Effect of mechanical alloying on the synthesis of Fe-TiC nanocomposite

3.1 Phase identification and XRD evaluation

The equilibrium phase diagram of Fe-Ti is shown in Figure 1. From Figure 1 and the semiquantitative analysis of ferrotitanium that has been achieved by XRD (Table 1), it can be assumed that FeTi and Ti phases should be in ferrotitanium at room temperature.

Figure 2 illustrates the XRD patterns of mixture and milled sample at room temperature for different milling times. The XRD pattern of unmilled (0 h) powders confirmed that FeTi (reference code: 19-0636) and Ti (reference code: 01-1197) are observable phases in ferrotitanium. Also, it should be noted that the reflection characteristic of carbon is present in the XRD patterns. Dyjak et al. [23] expressed that a peak of approximately 2θ≈26° in the TiC sample indicates that any carbon in the sample is amorphous. After 60 min of MA, carbon peaks disappear. This is attributed to the solid-state diffusion of the C atoms into the lattice of hcp-Ti, which have small atomic radius. At the same time, because of the formation of TiC and the reduction of iron, the FeTi peak intensity reduces and broadens continuously. The position of the peak for the milled sample shifted to higher diffraction angle compared to the initial mixed powder. These phenomena indicate that TiC nanoparticles have very large lattice distortion and low degree of crystallinity. With the progress of milling up to 90 min, the strongest peaks of TiC appeared, which imply the formation of Fe-TiC nanocomposite particles. By reaching to 5 h of milling, all of the raw materials’ peaks disappeared completely, and the visible peaks in the patterns correspond to TiC and α-iron phases (ferrite with reference code: 06-0696). By increasing the milling time to 20 h, the XRD pattern shows an observable peak broadening. The peak broadening is caused by grain refinement, the presence of the internal strain, and local correlated disorders. This is due to the large number of dislocations that resulted from heavy deformation caused by high-energy MA [24].

Figure 1:

Equilibrium phase diagram of Fe-Ti [22].

Figure 2:

XRD pattern of Fe-TiC nanocomposite at different times.

3.2 Thermodynamic studies

In a system containing titanium and carbon black powders, an MSR would not occur unless T_add>2500 K [25], although MSR was observed earlier in several cases of TiC formation [21, 26, 27]. Razavi et al. [28] calculated the adiabatic temperature for the formation of 1 mol TiC by an initial reactant of Ti and carbon black of approximately 3460.2 K. It seems that although the reaction of Ti+C=TiC has sufficient adiabatic temperature, as seen in Figure 3, because of no distinguishable change in temperature and pressure during milling, no evidence of combustion synthesis has found. Therefore, the existence of FeTi at ferrotitanium gives rise to a gradual diffusion mechanism of TiC synthesis contrary to TiC formation with combustion mechanism from titanium and carbon. Also, the adiabatic temperature was calculated at 2182 K for TiC formation from FeTi and carbon black [28]. Thus, as expected in the present work, the formation of Fe-TiC is governed by a gradual reaction mechanism during the milling of ferrotitanium and carbon.

Figure 3:

Dependence of the vial temperature and pressure on the ball milling time.

The equilibrium formation temperature of TiC from ferrotitanium can be calculated theoretically [29]. The Ellingham-Richardson diagrams of the formation reaction of TiC from two paths (reaction) with the amount of Gibbs free energy are shown in Figure 4. The Fe-TiC composite is more stable than TiC due to its lower Gibbs free energy function. At approximately 2044°C, free titanium reacts with carbon and produces TiC, and the reaction will continue until the reaction is completed for temperature of more than 2044°C. Mechanical activation can decrease the formation temperature of TiC. In fact, by MA, the raw materials are activated and reacted sooner than nonmilled samples.

Figure 4:

Ellingham-Richardson diagram for the formation reaction of TiC and Fe/TiC.

3.3 Structural evaluation

The grain size of TiC can be calculated by the Williamson-Hall method as shown in Table 2 and Figure 5. The results of these calculations are presented in Figure 6, which indicates that increasing milling times have reduced the crystallite size of TiC. After 20 h of milling, the crystallite size has decreased to 7 nm. For the case of lattice strain, the milling times up to 15 h have increased strain. However, after 15 h of milling, because of the strain release as a consequence of heat generated by collisions and friction between balls and particles during milling, strain has decreased [30].

Figure 5:

Williamson-Hall diagram of milled samples.

Figure 6:

Variation of crystallite size and strain in the Fe-TiC system by milling time.

3.4 Morphology of powders

The variations of size and morphology of the selected mechanically alloyed powders after milling for 1, 1.30, 3, 5, 10, and 20 h are shown in Figure 7.

Figure 7:

SEM images of (A) no milling and (B) 1.30 h, (C) 3 h, (D) 5 h, (E) 10 h, and (F) 20 h of milling.

Figure 7 shows that both ferrotitanium and carbon powders have an irregular shape. During high-energy milling, the powder particles are repeatedly flattened, cold welded, fractured, and rewelded. The powders are gradually strain hardened, whereas the fracture mechanism becomes dominant, which results in a general decrease in size with milling time. The powder size reduction will be stopped when the steady rates of cold welding and fracture are reached. In the early stage of milling, the fracture of large particles happens by entrapping within the milling media, especially using PCA and carbon black within ferrotitanium powder [31]. With further milling time, because the starting powder particles are soft, the powders are flattened as a flake after 1 h of milling and welded together due to the impacts of grinding balls to form larger particles, as shown in Figure 7B. As the milling time increased to 1.30 h, when, according to XRD patterns (Figure 2), TiC nuclei are formed gradually (Figure 7C), finer particles with irregular flake begin to shape. After 5 h of milling, when the TiC formation has been completed (with considering XRD pattern; Figure 2), TiC particles with size of less than 2 μm and wide distribution are visible.

The small hard brittle particles in the matrix act as milling agents, which reduces the steady-state milling time, until a steady state is established between the rate of fracturing and cold welding. As shown in Figure 7D–F, by increasing the milling time up to 15 h, well spreading and grain refinement of the powders are attained; at milling times of more than 15 h, this condition is not observed because of the agglomeration of the powders. Also, when the milling duration rises up to 20 h, the particles reach a value of 10 to 100 nm in size (Figure 7F).

3.5 Lattice parameter

The lattice parameter of TiC powders obtained from 5 and 20 h of milling were calculated by the Nelson-Riley method from the XRD analysis. The results are shown in Table 3 and Figure 8. The increasing milling time from 5 to 20 h cause the reduction of lattice parameter. Also, for both times, there is a deviation from the ideal size of lattice parameter (0.4327 nm). This deviation can be related to the increase in the amount of strain, formation of TiC with nonstoichiometric ratio, and lack of calibration of instruments [32].

Figure 8:

Changes of the lattice parameter in TiC system based on the Nelson-Riley function in the milling times of 5 and 20 h.