Construction and Analysis of Heat-and Mass Exchangers for Liquid Desiccant Systems

Construction and Analysis of Heat-and Mass Exchangers for Liquid Desiccant Systems Novel flat plate and tube-bundle heat and mass exchangers were designed, built, and experimentally examined in combination with an aqueous solution of lithium chloride as a desiccant. Prior to the construction of the components, two test rigs were built for the evaluation of several textiles and liquid desiccant distributors. The first test rig was used for the evaluation of liquid desiccant wicking performance of different fabrics. Several textiles with different compositions, thicknesses, and surface densities were tested at different desiccant solution mass fractions. The evaluation aimed to determine the textile with the highest absorption capacity and the best diffusion behavior. The best performing textile was implemented as a contact surface between the air and the desiccant solution in the heat and mass exchangers. The second test-rig was utilized for the evaluation of desiccant solution distribution over the exposed surface. Several tests were conducted for different geometry profiles and different solution flow rates. The experiments aimed to find a liquid desiccant distributor that is capable of facilitating the maximum uniform distribution of the liquid desiccant at low flow rates. Two plate-type liquid desiccant absorbers were implemented in internally-cooled and adiabatic dehumidification modes for air conditioning and drying applications, respectively. Also, two tube-bundles were implemented in an internally-heated mode for the regeneration of the liquid desiccant. Several experiments were conducted with the components that have been constructed in different processes; adiabatic dehumidification mode, internally-cooled absorber, and internally-heated regenerator. A parametric analysis was performed to evaluate the influence of some of the most important operational parameters on the components performance. The performance is represented by the moisture removal rate and the component effectiveness. The absorber performance was studied as a function of the desiccant solution flow rate, cooling water flow rate, air inlet humidity ratio, and air inlet temperature. Also, the liquid desiccant regenerator was studied as a function of the desiccant solution flow rate, desiccant solution inlet temperature, heating water inlet temperature, and air inlet temperature. The results indicate that water transfer rates increase with desiccant flow rate, but those results are not significant for the investigated range of desiccant flow rates. Increasing the cooling water flow rates also has a positive impact on water transfer rates between the desiccant and the air streams. Higher water temperatures increase the regeneration rate at the regenerator. Also, the results showed that increasing the air inlet temperature has a small effect on the moisture removal rate for both the absorber and regenerator experiments in the internally cooled or heated modes, respectively. The results of the internally cooled dehumidification showed a consistent reduction in the humidity ratio. In the experimental runs the change in the relative humidity ranges between 12 and 18 %-points accompanied with a change in the humidity ratio that ranges between 2.4 and 4.1 g/kg. For the given experimental setup, the conditions of the air leaving the internally cooled absorber lie within the thermal comfort zone according to European and American standards. Furthermore, a comparison between an internally cooled absorber with a cooling water temperature of 20°C and a conventional vapor compression system with an apparatus dew point temperature of 12 °C is performed. The results show that the dehumidification through the internally cooled absorber leads to savings in the dehumidification load of about 40% when compared to cooling the air below its dew point temperature in conventional air conditioning systems. The results of the liquid desiccant regeneration through the tube-bundle showed that heating-water inlet temperature (at constant mass flow rate of the heating water) has the highest impact on the moisture removal rate which was varied in the range between 50 °C and 90 °C, through which the diluted solution is regenerated with an increase of the LiCl-mass fraction in the regeneration of a value between 0.01 kg/kg and 0.06 kg/kg. Moreover, the air to solution mass ratio has a major impact on the performance of the regenerator. Furthermore, the plate-type absorber experimental results were compared with results of a numerical finite difference model that has been developed at the Institute of Solar and Engineering Systems. Significant disagreements are found when comparing the numerical model to the majority of the experimental results for the plate type absorbers. The model results over predict the heat and mass transfer compared to the experimental results beyond the experimental uncertainty. This is a result of ideal conditions and assumptions of uniform distribution of the circulated fluids. In the absorption experiments, the model yields moisture transfer rates of 11% to 35%-points than those observed in the dehumidification experiments performed in the laboratory. Also, the deviation of results was larger for the demonstration plant measurements compared to the laboratory measurements. However, the moisture transfer rate trends were similar. In the framework of the presented PhD study a liquid desiccant demonstration plant including instrumentation was developed and erected. The developed dehumidifier and regenerator were implemented in an open cycle liquid desiccant demonstration plant for drying hay bales. The components are installed in a 20 foot container at the Hessian State Domain Frankenhausen, Germany. Several adiabatic dehumidification experiments were performed in different seasons of the year by applying ambient air. The absorber aims at reducing the process air humidity ratio and heating up the air by few Kelvins above the ambient temperature. The performance of the absorber was studied as a function of the desiccant solution flow rate for a fixed air flow rate with varying ambient conditions. The experimental results showed a consistent increase in the process air temperature in the range of 3 to 8.5 K accompanied with a reduction in the air inlet humidity ratio in the range of 1.3 to 4.3 g/kg, depending on the desiccant flow rate and the ambient air conditions. The maximum change in the solution concentration was 5.7 % points for an air to desiccant mass ratio (G/L) of 82. This concentration spread yields a volumetric energy storage capacity of about 430 MJ/m3. This value is very low compared to the simulated value as a result of insufficient wetting of the absorber plates. In an additional test sequence, the desiccant solution was pumped in the absorber in an intermittent mode, in order to reach higher storage capacities and to improve wetting of the contact surface. In this measurement, a concentration spread of about 13 % points and an energy storage capacity of about 900 MJ/m3 (about 8 times more than for continuous flow with the same desiccant mass flow rate) was observed.