Development of a Gas-Particle Direct-Contact Trickle-Flow Heat-Exchanger for Application in Concentrating Solar Tower Systems

Kurzfassung

Ceramic particles represent a viable alternative as heat transfer and storage medium in concentrating solar tower systems. The particles are heated in solar-receivers close to 1,000 °C. To transfer the stored heat in the particles the principles of a trickle flow reactor, was considered to be superior compared to state-of-the-art technologies, like fluidized bed- or cyclone-HX. The trickle-flow heat exchanger (TFHX) provides a relative high power density at a relative low pressure drop, meeting the requirement for the use in a CSP tower system. To date, no design recommendation is known to identify for a given particle type an optimized packing structure that emphasizes gas-particle interaction by comprising a high total particle surface and a uniform spatial particle distribution of the trickling particles within the packing void. This is why this work aimed to develop and investigated a gas-particle direct-contact trickle flow heat exchanger for the envisioned use in a particle based solar tower system, using bauxite particles of 1 mm diameter. Based on observations in literature, different packed columns were designed and investigated numerically. In the DEM simulations the dimensionless total particle surface, the particle hold-up, such as the spatial particle distribution were analyzed. It was found that, for the used particles, bar elements with a flat cross-sectional area increase the particle hold-up and provide an even spatial distribution of the trickling solids. In a developed test-setup at ambient temperature, the packing geometry was refined experimentally. During those experiments it was found, that packing structures, designed to allow the accumulation of static particles can increase the hold-up of the trickling particles. Those static particles can absorb the kinetic falling energy of the trickling solids, and thereby reduce the particle sink velocity and conversely increase the total particle surface, or hold-up. The refined packing geometry was used in hot experiments to assess the heat transfer capabilities. Particles at am�bient temperature with varying flow rates were feed into the top of the column, while air at ambient pressure was induced at the bottom of the HX. Four air inlet temperatures up to 640 °C, such as five different media flow rates were defined, where the particle and air flow rates were chosen to achieve roughly equal heat capacity flow rates, since a high temperature change in both media flows was preferred. For the highest air inlet temperature, the power density of the investigated TFHX was determined over 1,000 kW/m3 for media flow rates above 2 kg/(m2⋅ s). The volumetric heat transfer coefficient ranges in the same flow conditions at approximately 15 kW/m3, with NTU numbers for the particle and air flow of approximately 5 and 8 respectively. In was shown and discussed, that the developed TFHX in this work is capable to increase the determined power densities, volumetric heat transfer coefficients, and NTU values by approximately 100 % compared to the available data in literature. It is assumed, that the increase in performance can be deduced by an increased surface of the trickling particles within the packed column, or conversely by a reduced sink velocity of the grains of approximately 0.1 m/s, compared to the determined velocities of roughly 0.2 m/s in the literature. This leads to the assumption that an optimized packing structure in a trickle flow reactor is capable to increase the surface that interacts with the gas flow and therefore can enhance the heat transfer capabilities substantially. Due to the analogy of heat- and mass�transfer, it is assumed that this finding can also be applied in chemical reactors.