Effect of Gd on the devitrification and magnetic properties of [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx metallic glasses

1 Effect of Gd on the devitrification and magnetic properties of [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx metallic glasses J. Torrens-Serra1,*, B. Kustov1, P. Bruna2,3,4 1Departament de Física, Universitat de les Illes Balears, Cra. de Valldemossa km 7.5, 07122, Palma de Mallorca. Spain 2Departament de Física, Universitat Politècnica de Catalunya, BarcelonaTech (UPC), Av. Eduard Maristany 16, 08019 Barcelona, Spain. 3Institut de Tècniques Energètiques (INTE), Av. Diagonal 647, 08028 Barcelona, Spain 4Barcelona Research Center in Multiscale Science and Engineering, Universitat Politècnica de Catalunya. BarcelonaTech (UPC), Av. Eduard Maristany 16, 08019 Barcelona, Spain Abstract The influence of microalloying with Gd on the thermal stability, crystallization and magnetic properties of [Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx (x=0,1,2,5) metallic glasses was examined. Various glass-forming ability criteria were calculated based on thermal characteristics obtained by differential scanning calorimetry for melt-spun ribbons and compared to the maximum size achieved of the alloys via direct rod casting. Although thermodynamic-based criteria predict larger glass-forming ability for the alloy with higher Gd content, x=5 alloy showed the lowest ability to form full glassy rods. The structural evolution after first crystallization event was analyzed by x-ray diffraction and Transmission Mössbauer spectroscopy. The observed changes in the precipitated phases in x=5 alloy compared to x=0,1,2 alloys are associated to the decrease in the glass-forming ability. The formation of * [email protected] 2 phases other than (FeCo)23B6 phase in x=5 alloy is responsible for not achieving fully glassy rods. Additionally, the magnetic properties of glassy ribbons are significantly affected by the addition of Gd. Saturation magnetization and Curie temperature decrease as Gd content increases. Keywords: Bulk Metallic Glasses; Mössbauer Spectroscopy; Glass-forming ability; Crystallization 1. Introduction Fe-based bulk metallic glasses are called to replace crystalline alloys in many applications due to their excellent functional properties. Despite these promising properties, the largescale industrial production has only been possible in form of ribbons of about 50 µm or pieces produced from amorphous powders by powder metallurgy. Many commercial applications require shapes or mechanical characteristics that only bulk metallic glasses can achieve. Since Inoue and co-workers [1] cast the first Fe-based bulk metallic glass (with a thickness larger than 1 mm) in 1995, many different alloy families have been reported in literature having high glass-forming ability (GFA) combined with excellent functional properties like high strength, high corrosion resistance and excellent soft magnetic properties such as low coercivity, high magnetic permeability and low core losses [2–8]. In recent years, considerable efforts have been directed towards the production of millimeter-sized full glassy alloys while maintaining their functional properties. Despite these extensive efforts, several challenges persist and remain unsolved. In a recent review, Li et al. [2] pointed to the following issues that would deserve further research for a future application and commercialization of Fe-based bulk metallic glasses: a) obtention of larger-sized Fe-based BMGs; b) improving the fabrication methods; c) enhancement of room temperature 3 toughness; d) attaining higher saturation magnetization; and e) cost reduction through the use of industrial raw materials and low vacuum conditions. In many alloys it has been observed that the large glass-forming ability is directly related to the formation of complex crystalline structures with a large unit cell as a primary crystallization product [9,10]. The high resistance to