Investigation into the Effect of Thermal Treatment on the Obtaining of Magnetic Phases: Fe5Y, Fe23B6, Y2Fe14B and αFe within the Amorphous Matrix of Rapidly-Quenched Fe61+xCo10−xW1Y8B20 Alloys (Where x = 0, 1 or 2)

The paper presents the results of research on the structure and magnetic properties of Fe61+xCo10−xW1Y8B20 alloys (where x = 0, 1 or 2). The alloys were produced using two production methods with similar cooling rates: Injection casting and suction casting. The alloy samples produced were subjected to isothermal annealing at 940 K for 10 min. The structure of the materials was examined using X-ray diffraction. Isothermal annealing has led to the formation of various crystallization products depending on the chemical composition of the alloy and the structure of the alloy in a solidified state. In two cases, the product of crystallization was the hard magnetic phase Y2Fe14B. However, the mechanism of this phase formation was different in both cases. The magnetic properties of alloys were tested using a vibrating sample magnetometer and a Faraday magnetic balance. It is found that the grain crystallite size of the crystalline phases have a decisive influence on the value of the coercive field (especially in the case of hard magnetic phases). It has been shown that privileged areas can already be created during the production process. Their presence determines the crystallization process.


Introduction
Modern materials are expected to fulfil increasingly demanding requirements to warrant their adoption in applications. Currently, the main requirements are efficiency and the protection of the natural environment. Many modern materials are being used in very specialist fields due to their desirable properties. Iron-based materials are used, amongst others, in the power and electrochemical industries. These materials are mostly magnetic alloys which feature soft magnetic properties. Materials exhibiting soft magnetic properties are characterized by a low-value coercive field, low core losses and high saturation magnetization. Such properties are exhibited, among others, by amorphous [1][2][3][4][5][6] and nanocrystalline iron-based alloys [7][8][9][10]. These parameters are especially important for such applications as magnetic cores, where efficient demagnetization is required. The magnetic properties of the material depend not only on its chemical composition, but also on the production process, and the purity of the component elements. One production method, suitable for obtaining materials of a titanium getter was undertaken, in order to remove any remaining oxygen and other impurities from the working chamber. The first melting of the material was undertaken using a lower electrode current to avoid material splatter. Each ingot was re-melted several times, in order to achieve a better homogeneity of the resulting alloy. After each melting process, the ingot was inverted by means of a manipulator, and this action further aided the mixing of the component elements. After this process, each ingot was divided into smaller fragments, and subjected to mechanical and ultrasonic bath cleaning.
The alloys used in this study were made by injection and suction-casting methods, in conjunction with a water-cooled copper die. These methods allowed the achievement of rapid cooling speeds of 10 1 -10 3 K/s. The casting process was performed under a stable argon pressure and using identical copper dies. The samples were cast in the form of rectangular plates with the following approximate dimensions: Width = 5 mm, length = 10 mm, and thickness = 0.5 mm. Five plate samples were made for each alloy composition.
Separate X-ray diffraction (Bruker, Billerica, MA, USA) studies were performed for each of the tested plates. Tests were carried out on the surface of the samples and after crushing. Then, powders were mixed for each of the examined alloys and X-ray examinations were performed. The obtained results coincided with high accuracy with the diffraction obtained for the samples from individual plates.
Following casting, the resulting samples were crushed using a low-energy process. The powdered samples were subjected to an isothermal annealing process in a vacuum at a temperature of 940 K for 10 min. The process of the thermal treatment of samples consisted of three stages: heating (8 min), annealing (10 min) and cooling (120 min). The samples enclosed in a quartz tube were placed in an oven (author's device, Częstochowa, Poland) preheated to 940 K. After 8 min, the sample reached oven temperature. According to the established parameters, the sample was annealed for 10 min. After this time, the oven chamber was opened and the sample was cooled to room temperature ( Figure 1). The heat treatment process designed in this way affects the relaxation processes which occur in the sample volume. to avoid material splatter. Each ingot was re-melted several times, in order to achieve a better homogeneity of the resulting alloy. After each melting process, the ingot was inverted by means of a manipulator, and this action further aided the mixing of the component elements. After this process, each ingot was divided into smaller fragments, and subjected to mechanical and ultrasonic bath cleaning.
The alloys used in this study were made by injection and suction-casting methods, in conjunction with a water-cooled copper die. These methods allowed the achievement of rapid cooling speeds of 10 1 -10 3 K/s. The casting process was performed under a stable argon pressure and using identical copper dies. The samples were cast in the form of rectangular plates with the following approximate dimensions: Width = 5 mm, length = 10 mm, and thickness = 0.5 mm. Five plate samples were made for each alloy composition.
Separate X-ray diffraction (Bruker, Billerica, MA, USA) studies were performed for each of the tested plates. Tests were carried out on the surface of the samples and after crushing. Then, powders were mixed for each of the examined alloys and X-ray examinations were performed. The obtained results coincided with high accuracy with the diffraction obtained for the samples from individual plates.
Following casting, the resulting samples were crushed using a low-energy process. The powdered samples were subjected to an isothermal annealing process in a vacuum at a temperature of 940 K for 10 min. The process of the thermal treatment of samples consisted of three stages: heating (8 min), annealing (10 min) and cooling (120 min). The samples enclosed in a quartz tube were placed in an oven (author's device, Częstochowa, Poland) preheated to 940 K. After 8 min, the sample reached oven temperature. According to the established parameters, the sample was annealed for 10 min. After this time, the oven chamber was opened and the sample was cooled to room temperature ( Figure 1). The heat treatment process designed in this way affects the relaxation processes which occur in the sample volume. The effect of the low-energy crushing on the possible segregation of alloying elements was determined using EDX analysis. A microscope was used, which was the Supra 25 Zeiss Detector SE (Oberkochen, Germany). The structure of the investigated alloys was studied by performing X-ray diffractometry on the powdered samples. A Bruker Advance D8 X-ray diffractometer (Bruker, Billerica, MA, USA) equipped with a CuKα radiation source, was used for this part of the investigation. The magnetic properties were studied by means of a LakeShore vibrating sample magnetometer (Carson, The effect of the low-energy crushing on the possible segregation of alloying elements was determined using EDX analysis. A microscope was used, which was the Supra 25 Zeiss Detector SE (Oberkochen, Germany). The structure of the investigated alloys was studied by performing X-ray diffractometry on the powdered samples. A Bruker Advance D8 X-ray diffractometer (Bruker, Billerica, MA, USA) equipped with a CuKα radiation source, was used for this part of the investigation. The magnetic properties were studied by means of a LakeShore vibrating sample magnetometer (Carson, CA, USA) using an external magnetic field of up to 1.7 T. Based on the recorded static hysteresis loops, the values of the coercivity and magnetization of the samples were determined. The thermal stability of the alloys was studied using a Faraday balance (AGH, Kraków, Poland). CA, USA) using an external magnetic field of up to 1.7 T. Based on the recorded static hysteresis loops, the values of the coercivity and magnetization of the samples were determined. The thermal stability of the alloys was studied using a Faraday balance (AGH, Kraków, Poland). Figure 2 presents SEM images for samples in the form of powder along with EDX analysis. EDX analysis was carried out for areas with a size of 10 μm. Crushing did not change the chemical composition and separation of alloy components. In Figures 3 and 4, X-ray diffraction patterns are presented for the samples of the investigated alloys using injection and suction casting in the as-cast state and following the thermal treatment (940 K/10 min). In Figures 3 and 4, X-ray diffraction patterns are presented for the samples of the investigated alloys using injection and suction casting in the as-cast state and following the thermal treatment (940 K/10 min).

