Influence of ultrasonication energy on the dispersion consistency of Al2O3–glycerol nanofluid based on viscosity data, and model development for the required ultrasonication energy density

Abstract

Achieving homogenised and stable suspensions has been one of the important research topics in nanofluid investigations. Preparing nanofluids, especially from the two-step method, is often accompanied with varying degrees of agglomerations depending on some parameters. These parameters include the physical structure of the nanoparticle, the prevalent particle charge, the strength of van der Waals forces of attraction and repulsiveness strength. Amongst the methods of deagglomeration, the use of ultrasonic vibration is most popular for achieving uniform dispersion. However, there are very few works related to its effect on the thermo-physical properties of nanofluids, and above all, standardising the minimum required ultrasonication time/energy for nanofluids synthesis. In this work, the optimum energy required for uniform and initially stable nanofluid has been investigated through experimental study on the combined influence of ultrasonication time/energy, nanoparticle size, volume fraction and temperature on the viscosity of alumina–glycerol nanofluids. Three different sizes of alumina nanoparticles were synthesised with glycerol using ultrasonication-assisted two-step approach. The viscosities of the nanofluid samples were measured between temperatures of 20–70 °C for volume fractions up to 5%. Based on the present experimental results, the viscosity characteristics of the nanofluid samples were dependent on particle size, volume fraction and working temperature. Using viscometry, the optimum energy density required for preparing homogenous nanofluid was obtained for all particle sizes and volume fractions. Finally, an energy density model was derived using dimensionless analysis based on the consideration of nanoparticle binding/interaction energy in base fluid, particle size, volume fraction, temperature and other base fluid properties. The model's empirical constants were obtained using nonlinear regression based on the present experimental data.