• A comprehensive 2D CFD model of a single-slope solar still was developed and validated. • Coupled heat, mass, and momentum transport in humid air was simulated using COMSOL Multiphysics. • Five new correlations were proposed for predicting the freshwater productivity of the still. • The second-degree polynomial correlation demonstrated the highest predictive accuracy, lowering the mean deviation to below −3.5 % relative to Dunkle’s classical model. • The numerical model also showed strong consistency with theoretical data, achieving an average deviation of −2.1 %. • A clear proportional relationship was identified between convective heat transfer and freshwater production. Solar stills are considered among the most thermally efficient and environmentally friendly technologies for producing potable water in sunny and arid regions. In this study, a single-slope solar still was simulated using COMSOL Multiphysics 6.1 to predict evaporative mass flux and hourly water productivity. Five new correlations-including Power, Exponential, and second, third, and fourth-degree polynomial forms-were also developed to provide simple predictive tools. All proposed models were validated using both theoretical and experimental datasets. The results show that, when compared with the classical Dunkle and Chilton-Colburn correlations, the second- and third-degree polynomial models consistently yield the lowest prediction errors across all cases. For instance, the second-degree model reduces the mean error relative to Dunkle to below −3.5% for theoretical data and to −5.11% and −12.38% for two experimental datasets. The numerical simulation also demonstrates strong accuracy, with average deviations of −2.1% from theoretical values and 2.61% from experimental measurements. Overall, both the CFD model and the new correlations provide reliable and practical tools for estimating solar-still productivity, with the variation in productivity following the trend of the convective heat-transfer coefficient
Abbasi et al. (2026) studied this question.