Structure and Properties of (Fe₈₀Ga₂₀)_99.8Ce_0.2 Alloy in Cast and Hot Rolled State

Abstract

FeGa alloys with small additions of rare-earth elements surpass binary alloys in magnetostriction and plasticity. For this reason, they are considered promising magnetostrictive materials for various electrical engineering applications. The alloy (Fe₈₁Ga₁₉)_99.8Ce_0.2 was prepared and investigated in this work. It was found that in the cast state, it has a magnetostriction of 3/2 λ about 100 ppm, saturation magnetization of 150 emu/g, tensile strength of about 300 MPa, and fracture strain of 3%. The microstructure, crystallographic texture, and behavior when heated of the alloy were investigated. Then the ingot was subjected to forging and hot rolling with a deformation degree of 90% at 1000 °C. The structure and mechanical properties of samples cut from a hot rolling sheet were studied. Their tensile strength and fracture strain increase compared to cast state up to 600 MPa and 4% correspondingly.

1. Introduction

Iron-gallium alloys have attracted increased attention for over twenty years due to their high magnetostriction, which can be varied over a wide range by controlling the phase [1] and chemical [2] composition. In addition to abnormal magnetostriction, the alloy has a relatively high magnetic permeability, good frequency [3] and temperature [4] stability of magnetic properties (compared to analogs), high saturation magnetization, and relatively low coercivity. All this makes it attractive for use in electrical devices operating in various conditions.

The problem with the Fe-Ga binary alloy is its low ductility, which complicates its mechanical processing [5]. In addition, the magnitude of magnetostriction, although superior to classical magnetostrictive alloys such as Fe-Co and Fe-Al and ones based on them, remains low compared to alloys of the Fe-Tb-Dy type [6]. To at least partially solve these problems, attempts have been made for over 15 years to improve the properties of the binary alloy by alloying with a third component. During this time, it was found that alloying the binary alloy with boron, carbon, aluminum, chromium, niobium, molybdenum, and several other elements can lead to improved plasticity [1,2,3]. This facilitates plastic deformation to obtain products of various shapes (sheets, rods, wires) from the alloy. However, the problem with this approach is that it results in a deterioration of magnetostriction [7], or, at best, it remains unchanged. In several works, it was found that, when alloying a binary alloy with small additions of rare earth metals (<0.5% at.), in some cases a simultaneous improvement in both magnetostriction and plasticity is observed [8]. In the scientific literature, there is a large number of experimental confirmations of this fact using the example of alloys mainly with the addition of Tb [8], as well as other rare earth elements such as Dy [9] and Y [10]. It is believed that mechanical properties are improved due to a decrease in grain size and a change in the features of dislocation motion in a two-phase material [8], and magnetostriction due to the simultaneous increase in magnetocrystalline anisotropy and softening of the tetragonal modulus, as well as the creation of additional local deformations [11]. Thus, this approach opens up new prospects for the use of Fe-Ga alloy for various practical applications since two of its main problems are solved simultaneously. For this reason, these materials are currently attracting increased attention from the scientific community. Their physical and mechanical properties [8,12,13], as well as phase transformations [14,15,16], are studied in detail in many papers.

The problem limiting the widespread production and use of such alloys may be the technical difficulties of working with rare earth elements, mainly the increased tendency to oxidize, as well as the increase in the price of the alloy. Thus, adding 0.2% Tb (the atomic percentage is used in this work) to the Fe₈₁Ga₁₉ alloy leads to an increase in the cost of a ton of such alloy by approximately 10–15% at prices at the beginning of 2024. In addition, the question of the formation of structure and properties in FeGaR alloys during thermomechanical processing remains open. The purpose of such treatments is to give the material the desired shape, as well as to create a crystallographic texture that is favorable from the point of view of magnetic properties. The first studies on this issue have only recently begun to appear [17].

