**1. Introduction**

With the dispersion of discontinuous micro or nanofibers, such as carbon nanotubes (CNTs), in a polymer matrix, it is possible to obtain composite materials with mechanical and electrical properties that open a vast range of applications. Depending on the composition, particle concentration and dispersion, these materials have an electrical conductivity ranging from electrical insulation (10−18–10−<sup>7</sup> Scm−1), to semi-conductivity (10−7–10−<sup>1</sup> Scm−1) or even metallic conductivity (above 10−<sup>1</sup> Scm−1), allowing for higher flexibility in applications when compared to plastics and metals that possess a unique and defined conductivity [1]. Electrically conductive polymer composites may be able to substitute metals in a variety of applications that allow for lightweight solutions and easy shaping, such as resistive or capacitive sensors, battery technology components, electromagnetic shielding or resistive heating elements [2–4].

The electrical properties in these functional composite materials arise due to a network of interconnected particles which ultimately allows for the flow of electrons. This network not only depends on the particle shape and properties, but also on all the processing steps of the material, from the compounding of matrix and particles up to the shaping of the final part. Therefore, the complete

history of the material a ffects the final electrical properties of the composite in the final parts [5–7]. Traditionally, the best way to reduce electrical resistivity is to increase the particle concentration, which entails additional costs and a ffects the processability due to higher viscosity [8].

The CNTs dispersive mixing operation consists of several stages-filler incorporation and the wetting and infiltration of the epoxy matrix, followed by dispersion, distribution and flocculation [9,10]. During the stage of filler dispersion, the large initial filler agglomerates are reduced in size up to the smallest dispersible unit. This can be mainly attributed to two coexisting mechanisms: rupture (a bulk phenomenon) and erosion (a surface phenomenon). The rupture mechanism breaks down agglomerates in a short time, related to erosion that happens slower by removing single or bundles of CNTs from the surface of the agglomerate [11]. This first step is the most di fficult and important one, since it determines the rate at which CNTs disperse in the polymer. The prevalence of a mechanism over the other depends on the shear stress: if it exceeds a certain threshold value (dependent on the filler) the rupture tends to be the dominant mechanism [12]. From the existing literature [13] it emerges that the ratio of applied shear stress and cohesive strength of the agglomerate defines the dominating mechanism. Dispersion by rupture is the fastest way to obtain small final agglomerates, but the sudden breaking of a whole cluster often implies a reduction in the aspect ratio of its constituent particles, resulting in a worsening of the macroscopic properties that the nanoparticles can give to the composite. To avoid the excessive breakages of single nanotubes, the mixing parameters should be ideally controlled in order to maintain the rupture mechanism at the lowest possible rate, promoting erosion that is slow but preserving the nanotube aspect ratio, and leading to a better infiltration of the epoxy matrix in the agglomerates [11].

Finally, this work presents a scalable method to investigate and produce nanocomposite samples with improved controlled electrical properties by manipulating the nanoparticles with electric fields during the shaping step [14,15]. This is possible due to the di fference in dielectric and electric properties between the particles and matrix, and particularly due to the high aspect ratio of carbon nanotubes [16,17]. According to the theory of dielectrophoresis, randomly dispersed particles under an electric field become polarized and orient, interacting due to Coulomb forces and assembling in conductive chains [16–18].

The experimental set-up hereby developed is inspired in resin-transfer-molding (RTM) [19] and allows for the application of an electric field on the electrically conductive composite resin during the curing. It was designed to be easily scaled-up for di fferent plate geometries and integrated with available composite technology, aiming at an industrial-relevant production. It is here demonstrated for a system of single-wall carbon nanotubes (SWCNTs) and epoxy. In a first step, the process of dispersion of SWCNT in epoxy is investigated in view of optimization of electrical resistivity. Then, this nanocomposite is cured under the influence of electrical fields, and the impact of di fferent mold release agents on the final resistivity is investigated.

#### **2. Materials and Methods**

The experimental procedure for the production of bulk epoxy nanocomposites was done in the following steps:


This section describes the materials and methodology used in the di fferent stages of the sample production.
