Advertisement

Metallurgical and Materials Transactions A

, Volume 46, Issue 6, pp 2415–2421 | Cite as

Crystallization Kinetics of Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1 at. pct) Bulk Amorphous Alloy

  • Hyo Yun Jung
  • Mihai StoicaEmail author
  • Seonghoon Yi
  • Do Hyang Kim
  • Jürgen Eckert
Symposium: Bulk Metallic Glasses XI

Abstract

The influence of Cu on crystallization kinetics of Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1 at. pct) bulk amorphous alloys was investigated by isothermal and isochronal differential scanning calorimetry combined with X-ray diffraction. The thermal analysis revealed that the crystallization of the amorphous matrix proceeds through at least two exothermic events. The Cu-free glassy alloy forms by primary crystallization the metastable Fe23C6 phase, while upon 0.5 at. pct Cu addition the primary crystallized phase is α-Fe. The activation energy for crystallization, calculated using both Kissinger and Ozawa methods, decreases from about 500 kJ/mol to about 330 kJ/mol. Further increase of Cu addition to 1 at. pct promotes the concomitant crystallization of several phases, as α-Fe, FeB, Fe3C, and Fe2P. In order to understand the crystallization behavior of the alloys as a function of Cu content, the Avrami exponent n, evaluated from the Johnson–Mehl–Avrami equation, was in details analyzed. The current study reveals that the minor Cu addition plays a crucial role at the initial stage of the crystallization. Among the studied alloys, the glassy samples with 0.5 at. pct Cu addition have the optimum compositional condition for the single α-Fe formation with a high nucleation rate.

Keywords

Amorphous Alloy Crystallization Kinetic Bulk Metallic Glass Glassy Alloy Nanocrystalline Alloy 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

In the last decade, the interests for the synthesis of soft magnetic nanocrystalline alloys, obtained upon controlled (nano) crystallization of an amorphous precursor with a relatively high glass-forming ability, have tremendously increased.[1, 2, 3, 4, 5] The soft magnetic nanocrystalline alloys usually show enhanced properties in regard with their amorphous counterparts, as for example higher magnetic saturation and higher permeability.[6,7] However, in order to achieve such properties, the first crystallized phase should have a higher saturation magnetization compared to that of the amorphous precursor, and the grains must be small enough to average out the magnetocrystalline anisotropy.[6,7] Some special alloys have been already extensively studied and commercialized under trade names as FINEMET,[6] VITROPERM,[8] or NANOPERM.[9] More recently, new nanocrystalline alloys, which combine high permeability of 25000 with low coercivity of 7 A/m and high saturation magnetic flux density of above 1.80 T, have been developed in the Fe-Cu-Si-B[10] and Fe-Si-B-P-Cu systems.[11] These nanocrystalline alloys require extremely high cooling rates (i.e., above 105 K/s) for production.[12,13] As a consequence, there is an inherent limitation of product dimensions-only ribbons or foils can be produced, as well as limitation on fabrication processes, i.e., the melt spinning or planar flow casting.

Further, tremendous efforts have been devoted to synthesize bulk-shaped nanocrystalline alloys. A worthy method to manufacture bulk components is the consolidation of amorphous powders,[14, 15, 16] powders which can be obtained for example upon ball-milling or gas-atomization. Moreover, with the progress of scientific and technological understanding of the glass-forming ability (GFA) of the metallic glasses, there have been developed several alloy systems having a GFA large enough to assure the preparation of bulk samples directly upon casting.[17, 18, 19, 20] Thus, the further development of nanocrystalline alloys using bulk metallic glass (BMG) precursors is expected to provide extension of practical application of these materials.

