Analysis of oxide (Al2O3, CuO, and ZnO) and CNT nanoparticles disaggregation effect on the thermal conductivity and the viscosity of nanofluids

Lee, J., Yoon, Y.J., Eaton, J.K., Goodson, K.E., and Bai, S.J., 2014, "Analysis of oxide (Al2O3, CuO, and ZnO) and CNT nanoparticles disaggregation effect on the thermal conductivity and the viscosity of nanofluids," International Journal of Precision Engineering and Manufacturing, " Vol. 15, pp. 703-710.

Oxide and CNT nanoparticles were prepared and characterized to understand the effect of disaggregation on the thermal conductivity and viscosity of nanofluids through experimental and theoretical analysis. The oxide nanofluids contained spherical oxide nanoparticles, including Al2O3, CuO, and ZnO, and the CNT nanofluids contained multiwalled carbon nanotubes in deionized water. Aggregations of numerous oxide nanoparticles were observed from Dynamic Light Scattering and Scanning Electron Microscopy. Ultrasonication and centrifugation were made to mechanically separate the agglomerated nanoparticles. Nanoparticle size decreased by 15% with one hour sonication for oxide nanoparticles with diameters in the range of 10∼100 nm and centrifugation disaggregation methods showed 36% and 40% reduction in size for the Al2O3/DI water and the CuO/DI water nanofluid, respectively. A chemical based disaggregation method for the oxide nanoparticles using surfactants and changing the pH were performed. A significant size reduction was achieved with a pH value of 4.2. The effects of agglomeration on the thermal conductivity and viscosity of nanofluids were examined based on the three-level homogenization model. The use of hydrodynamic particle nanofluid shows an effective thermal conductivity and viscosity, when the aspect ratio of the particle aggregation is below 4.1 and 2.5 for CuO/water and Al2O3/water nanofluid, respectively.

Related Projects

Nanofluids contain a small volume fraction of nanoparicles, nanowires, or nanotubes for targeted changes in thermal, rheological and/or other properties. Prior to Stanford's efforts in this...