**3. Results and Discussion**

Figure 1 presents the submicrometer agglomerates of primary <100 nm particles, which were observed in the amorphous specimen.

**Figure 1.** The SEM images of the amorphous Ti45Zr38Fe17.

The phase changes for the amorphous Ti45Zr38Fe17 during the heating in the temperature range from 200 ◦C to 800 ◦C recorded with the DSC are shown in Figure 2.

**Figure 2.** The DSC curve for the Ti45Zr38Fe17 amorphous powders was obtained by MA for 40 h.

The broad exothermic shoulder is visible between 200 ◦C and 700 ◦C. The energy introduced by mechanical alloying is released during heating, which is visible in the DSC shoulders.

Between 325 ◦C and 515 ◦C, the structure changes from amorphous to quasisrystalline with some addition of the crystal phases such as the FeZr3 and the Fe2(ZrTi)3. The minor peak appears at about 515 ◦C, which indicates that the quasicrystalline phase is fully formed.

Figure 3 shows the evolution of the neutron diffraction patterns for the amorphous structure during heating. As is apparent from Figure 3a, broad maxima at about 60◦ of 2θ is changing due to the start of thermal diffusion above roughly 300 ◦C. The change is connected with the gradual forming of crystalline and quasicrystalline (i) phases. The maxima broadens splitting into small reflections that at 500 ◦C develop rapidly into well-defined reflections of icosahedral and metastable FeZr3 [41] phases. Also at this temperature, the Fe2(ZrTi)3 phase appears roughly at 40◦ of 2θ. The DSC results are consistent showing maxima at 325 and 515 ◦C, which is understandable as the DSC measurement has a much higher temperature ramp than neutron diffraction measurements. The i-phase can be evidenced by several well-defined reflections, which appears at about 53.5◦, 58.8◦ of the 2θ angle. The results of the neutron diffraction experiment show the quasi-continuous character of the transition from the amorphous phase to the quasicrystalline phase. The icosahedral structure of the Ti45Zr38Fe17 starts to evolve into the w-phase above 525 ◦C. At higher temperatures (750 ◦C), an additional transition from the w-phase into the cubic phase was evidenced. The transitions at 525 ◦C and 750 ◦C are not strongly reflected in DSC as both are not associated with the significant rearrangement of the constituting atoms.

**Figure 3.** The transformation from the amorphous to the crystal phase for the Ti45Zr38Fe17 alloy as seen by neutron diffraction (**a**); the in-situ neutron diffraction patterns at some chosen temperatures (**b**).

The pattern for the cubic phase can be indexed in the same crystal phase as the wphase. The reflections are in similar positions; however, the primary reflections become significantly narrowed. The cubic phase is stable while cooling down to room temperature. It is worth noting that the Fe2(ZrTi)3 phase is stable to the highest investigated temperature of 900 ◦C and is also present after the cooling of the specimen.

The magnetic properties measured in the temperature range of 127 ◦C to 927 ◦C for the amorphous Ti45Zr38Fe17 powder are shown in Figure 4. A systematic decrease in the magnetization of Ti45Zr38Fe17 is observed as a function of increasing temperature, and the transition of the ferromagnetic phase into the paramagnetic phase is observed. The amorphous phase changes its structure into some crystal phases during the first heating.

The amorphous phase is not expected to be magnetic. The magnetic properties of the amorphous material originate probably from small amounts of the magnetic nanocrystals not successfully observed by neutron diffraction. The characteristic temperature with the Curie temperature (TC) was defined as the peak position in the dM/dT-T curve, as shown in Figure 4b. It is equal to 267 ◦C/300 ◦C, 352 ◦C, 468 ◦C, and 737 ◦C, respectively. From 127 ◦C to 267 ◦C/300 ◦C, a slight increase in the magnetic signal was observed, which can be explained by the ordering of the magnetic moments of the magnetic nanocrystallites in the amorphous matrix under the influence of magnetic field. The transformation of the amorphous phase started above 300 ◦C. The temperature value 352 ◦C obtained from the VSM measurement is correlated with the value of 325 ◦C derived from the DSC measurement, in the results of which one can observe the evolution of the amorphous phase to the i-phase. The formation of the quasi-phase is noticed both as peaks in the DSC (515 ◦C) or the VSM (468 ◦C) measurement, which can be additionally confirmed by the neutron diffraction at the temperature of 500 ◦C. The start of the formation of the i-phase is observed probably at 468 ◦C, and the end is observed at 515 ◦C. The Curie temperature (737 ◦C) is observed during the transition of the amorphous phase into the crystalline phases and during the subsequent heating of the crystalline phase. This temperature is close to the value of 770 ◦C known for pure iron [42]. This means that apart from an occurrence of the similar crystalline phase close to the w-phase, the sample also contains a small number of iron atoms that have not reacted.