crystallization from the undercooled liquid helps to stabilize the amorphous structure. In particular, Fe23B6 is the responsible for the superior GFA in (Fe,Co,Ni)-Si-B-Nb alloys [6,11–16], which has been reported to display excellent soft magnetic properties, as well as, high mechanical strength. However, the largest dimension successfully cast in fully amorphous state remains below 8 mm, which does not meet industrial expectations [9]. Many strategies have been used with the aim to enhance the GFA. Aside from other strategies like different casting techniques or processing routes [17,18], the appropriate addition of small amounts of different elements (microalloying) has been proven to be effective in enhancing both glass-forming ability and functional properties [19–28]. The presence of these alloying elements produce changes in the local atomic configuration [29] and create nanoscale structural heterogeneities [30] which affect glassforming ability, crystallization and functional properties of BMGs. However, the mechanisms underlying this effect are not fully understood. Li et al. [2] highlighted this issue as key point that should be solved. In some recent papers we have analyzed the role of different alloying elements on the GFA and crystallization path of Fe-B-Si-Nb-based alloys. The use of Transmission Mössbauer Spectroscopy (TMS) has enabled detail tracking of the changes in the environments of Fe atoms during the devitrification of the glass and to the determine the hyperfine parameters of the Fe23B6 phase formed in these alloys, which are influenced by the alloying elements. Moreover, this technique has allowed for quantitative analysis of the percentage of Fe atoms 4 forming this phase and differentiation between various environments generating distinct hyperfine magnetic fields [24,25]. The role of rare earth (RE) microalloying elements is of particular interest due to their impact on both glass formation and magnetic properties of BMGs. RE possess a magnetic moment originated from the localized 4f electrons. Several studies have demonstrated that the interaction between these 4f electrons and the 3d electrons from Fe atoms negatively affect the soft magnetic properties of Fe-B-Si-Nb-(Tb,Dy) BMGs, as well as, deteriorate their properties as glass formers [22,23]. Consequently, the use of RE as microalloyant is still controversial. In the present study, our purpose is to shed some light on the role played by Gd on the glass-forming ability, the devitrification process and magnetic properties of [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx (x=0,1,2,5) alloys. Several criteria predicting GFA and critical thickness have been calculated compared to the critical diameter obtained experimentally by direct casting of the alloys. Transmission Mössbauer spectroscopy, complemented by X-ray diffraction, have been employed to elucidate the structural changes in the amorphous alloys upon heat treatment. The study reveals that different phases precipitate depending on Gd content of the alloy, which is linked to the decrease in the glass-forming ability. Gd also has a detrimental effect on magnetic properties. 2. Experimental procedure Ribbons of [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx (x=0,1,2,5) compositions (from now on designed as Gd0, Gd1, Gd2 and Gd5, respectively) with a thickness of 50 µm and a width of 4 mm were prepared by single-roller melt spinning at a linear speed of 40 m/s from polycrystalline master alloys. The process for the preparation of master alloys can be found elsewhere [24]. All the ribbons show a regular and continuous shape except x=5 alloy which is produced in form of flakes of 2-3 cm. From x=5 alloy, rods of diameter 1.5 mm and 2 mm 5 were fabricated by injection casting in a water-cooled copper mold under protective highpurity argon atmosphere. The structure of the as-cast ribbons and rods, as well as, the evolution of the structure of the ribbons subjected to a different