Effective Thermal Conductivity of Nanofluids Containing Silicon Dioxide, Titanium Dioxide, Copper Oxide, Polystyrene, or Polymethylmethacrylate Nanoparticles Dispersed in Water, Ethylene Glycol, or Glycerol

The present study represents a continuation of our investigations on the effective thermal conductivity lambda eff of nanofluids by systematically varying the types of base fluids and particles. For the spherical nanoparticles with mean diameters between (20 and 175) nm, the metal oxides silicon dioxide (SiO2), titanium dioxide (TiO2), and copper oxide (CuO) as well as the polymers polystyrene (PS) and polymethylmethacrylate (PMMA) were selected to cover a broad range for the particle thermal conductivity lambda p from about (0.1 to 30) W & sdot;m-1 & sdot;K-1. The corresponding polar base fluids water, ethylene glycol, and glycerol allow to not only vary their thermal conductivity lambda bf by a factor of more than two, but also their dynamic viscosity by about three orders of magnitude. For the measurement of lambda eff of the twelve different particle-fluid combinations, i.e., TiO2 or CuO with all three liquids as well as SiO2, PS, or PMMA with water or ethylene glycol, a steady-state guarded parallel-plate instrument (GPPI) associated with an expanded (k = 2) relative uncertainty between 0.022 and 0.032 was used at atmospheric pressure over a temperature range from (283 to 358) K at varying particle volume fractions up to 0.31. The results for the thermal-conductivity ratio lambda eff<middle dot>lambda bf-1 are independent of temperature and show a moderate and relatively linear change as a function of the particle volume fraction. For similar ratios lambda p<middle dot>lambda bf-1, the experimental data for lambda eff<middle dot>lambda bf-1 are also very similar, which are above, close to, or below 1 if lambda p is larger than, comparable to, or smaller than lambda bf, respectively. For all nanofluids investigated, the Hamilton-Crosser model can describe the present measurement results and reliable experimental data reported in the literature for lambda eff<middle dot>lambda bf-1 typically within +/- 5 %. Overall, the measurement results from this work contribute to an extension of the database for lambda eff of nanofluids with respect to the investigated wide ranges of systems, temperature, and particle volume fraction.