In the context of the impact on the cost of the alloy, the issue of the possibility of choosing cheaper rare earth metals as alloying elements while maintaining the advantages they provide is relevant. In [11], the effect of various rare-earth elements on the magnetostriction of the Fe-Ga alloy obtained by melt spinning was studied in detail. It was shown that the addition of any of the studied rare-earth elements (La, Ce, Pr, Nd, Sm, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu) leads to an increase in the magnetostriction of melt-spun binary Fe₈₃Ga₁₇ alloy [11]. In this case, the best effect can be achieved by using as a dopant light rare-earth elements with a large negative quadrupole moment such as Ce and Pr. In addition, Ce is also one of the cheapest rare earth elements, which makes its use in FeGaR alloys potentially promising. In work [18] it was shown that even small additions of cerium can significantly increase the magnetostriction of a polycrystalline Fe-Ga sample. All this makes FeGaCe alloys potentially promising magnetostrictive materials. It can be used in devices operating under mechanical loads and high-frequency magnetization reversals, for example for the non-destructive testing based on ultrasonic guided waves. For this reason, these alloys require more thorough study not only in the context of the initial structure and properties but also in the context of deformability and features of structure and texture formation at different stages of thermo mechanical treatment. Deformation of the alloy is required for the production of various products from the alloy on an industrial scale. This work aims to study the structure, magnetic, and mechanical properties of the Fe-Ga alloy with the addition of 0.2% Ce for the first time, as well as to study the features of deformation of this material at high temperatures (hot rolling). Such deformation is one of the main stages of thermo mechanical treatment, implemented to obtain products of a given shape.

2. Results and Discussion

The actual chemical composition was controlled using the EDS method by averaging a set of area analyses from two samples cut from the top and bottom of the ingot. As a result, the following values with corresponding standard deviations were obtained: Fe = 79.47 ± 0.17%, Ga = 20.04 ± 0.26%, Ce = 0.32 ± 0.08% for the bottom part of the ingot, and Fe = 79.21 ± 0.58%, Ga = 20.51 ± 0.55%, Ce = 0.28 ± 0.17%. As can be seen, the distribution of elements is uniform, deviations between different sections of samples and different samples are relatively small. The actual composition differs slightly from the nominal, in particular, if all data are averaged, the actual Ga content is 20.3%, and Ce 0.3% instead of 20% and 0.2% correspondingly. In case of binary alloy, actual composition of the alloy is: Fe = 81.1 ± 0.16% and 18.9 ± 0.16%.

In the first stage of this work, the structure, physical and mechanical properties of the alloys in an as-cast state were studied. The material’s behavior when heated to 1000 °C was then analyzed for (Fe₈₀Ga₂₀)_99.8Ce_0.2 alloy using two different methods (DSC and VSM). Finally, hot rolling was carried out and the structure and mechanical properties of the hot-rolled plate were investigated.

2.1. As-Cast State

EBSD analysis was used for simultaneous analysis of structure and crystallographic orientations. It is seen, that small Ce addition leads to significant changes in alloy microstructure (Figure 1).