Results
Materials 2020, 13, 835 4 of 14 CA, USA) using an external magnetic field of up to 1.7 T. Based on the recorded static hysteresis loops, the values of the coercivity and magnetization of the samples were determined. The thermal stability of the alloys was studied using a Faraday balance (AGH, Kraków, Poland). Figure 2 presents SEM images for samples in the form of powder along with EDX analysis. EDX analysis was carried out for areas with a size of 10 μm. Crushing did not change the chemical composition and separation of alloy components. In Figures 3 and 4, X-ray diffraction patterns are presented for the samples of the investigated alloys using injection and suction casting in the as-cast state and following the thermal treatment (940 K/10 min). The production method resulted in an amorphous structure in two samples, as confirmed by the resulting X-ray patterns (Figure 3a,c). However, the sample of the Fe 62 Co 9 W 1 Y 8 B 20 alloy (Figure 3b), which was made under the same conditions, shows the presence of a small number of grains of the Fe 5 Y and αFe crystalline phases. After the isothermal annealing process, conducted at a temperature close to the crystallization temperature [31] of the alloy for 10 min, the X-ray diffraction patterns for all of the samples revealed sharp, relatively narrow, peaks, indicating the presence of crystalline phases. In the sample of the Fe 61 Co 10 W 1 Y 8 B 20 alloy ( Figure 3d) and the Fe 63 Co 8 W 1 Y 8 B 20 alloy (Figure 3f), the presence of two crystalline phases was detected: αFe and Fe 5 Y. This suggests that the crystallization process occurred at similar activation temperatures for these two samples. In the case of the sample of Fe 62 Co 9 W 1 Y 8 B 20 alloy (Figure 3b), which was made by the injection-casting method and partially crystallized during the production process, four crystalline phases can be seen after annealing (Figure 3e). In contrast with the other two samples, in this alloy, after the annealing process, grains of the crystalline phases Y 2 Fe 14 B and Fe 23 B 6 were observed. The first crystalline phase has been described in the literature as a magnetically hard phase [32], while the second phase is a meta-stable crystalline phase, which is often found during the initial stage of crystallization in the amorphous Fe-Co-B alloys. Figure 3. X-ray diffraction patterns for the alloys in the as-cast state (a-c) and following the thermal treatment at 940 K for 10 min (d-f) produced using the injection-casting method: Fe61Co10W1Y8B20, (a,d), Fe62Co9W1Y8B20, (b,e) Fe63Co8W1Y8B20 (c,f).