It is well known that the formation of nano-sized α-Fe phase from amorphous matrix may reduce the magnetic anisotropy of some Fe-based glassy alloys, which results in the dramatic enhancement of magnetic softness of the alloy.[7] Among various Fe-based BMGs, in the case of Fe-C-Si-B-P-Cu, the nanocrystallization of α-Fe phase has been reported.[21] The further development of nanocrystalline alloys based on this alloy system can be industrially attractive because of a relatively large GFA (critical thickness of about 2 mm), even when industrial raw materials as commercial cast iron and ferroalloys are used for making the master alloy.[21] In the case of mass production, nanocrystalline alloys based on this alloy system can provide improved material competitiveness stemming from the low cost of raw materials. However, although understanding of crystallization behavior of the amorphous alloy can provide useful insight for the effective process of the nanocrystallization, there has been little investigation up to now regarding the crystallization kinetics of these Fe-C-Si-B-P-Cu BMGs. In several nanocrystalline alloys, obtained from their amorphous precursors, the dimensions of α-Fe nano-sized grains are controlled through the presence of Nb atoms.[6,7] In the current studied alloy system, this is not possible because of the high C content, which certainly would trigger the formation of NbC and, therefore, decrease the GFA of the amorphous precursor. Hence, in the current work the crystallization behavior and crystallization kinetics of Fe76.5−x C6.0Si3.3B5.5P8.7Cu x BMGs were investigated. Taking in account that a control of compositional variation of the Cu content in steps of 0.1 at. pct would not be industrially viable, we considered only three compositions, i.e., x = 0, 0.5, and 1 at. pct. The variation of Cu content provokes dramatic changes of the crystallization product(s) as well as of the mechanism and it is revealed that 0.5 at. pct Cu addition plays a crucial role at the initial stage of the crystallization, being the major reason of the changes induced onto crystallization kinetics.

2 Materials and Methods

Master alloys with nominal compositions of Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) were prepared by arc melting in a Ti-gettered argon atmosphere. For that, mixture of cast iron (CI), industrial Fe-P alloy, and pure elements of Fe (99.7 pct purity), Si (99.99 pct purity), B (99.5 pct purity), and Cu (99.98 pct, purity) were used as raw materials. The actual compositions of CI and Fe-P alloy are listed in Table I. Cylindrical specimens having a diameter of 1 mm and length of 55 mm were prepared by suction casting. Structural analysis was carried out by X-ray diffraction (XRD) using Co radiation (Co Kα, λ = 0.179 nm). Thermal analysis was performed by differential scanning calorimetry (DSC, Perkin Elmer Instruments, PYRIS Diamond) and differential thermal analysis (DTA, Sinco, S-1600). Thermal stability and characterization temperatures of the amorphous rods including the glass transition temperature (T g), the onset temperature of crystallization (T x), and the peak temperature of the crystallization (T p) were measured using DSC under an Ar atmosphere at heating rates of 20, 30, and 40 K/min.
Table I

Chemical Composition (in Weight Percent) of Cast Iron (CI) and Ferro-phosphorus Alloy

 

Fe

C

Si

P

S

Ni

Ti

Cu

Cast iron

95.060

4.280

0.460

0.200

0.034

0.094

0.029

Fe-P

73.161

0.015

0.009

26.83

0.012

0.350

0.004

3 Results and Discussion

Figure 1 shows the XRD patterns of as-cast Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5 and 1.0 at. pct) rods with 1 mm diameter. The XRD patterns consist of broad peaks only, without any distinct sharp Bragg peak, indicating the formation of amorphous phase in the present alloys. As marked in Figure 1, it is interesting to note that the max positions of the broad maxima are shifted to the lower angles with increasing Cu contents. In order to minimize the possible errors, the curves from Figure 1 were fitted using a pseudo-Voigt function. The fitting errors are ± 0.01 deg. The Cu-free glassy samples have the main halo centered at max = 52.79 deg. The addition of 0.5 at. pct Cu shifts the main halo to max = 51.99 deg, i.e., 0.8 deg lower than the Cu-free alloy. Further increase of the Cu content to 1 at. pct leads to a further shift of the main halo, but much smaller, to max = 51.91 deg, which is only 0.07 deg lower that the value measured for 0.5 at. pct Cu. According to Ehrenfest relation,[22] the first maximum of the halo patterns is related to the largest interatomic distance (r max) between the near-neighbor atoms. In the case of transition-metal (TM)-metalloid (M) amorphous alloys, it is known that the r max evaluated through the analyze of max values can be considered as the average TM–TM nearest neighbor distance because the scattering intensity of M atoms is significantly weaker as compared to that of heavier TM atoms. Thus, one can consider that the increase of Cu contents in Fe76.5−x C6.0Si3.3B5.5P8.7Cu x BMGs results in an increase of average TM–TM nearest neighbor distance, i.e., the average distance between Fe and Cu atoms. Because Cu has a positive heat of mixing with Fe (ΔH mix = 13 kJ/mol[23]), it is obvious that the increase of Cu content leads to the enhancement of the repulsive forces between Fe and Cu atoms in amorphous structure, which has been believed as the origin of the formation of Cu clusters in amorphous structure.[6]
Fig. 1

