A detailed study of agglomeration in composite materials containing carbon nanotubes (CNT) is presented. Three dimensional samples with different degrees of agglomeration were created in three different ways, leading to a wider range of geometries available to study. Virtual charges are injected into the computer-generated samples and move through these samples according to a Monte Carlo hopping algorithm. Results show that there is an optimal level of agglomeration that is actually beneficial for charge transport at low volume concentrations, lowering the percolation threshold. It is found that near percolation, a more uniform CNT distribution (less agglomeration) leads to more conductive paths, but with a lower mobility. The optimum level of agglomeration comes from a trade off between these two properties. Beyond this optimum agglomeration state, it is observed that conductivity tends to decrease as dispersion increases at all concentrations studied here. At high concentration (percolated samples), where CNT clumps merge, conductivity seems to be less sensitive to agglomeration.

No abstract

No abstract