Analysis of these images allows us to draw three conclusions. Firstly, it is clear that regardless of composition ingots have multi-component texture without an obvious predominance of any orientations. Secondly, the grain structure of the (Fe₈₀Ga₂₀)_99.8Ce_0.2 alloy differs significantly from the binary one. The differences lie both in the reduction of grain size and in the change in grain morphology. The average grain size in (Fe₈₀Ga₂₀)_99.8Ce_0.2 is 150–200 µm, whereas in Fe₈₀Ga₂₀ it is 400–550 µm depending on the part of the ingot and the scanning area. The reason that limits grain growth is the formation of the Ce-rich second phase. It was previously established that only a small amount of Ce (about 0.05%) can dissolve in the α-Fe(Ga) lattice under normal conditions [18]. Excess Ce forms a Ce-rich phase, which precipitates along the grain boundaries of the main phase and inhibits their growth. For the analysis of the second phase, studies were carried out using EBSD and EDS methods at high magnification (up to ×10,000). The result is shown in Figure 2. To construct the phase map, the CeGa₂ phase was chosen for Kikuchi pattern fitting. According to the ternary diagram, this hexagonal P6/mmm phase is the most probable in the composition under study [18]. It can be seen that the particles of the second phase are distributed along the boundaries of α-Fe(Ga) grains and have a size of about 1 μm (green color in Figure 2a). Not all regions presumably corresponding to the CeGa₂ phase of the Kikuchi lines were assigned to this phase; there are undeciphered regions (black). However, the distribution of cerium (Figure 2b) shows that its increased content is observed in all these areas, which indicates the presence of a Ce-rich phase. There may be several reasons why not all points were identified as P6/mmm CeGa₂. Firstly, this may be due to the methodological features of sample preparation of two-phase materials for EBSD studies. The second phase, due to the difference in mechanical and chemical properties, may behave differently during grinding and electropolishing, for example, be etched, which complicates its detection. In addition, this may be because the Ce-rich phase in this case has either a modified or distorted P6/mmm lattice or some other one. The cerium content is too low to form a sufficient number of second-phase particles to be uniformly distributed along all grain boundaries of α-Fe(Ga). It is localized in separate areas, as shown in Figure 2. This is probably the reason for the curved grain boundaries and their irregular morphology, which are observed in Figure 1b. Both phases, which are present in the cast state, begin to solidify at approximately the same temperature (around 1400 °C), which can be concluded from the phase diagrams Fe-Ga [19] and Ga-Ce [20]. This also contributes to the observed structure. A lower temperature of formation of the RE-rich phase could lead to the release of this phase in the interdendritic space, as was observed, for example, in the works [3,8] for the Tb-rich phase which has a melting temperature of about 1250 °C [8].

Phase composition of two alloys was analyzed using XRD (Figure 3). Both alloys have only major peaks corresponding to α-Fe(Ga) disordered solid solution. The only difference is a slight shift of the peaks towards smaller angles in the (Fe₈₀Ga₂₀)_99.8Ce_0.2 alloy compared to the binary one. This is a consequence of the higher actual content of Ga and the limited dissolution of Ce in the host lattice. Both of these factors lead to an increase in the lattice parameter due to the size of Ga and Ce atoms compared to Fe. Minor peaks corresponding to CeGa₂ phase were not observed due to low content of this phase.

To evaluate the influence of Ce on the magnetic properties of the alloy, field dependency of magnetization at room temperature was analyzed (Figure 4). It is seen that the saturation magnetization M_S of both alloys is typical for alloys with a Ga content of about 20% [21]. The addition of Ce reduce it by about 2–3% because the CeGa₂ phase is paramagnetic at room temperature. It is interesting to note that both alloys have relatively high coercivity about 700 A/m, which is more than two times higher than in alloys with the addition of Tb and Y, obtained by solidification without casting [3]. Taking into account that the coercivity in binary and Ce-doped alloys does not differ, it can be concluded that in this case the dominant influence is the method of alloy production. Apparently, the conditions of ingot production, namely the casting parameters, lead to significant residual stresses, which is also confirmed by the broadening of the peaks on XRD (Figure 3). This is the main reason for the high coercivity in the studied samples, regardless of their composition.

To analyze magnetostriction, strain gauges were glued to samples cut from two ingots. The sample elongation along the field λ_// was recorded, then the sample was rotated by 90° concerning the field and the value of λ_⊥ was obtained. The result of measurements on samples in Figure 5.

The 3/2 λ_S (λ_// − λ_⊥) value for the as-cast state is about 100 ± 15 ppm depending on the sample on which it was measured. On average, this is significantly higher than the magnetostriction value measured on textureless polycrystals of a binary alloy with a gallium content of about 20% [22]. This is confirmed in our experiment also (Figure 5). The cause may be a combination of factors, such as a decrease in the tetragonal modulus ½(C₁₁–C₁₂) and an increase in the anisotropy constant due to increased spin-orbit interaction [11]. In addition, presence of hexagonal second phase between grains leads to a redistribution of stress fields in the material, which can lead to an increase in the number of 90° domains that contribute to the magnitude of magnetostriction.

In the context of the potential possibility of manufacturing various products from the alloy using thermomechanical treatment, it is important to analyze the mechanical properties of the ingot. For this purpose, dog-bone-shaped samples were cut from it. Results are shown in Figure 6. Similar samples were cut from a cast Fe-Ga sample, but a correct tensile test could not be performed in this case due to the fragility of this material. It fractures too quickly, which does not allow recording accurate values of fracture strain and tensile stress.