XRD patterns of as-cast Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) rods having a diameter of 1 mm

Figure 2 shows the DSC heating curves of the as-cast Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) rods with 1 mm diameter measured at a constant heating rate of 20 K/min. The glass transition temperature (T g), the onset temperature (T x) of crystallization, the extension of the supercooled liquid region (ΔT x = T x − T g), and the first crystallization peak temperature (T p) of the alloys are listed in Table II. All examined alloys show a clear supercooled liquid region followed by few exothermic peaks, indicating a multi-stage crystallization. It is interesting that there are dramatic changes of crystallization behavior upon the variation of the Cu content. Although the base alloy has a two-stage crystallization, with a very sharp second exothermic event, the alloy with 0.5 at. pct Cu addition shows three-stage crystallization. In turn, the 1.0 at. pct Cu alloy has almost single-stage crystallization, a very weak secondary crystallization stage being, however, visible.
Fig. 2

Continuous DSC heating curves of as-cast Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) rods having a diameter of 1 mm

Table II

Glass Transition Temperature (T g), Onset Temperature of the Crystallization (T x), Supercooled Liquid Region (ΔT x = T x − T g), and first crystallization peak temperature (T p1) of the as-cast Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) rods with a diameter of 1 mm

x

T g [K (°C)]

T x [K (°C)]

ΔT x (K/°C)

T p1 [K (°C)]

0

753 (480)

780 (507)

26

787 (514)

0.5

746 (491)

765 (492)

21

777 (504)

1.0

751 (478)

784 (511)

33

791 (518)

In order to investigate the crystallization product(s) related to the first crystallization event of the glassy alloys, parts of as-cast samples were heated up to temperatures just below the first crystallization peak temperature [i.e., 2 K (2 °C) below T p1] of the alloys as marked in Figure 2. The heating was immediately followed by cooling (i.e., no isothermal holding) and the entire thermal analysis was performed at 20 K/min. The cooling was carefully monitored and the thermograms did not show any additional crystallization events. Figure 3 shows the XRD patterns of partially crystallized specimens. As can be seen, all of them consist of sharp Bragg reflections superimposed on the amorphous hallo, indicating a composite structure. As primary precipitated phase of the base alloy one can identify the Fe23C6 type phase, while the alloy with 0.5 at. pct Cu shows the formation of a bcc cell. The lattice constant, calculated upon taking in account the position of the (110) and (200) reflections, is a = 0.286 nm, which matches perfectly the a = 0.286 nm, value of the lattice constant of bcc Fe.[24] Therefore, the first crystalline phase, which appears upon primary precipitation from amorphous matrix, is α-Fe.
Fig. 3

XRD patterns of partially crystallized Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) rods

