LETTER TO THE EDITOR

Journal of Magnetism and Magnetic Materials 257 (2003) L139–L145

Letter to the Editor

An original route for the preparation of hard FePt N.H. Haia,b, N.M. Dempseya, M. Veronc, M. Verdierc, D. Givorda,* b

a Laboratoire Louis N!eel, 25 Avenue des Martyrs, BP 166, 38042 Grenoble cedex 9, France Cryolab, Faculty of Physics, Vietnam National University, Hanoi, 334, Nguyen Trai, Thanh Xuan, Hanoi, Viet Nam c LTPCM-INPG, Domaine Universitaire, BP 75, 38402 St Martin d’H"eres, France

Received 16 September 2002

Abstract The preparation of FePt hard magnetic foils by an original procedure is described in this paper. The process associates cyclic co-deformation of Fe and Pt foils down to the nanometer scale (total thickness of multilayer E100 mm) followed by heat treatment in the temperature range 450–5501C. The formation of the high-anisotropy L10 FePt phase results from controlled diffusion and an ordering phase transformation. Coercivities as high as 0.9 T were measured in a VSM at room temperature following annealing at 4501C for 48 h. The coercivity of this sample was decreased by half when measured at 600 K while its energy product decreased from 100 kJ/m3 at 300 K to 25 kJ/m3 at 600 K. r 2002 Elsevier Science B.V. All rights reserved. PACS: 75.50.Ww; 81.05.Ys; 81.40.Ef Keywords: FePt magnets; Cold rolling; Nanostructured magnetic materials; Bulk multilayers; Micro-system magnets

1. Introduction Equiatomic FePt alloys may exist in a disordered face-centered-cubic phase (high-temperature phase) or an ordered face-centered-tetragonal (FCT) phase (low-temperature phase) [1]. Although the ordered phase is stable below about 13001C, the disordered phase may be stabilized by quenching bulk samples from high temperature or depositing thin film samples at room temperature. The ordered FCT phase, commonly known as the L10 phase, is of interest for permanent magnet *Corresponding author. Laboratoire Louis N!eel, 25 Avenue des Martyrs, BP 166, 38042 Grenoble cedex 9, France. Fax: +4-76-88-11-91. E-mail address: [email protected] (D. Givord).

applications due to its excellent intrinsic magnetic properties (m0 MS ¼ 1:43 T and K1 ¼ 6:6 MJ/m3 at 300 K; TC ¼ 750 K) and its corrosion resistance [2]. It can be formed by the controlled annealing of the disordered phase through a first-order phase transition [3–6]. Alternatively, the ordered phase can be directly prepared in thin-film form by deposition on substrates heated to the appropriate temperature [7]. Ordered FePt thin films have also been prepared by optimized heat treatments of Fe/ Pt multilayers with individual layer thickness of the order of 1–3 nm [8,9]. Bulk samples are magnetically isotropic owing to the three variants for the tetragonal distortion of the cubic phase. Under certain conditions, crystallographic texture may be obtained in thin films owing to the preferential crystallization along the materials

0304-8853/03/$ - see front matter r 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 8 8 5 3 ( 0 2 ) 0 1 2 8 4 - 2

LETTER TO THE EDITOR

L140

N.H. Hai et al. / Journal of Magnetism and Magnetic Materials 257 (2003) L139–L145

most dense crystallographic directions on polycrystalline substrates (fiber texture) or epitaxial growth on single crystalline substrates [5–7]. Room-temperature coercivity values of typically 0.5 T have been developed in bulk samples [3] while higher values are reported for thin-film samples (typically 1–2 T) [4–9]. In this paper we report on the novel use of a cyclic cold-rolling process to produce nanocomposite Fe/Pt multilayers which are then heat treated to produce hard magnetic FePt foils. This study forms part of a body of work on the use of classical mechanical deformation techniques (e.g. drawing, rolling and extrusion) to prepare nanocomposite materials by cyclic deformation involving sample re-assembly of macrocomposite materials [10–15].

2. Experimental A stack of iron and platinum foils {Fe(75 mm)/ Pt(100 mm)}10 (49 at% Fe—calculated from the measured masses; stack dimensions: 1.8  4.5  15 mm3) was inserted into a sheath (stainless-steel tube) and the ensemble was compressed in a hydraulic press under ambient conditions (Pmax ¼ 5 ton/cm2) so as to compact the sample vertically and produce parallel upper and lower surfaces. The thickness of the ensemble was then progressively reduced by multiple-pass cold rolling, the inter-cylinder spacing being slightly reduced for each new pass. In a given rolling cycle, involving about 100 passes, the total thickness was reduced by a factor 10 (i.e. tinit =tfinal E10). The low deformation rate per pass typical of cold rolling allowed progressive deformation without stress-relief heat treatment, a very important factor for multilayers consisting of metals which are miscible at the temperatures required for stress-relief heat treatment. The sample was then removed from the stainless-steel sheath by cutting off the edges of the sheath and simply lifting off the upper and lower steel layers. Following this, the multilayer sample was cut into short lengths, piled up to form a stack and inserted into a new sheath. The sample was submitted to four such rolling cycles (cumulative reduction

