Electrochemical Cooling: An Alternative to Vapor-Compression Refrigeration?

Abstract

With growing populations and economies, the worldwide cooling demand is expected to triple by 2050 [1]. Today, vapor-compression refrigeration cycles are state-of-the-art technology. However, vapor-compression cycles suffer from downscaling limits, environmentally harmful working fluids, and the significant need for maintenance. Thus, alternative technologies are greatly interesting. One particularly interesting electrochemical process combines parts of a classical vapor-compression refrigeration cycle with an electrochemical cell [2] (Figure 1). In this hybrid cycle, heat is still absorbed through evaporation of the circulating fluid (-mixture). However, heat rejection and re-condensation are realized in an electrochemical cell by changing the fluid composition in a reversible electrochemical reaction. Inducing the re-condensation by an electrochemical reaction shifts the needed compression step from the gas to the liquid phase and, thus, substantially reduces the high compression work of standard vapor-compression cycles. Various electrochemical reactions could be employed, e.g., proton- or hydroxide-ion exchange reactions. Here, we focus on proton-exchange reactions.In this work, we evaluate the efficiency potential of the hybrid electrochemical cooling process. For this purpose, we analyze how the efficiency depends on process and fluid parameters and estimate efficiency limits. Based on an equilibrium thermodynamic process model, seven pre-selected redox reactions with proton transfer are investigated in detail. From the results of this investigation, we deduce crucial fluid parameters for the efficiency of the electrochemical cooling process. These parameters are subsequently used in the screening for other promising reaction pairs. The screening is based on the extensive COSMO-RS [3] database, covering a wide range of potential molecules. The property analysis of the numerous found potential reaction pairs allows tightening the thermodynamic limits beyond the Carnot efficiency.The thermodynamic process model combines standard models of vapor-compression cycles with an equilibrium-based model for the electrochemical cell. Based on defined temperatures of heat sink and source, the model inputs are the initial concentration at the inlet of the cell at high pressure (Figure 1, state 1), the concentration change or conversion in the electrochemical cell, and the minimum approach temperature. The initial concentration and conversion are numerically optimized to maximize the coefficient of performance (COP). The optimizations are subject to various constraints to ensure process operation and thermodynamic consistency. For instance, a minimum conversion is required to ensure condensation of the fluid in the electrochemical cell. The minimum conversion depends on the thermodynamic properties of the reaction pair and the initial concentration.Our case study considers a cooling process between two heat reservoirs at 20°C (heat source) and 35°C (heat sink). For the seven pre-selected redox reactions with proton transfer, the simulations result in maximum COPs between 2.8 and 5.7. The reactions differ strongly regarding maximum COP and the optimal conversion. The reaction pair propylene glycol and hydroxyacetone yields the maximum COP of 5.7. The sensitivity analysis shows that the COP mainly depends on the evaporation enthalpy and the difference in the saturation temperatures of the pure components. The evaporation enthalpy has a major influence. Large evaporation enthalpies generally increase the absorbed heat in the evaporator and thus, yield higher cooling power and increased efficiencies. Mixtures with a high difference in the saturation temperatures of the pure components can condense in an electrochemical reaction at lower conversion, decreasing the required cell work.Based on the sensitivity analysis, the COSMO-RS [3] database was screened for reaction pairs with a difference in the number of hydrogen atoms. The screening yields 32099 possible reaction pairs theoretically able to exchange 2-4 protons, while both pure components are at a gas state at 15 °C. The screening did not identify reaction pairs with substantially larger evaporation enthalpies and difference in the saturation temperatures than the best reaction pair of the pre-selected reactions. The present analysis thus suggests that a COP of 5.7 is an upper bound for this electrochemical cooling process in this case study. Since the equilibrium-based thermodynamic model neglects losses in the electrochemical cell, such as kinetic or ohmic losses, practical COPs will be lower than 5.7. With these losses, we expect that COPs are, in practice, very similar to that of vapor-compression cycles with values of 3.0 – 4.0. Thus, the potential for efficiency gains through hybrid electrochemical cooling cells seems limited.