Heat transport in aqueous suspensions of alumina nanoparticles

Equilibrium molecular dynamics simulations are conducted to investigate heat transport in aqueous suspensions of alumina nanoparticles. The thermal conductivity of the suspension, calculated by the Green-Kubo relations, is studied for a wide range of volume fractions, particle sizes, and temperatures. The particle volume fraction is varied in the range of 1-9% and the particle size range of concern is 1-9 nm. The temperature varies between 300 and 370 K. The radial distribution function and the radial density profiles are utilized to estimate the thickness of the ordered base-fluid nanolayer adsorbed on the particle surface. Emphasis is placed on elucidating the relationship between the thermal conductivity enhancement and the nanolayer thickness. Results show that the effective thermal conductivity increases near-linearly as a function of volume fraction, whereas the slope of this function decreases with an increase in particle size. The nanolayer thickness is independent of the particle volume fraction. The effect of particle size on thermal conductivity is studied for a volume fraction of 5%, temperature of 300 K, and pressure of 1 atm. The nanolayer thickness remains almost constant with an increase in particle size. However, both effective thermal conductivity, and nanolayer thickness normalized with particle diameter decreases sharply with increasing particle size and attains an asymptotic value at particle diameter ~ 150 nm. The effect of temperature on the effective thermal conductivity is studied for a particle size of 3 nm and volume fraction of 5%. The thermal conductivity decreases steadily with increasing temperature, whereas the nanolayer thickness remains nearly constant. For particle sizes less than 10 nm, the enhancement in thermal conductivity is significantly greater than the predictions of existing theoretical models. A strong correlation between nanolayer properties and enhanced thermal conductivity of fully dispersed nanoparticle suspensions can be deduced from the results.