factor E104) and the final multilayered foil had a total thickness of about 100 mm. The nanostructured multilayer was then sealed under vacuum (105 mbar) in a quartz tube, annealed in a muffle furnace at temperatures in the range 450– 5501C (t=30 s to 48 h) and then water quenched. SEM images were taken with a LEO 1530 electron microscope equipped with a field emission gun and operated at 20 kV, the TEM images were taken with a 3010 FX Jeol electron microscope operated at 300 kV. XRD diffraction patterns were made with Cu Ka radiation and magnetization measurements were performed on either a VSM or an extraction magnetometer.

3. Structural analysis SEM images of the Fe/Pt multilayers after two and four rolling cycles are shown in Fig. 1. The thickness of the individual layers after two cycles is typically less than 1 mm and after four cycles it is of the order of some tens of nanometers, in agreement with the bulk reduction factor. X-ray diffraction analysis reveals that, after four deformation cycles, the Pt layers are (2 2 0) inplane textured, as expected for rolled FCC metals [16] (Fig. 2). The full-width at half-maximum of the rocking curve of the (2 2 0) peak is about 161 (inset of Fig. 2). The intensities of the Fe peaks are much lower than those of the Pt peaks owing to the low atomic number of Fe relative to Pt; nevertheless, the (2 0 0) texture expected for BCC metals is discernible. The diffraction spectra of the samples annealed for very short times (e.g. 30 s at

Fig. 1. SEM images of Fe/Pt multilayers after (a) two and (b) four rolling cycles.

LETTER TO THE EDITOR

N.H. Hai et al. / Journal of Magnetism and Magnetic Materials 257 (2003) L139–L145

(222) fct

(222) fcc

(113) fct

fwhm = 16°

(311) fct

(311) fcc

(220) Pt

Pt (220)

L141

5 min. 30

15 min. (222) Pt

(200) Fe

(311) Pt

20

48 hrs 60

70

80

90

100

80

82

84

86

2θ

92

(111) (110) s

(200) (002)

(001) s

5001C) are comparable to those of non-annealed samples. However, peaks of both the fundamental and superstructure reflections of ordered FCT FePt are observed for annealing times as short as 2 min at 5001C, indicating that both diffusion and ordering occur under these conditions. Only trace quantities of the FCC phase are evident for short annealing times (e.g. 5 and 15 mins at 4501C, Fig. 3) and there is no evidence of the phase after extended annealing times (e.g. 48 h at 4501C, Fig. 3). This indicates that the ordered L10 phase is formed almost immediately upon heating nanostructured Fe/Pt multilayers in this tempera( and a ture range. A lattice parameter aE3.86 A tetragonal distortion c=aE0.96 are estimated for all annealing conditions studied. The diffraction peak widths become narrower as the annealing time increases, which may be attributed to grain growth. The XRD spectrum of the sample annealed at 4501C/48 h is shown in Fig. 4. Partial crystallographic texture is identified as the intensities of the (0 0 1) and (0 0 2) peaks are twice as high as those expected for isotropic material (intensity values taken from the PDF file 431359). The degree of long range order in the L10 phase can be quantified by the chemical ordering

90

Fig. 3. High-angle section of the XRD spectra (Cu Ka radiation) of FePt foils produced by annealing Fe/Pt multilayers at 4501C for times ranging from 5 min to 48 h. The vertical dashed lines represent the 2y positions of peaks belonging to the disordered FCC structure, taken from PDF file 29-0717, and those of the ordered FCT FePt structure calculated using the lattice parameters estimated with lower angle peaks (taking ( l ¼ 1:54056 A).

counts

Fig. 2. XRD patterns (Cu Ka radiation) of Fe/Pt multilayer after four deformation cycles (powder diffraction intensities for Pt and Fe are represented by  and  , respectively). Inset: rocking curve of Pt (2 2 0) peak.

88

2θ

20

30

40

50 2θ

(220) (202)

50

(112) s

40

0 10 θ (°)

(201) s

30

(200) Pt

(110) Fe

(111) Pt

-30 -20 -10

60

70

80

Fig. 4. XRD patterns (Cu Ka radiation) of FePt foil produced by annealing Fe/Pt multilayer at 4501C/48 h, the superstructure reflections of the L10 phase are denoted by the letter ‘‘s’’ (experimental data are represented by solid lines; powder diffraction intensities for FCT FePt (L10) are represented by ).

