Calvin H. Li 1 Department of Mechanical, Industrial, and Manufacturing Engineering, University of Toledo, Toledo, OH 12180 Wesley Williams Jacopo Buongiorno Department of Nuclear Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139 Lin-Wen Hu Nuclear Reactor Laboratory, Massachusetts Institute of Technology, Cambridge, MA 02139 G. P. Peterson Department of Mechanical Engineering, University of Colorado, Boulder, CO 80309-0017 e-mail: bud.peterson@colorado.edu Transient and Steady-State Experimental Comparison Study of Effective Thermal Conductivity of Al 2 O 3 /Water Nanofluids Nanofluids are being studied for their potential to enhance heat transfer, which could have a significant impact on energy generation and storage systems. However, only limited experimental data on metal and metal-oxide based nanofluids, showing enhance- ment of the thermal conductivity, are currently available. Moreover, the majority of the data currently available have been obtained using transient methods. Some controversy exists as to the validity of the measured enhancement and the possibility that this en- hancement may be an artifact of the experimental methodology. In the current investiga- tion, Al 2 O 3 /water nanofluids with normal diameters of 47 nm at different volume frac- tions (0.5%, 2%, 4%, and 6%) have been investigated, using two different methodologies: a transient hot-wire method and a steady-state cut-bar method. The comparison of the measured data obtained using these two different experimental systems at room tempera- ture was conducted and the experimental data at higher temperatures were obtained with steady-state cut-bar method and compared with previously reported data obtained using a transient hot-wire method. The arguments that the methodology is the cause of the observed enhancement of nanofluids effective thermal conductivity are evaluated and resolved. It is clear from the results that at room temperature, both the steady-state cut-bar and transient hot-wire methods result in nearly identical values for the effective thermal conductivity of the nanofluids tested, while at higher temperatures, the onset of natural convection results in larger measured effective thermal conductivities for the hot-wire method than those obtained using the steady-state cut-bar method. The experi- mental data at room temperature were also compared with previously reported data at room temperature and current available theoretical models, and the deviations of experi- mental data from the predicted values are presented and discussed. DOI: 10.1115/1.2789719 Keywords: effective thermal conductivity, nanoparticle suspensions, nanofluids, transient hot-wire method, steady state cut-bar method Introduction There are two principal experimental methods typically used to measure the effective thermal conductivity of nanoparticle suspen- sions: transient methods and the steady-state methods. The most commonly used transient method utilized for the measurement of the effective thermal conductivity of nanoparticle suspensions, nanofluids, is the transient hot-wire method. Nagasaka and Na- gashima 1 first applied this method to measure the thermophysi- cal properties of electrically conducting liquids. In this approach, an electrically insulating coated platinum hot wire is suspended symmetrically in a liquid contained within a vertical cylindrical container. This hot wire serves as both a heating element, through electrical resistance heating, and as a thermometer, by measuring the temperature dependent change in the electrical resistance of the platinum wire. The thermal conductivity can be calculated from the relationship between the electrical and thermal conduc- tivity as Tt - T ref = q/4k ln 4K a 2 C t 1 where Tt is the temperature of the platinum hot wire in the fluid at time t, T ref is the temperature of the test cell, q is the applied electric power applied to the hot wire, k is the thermal conductiv- ity, K is the thermal diffusivity of the test fluid, a is the radius of the platinum hot wire, and ln C = g, where g is Euler’s constant. This relationship between T and lnt is linear. The data of T were valid only over a valid range of lnt, namely, between time t 1 and time t 2 , the thermal conductivity of the fluid can be calcu- lated as k = q 4T 2 - T 1  ln t 2 t 1  2 where T 2 - T 1 is the temperature difference of the platinum hot wire between times t 1 and t 2 . Recently, a number of investigators have utilized the transient hot-wire method to measure the effec- tive thermal conductivity of a number of different nanoparticle suspensions. These investigations included those of Choi 2, Eastman et al. 3, Lee et al. 4, Eastman et al. 5, Xuan and Li 6,7, and Xie et al. 8,9 and involve the effective thermal con- ductivities of a wide range of different nanofluids. However, the effective thermal conductivities of nanofluids obtained supported by steady-state methods are quite limited. The most widely refer- 1 Corresponding author. Contributed by the Heat Transfer Division of ASME for publication in the JOUR- NAL OF HEAT TRANSFER. Manuscript received January 22, 2007; final manuscript re- ceived March 5, 2007; published online March 18, 2008. Review conducted by Christopher Dames. Paper presented at the ASME 2006 Energy Nanotechnology International Conference ENIC2006, Boston, MA, June 26–28, 2006. Journal of Heat Transfer APRIL 2008, Vol. 130 / 042407-1 Copyright © 2008 by ASME Downloaded 18 Mar 2008 to 131.183.20.160. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm