Statistical Analysis of Thermal Conductivity Experimentally Measured in Ethylene Glycol - Based Nanofluids.

We collected literature data of thermal conductivity experimentally measured in ethylene glycol-based nanofluids and investigated the influence of concentration, temperature and nanoparticle size. We implemented statistical linear regression analysis of all data points and examined four separate nanoparticle materials - alumina, titania, copper oxide and carbon-nanotubes. We found that the statistical correlations are in good agreement with Maxwell's effective medium theory, despite large scatter in the data. The thermal conductivity increases linearly with concentration, and in the case of carbon-nanotubes with temperature, whereas the nanoparticle size shows significant influence for alumina and titania. The large scatter in the experimental data is one of the main problems. We suggest that there is a need for careful, detailed characterizations and measurements to quantify the potential of nanofluids.