In contrast, the 1.0 at. pct Cu alloy shows a mixture of α-Fe(Si) and various Fe-metalloid compounds as Fe3C, Fe2P, and FeB. The activation energy for every thermal event is strongly related to the phase transformation of amorphous alloy. Thus, for better understanding of these dynamic changes of crystallization behavior as a function of Cu contents, the activation energies of the glass transition and crystallization were further investigated using the well-known Kissinger[25] and Ozawa[26] method. Because the activation energies evaluated by both Kissinger and Ozawa model are strongly related to the peak temperatures of the DSC curve at different heating rate,[27] the continuous DSC scans of the alloys were measured at three different heating rates: 20, 30, and 40 K/min. As expected, the characteristic temperatures of the alloys are shifted to higher temperature side with increasing heating rate (not shown here). The Kissinger and Ozawa plots of Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) alloy were obtained by plotting ln(β/T a 2 ) vs 1/T a and ln(β) vs 1/T a, respectively. Thus, the apparent activation energies can be calculated from the slope of the linear fitting of Kissinger and Ozawa plots. The average apparent activation energy of the glass transition (E a,g) and crystallization (E a,x) are listed in Table III. The results from both models show similar tendency with variation of Cu contents. It is interesting to observe that 0.5 at. pct Cu addition decreases the E a,g while the alloy with 1.0 at. pct Cu addition shows increase of the E a,g from about 250 kJ/mol to about 320 kJ/mol. On the other hand, the E a,x of the glassy alloys largely decreases from about 500 kJ/mol for Cu-free samples to about 350 kJ/mol for 0.5 at. pct Cu, and further only slightly to about 310 kJ/mol upon 1 at. pct Cu addition. Because the activation energy calculated here represents the energy barrier against crystallization, this result indicates the strong influence of Cu additions on the crystallization kinetics of the alloy and on the primary precipitated phases. By comparing the effect of 0.5 at. pct Cu addition with 1.0 at. pct Cu addition, there it is only a marginal decrease of E a,x with the higher Cu contents, indicating the relatively less impact of higher Cu addition than 0.5 at. pct Cu addition.
Table III

Apparent activation energies for glass transition (E a,g) and crystallization (E a,x) of the Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0 to 1 at. pct) alloys calculated by Kissinger[25] and Ozawa model[26]

 

E a,g [kJ/mol]

E a,x [kJ/mol]

