CO₂-based heat pumps have gained considerable attention as a sustainable solution for space heating. However, their performance is significantly limited by high return-water temperatures, particularly in colder climates. Integrating an auxiliary cycle to manage return-water temperature is considered a promising approach to improve system efficiency. A key factor in this integration is the selection of a suitable working fluid for the auxiliary heat pump, as it directly affects thermal performance, safety, environmental impact, and system compatibility. While no single refrigerant meets all ideal criteria, trade-offs must be made to select the most effective working fluid. As a result, three low-GWP refrigerants, including R600a, R290, and R1234yf, were evaluated in the auxiliary cycle. In this study, a single-objective optimization was carried out using a genetic algorithm (GA) to maximize the COP of a two-stage CO₂ heat pump system with heat recovery, considering seven key operating variables. The analysis was conducted at fixed supply and return water temperatures of 75 °C and 55 °C, respectively. The results showed that R600a achieved the highest COP of 3.04, followed by R290 at 2.99 and R1234yf at 2.97. These represent improvements of 57.51%, 54.92%, and 53.89%, respectively, compared to the basic single-stage CO₂ system. This highlights the importance of working fluid selection in the auxiliary heat pump in such systems, with R600a emerging as the most promising refrigerant due to its superior COP improvement among the tested options. In addition, the results emphasize the critical role of heat recovery in overcoming performance limitations associated with high return-water temperatures.
