Optimization of experimental conditions of hot wire method in thermal conductivity measurements

Abstract: This work studied the hot wire method in measuring thermal conductivity at room temperature. The purpose is to find the optimized experimental conditions to minimize natural convection in liquid for this method, which will be taken as reference for high temperature thermal conductivity measurement of slag. Combining room temperature experiments and simulation with COMSOL Multiphysics 4.2a, the study on different experimental parameters which may influence the accuracy of the measured thermal conductivity was conducted. The parameters studied were the diameter of crucible, the position of wire in the liquid, including z direction and x-y plane position, diameter of the hot wire, and current used in the measurement. In COMSOL simulations, the maximum natural convection velocity value was used to evaluate the natural convection in the liquid. The experiment results showed after 4~5 seconds of the measuring process, the natural convection already happened. Also when current was fixed, the thinner the hot wire, the larger convection it would cause. This is because thinner wire generates more heat per unit surface area. Using higher current in measuring, more heat generation improved accuracy of result but also had earlier and larger effect on convection. Both simulation and experiments showed that with the height of the liquid fixed, the smaller diameter of the crucible (not small to the level which is comparable with hot wire diameter), the higher the position in z direction (still covered by liquid), the less natural convection effect existed. But the difference was not significant. The radius-direction position change didn’t influence the result much as long as the wire was not too close to the wall.