Base alloy

 Kissinger

243.67 ± 12.47

454.27 ± 28.10

 Ozawa

256.30 ± 12.38

517.61 ± 6.74

Cu 0.5 pct alloy

 Kissinger

209.09 ± 33.17

339.46 ± 33.42

 Ozawa

221.48 ± 33.17

352.51 ± 33.25

Cu 1.0 pct alloy

 Kissinger

314.26 ± 30.76

310.04 ± 33.42

 Ozawa

326.82 ± 30.79

323.35 ± 33.42

To further investigate the changes of crystallization kinetics of the Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) glassy alloys, the thermal behavior during isothermal annealing was analyzed by Johnson–Mehl–Avrami (JMA) model.[28] Figure 4 shows the isothermal DSC curves of the base alloy, Cu 0.5 at. pct and Cu 1.0 at. pct alloys at T x—20 K and T x—30 K. Except the curve measured at T x—20 K for the Cu 0.5 at. pct alloy, all curves show a single exothermic peak after certain incubation time (τ). Comparing the curves of the alloys obtained at T x—30 K, τ decreases with increasing Cu contents. The base alloy has backward-shifted crystallization shape, while the Cu 0.5 at. pct and Cu 1.0 at. pct alloy have a forward-shifted bell shape. Assuming that the fractional area of the peak is proportional with the crystallized volume fraction,[29] the shape of these curves at the initial crystallization stage can reflect the initial nucleation rate of the alloys. Thus, comparing the shape of the initial part of the curves, it can be deduced that the Cu 0.5 at. pct and Cu 1.0 at. pct alloys have distinct tendency, showing higher initial crystallization rate than the base alloy. In isothermal condition, the time dependence of the crystallized volume fraction (X) is given by the JMA equation:
$$ X(t) = 1 - \exp \left[ { - k\left( {t - \tau } \right)} \right]^{n} , $$
(1)
where X(t) is the transformed volume fraction after time t, τ is the incubation time for crystallization, n is Avrami exponent, and k is the reaction rate constant related to the activation energy for the process, E a, given by
$$ k = k_{0} \exp \left( { - \frac{{E_{\text{a}} }}{RT}} \right), $$
(2)
where k o is a constant. The Avrami exponent n reflects the nucleation and growth mechanisms of the crystallization:
$$ \ln \left[ { - \ln \left( {1 - X} \right)} \right] = n\ln k + n\ln \left( {t - \tau } \right) $$
(3)
Figure 5 shows the JMA plots. To minimize the experimental errors existed in the early and late stage of experiments, due in principal by the fact that these Cu-added BMGs have very short incubation time (see Figures 4(b) and (c)), only the data for 20 pct < X < 80 pct were considered. Although the JMA plot of the base alloy does not have a single slope, the Cu 0.5 at. pct and Cu 1.0 at. pct alloys show almost a straight line. The average Avrami exponent n of Cu 0.5 at. pct and Cu 1.0 at. pct alloys are 2.17 and 1.94, respectively. In order to analyze the details of crystallization behavior of the alloy as the crystallization proceeds, the so-called local Avrami exponent proposed by Calka et al.[30] can be evaluated. The local Avrami exponents can be computed using the first derivative of the Avrami plot against the crystallization fraction. Thus, the local Avrami exponent n(X) is given by
$$ n(X) = \frac{{\partial \ln \left[ { - \ln \left( {1 - X} \right)} \right]}}{{\partial \ln \left( {t - \tau } \right)}} $$
(4)
The variation of the local Avrami exponent of the base alloy, Cu 0.5 at. pct and Cu 1.0 at. pct alloys at T x—30 K is given in Figure 6. In the case of the Cu 0.5 at. pct alloy, the local Avrami exponent continuously decreases from 2.5 to 2.0, while local Avrami exponent of the Cu 1.0 at. pct alloy remains relatively constant, varying only slightly between 1.8 and 2.0. On the other hand, in the case of the base alloy, at the early stage of the crystallization (X ≈ 20 pct), the n is about 1.5, and continuously increases up to 4.3 (X ≈ 80 pct). According to the studies of Ranganathan and Von Heimendahl[31], the Avrami exponent n can be interpreted as
$$ n = a + bc $$
(5)
where a is a parameter showing the nucleation rate, which can be 0 for a zero nucleation rate, 0 < a < 1 for a decreasing nucleation rate with time, 1 for a constant nucleation rate and a > 1 for an increasing nucleation rate. b is a parameter showing the dimension of the growth, b = 1, 2, and 3 for one-, two-, and three-dimensional growth, respectively, and c is a parameter showing the type of growth (c = 1 for interface-controlled growth and c = 0.5 for diffusion-controlled growth). In the cases of present alloys, there has been a previous report[21] for the three-dimensional dendritic growth of crystallites and continuous formations of several phases having different composition than the amorphous matrix, which indicates that the crystallization of the present alloys is strongly related to the diffusion of atoms. Thus, supposing the alloys have diffusion-controlled growth (c = 0.5), one can assume that the alloys with 0.5 and 1.0 at. pct Cu addition have a decreasing nucleation rate with time (a = 0.5) and a nucleation rate of zero (a = 0), respectively. On the other hand, as shown in Figure 6, the base alloy has continuously changing n as the crystallization proceeds, which indicates a three-dimensional diffusion-controlled growth with an increasing nucleation rate.[32]
Fig. 4

Isothermal DSC curves of Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) rods, each sample annealed at T x—20 K and T x—30 K

Fig. 5

JMA plots of amorphous Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) rods

Fig. 6

Local Avrami exponents n(X) of Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) rods as a function of the crystallized volume fraction

