A method to simultaneously determine the resistivity, volume expansion and temperature relation of filled conductive polymers

Abstract

Polymer composites incorporated with conducting particles exhibiting the effect of positive temperature coefficient (PTC) of resistance were first reported by Frydman in 1945. To date, three models have been proposed to interpret this phenomenon [1–3]. All of these basically suggested that the volume expansion plays an important role in the PTC effect. However, the known methods determining the relationship of resistivity-temperature, voltage-current and dielectrics-temperature usually do not involve the volume expansion directly, and the experiments on the volume expansion of the composites have been reported rarely [4, 5]. In this paper, we present a method that can simultaneously measure the volume expansion and resistance of filled conductive polymer composites with temperature, in order to study the dependence of resistivity on the volume change with respect to weight fraction of the conductive fillers. A dilatometer normally used to determine the thermal volume expansion with temperature of a solid [6] was modified to measure the resistivity simultaneously (Fig. 1). The electrodes of the sample, being situated in a glass container full of pure silicone oil, were extended by two copper wires. The ends of the two wires were welded onto the electrodes and the other ends were connected to the multi-meter for resistivity measurement. The two wires run from a ceramic ring through the inner hole of the capillary to a movable PTFE ring, and were fixed between the two rings. The capillary was tightly joined with the container using a ground glass stop and hooks. The ceramic ring holding the two wires was placed inside the neck of the container, and the PTFE ring was movable outside the capillary. Moving the PTFE ring straightens the two parallel wires to enable an accurate volume calculation. The pure silicone oil was used as the liquid medium of the dilatometer because of its high electrical insulation and inertness to the composite samples and copper wires below 160 8C. A negligible interaction of silicone oil with the samples was proved by determining the change of the sample’s weight after a 7 h isothermal treatment in silicone oil. The composite sample was moulded into a cuboid (about 4:0 3 4:0 3 8:0 mm3). Two pieces of nickel nets were incorporated in opposite surfaces of the sample. To avoid deformation and gas bubbles as much as possible, the sample was first annealed in a vacuum oven at 150 8C for 2 h, and then compression-moulded at 50 kg cmy2 for another 2 h. Following a further anneal in N2 of 1 atm at 130 8C for a third 2 h period to eliminate stresses, the sample was cooled to 25 8C at 3 8C miny1. Finally, the two net electrodes were pre-soldered. The prepared sample was then inserted into the container. The dilatometer was fixed on a support rod and immersed in a large oil bath. The bath temperature was controlled to within 0.2 8C. The real volume expansion could be expressed with the ratio of the volume at temperature T to initial volume at 25 8C, namely V=V0. The resistivity can be calculated from the sample resistance by the assumption that the areas of the electrodes do not change when the sample expands. As is known, the temperature rate is a sensitive parameter in measuring the transition behaviour of semicrystalline polymers. In our experiment, the heating or cooling rate between any two tempera-