**Figure 4.** The temperature-dependent magnetization at 1 T for the amorphous phase (**a**); the derivative of the temperature-dependence used for the determination of the characteristic temperature (**b**); the thermal evolution of the magnetic coercivity (**c**); and the isothermal magnetization curves (**d**) before (red) and after heating (blue).

Figure 4c,d show the evolution of the magnetic coercivity vs. temperature and the magnetization vs. magnetic field curves measured at 20 ◦C before and after heating. The high values of the magnetic coercivity (HC = 46.5 mT) and the magnetic saturation (Ms) (MS = 4.9 emu/g at 1.5 T) are observed before heating; they can be associated with magnetic nanocrystals dispersed in an amorphous matrix. The abnormal behavior of the magnetic coercivity between 328 ◦C and 480 ◦C is related to the changes in the structure. It is correlated with the earlier results of the DSC and neutron diffraction measurements. After heating, the magnetic coercivity and the magnetic saturation change significantly (HC = 9.6 mT Oe, MS = 0.3 emu/g at 1.5 T).

The amorphous phase with some nanocrystals showed ferromagnetic properties. In the i-phase formation, most of the ferromagnetic behavior disappears due to a lack of long-range magnetic order in the quasicrystal structure. The w-phase and the new c-phase showed a weak magnetic signal, which came from the small addition of unreacted iron, which was confirmed by the Curie temperature after annealing.

After introducing hydrogen into its structure, the TiZrFe alloy remains amorphous. It creates an exciting opportunity to observe and research the behavior of hydrogen in the amorphous matrix.

Bringing hydrogen into an amorphous alloy leads to the creation of simple nanocrystalline hydrides. The hydrogen atoms are bonded to their particular constituents. The diffraction patterns for the base and hydrided alloy obtained from the XRD measurements are typical for amorphous materials. The broad central maxima observed for the hydrided sample are shifted toward the lower angles than for the base alloys (see Figure 5a), which can result from an increase in the mean metal–metal distances. Interestingly, we optimized the thermodynamic conditions so that formations of simple hydrides are not observed, which is favorable at higher temperatures or pressures.

**Figure 5.** The diffraction patterns for the amorphous phase before (black) and after (olive) hydrogenation (**a**); hydrogen mass absorption vs. time for the amorphous TiZrFe (**b**).

The isothermal hydrogen absorption kinetic curves for the TiZrFe compounds are shown in Figure 5b. Precisely at 163 ◦C, the investigated alloy can readily absorb hydrogen under initial 4 MPa H2 pressure and reach a hydrogen storage capacity of about 2.54 wt.%. The activation process energy for hydrogen atoms at some hydrogen storage alloys is a long process [43]. Hydrogen atoms must pierce the oxide layer surface of the nanoparticles before creating metal hydrides during the activation process. In the initial hydrogenation process, hydrogen atoms must penetrate the previously formed hydride layer to hold the hydrogenation reaction. Hydrogen atoms diffuse rapidly into the amorphous matrix from grain boundaries and phase interfaces between the amorphous phase and nanocrystals, which keep accelerating the hydrogenation process rate. After some time related to the processes mentioned above, the reaction is described with a single exponential function. It is probably related to hydrogen diffusion into the grains (bulk diffusion) after creating possible diffusion paths through the surface.

#### **4. Conclusions**

The amorphous phase is stable below 300 ◦C, and it is slowly evolving into the quasicrystal phase above this temperature., The i-phase structure is well formed close to 515 ◦C, with the addition of the FeZr3 and the Fe2(TiZr)3. The TiZrFe compound demonstrates the quasi-continuous character of the transformation from the amorphous into the quasicrystalline (+some other crystals) phase. The structure of the Ti45Zr38Fe17 is evaluated from the i-phase to the w-phase above 500 ◦C, close to 600 ◦C, at which temperature the reflections of the w-phase observed in the XRD diffraction patterns are well developed with some amount of the Fe2(ZrTi)3 with the FeZr3. The ferromagnetic signal of the amorphous TiZrFe comes from magnetic nanocrystallites in the amorphous matrix. The Curie temperature is close to 740 ◦C, and it shows a small number of iron atoms that have not reacted. The amorphous phase with a hydrogen capacity exceeding 2.54 wt.% is still stable after the hydrogenation process, which is exciting for the materials designed for hydrogen storage.

**Author Contributions:** Conceptualization, A.Z.; Data curation, Ł.G. and J.C.; Formal analysis, A. ˙ Z.; ˙ Resources, A.T.; Visualization, N.B.S.; Writing—original draft, A.Z.; Writing—review & editing, Ł.G. ˙ and P.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.