Results
The production method resulted in an amorphous structure in two samples, as confirmed by the resulting X-ray patterns (Figure 3a,c). However, the sample of the Fe62Co9W1Y8B20 alloy (Figure 3b), which was made under the same conditions, shows the presence of a small number of grains of the Fe5Y and αFe crystalline phases. After the isothermal annealing process, conducted at a temperature close to the crystallization temperature [31] of the alloy for 10 min, the X-ray diffraction patterns for all of the samples revealed sharp, relatively narrow, peaks, indicating the presence of crystalline phases. In the sample of the Fe61Co10W1Y8B20 alloy ( Figure 3d) and the Fe63Co8W1Y8B20 alloy ( Figure  3f), the presence of two crystalline phases was detected: αFe and Fe5Y. This suggests that the crystallization process occurred at similar activation temperatures for these two samples. In the case of the sample of Fe62Co9W1Y8B20alloy (Figure 3b), which was made by the injection-casting method and partially crystallized during the production process, four crystalline phases can be seen after annealing (Figure 3e). In contrast with the other two samples, in this alloy, after the annealing process, grains of the crystalline phases Y2Fe14B and Fe23B6 were observed. The first crystalline phase has been described in the literature as a magnetically hard phase [32], while the second phase is a meta-stable crystalline phase, which is often found during the initial stage of crystallization in the amorphous Fe-Co-B alloys. For all the investigated alloy samples made using the suction-casting method and still in the ascast state, the X-ray diffraction patterns revealed only a single broad maximum in close proximity to the 45° 2θ angle; independently of the angular position, the background could be observed. After these samples had been subjected to the thermal treatment at 940 K for 10 min, high-intensity X-ray diffraction rings originating from four different crystalline phases could be seen. For the For all the investigated alloy samples made using the suction-casting method and still in the as-cast state, the X-ray diffraction patterns revealed only a single broad maximum in close proximity to the 45 • 2θ angle; independently of the angular position, the background could be observed. After these samples had been subjected to the thermal treatment at 940 K for 10 min, high-intensity X-ray diffraction rings originating from four different crystalline phases could be seen. For the Fe 61 Co 10 W 1 Y 8 B 20 alloy ( Figure 4d) and the Fe 62 Co 9 W 1 Y 8 B 20 alloy (Figure 4e), within the volume of each sample, crystalline grains of three crystalline phases could be found: Fe 5 Y, αFe and one phase which could not be identified. For the third sample of Fe 63 Co 8 W 1 Y 8 B 20 , (Figure 4f), the annealing process led to the creation of an additional magnetically hard crystalline phase, namely Y 2 Fe 14 B. Based on the Scherrer equation, the average sizes of the crystalline phase grains were determined (please see Table 1).
In the case of the Fe 62 Co 9 W 1 Y 8 B 20 alloy in the as-cast state, the application of the Scherrer equation is more difficult, as this method should be used to estimate the size of crystallite with a diameter much larger than a few nanometers. For the isothermally annealed samples made by both methods, the grain crystallite size of the Fe 5 Y phase is similar. This means that, in this case, the crystallite size is independent of the production method. In the case of the αFe phase, there is a relationship between the iron content and the crystallite size: with increasing iron content in the alloy, there is an increase in the average crystallite size of the αFe phase. In addition, it could be stated that, in this case, the production method itself influences the crystallite size. A smaller average size of the αFe crystalline phase was observed for the sample made by the suction-casting method. The creation of the magnetically hard Y 2 Fe 14 B and soft Fe 23 B 6 crystalline phases is very interesting. There is a difference in the creation of the Y 2 Fe 14 B phase between the samples made by the injection and suction-casting methods. It turns out that, despite slight changes in the chemical composition and a similar cooling rate in the volume of alloys, there is a certain segregation of atoms. This arrangement is difficult to predict. The creation of this phase is related to the method of placing the liquid alloy in a copper form. This may result in the formation of various crystalline phases in the annealing process despite the identical composition of the alloy. For the as-cast sample of Fe 62 Co 9 W 1 Y 8 B 20 made by injection-casting, the size of the Fe 5 Y andαFe crystalline grains was found to be very small. As suggested by many Mössbauer studies of FeCoB [26,33] alloys, there are areas of differing iron concentrations present within the volume of these materials; that is, within the amorphous matrix, some areas are more favorable for the creation of crystalline grains. It can be assumed from the iron-rich amorphous matrix (the low-field component from the Mössbauer spectra) that iron and yttrium were precipitated to create the Fe 5 Y and αFe phases. The re-grouping of the iron and yttrium atoms within the alloy volume in certain preferred regions contributed towards the formation of areas where the atomic arrangement favors the creation of the different phases (which could be the high-field component in the Mössbauer spectra [26,34]). As a result of the annealing process, atoms diffuse in the volume of alloys. This creates areas with lower internal energy: The nuclei of crystalline phase grains.