thermal treatments was studied using a Philips PW 1050 x-ray diffractometer (XRD) with Co Kα radiation (λ=1.7888 Å) in Bragg-Brentano geometry. The thermal characterization of the amorphous ribbons was performed by differential scanning calorimetry (DSC) using a NETZSCH DSC 404 at a heating rate of 20 K/min under a flow of high purity Ar. Transmission Mössbauer spectra were obtained at room temperature and pressure using a conventional constant acceleration spectrometer with a 25 mCi source of 57Co in Rh matrix. The spectra were recorded in a multichannel analyzer using a velocity range of ±7.5 mm/s and were subsequently fitted with the NORMOS software [31], considering one magnetic hyperfine field distribution with linear correlation between the isomer shift and the magnetic field for the amorphous structures and one or more discrete spectra (singlets, doublets or sextets) for the crystalline phases. The fitted parameters are the isomer shift (δ), always expressed relative to the isomer shift of the bcc-Fe measured at room temperature and pressure, the quadrupole splitting (Δ) and the hyperfine magnetic field (BHF). In all cases the area expressed in % corresponds to the fraction of Fe atoms in a particular environment with respect to the total amount of Fe atoms and the magnitudes inside parentheses are the standard deviation. Saturation magnetization of the as-spun ribbons was obtained from the Magnetization vs. applied field (M-H) hysteresis loops recorded with a Lakeshore 735 vibrating sample magnetometer (VSM) at 305 K, for magnetic fields up to 1.5 T. The Curie temperature, TC, of the amorphous ribbons was determined from magnetization vs temperature curves, M(T), recorded using a home-made Faraday magnetometer. Samples subjected to a fixed magnetic field of 5.5 kOe were heated at a constant rate of 20 K/min up to the experimental limit of 6 the device, 973 K. Experimental data was then analyzed using the method proposed by Herzer [32]. Close to TC, the saturation magnetization can be written as: 𝑀𝑀𝑆𝑆(𝑇𝑇) = 𝑀𝑀(0) �1−𝑇𝑇 𝑇𝑇𝑐𝑐�0.36 (1) The experimental M(T) curves are plotted as 𝑀𝑀𝑠𝑠10.36 � vs T and the Curie temperature is considered the temperature where 𝑀𝑀𝑠𝑠10.36 � deviate from linearity. The coercivity Hc was measured using a Foerster Coercimat under an applied field high enough to saturate the samples. All magnetic properties were measured under DC magnetic field. 3. Results and discussion 3.1 Glass formation and thermal stability Figure 1(a) presents the X-ray diffraction (XRD) patterns of the as-spun ribbons, all of which exhibit a broad halo indicative of amorphous structures. Figure 1. (a) XRD patterns of as-spun ribbons for all four alloys, (b) XRD patterns for the as-casted rods of diameter d=1.5 and d=2 mm of Gd5 alloy. 7 Figure 2 displays the DSC thermograms recorded at a heating rate of 20 K/min for all four alloys. In Figure 2(a), the devitrification pathway of Gd5 alloy is compared to Gd0, Gd1, and Gd2 ones, already published in [24] and included in the plot to assist the reader. DSC curves are characterized by an endothermic change typical of the glass transition and multiple exothermic peaks. Gd5 alloy exhibits a more complex crystallization path, with at least three overlapping peaks during the initial crystallization stage, followed by two additional peaks at higher temperatures in contrast by the three distinct crystallization peaks found in Gd0, Gd1, and Gd2 alloys. The corresponding values of glass transition temperature (Tg), onset crystallization temperature (Tx), and peak temperature (Tp1, Tp2 and Tp3) are listed in Table 1. Glass transition temperature does not show a monotonous increase with the amount of Gd present in the alloy. The addition of 1% of Gd to the alloy decreases the glass transition temperature in respect to the Gd-free alloy (Gd0), yet this temperature increases again with larger Gd alloys. On the contrary, the onset crystallization temperature is shifted dramatically to higher