Discover nano · ·Added 13 Nov 2026

Thermodynamic analysis of electroosmotic transport in a williamson ternary nanofluid over a riga surface with coupled heat and reactive mass transfer.

Torikul Islam, B M Jewel Rana, Md Sondhi Sheikh, Md Yousuf Ali, Mohammad Afikuzzaman

DOI ↗

Abstract

Accurate prediction of coupled momentum, heat, and mass transport in chemically reactive electrokinetic systems is important for the design of microscale thermal and electrochemical technologies. This study investigates the coupled transport characteristics of a Williamson ternary nanofluid over a Riga surface embedded in a Darcy-Forchheimer porous medium, incorporating electroosmotic effects, nonlinear thermal radiation, volumetric heat generation, homogeneous chemical reaction, and pollutant transport. The resulting nonlinear similarity equations are solved using MATLAB bvp4c solver, and a Bayesian-regularized artificial neural network (BR-ANN) is employed as a surrogate model for predicting principal transport quantities. The results reveal that electroosmotic forcing and electromagnetic actuation significantly enhance momentum, heat, and mass transport, whereas Darcy-Forchheimer resistance suppresses fluid motion. Nonlinear thermal radiation and volumetric heat generation increase thermal field, while homogeneous chemical reaction substantially alters concentration distribution and associated mass-transfer characteristics. Increasing ternary nanoparticle volume fraction to 3% produces marked enhancements in the transport performance, with skin-friction coefficient, Nusselt number, and Sherwood number increasing by 215.56%, 75.13%, and 10.77%, respectively. These results demonstrate the pronounced influence of ternary nanoparticle loading on momentum, heat, and mass transfer in the coupled electrokinetic system. Furthermore, the BR-ANN exhibits excellent predictive capability, achieving a correlation coefficient of R = 0.99999, with mean squared errors ranging from 10⁻⁷ to 10⁻¹⁰ for the predicted transport quantities. The combined numerical-ANN framework provides a reliable and computationally efficient approach for analyzing and optimizing chemically reactive electrokinetic nanofluid systems with potential applications in microreactors, electrochemical devices, catalytic microfluidic systems, lab-on-a-chip technologies, and microscale thermal management.

Abstract as supplied by the indexing source. © the authors / publisher.