The migration of large atoms of yttrium within the amorphous alloy causes the collective movement of different, smaller atoms; in particular, boron.
A good example illustrating the impact of the production method on the structure and magnetic properties of amorphous alloys is its effect on the Curie temperature. In amorphous alloys produced using the injection-casting method, Tc is usually a few degrees Kelvin lower than in the same alloys produced by the suction-casting method [35,36]. Despite the fact that both methods give the possibility of producing amorphous material, these materials have a different magnetic structure. This means that the way of introducing the sample into the copper mold affects the distribution of magnetic atoms in volume. This reasoning confirms that, in the volume of amorphous samples, various clusters can be formed, which are conducive to forming various types of crystalline phases.
Therefore, it is possible that, as a result of the thermal treatment of the Fe 62 Co 9 W 1 Y 8 B 20 alloy, the aforementioned migration occurs, and the Y 2 Fe 14 B and Fe 23 B 6 phases are created. The Fe 23 B 6 phase is the so-called pre-natal, metastable phase, in which nuclei or clusters may be created during the rapid solidification process [36]. However, due to their small sizes, they cannot always be identified during X-ray diffractometry studies. In the case of the Fe 63 Co 8 W 1 Y 8 B 20 alloy produced by the suction-casting method, the mechanism for the creation of this phase is similar. However, it has to be highlighted that the alloy in the as-cast state has an amorphous structure, and therefore there is lack of preferred areas. In this case, the rapid solidification process has not resulted in the nucleation or creation of clusters of the Fe 23 B 6 phase. The thermal stability of the magnetic saturation polarization is a very important parameter for describing soft magnetic properties. From the analysis of the µ 0 M s -T curves (measured at a constant magnetic field of 0.7 T), it could be seen that the individual phases which are present in the volume of the sample affect its thermal stability. In Figures 5 and 6, the thermomagnetic curves, measured over the temperature range from room temperature up to 850 K, are presented. The heating speed was 10 K/min and curves were recorded in both directions. The shapes of the magnetic saturation polarization curves for all of the samples in the as-cast state are similar. However, for the samples of the injection-cast Fe 62 Co 9 W 1 Y 8 B 20 alloy and the suction-cast Fe 63 Co 8 W 1 Y 8 B 20 alloy, the area under the curve (with a bend on the curve at around 580 K) is significantly higher than for all the other investigated samples. These small changes in the shape of the thermomagnetic curves are related to the setup of the amorphous structure. As has been mentioned previously, the conglomerates become the areas which favor the creation of crystalline phases.
Materials 2020, 13, 835 7 of 14 possibility of producing amorphous material, these materials have a different magnetic structure. This means that the way of introducing the sample into the copper mold affects the distribution of magnetic atoms in volume. This reasoning confirms that, in the volume of amorphous samples, various clusters can be formed, which are conducive to forming various types of crystalline phases. Therefore, it is possible that, as a result of the thermal treatment of the Fe62Co9W1Y8B20 alloy, the aforementioned migration occurs, and the Y2Fe14B and Fe23B6 phases are created. The Fe23B6 phase is the so-called pre-natal, metastable phase, in which nuclei or clusters may be created during the rapid solidification process [36]. However, due to their small sizes, they cannot always be identified during X-ray diffractometry studies. In the case of the Fe63Co8W1Y8B20 alloy produced by the suction-casting method, the mechanism for the creation of this phase is similar. However, it has to be highlighted that the alloy in the as-cast state has an amorphous structure, and therefore there is lack of preferred areas. In this case, the rapid solidification process has not resulted in the nucleation or creation of clusters of the Fe23B6 phase. The thermal stability of the magnetic saturation