MechanicalamorphizationofFe Nb B powder:Microstructuralandmagneticcharacterization Intermetallics

Mechanical amorphization of Fe ₇₅ Nb ₁₀ B ₁₅ powder: Microstructural and magnetic characterization

J.J. Ipus

, J.S. Bla´zquez

, V. Franco

, A. Conde

, L.F. Kiss

aDpto. Fı´sica de la Materia Condensada, ICMSE-CSIC, Universidad de Sevilla, P.O. Box 1065, 41080 Sevilla, Spain

bResearch Institute for Solid State Physics and Optics, Hungarian Academy of Sciences, P.O. Box 49, 1525 Budapest, Hungary

a r t i c l e i n f o

Article history:

Received 13 August 2009 Received in revised form 30 September 2009 Accepted 5 October 2009 Available online 29 October 2009

Keywords:

B. Magnetic properties

C. Mechanical alloying and milling

a b s t r a c t

The evolution of the amorphous fraction developed during the mechanical alloying of a mixture of pure 75 at.% Fe, 10 at.% Nb and 15 at.% B,X^Am, has been followed by different techniques: X-ray diffraction, Mo¨ssbauer spectroscopy and magnetic permeability measurements; in order to compare their sensitivity in the detection of small fractions. The values obtained for the amorphous fraction from the three techniques show a roughly linear correlation abovew20%. The most sensitive technique was Mo¨ssbauer spectroscopy (X^Amobtained from the low hyperfine field contributions) and the less sensitive technique was X-ray diffraction. The Curie temperature of the amorphous phase increases with the milling time due to a slow and progressive incorporation of boron into this phase and the nanocrystals from the boron inclusions.

Ó2009 Elsevier Ltd. All rights reserved.

1. Introduction

Amorphous and nanocrystalline materials are very interesting due to their physical properties (mechanical and magnetic)[1,2]

and their technological applicability[1,3]. Ball milling has become a very suitable and versatile technique to directly produce meta- stable microstructures (amorphous, nanocrystalline, supersatu- rated solid solutions, etc.) in a wide compositional range[1]from elemental powders or alloys. The study of microstructure evolution is commonly followed by several techniques as X-ray diffraction (XRD), transmission electron microscopy (TEM) and Mo¨ssbauer spectroscopy, characterizing phase evolution, structural parame- ters and composition.

The Nanoperm type alloys[4]have received much attention in the last years due to their excellent soft magnetic properties, leading to an extensive use in many commercial applications[2]. In fact, the microstructure developed in this type of alloys, (ferro- magnetic bcc Fe type nanocrystals embedded in an amorphous matrix which is also ferromagnetic) is made responsible for the soft magnetic properties. Therefore, the microstructural characteriza- tion is very important in order to understand the behaviour of the system and to predict its possible technological capabilities.

Previous studies [5] show that the power released during milling scales with the cube of the frequency of the main disk,U, in

a planetary mill. This law has been tested for the microstructural and magnetic properties evolution of Fe–Ge–Nb alloys[5]and for magnetization, Curie temperature and magnetocaloric effect of Fe–Nb–B alloys from vibrating sample magnetometry[6].

In this work, the sensitivity in the detection of small amount of amorphous phase developed during the mechanical amorphization of Fe₇₅Nb₁₀B₁₅ powders is studied using different experimental techniques: X-ray diffraction (XRD), Mo¨ssbauer spectrometry and magnetic permeability. Results on samples prepared at two milling frequencies,U¼150 rpm[7]andU¼350 rpm, were used in order to test theU³law.

2. Experimental

Fe75Nb10B15 alloy was prepared by mechanical alloying from pure powders in a planetary ball mill Fritsch Pulverisette 4 Vario.

Manipulation of powder was done under argon atmosphere in a Saffron glove box. The milling process was performed using two frequencies,U¼150[7]and 350 rpm, in order to study the appli- cability of the predicted U³ law for the power released during milling[5]. In this frame, results will be presented as a function of an equivalent time defined as teq ¼ ðU=U₀Þ³t, referenced to the milling performed at U₀¼150 rpm. More details of milling parameters can be found elsewhere[7]. The contained phases and the microstructural evolution were studied with XRD, using Cu Ka radiation, and Mo¨ssbauer spectroscopy. Mo¨ssbauer spectra were recorded at room temperature in a transmission geometry using a ⁵⁷Co(Rh) source. Values of the hyperfine parameters were

*Corresponding author. Tel.:þ34 954559541; fax:þ34 954612097.

