**3. Results and Discussion**

#### *3.1. Morphological Analysis*

The initial powder has a sponge-like structure with a high specific surface and lots of cavities (Figure 2a). Based on the SEM measurements, the average particle size (APS) of the initial powders was about 180 μm. After ball milling for 20 h, the particles' size and morphology changed to very fine and flake-like with 3 μm APS (Figure 3b). By conducting a wet chemical analysis (ICP-OES Varian 720 ES, Palo Alto, CA, USA) (Table 2), the chemical composition of the initial powder was assessed and compared to the milled powders. Table 2 shows the results. It can be concluded that the composition of the initial and the milled powder slightly differs due to contamination derived from the milling equipment. The amount of contaminant elements has increased due to the long-term high-energy milling, which was, however, an inevitable phenomenon. If we consider that the maximum milled content of the produced sample is 10 wt.%, this impurity overall is

not significant (max. 0.5 wt.%). Moreover, due to the applied rapid and low-energy mixing, the contamination level is not further increased.

**Figure 2.** (**a**,**b**) Show the morphology of the initial and the milled Ti powders, respectively.

**Figure 3.** (**a**) Morphology of the typical particles after mixing the initial powder with 2 wt.%, 6 wt.% and 10 wt.% milled powder; (**b**) typical SEM images of the green sample which contain 10 wt.% milled powder.


**Table 2.** Chemical composition of initial and milled Ti powder (wt.%).

Due to the intensive plastic deformation caused by high-energy ball milling, not only the particle size, but also the crystallite size, as determined by X-ray diffraction (XRD), was decreased from 180 nm to 4 nm. This value was determined in previous research [35].

The microhardness of the milled powder shows a four-fold increase compared to the initial powder, namely from 200 to 800 HV0.025. During the mixing, these hard and fine particles were trapped within the soft and big initial particles, as can be seen in Figure 4a, wherein different amounts of milled powder content have been considered (2, 6 and 10 wt.%).

**Figure 4.** (**a**,**b**) SEM images of the samples containing 10 wt.% milled powder after sintering at 850 ◦C, 1 h at different magnifications.

Based on the SEM images (Figure 3a), it is evident that during the mixing process, the fine, milled powder fills out the pores of the large Ti spongy particles first; afterwards, the surface becomes covered with the fine Ti particles. This structure is visible in the cross-section of the green sample (Figure 3b). Some pores can be seen on Figure 3b, which are mainly closed porosity inside the coarse unmilled particles or voids inside the fine grain shell structure. Figure 4 shows SEM images of a sintered sample with 10 wt.% of ball-milled powders at different magnifications. This sample was sintered at 850 ◦C for 1 h. The images display both at low and high magnification that the dual-scale structure was not altered during the applied sintering process. The fine milled particles formed a 3D shell structure around the coarse unmilled grains. The thickness of this boundary layer is not constant. It varies between ~1 and ~20 μm. This is a consequence of the different sized voids of the spongy initial powder. However, the APS value of the initial powders was 180 μm, while the size of the coarse grains covered with the fine shell layer was less than ~100 μm in the case of the sintered sample. This means that during the mixing process, the fine particles managed to fill in not just the surface cavities but also the internal ones. The investigated surface has two incisions made by plasma ablation. The aim of these incisions was to proves that this dual-scale structure is truly a 3D structure.