It is well known that in cast polycrystalline binary Fe-Ga alloys there is virtually no plastic deformation at room temperature. The value of tensile strength, according to various sources, ranges from 150 to 450 MPa, and fracture strain is up to 1% for alloys with 17–19% Ga [8,10,23,24]. The binary alloy is characterized by a typical brittle intergranular cleavage. We also observed this in the fracture image of binary alloy (Figure 6c).

It is known that the addition of a small amount (up to 2%) of various elements, both rare earth and others, changes the nature of the destruction to transgranular quasi-cleavage, as has been shown in a large number of experiments [5,10,24]. As we shown here, the addition of 0.2% Ce has the same effect on the mechanical properties and fracture mode to transgranular (Figure 6b). There are cleavage facets characteristic of this type of fracture. The fracture has mixed ductile-brittle character.

2.2. Heating Experiments

In this paper, the possibility of hot rolling is investigated. The temperature of forging and hot rolling is chosen to be 1000 °C, as it was done in many works on the deformation of Fe-Ga [5,25]. It is important to analyze the behavior of the alloy up to the deformation temperature since the presence of Ce and Ce-rich phases can lead to some transformations during heating, which can affect the processes of structure formation. To analyze the heating behavior, two methods are used, namely VSM and DSC. The results of the study by these methods are shown in Figure 7.

On the temperature dependence of magnetization, only one peak is visible at a temperature of about 600 °C (during heating), which is apparently due to a partial ordering A2/D0₃—L1₂. This is due to the high gallium content, which leads to the fact that the number of gallium atoms becomes sufficient for the local formation of the L1₂ phase, in an amount that can affect the magnetization. As is known, the magnetization of this phase is higher at this temperature [4], therefore, its formation leads to a peak on the M-T curve. A similar phenomenon for alloys with a Ga content of about 20% was discovered earlier [26,27].

This is also confirmed by the results of the DSC analysis. There are two peaks on the curve corresponding to heating, one of them (at a temperature of 674 °C) corresponds to the Curie temperature and the second to the partial transformation into the L1₂ phase. The position of these peaks coincides with the M-T curve despite different methods and different heating rates (2 °C/min at VSM and 10 °C/min at DSC). Since the number of gallium atoms is less than required by the stoichiometry of Fe₃Ga to form 100% L1₂, and also because phase transformations in alloys with a gallium content of about 20% have low kinetics [28] a very small amount of this phase is formed, which appears as a small peaks on the DSC and M-T curves. Apart from this, no transformations were detected that could be associated with the presence of Ce in the composition.

2.3. Hot-Rolled State

Samples cut from the ingot were forged and hot rolled. Hot rolling is an important stage of thermomechanical treatment in the manufacture of thin sheets. The structure formed at this stage has a significant effect on the final structure. The mechanical properties in the hot-rolled state determine the behavior of the material during further cold or warm rolling. Therefore, this stage is studied in detail in this work. To study the structure and texture of the hot-rolled sheet, a section was prepared from its side surface. EBSD analysis was performed over almost the entire thickness of the sheet in several areas along its length. The results of the analysis of three adjacent areas, compiled into one image, are shown in Figure 8.

It can be seen that hot rolling produces a texture and grain morphology characteristic of this type of deformation. In the middle of the sheet, elongated grains with a predominant orientation <111>//ND (γ-fiber) are observed; closer to the surface, grains of other orientations are formed. The grains significantly change their morphology and size in comparison with the as-cast state despite the presence of the second phase.

It can be seen that some grains have grown significantly as a result of dynamic (during hot rolling) and static (during heating between passes) recrystallization. The calculated fraction of the area of recrystallized grains is about 25%.

Since it has been established that the particles of the Ce-rich phase are stable up to a temperature of 1000 °C, it can be concluded that grain growth became possible not due to their dissolution, but due to their insufficient pinning energy under temperature and deformation impact. The fact that the second phase particles do not dissolve but only redistribute during hot rolling is confirmed by the undiminished coercive force, which is still close to 1000 A/m, despite the coarse-grained structure, which should contribute to its reduction.

