed and operated according to Glicksman’s simplified scaling laws. In this cold model, fine bronze particles are fluidized with ambient air to fluid-dynamically resemble conditions representative of industrial-scale thermochemical fuel conversion applications. The industrial unit being modeled features a bed channel with a cross-sectional width of 0.92 m and a transport loop length of 10.35 m, in which coarse, sand-like (Geldart B-type) particles are fluidized with flue gas at approximately 800°C.
Four measurement methods for quantifying solids circulation were evaluated—namely, integral mass accumulation, differential mass accumulation, thermal tracing, and magnetic solids tracing—with the latter proving to be the most precise and robust. Using this method, five solids-conveying configurations based on different fluid-dynamical mechanisms were tested: (a) free solids splashing, which relies on bubble bursts to eject particles; (b) confined solids splashing, whereby turbulent fluidization creates particle transport that is dominated by eddies and bubble buoyancy; (c) slugging, whereby gas slugs drive particle movement in vertical ducts; (d) solids entrainment, which is achieved by elutriation at high gas velocities; and (e) directed gas injection, which imparts lateral momentum through angled nozzles. Conveying solids under a controlled bubbling fluidization regime was found to be the most efficient configuration for promoting horizontal transport of solids in the system.
The horizontal solids flow established in the channel was evaluated under various operational conditions using a combination of experimental and modeling approaches, ranging from reduced-order descriptions to Eulerian–Eulerian computational fluid dynamics (CFD) simulations. A positive linear relationship was observed between the solids dispersion coefficient and the mean solids velocity, both evaluated in the streamwise direction. This was further explained by the CFD simulations, which indicated that at low crossflow rates, the solids flow organizes into coherent, counter-rotating vortices along the bubble paths. In contrast, at high crossflow rates, these structures are disrupted, resulting in less streamlined (stronger mixing) and more elongated (longer characteristic lengths) flow patterns. In addition, microscale dispersion was found to be dominated by bubble- and eddy-induced mixing, rather than by random particle motion or collisions. Rheological analysis revealed that the bed exhibits shear-thinning behavior and that single-phase models for non-circular open-flow channels underestimate the influence of geometry on gas-solids flows.
Analysis of the transport behavior of large, light particles added to the bed as a lean phase showed that, similar to bulk solids, a positive correlation exists between dispersion and convection, with sensitivity strongly dependent on the degree of fluidization. The dispersion of the lean particles was greater than that of the bulk solids; it remained at a similar level for both low and high fluidization velocities. However, compared to the bulk solids, the lean particles were transported at lower convection velocities under high fluidization levels. In contrast, at low fluidization levels, the particles formed a layer above the dense bed and were conveyed in a plug-like manner, exhibiting much higher horizontal velocities than the bulk solids.
