Detailed description of the process of NPC

Detailed description of the process of NPC samples fabrication from the carbides of boron, zircon, titanium and molybdenum and typical parameters of the product structure were disclosed in [53–55]. The structural parameters substantially differed for the samples of different origin. Thus, the mean pore size for the products of B4C and Mo2C chlorination was approximately 5 times greater than that for ZrC and TiC derivatives. As a result, the apparent density in the latter case was almost two times greater, while the pycnometric density (i.e. the density of carbon skeleton, with pore volume excluded from consideration) was close to the density of graphite for all the samples. Silicon carbide was also used for fabricating NPC samples for the described experiments. As it has been showed in [55], the pore size for these samples varied in wider range than for the derivatives of other carbides. Chlorination temperature had significant effect on the pore size distributions. In the samples prepared at low temperatures (700–900 °C), the pore size was close to 1 nm. If the chemical treatment temperature was higher (1200°), larger pores appeared and the sample densities (both apparent and pycnometric ones) decreased, presumably due to higher chlorine etching efficiency with respect to carbon atoms. After treatment at 2000 °C, only large pores (meso-pores) remained in the sample; the carbon skeleton lost more than a half of its mass and underwent a reconstruction with dramatic reduction of the specific surface area. The morphology of the studied NPC samples may be overviewed from microscopic images. All overview images (with relatively low resolution and wide field of view) acquired with a scanning jak stat inhibitor microscope were similar to the one presented in Fig. 1. The μm- and sub-μm-sized irregular NPC particles presented in the image inherit the shapes of the original carbide powder grains. Atomic-scale images of small spots of NPC samples obtained by the transmission electron microscopy (HR TEM) are shown in Fig. 2a–c. They demonstrate that the studied materials included both amorphous (Fig. 2c) and ordered domains. The latter were comprised of faceted (Fig. 2a) or curved (Fig. 2b) atomic layers. The distances between atomic planes (∼0.35 nm) correspond to the graphite lattice period. Domain sizes between 20 and 100 nm were the most typical, though sometimes larger faceted crystallites were also observed (Fig. 2d). As much as diencephalon can be seen from microscopic data, the relative part of ordered domains in NPC samples derived from the same carbide usually grew with the increase of chlorination temperature [52,55].
Emission properties Emission properties of NPCs were tested in the layout described previously in [56]. Emitter samples were prepared by depositing NPC powders over plane metal substrates. The electric field was applied between the sample and a cylindrical tungsten anode with a flat end side in vacuum better than 10–6 Torr. The field gap width between the anode and cathode was 0.50–0.75 mm, the anode diameter was 6 mm. During the experiments, the samples could be heated up to 400 °C to remove volatile contaminants and perform thermo-field activation procedures [56]. All the investigated NPC coatings demonstrated comparable emission properties. Emission characteristics of NPC samples fabricated from different carbides are compared in Fig. 3. The NPC samples produced from silicon carbide by chlorination at different temperatures also demonstrated similar emission properties (Fig. 4), excluding the case of the highest temperature of 2000 °C. The current dependencies measured in the emission tests usually demonstrated exponential character (for the field values well above the emission threshold): presented in FN coordinates (Figs. 3b and 4b), the emission plots were approximately linear. In accordance with the classical field-emission model, their slope angles are determined by the combined parameter (ϕ3/2/β), where ϕ is the work function of the emitter and β is the field enhancement factor for the emitting spot. For the presented plots, this parameter varied between 2.0 × 10−3 and 6.5 × 10−3 eV3/2. Taking ϕ = 4.5 eV as a typical value for various forms of carbon [28,38,57], we obtain an estimate for the field enhancement factor β ≈ 1500–4500. These values obviously disagree with the surface topography revealed by the microscopic studies (see Figs. 1 and 2). A more plausible value β = 10 gives us the work function estimates of 0.07–0.16 eV. For a material with such a little work function, heating to 400 °C would result in thermal emission with space-charge-limited current density of more than 10 A/cm2. No such emission current weakly depending on the applied field was ever observed in the experiments. Thus, we can conclude that the experimental data apparently disagree with the classical emission theory, which is not unusual for many forms of carbon.