**1. Introduction**

Tungsten disulfide inorganic fullerene-like particulates (IF-WS2), a dichalcogenide with distinct physical and chemical properties [1,2] that presents a hollow cage structure with potential uses as lubricant, component in batteries, supercapacitors or catalyst, among others [3–9], was introduced a few years ago as a material with shock resistance properties [10–13]. The shock absorbing ability of IF-WS2 particulates allows them to endure pressures up to 25 GPa, with concomitant temperatures of up to 1000 °C, without structural degradation or phase change [12], a characteristic that opens an exciting window of possibilities for protective systems applications. Recent reports have further explored the features of such WS2 nanostructures and some studies of their inclusion in polymeric matrices, along with the composites mechanical properties, have been published [14–24]. Despite those outstanding characteristics, the WS2 density might be considered a drawback if it is to be used as component in protective gear that should be lightweight, able to withstand high temperatures and be flame resistant. A possible compromise between all those requirements might be made by joining the sought-after shock absorbing attribute of IF-WS2 with lighter materials.

Due to their strength and light weight, tridimensional carbon assemblies and porous carbon structures such as nanotubes, foams or intertwined nanofibers [25–37], along with two-dimensional structures, such as Graphene [38–40], have been the focus of attention for energy absorbing applications. However, despite the advances in the field, the performance of personal protective equipment or sporting gear composed of carbon nanostructures can still be improved by the use of materials that could divert, distribute or dissipate the energy of impacts and the shock waves associated with them, in a more efficient manner.

Given the shock resistance characteristics of IF-WS2 and the energy absorption of carbon nanostructures mentioned above, the combination of those two types of materials seems as a natural next step in the design of protective systems. Recent work that has successfully explored the possibility of merging the carbon component with IF-WS2 using polymeric matrices [16,18,41–45]. However, the energy dissipation characteristics of those products could be attributed, to an extent, to the role of the viscoelastic polymeric matrices. Conversely, new reports in the generation of carbon fiber tridimensional structures have proven that CNF can be grown to form a macroscopic foam with viscoelastic properties without the need of a polymeric component [46]. Adding the known shock absorbing characteristics of IF-WS2 to such type of carbon structures will be highly desirable and is the focus of this investigation.

The development of a protocol to combine the IF-WS2 characteristics with those of the carbon structures in the absence of a polymer, to generate a random distribution of those two phases at the nanoscale (*ca.* a hybrid made of solely inorganic components, thus, avoiding some of the polymer drawbacks such as aging but gaining in terms of lightweight and thermal stability), remains as one of the major challenges.

Experimentation by our group has shown that adding IF-WS2 to an already existing 3D carbon structure, even when using solvents to achieve the mixture, renders an inhomogeneous solid. The present work aims to produce a well-dispersed hybrid system composed of a carbon solid (Carbon Nanofiber or Graphene) and low loading levels of IF-WS2. We found that to produce a well dispersed 3D structure of carbon nanofibers and IF-WS2 it is necessary to employ the two stage *in situ* protocol described in the next section. In contrast, to produce a hybrid of Graphene (2D layered structure) with IF-WS2, the use of the *in situ* protocol did not show significant improvement in the phase distribution when compared to physically mixing the components.

As an extension of the primary goals, we studied the mechanical properties of epoxy composites based on hybrid CNF/IF-WS2 and G/IF-WS2 made by *in situ* routes, and contrast the latter to those created from physical mixtures of the components.