In the case of conventional soft magnetic nanocrystalline alloys, it is well known that the Cu addition is effective to lead the nanocrystallization of Fe-based amorphous alloy. In the case of FINEMET[6] alloy, Hono et al.[33] revealed by the three-dimensional atom probe (3DAP) study the formation of Cu clusters prior to the primary crystallization. The Cu clusters develop an fcc-type structure and provide heterogeneous nucleation sites for the primary crystallization of α-Fe. In present results, it is interesting to note that the α-Fe became a primary precipitated phase of the alloy with 0.5 at. pct Cu addition while the primary phase of the base alloy was the Fe23C6 phase. The Fe23C6 phase has been frequently reported in Fe-based BMGs having relatively high GFA.[34] However, minor additions of Cu may change the crystallization behavior of Fe-based BMG. For example, in (Fe, Co)-B-Si-Nb BMGs with Cu additions, Stoica et al.[35,36] reported the formation of α-Fe phase, while in the base alloy typically the complex Fe23B6 phase precipitates upon heating.[35] Ohkubo et al.[37] observed special orientation relationship between fcc-Cu and α-Fe primary phase in Fe-Zr-B-Cu amorphous alloy. They insist that the special orientation relationship reduces the activation energy for the nucleation of α-Fe phase. In the case of the current study, the effect of Cu addition on the activation energy for the crystallization was investigated. Considering the evaluated activation energy related to the primary crystallization of the base and Cu 0.5 at. pct alloys, it can be concluded that the activation energy for the formation of the α-Fe is lower than that of Fe23C6 phase. Since the activation energy for the crystallization is the energy barrier necessary to be overcome for nucleation, lower activation energy means easier nucleation. Thus, the formation of α-Fe could be more preferred than that of Fe23C6 phase. Moreover, comparing the crystal structure of the α-Fe and Fe23C6 phases, the formation of α-Fe is energetically more advantageous than that of Fe23C6 phase because of its simpler atomic structure. Regarding the kinetics of the phase formation, the Fe23C6 phase having giant lattice (a = 1.059 nm) and complex structure (space group Fm-3m[24]) inevitably requires longer range atomic re-arrangements than the α-Fe phase having the simpler bcc structure (a = 0.286 nm, space group Im-3m).

In FINEMET-type alloy, Ohnuma et al.[38] studied the kinetics of the Cu-clustering with different Cu contents. They compared the two types of FINEMET alloy, and revealed that the optimum Cu content of the original FINEMET alloy (Fe74.5−x Si13.5B9Nb3Cu x ) is 1 at. pct (x = 1) and that of the modified FINEMET alloy (Fe77Si11B9Nb3−x Cu x ) is 0.6 at. pct (x = 0.6). Moreover, they insisted that the higher Cu contents within same series of alloy lead the faster Cu clustering. In the case of current study, during continuous heating process, the Cu 1.0 at. pct alloy shows almost single crystallization stage having simultaneous formation of various compounds and α-Fe phase. This result indicates that there are strong competitions between several crystalline phases. In consideration of the results of JMA analysis showing that the Cu 1.0 at. pct alloy has a massive nucleation at the initial stage of crystallization, the strong crystallization tendency of the Cu 1.0 at. pct alloy for the simultaneous crystallization can be attributed to the enough nucleation sites, which are able to provide heterogeneous nucleation sites to various crystalline phases (α-Fe, Fe3C, and Fe2P). Therefore, when a number of heterogeneous nucleation sites are provided in the present alloy, the almost simultaneous formations of various phases can be much easier than in the case of FINEMET alloys having wide process window to form single precipitation of α-Fe type phase.[6] Thus, in the case of present alloy, the Cu 0.5 at. pct addition can be considered as an optimum Cu content for the nanocrystallization of single α-Fe phase.

4 Conclusions

We have studied the effect of Cu addition on crystallization kinetics of Fe76.5−x C6.0Si3.3B5.5P8.7Cu x (x = 0, 0.5, and 1.0 at. pct) bulk amorphous alloy. It was found that the crystallization process of the alloys strongly depends on the Cu contents, which is probably related to the amount of the nucleation sites supplied by Cu clusters. The activation energy for the primary crystallization was calculated using Kissinger and Ozawa model, which revealed that the formation of primary α-Fe phase requires lower activation energy than that of complex Fe23C6 phase. The addition of 1 at. pct Cu stimulates the almost simultaneous formation of various compounds together with the primary α-Fe. Thus, the present results revealed that the Cu 0.5 at. pct addition is the optimum content for the nanocrystallization of α-Fe with a high nucleation rate.

Notes

Acknowledgments

This work was supported by the Global Research Laboratory Program of the Korea Ministry of Science and Technology. The support of the German Science Foundation (DFG) through the Grants STO 873/2-1 and STO 873/2-2 is also acknowledged.

