Department of Mechanical Engineering
Kenneth E. Goodson
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 research area, several research groups reported improvements in the thermal conductivity of nanofluids that far exceeded even the most generous predictions from effective medium models. A variety of physical mechanisms were invoked to explain these provocative data, including nanoparticle Brownian motion, thermally-induced diffusion, and molecular ordering of the nanofluid around nanoparticles.
We established a rigorous set of experiments to determine the impact of variety of nanofluids containing both nanoparticles and carbon nanotube on the stationary fluid thermal conductivity and the thermal convection coefficient in heat exchangers. Thermal conductivity measurements used diffraction-limited cross-sectional IR imaging, which provided temperature gradients and effective thermal conductivities over a broad range of temperatures in a given sample and was used to track the temporal impact of aggregation and diffusion. We used both instrumented MEMS microchannels and metal microtubes to measure the heat transfer coefficients for microfluidic heat exchangers, and larger diameter tubes were used to study the heat transfer coefficient in developing flow. The data were analyzed considering effective viscosity measurements and light-scattering data for particle distributions, and the potential improvements and problems for heat exchanger design were assessed. While particle based nanofluids show little promise for heat exchanger design, there is some indication that CNT-based nanofluids may offer improvements in thermal conductivity that outpace the impact of the viscosity increase.