polarization is a very important parameter for describing soft magnetic properties. From the analysis of the μ0Ms-T curves (measured at a constant magnetic field of 0.7 T), it could be seen that the individual phases which are present in the volume of the sample affect its thermal stability. In Figures 5 and 6, the thermomagnetic curves, measured over the temperature range from room temperature up to 850 K, are presented. The heating speed was 10 K/min and curves were recorded in both directions. The shapes of the magnetic saturation polarization curves for all of the samples in the as-cast state are similar. However, for the samples of the injection-cast Fe62Co9W1Y8B20 alloy and the suction-cast Fe63Co8W1Y8B20 alloy, the area under the curve (with a bend on the curve at around 580 K) is significantly higher than for all the other investigated samples. These small changes in the shape of the thermomagnetic curves are related to the setup of the amorphous structure. As has been mentioned previously, the conglomerates become the areas which favor the creation of crystalline phases.  Due to the fact that amorphous alloys are characterized by a chaotic ordering of atoms, volume areas with different concentrations of atoms can be formed. Such a state may even be associated with the formation of several different magnetic phases in the amorphous structure.
A higher value of saturation magnetic polarization for the injection-cast Fe 62 Co 9 W 1 Y 8 B 20 alloy and the suction-cast Fe 63 Co 8 W 1 Y 8 B 20 alloy can be explained by the presence of additional magnetic ordering in the volume of these alloys. This ordering is a contribution to the creation phase Y 2 Fe 14 B after the annealing process.
Materials 2020, 13, 835 8 of 14 Due to the fact that amorphous alloys are characterized by a chaotic ordering of atoms, volume areas with different concentrations of atoms can be formed. Such a state may even be associated with the formation of several different magnetic phases in the amorphous structure.
A higher value of saturation magnetic polarization for the injection-cast Fe62Co9W1Y8B20 alloy and the suction-cast Fe63Co8W1Y8B20 alloy can be explained by the presence of additional magnetic ordering in the volume of these alloys. This ordering is a contribution to the creation phase Y2Fe14B after the annealing process. In the studied temperature range, as a result of the supply of thermal energy, atomic diffusion occurs within the system, which is evident as a difference in the shape of the μ0M-T curves, taken in the directions of the heating up and cooling down of the sample. There is a major difference in the shape of the thermomagnetic curves for the samples subjected to thermal treatment. The crystallization products-in the form of crystalline grains-affect the shape of the curve. It was noticed that, in the samples of the injection-cast Fe62Co9W1Y8B20 alloy and the suction-cast Fe63Co8W1Y8B20 alloy, the contribution of the hard-magnetic phase increased at the expense of the amorphous matrix. On the assumption that the investigated magnetic samples fulfil Heisenberg assumptions, and with a critical coefficient of β = 0.36, the Curie temperatures for the investigated alloys were calculated (limited to 850 K by the equipment capability). Inserts show the linear fits of the μ0M (1/β) curves and determined TC (inserts to Figures 5 and 6). In the studied temperature range, as a result of the supply of thermal energy, atomic diffusion occurs within the system, which is evident as a difference in the shape of the µ 0 M-T curves, taken in the directions of the heating up and cooling down of the sample. There is a major difference in the shape of the thermomagnetic curves for the samples subjected to thermal treatment. The crystallization products-in the form of crystalline grains-affect the shape of the curve. It was noticed that, in the samples of the injection-cast Fe 62 Co 9 W 1 Y 8 B 20 alloy and the suction-cast Fe 63 Co 8 W 1 Y 8 B 20 alloy, the contribution of the hard-magnetic phase increased at the expense of the amorphous matrix. On the assumption that the investigated magnetic samples fulfil Heisenberg assumptions, and with a critical coefficient of β = 0.36, the Curie temperatures for the investigated alloys were calculated (limited to 850 K by the equipment capability). Inserts show the linear fits of the µ 0 M (1/β) curves and determined T C (inserts to Figures 5 and 6).
