Adsorption for desalination, cooling and generation of electricity from low-grade heat
Olkis, Christopher Jan
Two thirds of all primary energy are converted into waste heat causing thermal pollution and economic loss. The heat is emitted at different temperatures, but most of it is 100 °C and below, where it is more difficult for another process to use it. This work concerns the utilisation of waste heat < 100 °C for desalination and the generation of electricity in the Reverse Electrodialysis Heat-to-Power process. The process combines a Reverse Electrodialysis membrane stack RED with a thermal desalination system in a closed loop. Two salt solutions at different concentrations circulate between them; the RED unit generates electricity, while the thermal desalination unit restores the salt concentration difference between the two solutions. The thermodynamics of salt solutions are fundamental to the process and were experimentally assessed using a Barker’s cell to determine saturation pressures and temperatures. Novel salt solutions for K, Li and Cs acetate salts are measured and the results have been presented in a joint publication with the Università degli Studi di Palermo, Italy. The results were fitted to the Pitzer model for further process modelling of both RED and desalination. The salt measurements were conducted for T = 10-90 °C in increments of 10 °C and concentrations 2-8 mol/kg resulting in boiling point elevations up to 12 °C. Different common desalination methods were identified and have been modelled in Matlab and UniSim to compare their performance. The Matlab models used the Pitzer parameters from either the literature or the previous investigation on the novel salt solutions. UniSim was used in combination with the OLI electrolytes package allowing UniSim to access precise salt properties for established salts. Multi-effect distillation is the best performing, benchmarking desalination method as the Specific Thermal Consumption increases with the number of effects ranging from 25 kWh/m3 for large systems (27 effects) to 150 kWh/m3 for compact systems (5 effects). Absorption vapour compression desalination was identified as another promising desalination method with a simpler process design than multi-effect distillation. A UniSim model has shown that the performance of the system is constant at 250 kWh/m3 and almost independent of the salt concentration. The focus of this work is adsorption desalination as it has the simplest process scheme of all desalination methods with a minimum of moving parts and pumps. Experimental systems in the literature have proven the feasibility of the process, but all systems are very large in size with the best performing system using 144 kg of silica gel. The large size makes it difficult to test novel adsorption materials and system components, which is necessary to advance the technology. Thus, an experimental adsorption desalinator was designed and build with 0.025 - 0.4 kg adsorption material capacity. In addition, the system can also be used as an adsorption chiller. The first set of experiments was conducted using silica gel showing that the small size is not detrimental to the system performance. A maximum Specific Daily Water Production SDWP of 10.9 kgw/(kgsgd) was experimentally obtained at 80 °C, which is one of the best results ever reported for silica gel. A full characterisation of the test rig using silica gel was presented in two publications highlighting the novelty of the small scale, novel analysis methods (e.g. cycle analysis, thermal response) and regeneration temperatures as low as 40 °C. In addition, an ionogel adsorption material was analysed in a water adsorption process for the first time. The novel material is composed of a silica gel support structure impregnated with an ionic liquid. The ionic liquid was chosen based on an equilibrium data screening that identified 1-Ethyl-3-methylimidazolium acetate as best for the application. Experiments in the test rig showed that the ionogel can be regenerated using a waste heat source of 25 °C achieving SDWP = 6.7 kgw/(kgsgd). An increase of the regeneration temperature to 45 °C improves the SDWP to 17.5 kgw/(kgsgd). The material appears stable in terms of performance and even improved by 30 % due to thermal swings in the test rig, but minor leakages of the ionic liquid from the support structure were observed. The results of the experiments on the test rig provided a better understanding of the process and led to the development of Adsorption Reverse Electrodialysis process (ADRED). The ADRED process integrates adsorption desalination into the closedloop of the RED Heat-to-Power process. A thermodynamic analysis assesses different adsorption materials and salts to identify the best combination having the highest efficiency. In addition, the results show that the energy consumption of adsorption desalination is independent of the salt concentration. Thus, ADRED is not limited by the salt concentration on the regeneration side. A dynamic ADRED model was developed and validated. The AD side is validated on the results of the experimental test rig for regeneration temperatures 40-60 °C, while a validated RED model was adapted from literature. A case study was performed for a system powered by 700 MW of waste heat emitted by a power plant at 40 °C. The simulations show a net electricity output of 2.3 MW with a net exergy efficiency of 7.1 % and net energy efficiency of 0.33 %.