Non-uniform heat generation in modern electronic components leads to localized hotspots that can degrade performance and reduce operational lifetime, highlighting the need for effective, targeted cooling solutions. This work introduces a novel hybrid heat sink (HHS) combining precision cooling via jet impingement with localized convection enhancement using pin fins placed only around high heat-flux regions. Unlike jet-only or fully finned heat sinks, the HHS selectively targets hotspots, improving local thermal control without excessive hydraulic penalties. The performance of the HHS is numerically evaluated and compared with two baseline configurations: a pin-fin heat sink (PHS) and a jet-impingement heat sink (JHS). Simulations are conducted at low and high flow rates under a total non-uniform heat input of ≈250W. Results show that the HHS effectively controls hotspot temperatures and enhances local heat transfer compared with the PHS and JHS. It achieves up to ≈45%reduction in local hotspot peak temperature relative to the JHS, making it the most effective configuration for critical temperature management. This improvement is attributed to the synergistic effect of jet impingement and localized pin-fin structures, which combine high-precision cooling of jets with conventional pin-fin cooling, suppress recirculation zones, and maintain efficient heat removal downstream of stagnation regions. In addition to reducing local peak temperatures, the HHS provides up to ≈40%improvement in global heat transfer with only a modest increase in pumping power, demonstrating its ability to significantly control hotspot peak temperatures in electronic systems under non-uniform heat flux conditions.

Localized thermal hotspots can create steep temperature gradients within microprocessors, significantly reducing their performance and accelerating failure. This study presents a numerical investigation of a novel two-inlet heat sink (TIHS) incorporating a gyroid triply periodic minimal surface (TPMS) structure to mitigate such hotspots. The heat sink features two inlets and two outlets, with the outlets positioned diagonally opposite to the inlets. The TIHS consists of two independent flow channels that exchange heat through the TPMS walls. To replicate realistic thermal loading conditions, three non-uniform heating schemes—with five, three, and two randomly distributed hotspots—were applied to the bottom surface of the heat sink. The two-inlet configuration significantly reduced hotspot intensity and improved temperature uniformity at the outlet, because of convoluted flow paths and the large heat transfer surface area provided by the TPMS structure. Key thermal performance indicators such as maximum temperature rise, mean temperature deviation, and thermal resistance all decreased with increasing flow rate, indicating enhanced heat dissipation, although this required slightly higher pumping power. The temperature gradient along the streamwise direction indicated efficient bidirectional heat transfer between the fluid and solid regions of the TPMS structure. The heat sink maintained a uniform temperature gradient beyond a certain height, with only minor non-uniformities in localized zones, commonly referred to as “dead zones” within the TPMS. Compared to a non-TPMS model, the two-inlet gyroid-based heat sink dissipated up to 40 times more heat. These results underscore the potential of this design for efficient thermal management in power electronic systems.

URANS and LES simulations of a test facility have been performed to investigate the thermal hydraulic phenomena during a step-down thermal transient, which is typical of some important accident scenarios in sodium-cooled fast reactors (SFRs). Wall-modelled URANS simulations were used to study the entire transient duration, while LES was used to provide detailed information about its early stages, with both validating well against experimental data from the facility. The results showed thermal stratification developed above the outlet, with the lower parts of the test section becoming well-mixed after a short period, mainly due to the presence of a mean flow circulation and the unsteadiness of the jets, which resulted from a feedback loop between the flow from the inlets and the recirculating flow. Comparing different URANS models with LES showed that all tested could capture the basic flow behaviour, but only the more advanced turbulence model could capture the unsteadiness of the jets or the irregular stratification interface. It was also found that for the present case, different formulations of turbulent Prandtl number did not significantly change the overall transient response.

A numerical analysis of thermal stratification in the upper plenum of the European SCAled Pool Experiment (E-SCAPE) facility was conducted using Large Eddy Simulation (LES) to enhance the understanding of thermal hydraulics phenomena in liquid metal fast reactors (LMFR) under mixed convection conditions. The geometric complexity in the above-core structure region was represented using a porous medium model. The results provide insights into the overall flow phenomena in the upper plenum region, the thermal instability in the above-core structure region, as well as the characteristics of rounded jets that emerge from the barrel walls. Under mixed convection conditions (low flow rate conditions), the strong buoyancy causes hot lead–bismuth eutectic (LBE) to accumulate at the top first, then flowing downwards, and then exiting the region through the upper set of barrel holes. Conversely, unmixed cold LBE spreads through the lower set of barrel holes. This results in a stratified temperature distribution, with lower temperatures at the bottom, slightly higher temperatures in the middle, and the highest temperatures at the top, demonstrating thermal stratification in the upper plenum region. This stratification occurs because the jets are weak in strength, resulting in poor mixing in the upper plenum region. Flow movements are confined to regions close to the jets, while areas away from the jets experience almost no movements or negligible movements, referred to as dead zones. It is useful to note that strong circulations are observed in one of our previous studies in the same facility under forced condition, which results in good mixing and no thermal stratification. In the above-core structure region, large-scale temperature fluctuations in the form of Kelvin-Helmholtz (KH) instabilities and mixing layers have been observed when the hot fluid backflows and interacts with the cold fluid. The behaviour of the flows in the upper plenum region is dominated by the influence of different types of jets, including horizontally issued jets, jets angled upwards, and jets impinging on the components of the upper plenum. Under mixed convection conditions, the jets behave similarly to negatively inclined positively buoyant jet with minimal interactions between the top and middle jets and between the middle and bottom jets, occurring only in the vicinity of the jets. The latter stages of the jets indicate very little or almost negligible background flow movement. Additionally, high turbulence is observed in the shear layer of the jet orifice, which transitions into mixing layers after six jet diameters along the trajectory for the upper set of jets.