In the alloys in which the hard-magnetic phase was absent, over the studied temperature range, only the Curie temperature of the amorphous matrix was determined. In the cases of the further discussed alloys (Fe 62 Co 9 W 1 Y 8 B 20 and Fe 63 Co 8 W 1 Y 8 B 20 ), the Curie temperature of the Y 2 Fe 14 B phase was also determined. It has to be noted that, in the case of the amorphous materials, it is only possible to talk about the average value of the Curie temperature, as the process occurs within a narrow temperature range. In the case of the as-cast samples, it is evident that the production method itself influences the value of T C ( Table 2). For the injection-cast samples, the T C was found to be a few degrees Kelvin lower than for the suction-cast samples [36,37]. The chemical composition of the alloy also affected the value of T C . With an increase in Co content, the Curie temperature of the alloy was slightly higher. For the Fe 62 Co 9 W 1 Y 8 B 20 alloy, the T C values related to the Y 2 Fe 14 B and Fe 23 B 6 crystalline phases were determined: they were found to be 580 K, and 680 K, respectively. For the suction-cast Fe 63 Co 8 W 1 Y 8 B 20 alloy, the T C corresponding to the phase was found to be 595 K. Analyzing further the µ 0 M (1/β) curves for the samples of the Fe 62 Co 9 W 1 Y 8 B 20 and Fe 63 Co 8 W 1 Y 8 B 20 alloys, the T C for the bend on the curve above 700 K has to be considered. Neither of the aforementioned crystalline phases has a T C of this value. The degree of crystallization of these samples is quite substantial (Figures 3e and  4f), and neither of the identified crystalline phases contain Co. This could suggest that the described T C in the vicinity of 700 K is the Curie temperature of the remaining amorphous matrix, which is rich in Co. Figures 7 and 8 reveal the static hysteresis loops for the studied samples. Based on the analysis of the static hysteresis loops, saturation magnetic polarization and coercivity values were determined, and these are given in Table 3. After further analysis, the differences in the magnetization curves depending on the "direction" (magnetizing/demagnetizing) of the external magnetic field were highlighted, and these are shown in the respective inserts in Figures 7a-c and 8a-c. In the literature, this shape of hysteresis loop is called the "wasp" shape [38,39]. In the initial stages of growth of the magnetically hard phases, the widening of the hysteresis loop at the beginning of the M-H system was not observed. However, the presence of the phases affected the width of the loop at higher values of the external magnetic field.  For the samples in the as-cast state, there is generally a visible increase in the value of the coercive field with decreasing Co content in the alloy; the injection-cast Fe 62 Co 9 W 1 Y 8 B 20 alloy is an exception. The decrease in the value of the H C is related to the presence of small crystalline grains of the Fe 5 Y and αFe phases (Figure 3b).
In soft magnetic materials, the crystallite size is decisive in the case of the saturation magnetic polarization or coercive field [40]. The presence of small grains does not always lead to improved soft magnetic properties in nanocrystalline ferromagnetics. The results of the study published in [41] prove that the presence of a hard (or semi-hard magnetic phase) in relation to the soft magnetic phase can reduce the coercive field value.
The annealing process caused a major increase in the value of coercive field. Such a high coercive field value compared to the other samples tested is associated with the presence of Y 2 Fe 14 B hard magnetic phase grains in the alloy volume. In the sample of Fe 62 Co 9 Y 8 W 1 B 20 alloy produced by the injection method and subjected to the annealing process, the crystallites of this phase were characterized by a smaller average dimension. In addition, the presence of other crystalline phases was identified in this sample, including two with soft magnetic properties, which resulted in a lower total coercive field value despite the presence of a hard magnetic phase. For the samples in the as-cast state, there is generally a visible increase in the value of the coercive field with decreasing Co content in the alloy; the injection-cast Fe62Co9W1Y8B20 alloy is an exception. The decrease in the value of the HC is related to the presence of small crystalline grains of the Fe5Y and αFe phases (Figure 3b).
In soft magnetic materials, the crystallite size is decisive in the case of the saturation magnetic polarization or coercive field [40]. The presence of small grains does not always lead to improved soft magnetic properties in nanocrystalline ferromagnetics. The results of the study published in [41] prove that the presence of a hard (or semi-hard magnetic phase) in relation to the soft magnetic phase can reduce the coercive field value.
The annealing process caused a major increase in the value of coercive field. Such a high coercive field value compared to the other samples tested is associated with the presence of Y2Fe14B hard magnetic phase grains in the alloy volume. In the sample of Fe62Co9Y8W1B20 alloy produced by the injection method and subjected to the annealing process, the crystallites of this phase were characterized by a smaller average dimension. In addition, the presence of other crystalline phases was identified in this sample, including two with soft magnetic properties, which resulted in a lower total coercive field value despite the presence of a hard magnetic phase.