LETTER TO THE EDITOR

L142

N.H. Hai et al. / Journal of Magnetism and Magnetic Materials 257 (2003) L139–L145

parameter S; defined as S ¼ ðrFe  xFe Þ=yPt ¼ ðrPt  xPt Þ=yFe ; where xFeðPtÞ is the atomic fraction of Fe(Pt) in the sample, yFeðPtÞ the fraction of Fe(Pt) sites, and rFeðPtÞ the fraction of Fe(Pt) sites occupied by the right atomic species [17,7]. It can be estimated by comparing the integrated intensities of the (0 0 1) superstructure peak and the (0 0 2) fundamental peak [7]. SE0:8 was estimated for the sample annealed at 4501C/48 h (Fig. 4), which is comparable to the value, 0.970.1, reported for epitaxial films [7]. The fact that this value is less than 1 may be attributed to incomplete order in the FCT phase (possibly due to the relatively low annealing temperature) and/or a variation in composition away from the 50:50 stoichiometry of the perfectly ordered L10 phase. Plane-view TEM analysis of a heat-treated sample (4501C/48 h), reveals many interesting microstructural features (Figs. 5 and 6). The in-plane dimensions of the FePt grains vary from tens of nanometers to some hundreds of nanometers; an area with relatively large grains is shown in Fig. 5. Diffraction patterns taken on individual grains reveal superstructure reflections,

confirming that the FePt is ordered. A band-like structure, with bandwidths of the order of 10 nm, is observed in many grains. Diffraction analysis of an area of a large grain containing these bands (Fig. 6) reveals that the area has single crystalline structure, i.e. these bands are not twins, and that the bands are roughly perpendicular to the / 2 2 0S direction. Twinning is observed in the diffraction pattern of another area of this grain, in which the orientation of the bands changes along linear fronts (Fig. 6). This band structure is not well understood and may simply be due to heavy faulting. However, the fact that the bands are perpendicular to the /2 2 0S direction, a direction along which Fe and Pt atoms form alternate layers in the ordered FCT FePt structure, may suggest that the bands are related to an antiphase-domain structure. Moire! patterning and boundary traces due to the superposition of grains are also clearly visible. Cross-sectional observations are under way to determine the thickness of the grains. A detailed investigation of the microstructure of as-rolled and annealed Fe/Pt structures will be reported elsewhere.

4. Magnetic properties

Fig. 5. TEM image and diffraction pattern of a section of an FePt foil, produced by annealing an Fe/Pt multilayer (4501C/ 48 h), containing a number of grains with in-plane dimensions in the range 100–300 nm.

Room-temperature VSM magnetization measurements reveal that coercivity is developed even after very short annealing times, demonstrating the early onset of ordering into the high-anisotropy L10 phase (Fig. 7). Coercivity continues to increase with annealing time, which can be attributed to progressive changes in the complex microstructure of this system. The optimum coercivity value obtained in these foils (m0 H ¼ 0:9 T) is significantly higher than that usually obtained in bulk FePt, which can be related to the foil’s specific nanostructure. A detailed study of the relationship between coercivity and nanostructure is under way. Hysteresis loops of the sample with the highest observed value of coercivity (annealed at 4501C/48 h), measured in three orthogonal directions in an extraction magnetometer, are shown in Fig. 8 (for a given sample the value of coercivity measured in the VSM was higher than that measured in the

LETTER TO THE EDITOR

N.H. Hai et al. / Journal of Magnetism and Magnetic Materials 257 (2003) L139–L145

L143

Fig. 6. TEM image of one large grain of an FePt foil, produced by annealing an Fe/Pt multilayer (4501C/48 h), displaying a band structure (see text). The diffraction pattern of the left-hand side of the grain is characteristic of a single crystalline structure (top left inset) while that of the right-hand side is characteristic of a twinned structure (top right inset).

1 0.8

500°C

0.6

550°C 1

0.4 0.2

0.1

1

Mr

0.6 0.4 0.2 0

0 0.001 0.01

Hc

0.8 µ0Hc, µ0Mr (T)

µ0HC (T)

450°C

0 100 200 300 400 500 600 700 T (K)

10

100

1000

104

annealing time (minutes) Fig. 7. Variation of room-temperature coercivity of FePt foils prepared by annealing Fe/Pt multilayers plotted as a function of annealing conditions (VSM measurements); inset: temperature dependence of coercivity and remanent magnetization of FePt foil produced by annealing Fe/Pt multilayer at 4501C/48 h (extraction magnetometer measurements).

