In circulating fluidized bed (CFB) boilers, provision of required amounts of heating surfaces inside the combustor is an important design issue especially when the capacity of the boilers increases. The enclosing wall, which constitutes the principal heat transfer surfaces, needs to be supplemented with additional heat transfer surfaces. Wing wall is the most common additional heat transfer surface used in the CFB-boiler furnace at present. Despite its frequent utilization, there is no reported study available on hydrodynamics and heat transfer mechanism on wing wall in the literature.

The thesis investigates the flow dynamics and heat transfer mechanism on wing wall in both pilot plant and commercial units. It also investigates the potential of using two new heat transfer surfaces;​ walls of standpipe and cavity- type inertial separators in the CFB loop. In the experimental part of this thesis, measurements on temperature, pressure, axial and lateral solids fluxes were carried out to understand the flow hydrodynamics and heat transfer mechanism on above surfaces. Experiments were performed in a pilot scale CFB riser (1 m × 0.5 m in cross section and 5 m in height) with sand particles. In the theoretical part of this study, mechanistic models were proposed to predict the heat transfer coefficients on all those surfaces. A correlation on fractional wall coverage by the cluster on the enclosing wall was developed to improve the cluster renewal model to estimate the heat transfer coefficients on the enclosing wall. Two empirical correlations for both enclosing water walls and wing walls were also developed by relating heat transfer coefficients, which are averaged over the entire height of the absorbing walls with suspension density and average bed temperature.

Analysis of data from a 170 MWe commercial unit showed that heat transfer coefficients on the wing walls were always smaller that that on enclosing water wall. Experiments conducted in the pilot plant revealed that hydrodynamics condition on the wing wall are different from the enclosing water wall. Gas convection dominates the convective heat transfer to the wing wall whereas particle convection dominates on the wall layer. Exploratory research on new heat transfer surfaces i.e., standpipe and cavity-type inertial separators, shows an encouraging result for using these types of surfaces in the CFB loop. Use of in-furnace cavity-type inertial separator increases the overall solids hold up in the riser and hence increase the heat transfer coefficients of the enclosing wall.

Mechanistic models developed for all the surfaces predict the heat transfer coefficients within ±10% error. Empirical correlations developed from the data on commercial units predict heat transfer coefficients within ±15% error which is of the same order as the experimental error.