Numerical investigation of entropy generation to predict irreversibilities in nanofluid flow within a microchannel: Effects of Brownian diffusion, shear rate and viscosity gradient

- Irreversibilities in nanofluid flow are evaluated considering particle migration.
- Effects of viscosity gradient, shear rate, and Brownian diffusion are considered.
- Brownian impact reduces at higher concentrations compared with other factors.
- Total entropy generation in the microchannel decreases by particles enlargement.
- An ANN model for entropy generation rates is developed using the numerical data.

In the present contribution, irreversibilities caused by heat transfer and friction for the water-TiO2 nanofluid flow in a circular microchannel are investigated by evaluating entropy generation rates. The effects of viscosity gradient, non-uniform shear rate and Brownian diffusion on particle migration are taken into account in order to examine the effect of nanoparticle arrangement on entropy generation rates. The results show that nanoparticle migration alters concentration distribution and consequently, changes entropy generation rates. Nanoparticle migration increases concentration of the particles in central regions, and this migration is more noticeable for higher mean concentrations and larger particles. Thermal entropy generation rate intensifies with increasing wall heat flux and particle size while decreases with increasing concentration. Frictional entropy generation rate increases by concentration increment and decreases by particles enlargement, while it changes trivially by increasing wall heat flux. Frictional entropy generation rate is larger than thermal entropy generation rate in the microchannel under study and therefore, total entropy generation mostly stems from friction. Thus, total entropy generation rate decreases by particles enlargement, which is a positive result according to second law of thermodynamics. Eventually, a model for entropy generation rates is developed using the numerical data by means of Artificial Neural Network (ANN).