E-mail address:jsebas@us.es(J.S. Bla´zquez).

Contents lists available atScienceDirect

Intermetallics

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / i n t e r m e t

0966-9795/$ – see front matterÓ2009 Elsevier Ltd. All rights reserved.

doi:10.1016/j.intermet.2009.10.007

Intermetallics 18 (2010) 565–568

obtained by fitting with NORMOS program[8]. Isomer shift was quoted relative to that ofa-Fe at room temperature.

Permeability measurements were performed using the induc- tion method[9], in which the sample was placed between two coils to determine the relative permeability (since demagnetization factor is unknown). The measurements were done in a temperature range from 77 K to 300 K. The applied magnetic field (w1 A/m) was low enough to assure initial permeability measurements and the frequency was 6 kHz.

3. Results and discussion

The microstructural evolution of milled samples was followed at room temperature by XRD and Mo¨ssbauer spectroscopy and some examples are shown inFigs. 1 and 2, respectively. This evolution can be rescaled using the equivalent milling time defined above derived from theU³law of the power released during milling. At intermediate equivalent milling times, the formation of a super- saturated bcc Fe(Nb,B) solid solution was observed, whose hyper- fine field distribution consists of a huge fraction of Fe atoms with high hyperfine magnetic fields representing pure bcc and interface environments. Moreover, at these equivalent milling times a refinement of crystal size down to w5 nm and an increase of microstrain up tow2% are also observed, in agreement with the results observed after milling at a single frequencyU¼150 rpm[7].

For larger equivalent milling times, besides the supersaturated solid solution, an amorphous phase was also developed, without significant changes in the microstructure of the solid solution[7].

From hyperfine field distributions, an important increase in low

field contributions with the milling time was observed, even appearing a zero field contribution (doublet in the Mo¨ssbauer spectra), which becomes the main contribution at equivalent long milling times. This microstructure agrees with transmission elec- tron microscopy results[10], observing a refinement of crystal size with the milling time. Moreover, bright crystalline B inclusions larger than the bcc nanocrystals were detected in the studied samples after the longest milling time explored; hence, at least some B remains undissolved in the Fe–Nb matrix after long milling times[10].

The amorphous fraction evolution with the equivalent milling time can be followed from XRD patterns, where this fraction,X^XRD, is obtained as the area of the broad Gaussian contribution (amor- phous halo) divided by the whole area of the experimental diffraction peak composed by this amorphous contribution and a narrower Lorentzian contribution ascribed to the bcc nanocrystals.

The fitting of Mo¨ssbauer spectra of such complex systems as the one studied here prevents a clear distinction between ferromag- netic contributions with low HF and paramagnetic ones (HF¼0 T).

Therefore, and after previous VSM results[6], contributions with these low hyperfine field values (doublet in the Mo¨ssbauer spectra and low field distribution, withCHFDw2.5 T) were considered as the fraction of Fe atoms in paramagnetic sites and identified with the amorphous phase fraction,X^MS.

Fig. 3shows the correlation between the values of amorphous fraction calculated from XRD and from Mo¨ssbauer spectra. For short milling times, although, by XRD, the amorphous phase is not detected in the diffraction patterns, an increase of low field contributions in Mo¨ssbauer spectra is observed. For larger milling times, the correlation between both XRD and Mo¨ssbauer measurements is good and follows the linear trendX^XRD¼fX^MS withfw1. In fact, each technique does not detect exactly the same

40 50 60

Fe

Fe

intensity [a.u.]

Nb

Nb 1 h

2θ [degrees]

25 h 50 h 127 h 300 h 508 h

Fig. 1.XRD patterns of powder samples after different equivalent milling times.

