Design of Sustainable Refrigerants by multi-scale Modeling

Identifier:  TDX:4517

Handle: https://hdl.handle.net/20.500.11797/TDX4517

Authors:  Albà I Garriga, Carlos

Abstract:
As Europe enforces mandates to substantially phase down the emission of high global warming potential refrigerants, a pressing challenge emerges in the refrigeration and air conditioning industry: the development of environmentally sustainable alternatives to hydrofluorocarbons. In response, this thesis focuses on the implementation of multiscale modelling tools for the development of a consistent methodology to identify new refrigerants with lower emissions. The proposed approach relies on the robust polar soft-SAFT equation of state to predict thermodynamic properties required for their technical evaluation at conditions relevant for cooling applications, in combination with artificial intelligence neural networks integrated using molecular descriptors through COSMO-RS. Overall, the strength of this methodology lies in the development of accurate coarse-grain models that provide the required data for the rational design of new refrigerants, without the need of further experiments. Based on these data, an energy, exergy, economic and environmental (4E) analysis is conducted to minimize retrofitting costs of existing systems, in order to address data gaps and enhance the accuracy of predictions for thermodynamic properties and system performance. This framework is applied across a wide range of operating conditions and system configurations to ensure robustness and accuracy in the context of waste-heat recovery, finding a potential blend [(60/40) wt.% R1243zf + R1234ze(E)] that can effectively replace R134a. A multi-objective optimization method using statistical tools has been developed to assess the impact of design factors on potential enhancements in cooling cycle performance, finding annual cost savings and reduction in CO2 emissions. Additional analysis of environmental impact and projected cost is included to quantify the impact associated with their use and emissions, aiding in the identification of appropriate drop-ins from a holistic techno-environmental-economic perspective. Finally, this thesis explores promising retrofitting alternatives to CO2, overcoming its safety limitation in sub-critical cascade cycles, extending the methodology presented to non-fluorinated compounds.