References

  1. 1.
    [1] T.-S. Chin, C.Y. Lin, M.C. Lee, R.T. Hunag, and S.M. Huang: Intermetallics, 2008, vol. 16, pp. 52-57.CrossRefGoogle Scholar
  2. 2.
    [2] A. Inoue, and B. Shen: Mater. Trans., 2002, vol. 43, pp. 766-769.CrossRefGoogle Scholar
  3. 3.
    [3] A. Chrobak, M. Karolus, and G. Haneczok: Solid State Phenomena, 2010, vol. 163, pp. 233-238.CrossRefGoogle Scholar
  4. 4.
    [4] Z. Long, Y. Shao, F. Xu, H. Wei, Z. Zhang, P. Zhang, and X. Su: Mater. Sci. Eng. B, 2009, vol. 164, pp. 1–5.CrossRefGoogle Scholar
  5. 5.
    [5] A. Makino, X. Li, K. Yubuta, C. Chang, T. Kubota, and A. Inoue: Scripta Mater., 2009, vol. 60, pp. 277–280.CrossRefGoogle Scholar
  6. 6.
    [6] Y. Yoshizawa, S. Oguma, and K. Yamauchi: J. Appl. Phys., 1988, vol. 64, pp. 6044-6046.CrossRefGoogle Scholar
  7. 7.
    [7] G. Herzer: IEEE Trans. Magn,. 1990, vol. 26, pp. 1397-1402.CrossRefGoogle Scholar
  8. 8.
    [8] J. Petzold: J. Magn. Magn. Mater., 2002, vol. 242–245, pp. 84-89.CrossRefGoogle Scholar
  9. 9.
    [9] K. Suzuki, A. Makino, N. Kataoka, A. Inoue, and T. Masumoto: Mater. Trans. JIM, 1991, vol. 32, pp. 93-102.CrossRefGoogle Scholar
  10. 10.
    [10] M. Ohta and Y. Yoshizawa: Jpn. J. Appl. Phys., 2007, vol. 46, pp. L477-L479.CrossRefGoogle Scholar
  11. 11.
    [11] A. Makino, H. Men, T. Kubota, K. Yubuta, and A. Inoue: Mater Trans., 2009, vol. 50, pp. 204-209.CrossRefGoogle Scholar
  12. 12.
    [12] T. Kulik: J. Non-Cryst. Solids, 2001, vol. 287, pp. 145-161.CrossRefGoogle Scholar
  13. 13.
    [13] K. Suzuki, A. Makino, A. Inoue, and T. Masumoto: J. Appl. Phys., 1993, vol. 74, pp. 3316-3322.CrossRefGoogle Scholar
  14. 14.
    [14] Y. Kawamura and M. Takagi: Mater. Sci. Eng., 1988, vol. 98, pp. 415-418.CrossRefGoogle Scholar
  15. 15.
    [15] A. Kojima, H. Horikiri, Y. Kawamura, A. Makino, A. Inoue, and T. Masumoto: J. Magn. Magn. Mater., 1996, vol. 162, pp. 95-102.CrossRefGoogle Scholar
  16. 16.
    [16] W.L. Johnson: J. Mater., 2002, vol. 54, pp. 40-43.Google Scholar
  17. 17.
    [17] W.H. Wang, C. Dong, and C.H. Shek: Mater. Sci. Eng. R, 2004, vol. 44, pp. 45-89.CrossRefGoogle Scholar
  18. 18.
    [18] A. Inoue, and A. Takeuchi: Mater. Sci. Eng. A, 2004, vol. 375–377, pp. 16-30.CrossRefGoogle Scholar
  19. 19.
    [19] Y. Zhang, Y. Li, H. Tan, and G.L. Chen: Intermetallics, 2007, vol. 15, pp. 744-748.CrossRefGoogle Scholar
  20. 20.
    Z.P. Lu and C.T. Liu: Phys. Rev. Lett., 2003, vol. 91, pp. 115505(1)–5(4).Google Scholar
  21. 21.
    [21] H.Y. Jung and S. Yi: J. Alloys Compd., 2013, vol. 561, pp. 76-81.CrossRefGoogle Scholar
  22. 22.
    P. Ehrenfest: Proc. Sec. Sci. Kon. Akad. Weten. Amsterdam, 1915, vol. 17, pp. 1184-90.Google Scholar
  23. 23.
    [23] A Takeuchi and A Inoue: Mater Trans., 2005, vol. 46, pp. 2817-2829.CrossRefGoogle Scholar