-8 -6 -4 -2 0 2 4 6 8 0 10 20 30

1

velocity [mm/s] HF [T]

1 h

0.12 25 h

50 h 0.12

relative transmission

0.04

probability

300 h

127 h

0.15 0.08 508 h

Fig. 2.Mo¨ssbauer spectra and hyperfine field distributions of powder samples after different equivalent milling times.

J.J. Ipus et al. / Intermetallics 18 (2010) 565–568 566

observable, XRD measures the amount of phase existing in the sample, while Mo¨ssbauer spectroscopy measures the amount of Fe atoms in a common environment (phase), being necessary to correct the diffraction areas with the scattering power of each phase for XRD and to take into account the composition of each phase to correct the fraction of Fe atoms for Mo¨ssbauer spectros- copy. Nevertheless, taken into account the actual results, the observed value offw1 could suggest a uniform distribution of Fe atoms in both (amorphous and crystalline) phases. The maximum amorphous fraction achieved was 85%. It is worth mentioning that some works in the literature show that the amorphous fraction could not reach 1 due to recrystallization phenomena[1,11].

Fig. 4shows the relative permeability,m, measured from liquid nitrogen temperature to room temperate for samples after different milling times. It can be observed that, as temperature increases, mdecreases due to the ferro-paramagnetic transition of the amor- phous phase and, as milling time increases, this decrease becomes more evident, ascribed to an increase of this amorphous phase.

From the permeability curves it is also possible to estimate the amorphous fraction as the ratio between twice the decrease of

mfrom 77 K,m(77 K), to the inflexion point,m(ip), and the perme- ability value at 77 K,

X^m ¼ 2ð

m

m

m

as illustrated in the scheme shown inFig. 5.

In Fig. 6 these values of amorphous fraction are shown as a function of those obtained from Mo¨ssbauer spectra. Similarly to the XRD case, for short milling times, permeability measurements cannot detect evolution in the amorphous fraction. However, the deviation fromX^m¼0 is found at a smaller amount of amorphous fraction than for XRD (X^MS¼0.17 and 0.3, respectively). For larger milling times, a roughly linear trend between these two parameters is observed (the obtained slope is 0.82). This result would suggest that the magnetic response of Fe atoms in the amorphous phase is lower than that corresponding to Fe atoms in the crystalline phase.

Comparing the different values obtained for the evolution of amorphous phase from the different techniques used, it can be suggested that Mo¨ssbauer spectroscopy is the most sensitive technique in order to detect small amounts of non-pure crystalline phases XRD being the less sensitive technique among the three studied here.

0,0 0,2 0,4 0,6 0,8 1,0

0,0 0,2 0,4 0,6 0,8

1,0 amorphous fraction

X XRD

X ^MS

150 rpm 350 rpm

Fig. 3.Amorphous fraction from XRD patterns as a function of the amorphous fraction from Mo¨ssbauer spectra. The solid symbols correspond to samples milled at 150 rpm and hollow symbols correspond to samples milled at 350 rpm.

80 120 160 200 240 280 320

40 60 80 100

508 h 381 h 254 h 127 h 76 h 50 h 32 h

μ [a.u.]

T [K]

Fig. 4.Magnetic permeability as a function of temperature for samples after different equivalent milling times.

phase B phase A

μ=0 μ(ip) μ(77 K)

T(ip) 77 K

Fig. 5.Schematic representation ofX^mevaluation in a material with two magnetic phases.

0.0 0.2 0.4 0.6 0.8 1.0

0.0 0.2 0.4 0.6 0.8 1.0

150 rpm 350 rpm amorphous fraction

Xμ

X^MS

Fig. 6.Amorphous fraction from permeability measurements as a function of amor- phous fraction from Mo¨ssbauer spectra. The solid symbols correspond to samples milled at 150 rpm and hollow symbols correspond to samples milled at 350 rpm.

J.J. Ipus et al. / Intermetallics 18 (2010) 565–568 567

Permeability curves also allow us to determine the Curie temperature of the amorphous phase as the temperature at the inflexion point. The Curie temperature of the amorphous phase increases as the amount of this phase increases as is presented in Fig. 7, indicating a compositional change in this phase. This evolution of Curie temperature is compared with values from VSM measurements [6] observing a good quantitative agreement between both measurements. It is worth mentioning that perme- ability is a more sensitive technique in order to measure Curie temperatures of phases with low fraction in the sample.

