Dielectric and physicochemical properties of a composite material prepared by incorporating carbon black particles into a polymer matrix were investigated. Two types of carbon blacks, having very different structures of aggregates, were used. The volume fraction of the carbon blacks ranged from 0.2% to 7%, i.e. below and above the percolation threshold concentration observed from the measurements of dc conductivity. The composite samples were characterized in terms of: swelling by a compatible solvent, electron paramagnetic resonance (EPR) response, and frequency variation of permittivity. First, the article attempts to evaluate the diffusion coefficient of an appropriate solvent in these materials. Sorption kinetics experiments with toluene indicate that the initial uptake of solvent exhibits a square root dependence in time as a consequence of Fick’s law and permit to evaluate the effective diffusion coefficient in the range 10−11–10−12 m2 s−1 depending on the volume fraction of the carbon black in the sample. Second, the analysis of the carbon black concentration dependence of the intensity and linewidth of the EPR signals indicates that EPR is an important experimental probe of the structure of the elasticity network. The most notable feature of the present work is that we find a correlation of the percolation threshold concentration which is detected from the dc electrical conductivity with moments of the EPR lines. The conclusions on the elasticity networks deduced from swelling measurements are confirmed by EPR data carried out on swollen samples. On qualitative grounds the role of the specific surface of carbon black is further analyzed. It is suggested that the elasticity network is mainly controlled by secondary (respectively primary) aggregates for samples containing low (respectively high) specific surface carbon blacks. Last, the article reports precise experimental data on the permittivity of these composite materials as a function of frequency. Thanks to a sensitive measurement technique using an impedance analyzer, we are able to measure the complex permittivity and permeability values of the samples in the frequency range from 108 to 1010 Hz. It is found that the real part of the permittivity is a function of frequency f, via a power law expression ε′=af−b, where a and b are two parameters depending upon carbon black concentration, in the range of frequency investigated. The data analysis reaffirms the result that percolation threshold is a key parameter for characterizing the topological arrangement in these structures.

Electron paramagnetic resonance (EPR) has now become firmly established as one of the methods of choice for analyzing the carbon network over a range of different volume fraction of the carbon black in the composite, i.e., below and above the respective conduction threshold concentration. In the present article, two types of carbon blacks, having very different primary structures, surface areas, and percolation thresholds, were used; Raven 7000 (of high surface area and high percolation threshold volume fraction) and Y50A (of low surface area and low percolation threshold volume fraction). A semiquantitative image analysis of the microstructure from transmission electron microscopy reveals information about the spatial distribution of the carbon aggregates and agglomerates inside the composite. We observe that the apparent surface of agglomerates increases significantly with increasing carbon black content for the two types of blacks investigated. Adsorbed oxygen on the carbon black cristallites and dynamic coalescence under mixing conditions can be responsible for the broadening of the dispersed phase surface distribution. The interagglomerate distance in two samples of concentrations f<fc and f≅fc of Raven 7000 are nearly identical indicating that the dc condition threshold can therefore be almost entirely attributed to the coalescence of smaller aggregates. Line shape simulation showed that the changes in the absorption EPR spectra, at temperatures between 105 and 300 K, of the composite samples containing Raven 7000 can be described by a linear superposition of two distinct Lorentzian (one broad and the other narrow) resonance lines and a single (narrow) Lorentzian resonance line for composite samples containing Y50A. The spins giving rise to the EPR signal reside in the carbon black particles. In Raven 7000, the significant difference in linewidth between the two signals demonstrates a different environment where the restriction of the motion of the paramagnetic centers varies. The narrower line was assigned to spin probes with high mobility (carbon black aggregates) and the broad one to probes with restricted mobility incorporated in carbon black agglomerates. In Y50A, only the sites with high mobility were detected. When the temperature is increased the data demonstrate that the EPR signal intensity, which is the double integral in arbitrary units divided by the mass of the carbon black contained in the sample, decreases slowly in the temperature range 105–300 K. The various phenomena observed are attributed mainly to the aggregates and agglomerates structure in the composite samples. The temperature dependence of the paramagnetic susceptibility deduced from the EPR integrated intensity is discussed in terms of Adriaanse et al.’s model [L. J. Adriaanse, J. A. Reedijk, P. A. A. Teunissen, H. B. Brom, M. A. J. Michels, and J. C. M. Brokken-Zijp, Phys. Rev. Lett. 78, 1755 (1997)]. The magnetic susceptibility of the composite samples is also measured with a superconducting quantum interference device magnetometer, operating at an applied magnetic field of 0.5 T, from 2 K to room temperature. The observed temperature dependence of the spin susceptibility is discussed and suggests that morphology heterogeneity is of overwhelming importance to understand the magnetic properties of these materials.

Sorption kinetics in excess toluene and electron paramagnetic resonance (EPR) experiments on carbon black polymer composites show that these systems exhibit remarkable behaviours. Two types of carbon blacks, having very different primary structures, were used. The swelling ratio expressed as the mass of absorbed solvent divided by the mass of the dry sample in excess toluene has been measured over a wide range of carbon black concentration. The absorbency of these materials is strongly affected by the surface area of the carbon black and is related to the characteristics of the microstructure (porosity and tortuosity). The EPR spectra depend significantly on the structure of the carbon black aggregates. The peak-to-peak amplitudes and the line widths of the EPR spectra are suggested to be significantly affected by the surface area and the morphology of the carbon black. For Y50A samples (low-surface-area carbon black), the experimental observations are explained within the context of a heterogeneous gel: finely dispersed aggregates of submicrometre size dispersed in the polymeric matrix may be considered as additional cross links. For Raven 7000 samples (high-surface-area carbon black), it is suggested that much of the observed behaviour at low concentration can be explained in terms of the agglomerated structure increasing the tortuosity of the material and again by the heterogeneous gel picture for high concentrations.

In this article we describe the results of an experimental study of the dielectric and physicochemical properties of a composite material prepared by incorporating carbon black particles into a polymer matrix. The analysis of the carbon black concentration dependence of the amplitudes and linewidths of the ESR signals indicates that ESR is an important experimental probe of the structure of the elasticity network. Sorption kinetics experiments with toluene indicate that the uptake of solvent exhibits a square root dependence in time as a consequence of Fick's law and permit to evaluate the effective diffusion coefficient in the range 10-lo-lO-ll cm2s-l depending on the mass fraction of the carbon black in the sample.

No abstract