There is a growing interest in a fluidized bed particle receiver that directly irradiates sunlight to particles in the fluidized bed as a solar thermal collector for heating. Thermal performance of directly-irradiated fluidized bed gas heater is strongly affected by the physical properties of the particles. The effect of SiC particle size on heat transfer characteristics in the solar fluidized bed gas heater (50 mm-ID × 100 mm high) has been determined. The outlet gas temperatures showed a maximum value with increasing gas velocity due to the particles motion by bubble behavior in the bed, and the maximum values were found at 3.6 times of Umf for fine SiC and less than 2.0 times of Umf for coarse SiC. Heat absorption from the receiver increased with increasing gas velocity, showing with maximum 18 W for the fine SiC and 23 W for the coarse SiC at 4.5 times of Umf. The thermal efficiency of the receiver increased with increasing gas velocity, but was affected by the content of finer particles. The maximum thermal efficiency of the receiver was 14% for fine SiC and 20% for coarse SiC within the experimental range, but showing higher for the fine SiC at the same gas velocity. A design consideration was proposed to improve the thermal efficiency of the system.
Fluidized beds have been utilized for various chemical and physical applications including heat transfer such as the gas–solid heat exchanger. It is advantageous to use carbon nanotubes (CNTs) with high thermal conductivity as bed materials for heat transfer enhancement in a direct gas–solid contacting heat exchanger. However, the poor fluidization of CNTs is the biggest challenge due to the strong cohesive force between the particles. A control over the macroscopic shapes of CNT powders is required for their application. A preparation method of CNT microbeads has been proposed to be suitable for fluidized bed applications. The method is characterized by using m-cresol known as processing solvents for fabrication of the CNT microbeads. Multiwalled CNT powders were directly mixed with m-cresol to yield a thick paste-like material. The paste droplets were rolled into round particles with in pure water with and without surfactant. The obtained particles were dried in a vacuum oven. The obtained microbeads have diameters ranging 300–2200 μm and apparent particle density of 350–400 kg/m3, which corresponds to Geldart group B in the fluidization classification. The micrograph of the CNT microbeads exhibited stacked nanotubes array on the surface, indicating obvious densification of the raw CNT powders. The microbeads prepared in water containing surfactant have better shape factor such as circularity and solidity. The thermal conductivity of the microbeads is about 1.18 W/mK in a bulk state, which is much higher than raw CNT powder (0.032 W/mK). The flowability and fluidization characteristics of the multiwalled CNT (MWCNT) microbeads showed a possibility as promising bed material suitable for the fluidized bed heat exchanger.
A measurement method of the apparent particle density of the carbon nanotube (CNT) particles, characterized by enveloped volume formed by loosely entangled nanotubes, has been proposed for the CNT fluidized bed application. The method is characterized by obtaining the enveloped volume from the CNTs imaging under the free falling condition similar to the fluidized bed. The shape of the falling CNT particles in a column (0.1 m long × 0.012 m wide × 0.60 m high) was photographed using a high-speed camera under the sedimentation condition, and the apparent CNT particle density was calculated from the enveloped volume obtained by image-processing for the particles images. The apparent densities and solid holdups by the imaging method at various conditions were compared with those by the previous Hg-porosimetry method for the two types of CNTs (a vertically aligned CNT and two entangle CNTs) and the nonporous polycarbonate particle (a reference particle). The imaging method reflects well the packed bed and fluidized bed phenomena observed in the experiments with reasonable solid holdups, compared with the Hg-porosimetry method showing high densities and low holdups. The sizes of CNT particles predicted with the density by the imaging method were in good agreement with the measured mean particle sizes when calculated based on the Richardson–Zaki equation, indicating the imaging method represented well the enveloped volume and shape formed by entangled nanotubes on the CNTs.