Absorption of ammonia in an ammonia - ionic liquid solution

Abstract

Excessive consumption of non - renewable resources in the recent times has negative effects on the environment, one of the important of which include global warming. This suggests an urgent need for adaption to the sustainable use of the resources, for which the thermal technologies show a lot of scope. Absorption heat pump is a type of heat pump which uses a thermal compressor instead of a mechanical compressor as in the case for the conventional heat pumps. Thus, such heat pumps are indeed an alternative energy conversion technology. This thesis focuses on the use of an inventive fluid as an absorbent in the absorption thermodynamic cycles: ionic liquids. In this thesis, the process of absorption of NH₃ into an NH₃ + ionic liquid solution is studied. [emim][SCN] was selected as the ionic liquid to be used as an absorbent based on its efficiency, availability and the beneficial economic factors. In the first part of this thesis, an existing numerical vapor - liquid equilibrium model for the working pair of NH₃ - [emim][SCN] developed by Wang (2019) was validated. This is done by comparing the results of the model with the experimental data obtained by Yokozeki and Shiflett (2007) for the same working pair. The second part of this thesis includes experimental and numerical studies of the absorption process in question. For the experimental study, the absorption setup located in Process and Energy laboratory at TU Delft was used. This setup does not accommodate the complete absorption heat pump thermodynamic cycle, but just the absorption process. Firstly, H₂O - H₂O convective heat transfer experiments were performed inside the setup, for both the lower range Reynolds numbers (40-140), and for the higher range Reynolds numbers (150-360). It was ascertained that the derived correlations governing the heat transfer for both of these ranges, along with the average correlation covering all the ranges of the Reynolds number, agree with the previous literature, to ascertain that the setup worked correctly. Consecutively, the absorption experiments were performed with NH₃ - [emim][SCN] working pair. The amount of vapor flow recorded during these experiments was quite low (maximum upto 0.1\%), suggesting that most of the heat was transferred to the cooling water by means of the sensible heat transfer. A numerical model for the absorption + sensible heat transfer process developed by Wang (2019) was used as a reference to design a similar numerical model in the present work. The data obtained from the experiments was used as an input to the numerical model. It was observed that the overall heat transfer coefficient for the absorption process is indeed more than that for the convective heat transfer process. The numerical results agree with the experimental results including a moderate error margin. Therefore, the numerical model was considered validated. The conceptual reasoning behind the error margin between the numerical and the experimental results are also presented. Little can be said about the validity of the empirical correlations governing the absorption process estimated by Wang (2019), since absorption only took place during a small part of the length of the absorber (maximum 10.5\%). However, it is expected that these correlations work satisfactorily for those cases where the solution is subcooled before entering the absorber. It is concluded based on the numerical results that the correlation governing the convective heat transfer derived from the water - water experiments is considered acceptable for the present case. Finally in the last part of this thesis, an attempt was made to predict the accurate value for the viscosity of NH₃ + [emim][SCN] solution by an experimental study. A separate set of convective heat transfer experiments was performed with NH₃ + [emim][SCN] solution and the cooling water. The experimental data obtained was analyzed using the correlations obtained from the H₂O - H₂O convective heat transfer experiments. The results of this analysis were compared with the solution viscosity values predicted by Wang (2019) with a logarithmic correlation, wherein the solution viscosity is a logarithmic function of the viscosities of the individual fluids present in the solution. The reasonable agreement (with on an average 32.05 \% accuracy, as was estimated by Wang (2019)) between the viscosity values obtained by the analysis and the values predicted by Wang (2019) confirmed the validity of this logarithmic function. Recommendations are presented in the last part of this thesis for future studies.