In this work, an analysis of mechanical properties on cast samples has already been carried out, however, this approach to the analysis of mechanical properties has disadvantages. The fact is that the number and size of pores in samples cut from an ingot can significantly depend on the method of melting and pouring, the shape and size of the mold, and the place from which they were cut. If a pore of sufficient size is found in any area of the tensile test specimen, it may serve as a crack initiation site and have a significant impact on the results obtained. Therefore, for a more reliable analysis of mechanical properties, it is better to use hot-rolled samples, in which there are no pores. Dog-bone-shaped samples were cut from a hot-rolled plate along a rolling direction. The results of the tensile test for one of them is shown in Figure 9.

In this case, significantly higher tensile strength can be observed (average about 600 MPa). Fracture strain is also higher than in the as-cast state (about 4%). These results allow to expect relatively good deformability of this alloy, which opens up the potential for manufacturing products of various shapes from it in industrial conditions. In combination with higher magnetostriction than in the binary alloy, as well as low cost, this makes the studied alloy a promising alternative to the classical Fe-Ga.

3. Materials and Methods

An alloy of nominal composition (Fe₈₀Ga₂₀)_99.8Ce_0.2 was made by alloying Fe (99.95% purity), Ga (99.999% purity) and Ce (99.9%). For this purpose, induction melting was carried out in a protective argon atmosphere, followed by pouring into a cast iron mold. To compensate for Ga losses during melting, its amount was increased by 0.8%. The ingot weighing about 1 kg had the shape of a parallelepiped with dimensions of 25 × 50 × 100 mm. In addition, an ingot with Fe₈₀Ga₂₀ composition was prepared under the same conditions as a reference material. To analyze the structure and properties, samples of different sizes and shapes were cut from different parts of the ingots. One of the parts of the (Fe₈₀Ga₂₀)_99.8Ce_0.2 ingot was forged with a deformation degree of 50% and hot rolled with a deformation degree of 90% (from 10 to 1 mm) in 5 passes. The forging and hot rolling temperature was 1000 °C. Heating was carried out after each pass. The structure, texture, and chemical composition were analyzed using a Tescan Mira scanning electron microscope equipped with EBSD and EDS attachments. EBSD results were post-processed using Analysis Tools for Electron and X-ray diffraction (ATEX) [29]. The calculation of the proportion of recrystallized grains was carried out based on disorientations with an average intragranular disorientation of 2° and a maximum intragranular disorientation of 3°. Phase composition was analyzed using Panalytical Empyrean in Cu Kα radiation. Magnetic properties at room temperature and when heated were measured using a LakeShore 7407 vibration sample magnetometer on cubic samples 3 × 3 mm for as-cast state and 1 × 1 mm for hot rolled state. Heating was carried out to a temperature of 1000 °C at a rate of 2 °C/min in a field of 5 kOe. To study the material’s behavior during heating, DSC analysis was also used using an STA 449 F3 Jupiter (Netzsch). Heating was also carried out up to 1000 °C but at a rate of 10 °C/min in an argon atmosphere. The mechanical properties were analyzed using tensile experiments (Instron 5982) at a strain rate of 1 mm/min. Magnetostriction was measured using strain gauges in a magnetic field up to 3 kOe.

4. Conclusions

In this work, the structure, magnetic, and mechanical properties of the alloy were investigated for the first time (Fe₈₀Ga₂₀)_0.98Ce_0.2. It was shown that the addition of Ce leads to the precipitation of small particles of the CeGa₂ phase along the grain boundaries of the solid solution. This alloy has a good combination of mechanical properties and magnetostriction. It has a fracture strain of 3–4% depending on the state (as-cast or hot-rolled) and magnetostriction 3/2 λ_S about 100 ppm. In addition, the behavior of the alloy was studied when heated up to 1000 °C, and it was shown that the addition of Ce does not lead to any transformations in this temperature range. The alloy was successfully subjected to hot rolling. It was shown that due to good plasticity in this state, it can be used as a starting material for further deformation by various methods for shape change.