  24. 24.
    W.B. Pearson: A Handbook of Lattice Spacings and Structures of Metals and Alloys, 6th ed., vol. 1, Pergamon Press, Oxford, 1964, p. 625.Google Scholar
  25. 25.
    [25] H.E. Kissinger: J. Res. Nat. Bur. Stand., 1956, vol. 57, pp. 217-220.CrossRefGoogle Scholar
  26. 26.
    [26] T. Ozawa: Bull. Chem. Soc. Japan, 1965, vol. 35, pp. 1881-1882.CrossRefGoogle Scholar
  27. 27.
    [27] K. Matusita, T. Komatsu, and R. Yokota: J. Mater. Sci., 1984, vol. 19, pp. 291–296.CrossRefGoogle Scholar
  28. 28.
    [28] W. A. Johnson and R. F. Mehl: Trans. Am. Inst. Mining Met. Engrs., 1939, vol. 135, pp. 416-443.Google Scholar
  29. 29.
    [29] M. G. Scott: J. Mater. Sci., 1978, vol. 13, pp. 291-296.CrossRefGoogle Scholar
  30. 30.
    [30] A. Calka and A. P. Radinski: J. Mater. Res., 1985, vol. 3, pp. 59-63.CrossRefGoogle Scholar
  31. 31.
    [31] S. Ranganathan and M. Von Heimendahl: J. Mater. Sci., 1981, vol. 16, pp. 2401-2404.CrossRefGoogle Scholar
  32. 32.
    [32] Y. Ouyang, J. Zhang, H. Chen, S. Liao, and Z. Zhong: J. Alloys Compd., 2008, vol. 454, pp. 359-363.CrossRefGoogle Scholar
  33. 33.
    [33] K. Hono, D.H. Ping, M. Ohnuma, and H. Onodera: Acta Mater., 1999, vol. 47, pp. 997-1006.CrossRefGoogle Scholar
  34. 34.
    A. Hirata, Y. Hirotsu, K. Amiya, and A. Inoue: Phys. Rev. B, 2008, vol. 78, pp. 144205(1)-144205(6).Google Scholar
  35. 35.
    [35] M. Stoica, R. Li, A.R.Yavari, G. Vaughan, J. Eckert: N. van Steenberge, and D. Ruiz Romera: J. Alloys Compd., 2010, vol. 504S, pp. S123-128.CrossRefGoogle Scholar
  36. 36.
    [36] M. Stoica, R. Li, S. Roth, J. Eckert, G. Vaughan, and A.R. Yavari: Metall. Mater. Trans., 2011, vol. 42A, pp. 1476-1480.CrossRefGoogle Scholar
  37. 37.
    [37] T. Ohkubo, H. Kai, D.H. Ping, K. Hono, and Y. Hirotsu: Scripta Mater., 2001, vol. 44, pp. 971- 976.CrossRefGoogle Scholar
  38. 38.
    [38] M. Ohnuma, K. Hono, S. Linderoth, J. S. Pedersen, Y. Yoshizawa, and H. Onodera: Acta Mater., 2000, vol. 48, pp. 4783-4790.CrossRefGoogle Scholar

Copyright information

© The Minerals, Metals & Materials Society and ASM International 2014

Authors and Affiliations

  • Hyo Yun Jung
    • 1
  • Mihai Stoica
    • 1
    • 2
    Email author
  • Seonghoon Yi
    • 3
  • Do Hyang Kim
    • 4
  • Jürgen Eckert
    • 1
    • 5
  1. 1.IFW DresdenInstitute for Complex MaterialsDresdenGermany
  2. 2.POLITEHNICA University of TimisoaraTimisoaraRomania
  3. 3.Department of Materials Science and Metallurgical EngineeringKyungpook National UniversityDaeguSouth Korea
  4. 4.Center for Non-crystalline Materials, Department of Metallurgical EngineeringYonsei UniversitySeoulSouth Korea
  5. 5.Institute of Materials ScienceUniversity of Technology DresdenDresdenGermany

Personalised recommendations