temperatures with the addition of Gd. Consequently, the supercooled liquid range (SLR), defined as ∆𝑇𝑇𝑥𝑥=𝑇𝑇𝑥𝑥−𝑇𝑇𝑔𝑔 , shows a large enhancement, from the 37 K found in Gd0 alloy to 60 K in Gd5. The first two crystallization stages also show a shift towards larger temperatures with the Gd content. However, the third one seems to be reduced in Gd5 alloy and does not follow the same trend. The melting and the subsequent solidification of the alloys can be found in figure 2(b). The increase of Gd in the composition expands the temperature range in which the coexistence of solid and liquid phases is found. The characteristic melting temperature (Tm), liquidus temperature (Tl) and solidus temperature (Ts), for all alloys are also listed in Table 1. The addition of Gd decreases the stability of the liquid shown by a shift of liquidus and solidus temperatures to larger values. Also, the 8 different melting peaks tend to be more separated as more Gd is in the composition. This indicates that the Gd is moving the alloys far from the eutectic point. Figure 2. DSC thermograms of as-spun ribbons showing (a) the crystallization and (b) the melting and solidification, for all four alloys. Characteristic temperatures are signed by an arrow. Table 1. Glass transition temperature (Tg); onset temperature of the first crystallization peak (Tx); supercooled liquid range (ΔTx); first, second and third crystallization peak temperatures (Tp1, Tp2, Tp3); melting temperature (Tm); liquidus temperature (Tl); solidus temperature (Ts) and primary crystallization enthalpy (ΔHx1), for all the studied alloys. The glass forming ability can be directly measured by two parameters: the critical cooling rate (Rc) and the maximum diameter (or dimension) of fully amorphous alloy (dc). However, these magnitudes are difficult to measure and depend strongly on the fabrication method. In T g (K) T x (K) ΔT x (K) T p1 (K) T p2 (K) T p3 (K) T m (K) T l (K) T s (K) ΔH x1 (J/g) Gd0 821 860 39 869 981 1062 1326 1421 1368 33±2 Gd1 815 867 52 879 1039 1110 1318 1410 1376 30±2 Gd2 833 894 56 903 1047 1127 1328 1452 1431 37±2 Gd5 907 967 60 973 1072 1097 1340 1451 1425 51±2 9 the literature several criteria based on the characteristic temperatures easily measurable by thermal analysis have been proposed to predict the glass forming ability of the alloys and many papers are devoted to evaluating which is the best criterion [33,34]. The supercooled liquid range (SLR) is commonly used to evaluate the stability of the glass (resistance to crystallization), whereas the reduced glass transition temperature introduced by Turnbull [35], defined as Trg=Tg/Tl, indicates that the smaller is the interval between glass transition and liquidus temperature, the easier is to avoid nucleation of crystalline phases during cooling. More recently, the new criteria incorporate both considerations and basically depend on the three characteristic temperatures – Tg, Tx and Tl. In order to test some of these criteria in our alloys, we have calculated the values of the 𝛼𝛼=𝑇𝑇𝑥𝑥 𝑇𝑇𝑙𝑙 [36], 𝛽𝛽=�𝑇𝑇𝑥𝑥 𝑇𝑇𝑔𝑔+𝑇𝑇𝑔𝑔 𝑇𝑇𝑙𝑙� [36], 𝛾𝛾=𝑇𝑇𝑥𝑥 𝑇𝑇𝑔𝑔+𝑇𝑇𝑙𝑙 [37], 𝛾𝛾𝑚𝑚=2𝑇𝑇𝑥𝑥−𝑇𝑇𝑔𝑔 𝑇𝑇𝑙𝑙 [34], 𝜔𝜔=2𝑇𝑇𝑥𝑥−𝑇𝑇𝑔𝑔 𝑇𝑇𝑙𝑙+𝑇𝑇𝑥𝑥 [38], and 𝜛𝜛=(𝑇𝑇𝑥𝑥−𝑇𝑇𝑔𝑔)𝑇𝑇𝑔𝑔𝑇𝑇𝑥𝑥 (𝑇𝑇𝑙𝑙−𝑇𝑇𝑥𝑥)3 [39] criteria from the measured characteristic temperatures and compared them to the maximum diameter of the fully amorphous rod casted experimentally. We have also determined the critical cooling rate (Rc) and the critical section thickness (Zc) with the empirical formula proposed by Lu [37]: 𝑅𝑅𝑐𝑐= 5.1 × 1021exp(−117.19𝛾𝛾) and 𝑍𝑍𝑐𝑐= 2.8 × 10−7exp(41.7𝛾𝛾) , respectively. In a recent paper we discussed the influence of some minor additions to the glass forming ability of Febased BMGs. There, we showed that rods of compositions Gd0 and Gd2 can be obtained in fully amorphous state up to 1.5 mm of diameter, but it was not possible for Gd1 alloys [24]. To complete this study and unambiguously determine if this rare earth addition may be beneficial to GFA or not, 1.5 and 2 mm diameter rods of Gd5 composition have been cast. The XRD measurements (Figure 1(b)) show that none of them is fully glassy. Some conclusions can be extracted from the comparison between the GFA criteria and the results of the casting. Both SLR and reduced glass transition criteria cannot describe the GFA of our 16 average magnetic hyperfine field of the hyperfine field distribution. at% Fe corresponds to the atomic percentage of Fe atoms in each phase. Phase As-quenched 1st. trans. Amorphous δ (mm s-1) 0.03 (2) 0.02 (1) Δ (mm s-1) -0.04 (1) 0.01 (1) BHFavg(T) 18.1 (1) 20.0 (2) at.