The increase of Curie temperature as the milling (and amorph- ization) progresses can be explained by an increase of B content in the amorphous phase with the milling time due to the progressive but slow degradation of B inclusions. In fact, an enrichment of 3 at.%

B could explain the enhancement of Curie temperature for amor- phous fractions larger than 0.3 (seeFig. 7)[12,13]. These values of Curie temperature are lower than those of completely amorphous samples with similar composition [14]. The presence of small amounts of Cr due to contamination from the milling media would also reduce the Curie temperature of the amorphous phase in comparison with Cr free compositions but in the studied case it does not affect the dependence with milling time.

4. Conclusions

The main conclusions derived from this study were as follows.

Mo¨ssbauer spectroscopy is more sensitive to detect small amount of phases compared with permeability measurements and XRD. The latter technique is shown to be the less sensitive one.

The Curie temperatures of amorphous phase increase as the amorphous fraction increases, which could be ascribed to a progressive enrichment of B in the amorphous phase with the milling time. Permeability measurements allow us to measure the Curie temperature of the amorphous phase even for amorphous fraction below 0.1.

The cubic law of the power released during milling with the frequency is fulfilled for both parameters studied in this work:

amorphous fraction and Curie temperature.

Acknowledgements

This work was supported by the Ministry of Science and Innovation (MICINN) and EU FEDER (project no. MAT2007-65227), the PAI of the Regional Government of Andalucı´a (project no. P06- FQM-01823) and the Hungarian Scientific Research Fund (grant no. OTKA-NKTH 68612). J.J.I. acknowledges a fellowship from MICINN. J.S.B. acknowledges a research contract from Junta de Andalucı´a.

References

[1] Suryanarayana C. Prog Mater Sci 2001;46:1–184.

[2] McHenry ME, Willard MA, Laughlin DE. Prog Mater Sci 1999;44:291–433.

[3] Gleiter H. Acta Mater 2000;48:1–29.

[4] Suzuki K, Kataoka N, Inoue A, Makino A, Matsumoto T. Mater Trans JIM 1990;31:743–6.

[5] Ipus JJ, Bla´zquez JS, Franco V, Milla´n M, Conde A, Oleszack D, et al. Interme- tallics 2008;16:470–8.

[6] Ipus JJ, Bla´zquez JS, Franco V, Conde A, Kiss LF. J Appl Phys 2009;105:

123922.1–6.

[7] Ipus JJ, Bla´zquez JS, Franco V, Conde A. Intermetallics 2008;16:1073–82.

[8] Brand RA, Lauer J, Herlach MD. J Phys F Met Phys 1983;13:675–83.

[9] Zijlstra H. Experimental methods in magnetism. V2. Amsterdam: North- Holland; 1967. p.133.

[10] Ipus JJ, Bla´zquez JS, Lozano-Pe´rez S, Conde A. Philos Mag 2009;89:1415–23.

[11] Sharma S, Suryanarayana C. J Appl Phys 2007;102:083544.1–7.

[12] Wijn HPJ. Landolt-Bo¨rnstein: magnetische eigenschaften von metallen, vol. 19.

Berlin: Springer; 1991. p. 92–4.

[13] Chien CL, Unruh KM. Phys Rev 1981;B24:1556–8.

[14] Wijn HPJ. Landolt-Bo¨rnstein: Magnetische Eigenschaften von Metallen, vol. 19.

Berlin: Springer; 1991. p. 103.

0.0 0.2 0.4 0.6 0.8 1.0

120 160 200 240

permeability measurements 150 r.p.m.

350 r.p.m.

VSM measurements 150 r.p.m.

350 r.p.m.

TC [K]

X ^MS

Fig. 7.Curie temperature calculated from permeability (square) and VSM (circles) measurements as a function of the fraction of amorphous phase from Mo¨ssbauer spectra. The solid symbols correspond to samples milled at 150 rpm and hollow symbols correspond to samples milled at 350 rpm.

J.J. Ipus et al. / Intermetallics 18 (2010) 565–568 568