Design criteria and performance optimization of high-power micro heat sinks

Abstract

A three-dimensional mathematical model of micro heat sinks was developed to achieve efficient thermal management in microelectronic devices. Comprehensive design criteria based on the theory of heat transfer enhancement at the micro-scale are also proposed. On this basis, the size of the microchannel structure is designed, considering a fixed heat transfer area and heat flux. Then, a combination of a response surface approximation, an non-dominated sorting genetic algorithm, and k-means clustering are used to optimize the width and height of each microchannel. The designed structure size is combined with supercritical carbon dioxide (SCO₂) working fluid to optimize the thermal performance of micro heat sinks. The optimization results demonstrated that the clustering point I of the evaluation factor j/f_ave increased by 4.11 %, while the wall temperature T_w decreased by 4.69 %. Compared to the SCO₂ scenario, the pump power and total entropy generation were respectively 61.63 % and 6.9 % lower than those of water with a mass flow rate of 6000 kg/m²·s and an inlet temperature of 293 K. For inlet temperatures ranging from 303 K to 307 K, the evaluation factor values reported were 0.2405, 0.2018, 0.1045, 0.1453, and 0.1747 under a pressure of 7.6 MPa and flow rate of 4000 kg/m²·s. For mass flow rates ranging from 3000 kg/m²·s to 6000 kg/m²·s, values of j/f_ave were 0.0591, 0.1045, 0.1515, and 0.2084, indicating good thermal performance at relatively high mass flow rates. It was noted that as the distance from the critical point of the channel increases, the overall heat transfer performance is improved when the inlet temperature is less than the critical temperature.