% Fe 100 87.0 (5) Fe2Nb δ (mm s-1) - -0.20 (1) Δ (mm s-1) - 0.34 (3) at.% Fe - 5.5 (5) ε-FeSi δ (mm s-1) - 0.18 (1) Δ (mm s-1) - 0.46 (2) at.% Fe - 7.5 (5) Figure 4. Experimental Mössbauer spectra (blue dots) and their fit (red line) for the Gd5 composition in the as-quenched state (a), after the first transformation (c) and in the fully crystalline state (d). The colored lines correspond to the several subspectra used for the fitting and explained in text. (b) The hyperfine field distributions of the as-quenched ribbon and after the first transformation. The inset shows the evolution of the average hyperfine field of the as-quenched ribbons for all the compositions. 17 Table 4. Hyperfine parameters for the fully crystalline Gd5 sample. δ is the isomer shif, Δ is the quadrupole splitting and BHF is the magnetic hyperfine field in T. at% Fe corresponds to the atomic percentage of Fe atoms in each phase. bcc-(FeCo) Fe23B6 FeB - I FeB - II Fe3B – I Fe3B – II Fe2Nb ε-FeSi δ (mm s -1 ) 0.01 (1) -0.26 (1) 0.1 (1) 0.07 (1) 0.15 -0.04 (2) -0.1 0.28 Δ (mm s-1) 0.02 (1) 0.4 (1) 0.05 (2) -0.3 (1) -0.2 (2) 0.10 (3) 0.56 0.49 (3) BHF (T) 36.7 (1) 16.6 (1) 9.8 (1) 13.1 (1) 21.4 (1) 25.5 (1) - - at.% Fe 30.7 (1) 10.6 (1) 12.2 (1) 16.9 (1) 15.2 (1) 6.5 (1) 5.2 (1) 2.7 (1) Figure 5. Variation of the atomic percentage of the different phases in the fully crystalline samples. bcc-(FeCo) and (FeCo)23B6 include the corresponding Co-free phases. 3.3 Magnetic properties 18 Soft magnetic alloys should have high Curie temperature to enlarge the maximum service temperature, large saturation magnetization to maximize the magnetic flux, large initial permeability and minimal coercivity to reduce the hysteresis losses. In Figure 6 the M(H) loops for as-spun ribbons are presented for all four compositions. The shape of the curve exhibits a very sharp increase of magnetization at low applied fields, hallmark of a high initial permeability, reaching nearly the saturation value below applied fields of 𝜇𝜇0𝐻𝐻=0.2 T. Furthermore, saturation magnetization (Ms) values can be estimated as the magnetization at the highest applied field (1.5 T) provided the slope of the M-H curve is 𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 ≈0. The evolution of the normalized magnetization, 𝑑𝑑𝑠𝑠(𝑇𝑇) 𝑑𝑑𝑠𝑠(298 𝐾𝐾) with temperature of the as-spun ribbons is plotted in Fig. 7. A drop to zero magnetization demonstrates the ferromagnetic to paramagnetic transition of the amorphous structure. The Curie temperature of the as-spun ribbons for the different compositions has been calculated using Herzer method described in the experimental section. Figure 8 compares (data also reproduced in Table 5) the values of saturation magnetization and Curie temperature in function of Gd content. Gd-free alloys has the largest values with a Curie temperature of 676 K and Ms=113 Am2/kg, which are values consistent with those found by other authors in alloys with close composition [47]. The addition of Gd has a detrimental effect on saturation magnetization and Curie temperature. Fig. 8 shows that the shift of the Curie temperature to lower values is not linear, but the slope of the curve gets stepper as Gd content becomes larger. A similar trend is found in the saturation magnetization behaviour. Both magnitudes depend on the exchange coupling between TM atoms (Fe and Co), where the magnetic moments carried by 3d electrons