Conclusions
In this work, the results of an investigation into the structure and magnetic properties of rapidlyquenched iron-based alloys are presented. Samples of the alloys Fe61+xCo10−xW1Y8B20 (where x = 0, 1 or

Conclusions
In this work, the results of an investigation into the structure and magnetic properties of rapidly-quenched iron-based alloys are presented. Samples of the alloys Fe 61+x Co 10−x W 1 Y 8 B 20 (where x = 0, 1 or 2) were made using two different production methods. The obtained materials were subjected to thermal treatment at a temperature of 940 K for 10 min. Each of the investigated alloys was made several times, and the results of the studies have a high repeatability.
Based on previous work by the authors concerning alloys with similar chemical compositions [42], it was found that their crystallization process is primary. This means that, after the crystallization process, there are at least two different crystallization products-the amorphous phase and the crystalline phase-with different chemical compositions.
Based on the aforementioned results, there are certain conclusions which can be drawn: • The injection and suction-casting methods (into a copper die) allow the manufacture of bulk amorphous alloys with the chemical composition Fe 61+x Co 10−x W 1 Y 8 B 20 (x = 0, 2).

•
In the case of the alloy Fe 62 Co 9 W 1 Y 8 B 20 in the as-cast state, the presence of a small amount of the phase Fe 5 Y was observed; this could be related to the minor difference in the process time between the two production methods: for the injection-casting method, the time is marginally longer, which might lead to the formation of crystallites during the production process.

•
The isothermal annealing process, carried out at a temperature close to the crystallization temperature (940 K/10 min), led to the partial crystallization of the alloys.

•
Based on the obtained magnetic test results-i.e., the thermomagnetic curves and static hysteresis loops-it was found that the tested alloys were magnetically heterogeneous materials.

•
The Curie temperature values were found to be slightly higher for the suction-cast alloys, which could be related to the different degrees of disorder in the structure.

•
The magnetic properties of the investigated alloys in the as-cast state were found to be similar, regardless of the chosen production method.

•
The low value of coercivity for the injection-cast Fe 62 Co 9 W 1 Y 8 B 20 alloy is related to the presence of grains of the Fe 5 Y crystalline phase.

•
The carefully designed parameters of the production process for the rapidly-quenched alloys have an influence on the obtaining of fine grains of crystalline phases within the amorphous matrix during the solidification process.

•
An increase in the Fe content, at the expense of the Co content, resulted in a decrease in the Curie temperature.

•
The size of the crystalline grains was found to exert a strong influence on the value of the coercive field in terms of an increase in the crystallite size hindering the magnetization process, which was visible from the high coercivity value for the suction-cast Fe 63 Co 8 W 1 Y 8 B 20 alloy. Despite the presence of the Y 2 Fe 14 B phase in the volume of the injection-cast Fe 62 Co 9 W 1 Y 8 B 20 alloy, a low coercivity was achieved after annealing. This is connected with the relatively small crystallitesize of this phase and the presence of the Fe 23 B 6 soft magnetic phase.