Experimental benchmarking and refinement of nucleate pool boiling correlations to support low-grade waste heat recovery with heat pipes

Low-grade waste heat, typically within the temperature range of 30–70 °C, represents a substantial energy resource that can be effectively recovered using heat pipes (HPs). However, the absence of validated boiling heat transfer models at these temperatures significantly impedes accurate system design and performance prediction. In this study, an experimental setup was developed to systematically investigate nucleate pool boiling across a wide range of surface heat fluxes at saturation temperatures between 30 °C and 70 °C. Four working fluids commonly employed in low-temperature HP applications—water, methanol, acetone, and R141b—were tested on a polished copper surface, yielding an extensive and internally consistent dataset with a heat transfer coefficient (HTC) measurement uncertainty of ±4.2 %. The results indicated that surface heat flux is the dominant factor governing HTC, while increasing saturation temperature enhances HTC due to intensified bubble dynamics. Comparisons with classical empirical correlations—Rohsenow's, Imura's, Stephan–Abdelsalam's, and Cooper's—revealed substantial prediction errors. These discrepancies are attributed to various limitations, including the underestimation of heat flux dependence, strong sensitivity to empirical constants, and mismatches between original validation conditions and the present experimental regime. To address these issues, new boiling correlations were developed through dimensionless regression analysis, incorporating the effects of surface heat flux, saturation temperature, and fluid thermophysical properties. The resulting correlations demonstrated excellent predictive accuracy, with mean absolute percentage errors below 4 % and maximum deviations within ±10 %, thus offering a robust tool for the design and optimization of HP-based waste heat recovery systems operating in the low-temperature range.