couple ferromagnetically. In contrast, the Gd atoms possess 4f electrons, which interact with localized 3d electrons of Fe and Co antiferromagnetically producing the reduction of the 19 overall magnetic moment. Buschow and co-workers [48] calculated the normalized magnetic moment in amorphous Gd-Fe and Gd-Co alloys using a model based on the probability of Fe (or Co) having n Fe atoms as nearest neighbours. They found a good correlation with experimental data. The magnetic moment per unit of atom has a quasi-linear change with Gd concentration at intermediate concentration but tends to smooth down as the concentrations tends to pure Fe (or Co). Moreover, the observed change of the mean hyperfine field in the amorphous by Mössbauer spectroscopy (inset Fig 4(b)) corresponds perfectly to the magnetic magnitudes and therefore confirm this interpretation based on the idea that magnetic properties are determined by the exchange interactions between Fe (or Co) atoms and their closest environments. The value of coercivity of as-spun ribbons is below 10 A/m which indicates that are excellent soft magnetic alloys, even with no further annealing. Li and coworkers reported an increase of coercivity with the addition of a RE like Tb and they attributed it to the larger anisotropy constant [22]. It is worth noting, that the Hc for Gd5 alloys could not be measured because of the limitations of the experimental device due to the flaky-shape of the ribbon. Table 5. Curie temperature (TC), saturation magnetization (Ms) and coercivity (Hc) for [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx (x=0,1,2,5) alloys. T C (K) H c (A/m) M s (Am2/kg) Gd0 676±1 2.2±1.1 113±1 Gd1 666±1 10.7±0.5 108±1 Gd2 624±3 3.6±1.3 96±2 Gd5 555±3 - 68±1 20 Figure 6. M-H curves measured at T=305 K for [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx (x=0,1,2,5) alloys. 21 Figure 7. Normalized magnetization in function of temperature for [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx (x=0,1,2,5) alloys. Figure 8. Curie temperature (Tc) and saturation magnetization (Ms) plotted against Gd concentration. 22 Conclusions To sum up, the role of Gd microalloying in [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Gdx (x=0,1,2,5) alloys has been investigated. On the one hand, the larger thermal stability of the amorphous phase up to higher temperatures and the improvement of the GFA predicting parameters with the addition of Gd point to an enhancement of glass forming ability with the addition of Gd. On the other hand, the experimental casting of the alloys shows a critical diameter below 1.5 mm in the Gd5 alloy. The primary crystallization of glassy ribbons has been investigated using x-ray diffraction and Transmission Mössbauer spectroscopy. Whereas in alloys with low amount of Gd (x=0,1,2) the main crystallization phase is Fe23B6-type boride, in the alloy with larger amount of Gd (x=5), the primary crystallization proceeds through the precipitation of multiple phases like Fe3B, FeB, bcc-Fe(Co), Fe2Nb, Fe2Gd and FeSi. Therefore, we conclude that the deterioration of the GFA in the latter alloy is related to a different devitrification path in respect to Gd0, Gd1 and Gd2 alloys, especially to the formation of phases with smaller unit cell than Fe23B6-type boride. Magnetic properties are also influenced by the addition of Gd. Both Curie temperature and saturation magnetization of the metallic glasses are reduced with Gd. Acknowledgements Part of the experimental work reported in this paper was conducted at IFW-Dresden by J. Torrens-Serra during a post-doctoral stay financially supported from Generalitat de Catalunya through a “Beatriu de Pinós” grant (Nº 2009 BP-A 00138). Prof. J. Eckert and Dr. M. Stoica are acknowledged for fruitful discussions and assistance. J. Torrens-Serra and B. Kustov also acknowledges financial support from the Comunitat Autònoma de les Illes 23 Balears through the Servei de Recerca i Desenvolupament and the Conselleria d'Educació i Universitats with funds from the Tourist Stay Tax Law ITS (PDR2020/39-2 - ITS2017-006). P. Bruna acknowledges financial support from Grant PID2023-146623NB-I00 funded by MICIU/AEI/10.13039/501100011033 and by ERDF/EU, Generalitat de Catalunya AGAUR grant 2021-SGR-00343. This work is part of Maria de Maeztu Units of Excellence Programme CEX2023-001300-M / funded by MCIN/AEI / 10.13039/501100011033. References [1] A. Inoue, Y. Shinohara, J. Gook, Thermal and magnetic properties of bulk Fe-based glassy alloys prepared by copper mold casting, Mater. Trans. JIM. 36 (1995) 1427– 1433. [2] H.X. Li, Z.C. Lu, S.L. Wang, Y. Wu, Z.P. Lu, Fe-based bulk metallic glasses: Glass formation, fabrication, properties and applications, Prog. Mater. Sci. 103 (2019) 235–318. https://doi.org/10.1016/j.pmatsci.2019.01.003. [3] C. Suryanarayana, A. Inoue, Iron-based bulk metallic glasses, Taylor & Francis, 2013. https://doi.org/10.1179/1743280412Y.0000000007. [4] W. Yang, H. Liu, Y. Zhao, A. Inoue, K. Jiang, J. Huo, H. Ling, Q. Li, B. Shen, Mechanical properties and structural features of novel fe-based bulk metallic glasses with unprecedented plasticity, Sci. Rep. 4 (2014) 6233. https://doi.org/10.1038/srep06233. [5] V. Ponnambalam, J.S. Poon, G.J. Shiflet, Fe-based bulk metallic glasses with diameter thickness larger than one centimeter, J. Mater. Res. 19 (2004) 1320–1323. https://doi.org/10.1557/JMR.2004.0176. [6] A. Inoue, B.L. Shen, C.T. Chang, Super-high strength of over 4000 MPa for Febased bulk glassy alloys in [(Fe1−xCox)0.75B0.2Si0.05]96Nb4 system, Acta Mater. 52 (2004) 4093–4099. https://doi.org/10.1016/j.actamat.2004.05.022. [7] A. Inoue, B.L. Shen, A.R. Yavari, A.L. Greer, Mechanical properties of Fe-based bulk glassy alloys in Fe-B-Si-Nb and Fe-Ga-P-C-B-Si systems, J. Mater. Res. 18 (2003) 1487–1492. https://doi.org/10.1557/JMR.2003.0205. [8] Z.P. Lu, C.T. Liu, J.R. Thompson, W.D. Porter, Structural amorphous steels, Phys. 24 Rev. Lett. 92 (2004) 245503. https://doi.org/10.1103/PhysRevLett.92.245503. [9] A. Inoue, F.L. Kong, Q.K. Man, B.L. Shen, R.W. Li, F. Al-Marzouki, Development and applications of Feand Co-based bulk glassy alloys and their prospects, J. Alloys Compd. 615 (2015) S2–S8. https://doi.org/10.1016/j.jallcom.2013.11.122. [10] M.J. Duarte, J. Klemm, S.O. Klemm, K.J.J. Mayrhofer, M. Stratmann, S. Borodin, A.H. Romero, M. Madinehei, D. Crespo, J. Serrano, S.S.A. Gerstl, P.P. Choi, D. Raabe, F.U. Renner, Element-Resolved Corrosion Analysis of Stainless-Type GlassForming Steels, Science (80-. ). 341 (2013) 372–376. https://doi.org/10.1126/science.1230081. [11] A. Inoue, B.L. Shen, A.R. Yavari, A.L. Greer, Mechanical properties of Fe-based bulk glassy alloys in Fe–B–Si–Nb and Fe–Ga–P–C–B–Si systems, J. Mater. Res. 18 (2003) 1487–1492. https://doi.org/10.1557/JMR.2003.0205. [12] A. Inoue, B. Shen, Soft magnetic bulk glassy Fe-B-Si-Nb alloys with high saturation magnetization above 1.5 T, Mater. Trans. 43 (2002) 766–769. https://doi.org/10.2320/matertrans.43.766. [13] M. Stoica, R. Li, A.R. Yavari, G. Vaughan, J. Eckert, N. Van Steenberge, D.R. Romera, Thermal stability and magnetic properties of FeCoBSiNb bulk metallic glasses, J. Alloys Compd. 504 (2010) S123–S128. https://doi.org/10.1016/j.jallcom.2010.04.013. [14] B. Shen, C. Chang, A. Inoue, Formation, ductile deformation behavior and softmagnetic properties of (Fe,Co,Ni)-B-Si-Nb bulk glassy alloys, Intermetallics. 15 (2007) 9–16. https://doi.org/10.1016/j.intermet.2005.11.037. [15] M. Imafuku, S. Sato, E. Matsubara, A. Inoue, Structural study of Fe90−xNb10Bx (x=10, 20 and 30) glassy alloys, J. Non. Cryst. Solids. 312–314 (2002) 589–593. https://doi.org/10.1016/S0022-3093(02)01790-8. [16] J. Fornell, S. González, E. Rossinyol, S. Suriñach, M.D. Baró, D. V. LouzguineLuzgin, J.H. Perepezko, J. Sort, A. Inoue, Enhanced mechanical properties due to structural changes induced by devitrification in Fe-Co-B-Si-Nb bulk metallic glass, Acta Mater. 58 (2010) 6256–6266. https://doi.org/10.1016/j.actamat.2010.07.047. [17] T. Bitoh, A. Makino, A. Inoue, A.L. Greer, Large bulk soft magnetic [(Fe0.5Co0.5)0.75B 0.20Si0.05]96Nb4 glassy alloy prepared by B2O3 flux melting and water quenching, Appl. Phys. Lett. 88 (2006) 10–12. https://doi.org/10.1063/1.2201900. [18] S. Di, Q. Wang, J. Zhou, Y. Shen, J. Li, M. Zhu, K. Yin, Q. Zeng, L. Sun, B. Shen, Enhancement of plasticity for FeCoBSiNb bulk metallic glass with superhigh strength through cryogenic thermal cycling, Scr. Mater. 187 (2020) 13–18. https://doi.org/10.1016/j.scriptamat.2020.05.059. [19] C. Chang, B. Shen, A. Inoue, Synthesis of bulk glassy alloys in the (Fe,Co,Ni)–B– Si–Nb system, Mater. Sci. Eng. A. 449–451 (2007) 239–242. https://doi.org/10.1016/j.msea.2006.02.253. [20] B. Shen, C. Chang, Z. Zhang, A. Inoue, Enhancement of glass-forming ability of FeCoNiBSiNb bulk glassy alloys with superhigh strength and good soft-magnetic 25 properties, J. Appl. Phys. 102 (2007) 023515. https://doi.org/10.1063/1.2757013. [21] Z. Long, Y. Shao, F. Xu, H. Wei, Z. Zhang, P. Zhang, X. Su, Y effects on magnetic and mechanical properties of Fe-based Fe-Nb-Hf-Y-B bulk glassy alloys with high glass-forming ability, Mater. Sci. Eng. B Solid-State Mater. Adv. Technol. 164 (2009) 1–5. https://doi.org/10.1016/j.mseb.2009.04.010. [22] J.W. Li, A.N. He, B.L. Shen, Effect of Tb addition on the thermal stability, glassforming ability and magnetic properties of Fe-B-Si-Nb bulk metallic glass, J. Alloys Compd. 586 (2014) S46–S49. https://doi.org/10.1016/j.jallcom.2012.09.087. [23] J. Li, W. Yang, M. Zhang, G. Chen, B. Shen, Thermal stability and crystallization behavior of (Fe0.75−xDyxB0.2Si0.05)96Nb4 (x=0–0.07) bulk metallic glasses, J. Non. Cryst. Solids. 365 (2013) 42–46. https://doi.org/10.1016/j.jnoncrysol.2013.01.033. [24] J. Torrens-Serra, P. Bruna, M. Stoica, J. Eckert, Glass-forming ability and microstructural evolution of [(Fe0.6Co0.4)0.75Si0.05B0.20]96-xNb4Mxmetallic glasses studied by Mössbauer spectroscopy, J. Alloys Compd. 704 (2017) 748–759. https://doi.org/10.1016/j.jallcom.2017.02.098. [25] S. Leila Panahi, P. Ramasamy, F. Masdeu, M. Stoica, J. Torrens-Serra, P. Bruna, S.L. Panahi, P. Ramasamy, F. Masdeu, M. Stoica, J. Torrens-Serra, P. Bruna, Evaluation of the effect of minor additions in the crystallization path of [(Fe0.5Co0.5)0.75B0.2Si0.05]100-xMx metallic glasses by means of Mössbauer spectroscopy, Metals (Basel). 11 (2021) 1–13. https://doi.org/10.3390/met11081293. [26] T. Bitoh, D. Watanabe, Effect of yttrium addition on glass-forming ability and magnetic properties of Fe–Co–B–Si–Nb bulk metallic glass, Metals (Basel). 5 (2015) 1127–1135. https://doi.org/10.3390/met5031127. [27] P. Rezaei-Shahreza, S. Hasani, A. Seifoddini, M. Nabiałek, P. Czaja, The crystallization process in a new multicomponent Fe-based bulk amorphous alloy: A kinetic study approach, Mater. Charact. 196 (2023). https://doi.org/10.1016/j.matchar.2022.112602. [28] J. Torrens-Serra, P. Bruna, M. Stoica, S. Roth, J. Eckert, Glass forming ability, thermal stability, crystallization and magnetic properties of [(Fe,Co,Ni)0.75Si0.05B0.20]95Nb4Zr1 metallic glasses, J. Non. Cryst. Solids. 367 (2013) 30–36. https://doi.org/10.1016/j.jnoncrysol.2013.02.019. [29] A. Hirata, P. Guan, T. Fujita, Y. Hirotsu, A. Inoue, A.R. Yavari, T. Sakurai, M. Chen, Direct observation of local atomic order in a metallic glass, Nat. Mater. 10 (2011) 28–33. https://doi.org/10.1038/nmat2897. [30] R. Gemma, M. to Baben, A. Pundt, V. Kapaklis, B. Hjörvarsson, The impact of nanoscale compositional variation on the properties of amorphous alloys, Sci. Rep. 10 (2020) 11410. https://doi.org/10.1038/s41598-020-67495-4. [31] R. Brand, NORMOS-98 Mössbauer Fitting Program Package, (1998). [32] G. Herzer, Grain structure and magnetism of nanocrystalline ferromagnets, IEEE Trans. Magn. 25 (1989) 3327–3329. https://doi.org/10.1109/20.42292. [33] A.F. Kozmidis-Petrović, Which glass stability criterion is the best?, Thermochim.