extraction magnetometer owing to the higher field sweep rate in the former; kinetic effects on magnetization reversal in this system were reported by Luo et al. [18]). The higher value of remanent magnetization in the out-of-plane loop is in agreement with the observation of partial (0 0 1) texture in the XRD pattern of this sample (Fig. 4). The temperature dependence of coercivity and remanent magnetization in the sample annealed at 4501C/48 h, measured in-plane in the range 10–700 K in an extraction magnetometer, are shown in the inset of Fig. 7. A significant hightemperature coercivity of 0.37 T was measured at 600 K. The energy product of this foil was estimated to be 100 kJ/m3 at 300 K and about 25 kJ/m3 at 600 K. No decrease in room-temperature coercivity was detected following repeated high-temperature measurements up to 700 K. This indicates that the microstructure is stable up to this temperature, which is very important if

LETTER TO THE EDITOR

N.H. Hai et al. / Journal of Magnetism and Magnetic Materials 257 (2003) L139–L145

L144

0.8

0.4 µ0M (T)

suitable candidates for magnetic micro-system applications [19].

x y z

Acknowledgements 0 1.2 0.8 0.4 0 -0.4 -0.8 -1.2 -10

-0.4

-0.8 -2

-1.5

-1

-0.5

0

0.5

1

-5

0

5

1.5

10

2

µ0H (T) Fig. 8. Room-temperature magnetization loops of FePt foils (4501C/48 h) measured: in-plane transverse to rolling direction (x), in-plane parallel to rolling direction (y) and out-of plane (z) (out-of-plane measurement corrected for demagnetizing field with demagnetizing factor N=1. Note that this simple classical treatment of demagnetizing field effects is valid at remanence but not in the vicinity of the coercive field [20]). Inset: MðHÞ measured in the z-direction with ðm0 Happl Þmax ¼ 10 T (extraction magnetometer measurements).

these foils are to be used in high-temperature applications.

5. Conclusions Cyclic cold rolling has been used to prepare textured Fe/Pt multilayer foils of total thickness 100 mm and individual layer thicknesses of the order of 10 nm. Ordered FCT FePt (L10) is rapidly formed upon annealing in the temperature range 450–5501C. Coercivities as high as 0.9 T at 300 K have been measured in annealed foils. This coercivity value is intermediate between the highest values reported for bulk and thin film samples. Coercivity decreases relatively slowly with increasing measuring temperature, attaining a value of 0.37 T at 600 K. Efforts are now under way to increase the texture of the hard foils obtained and thus further increase remanence/energy product values. Finally, the form and size of the samples produced (foils of thickness 100 mm) make them

This work was carried out within the framework of the European project for the development of high-temperature magnets ‘‘HITEMAG’’ (G5RD2000-00213) which is supported by the Commission of the European Union (D.G. XII). The authors would like to thank Dr. M. Venkatesan for fruitful discussions. Scanning electron microscope imaging was performed with Nanofab (CNRS Grenoble) facilities. N.H. Hai gratefully acknowledges support of the CNRS-PICS program (Nanomate!riaux) and the French Embassy in Vietnam.

References [1] O. Kubaschewski, Fe–Pt binary phase diagram, in: IronBinary Phase Diagrams, Springer, Berlin, 1982, 91pp. [2] R. Skomski, J.M.D. Coey, Permanent Magnetism, Institute of Physics Publishing, Bristol and Philadelphia, 1999, 269pp. [3] K. Watanabe, H. Masumoto, J. Jpn. Inst. Met. 48 (1984) 930. [4] J.A. Aboaf, T.R. McGuire, S.R. Herd, E. Klokholm, IEEE Trans. Magn. 20 (1984) 1642. [5] K.R. Coffey, M.A. Parker, J.K. Howard, IEEE Trans. Magn. 31 (1995) 2737. [6] R.A. Ristau, K. Barmak, L.H. Lewis, K.R. Coffey, J.K. Howard, J. Appl. Phys. 86 (1999) 4527. [7] A. Cebollada, D. Weller, J. Sticht, G.R. Harp, R.F.C. Farrow, R.F. Marks, R. Savoy, J.C. Scott, Phys. Rev. B 50 (1994) 3419. [8] B.M. Lairson, M.R. Visokay, R. Sinclair, B.M. Clemens, Appl. Phys. Lett. 62 (1993) 639. [9] J.P. Liu, C.P. Kuo, Y. Liu, D.J. Sellmyer, Appl. Phys. Lett. 72 (1998) 483. [10] F.P. Levi, J. Appl. Phys. 31 (1960) 1469. [11] F. Dupouy, S. Ask!enazy, J.-P. Peyrade, D. Legat, Physica B 211 (1953) 43. [12] K. Yasuna, M. Terauchi, A. Otsuki, K.N. Ishihara, P.H. Shingu, J. Appl. Phys. 82 (1997) 2435. [13] F. Wacquant, S. Denolly, A. Gigu"ere, J.P. Nozi"eres, D. Givord, V. Mazauric, IEEE Trans. Magn. 35 (1999) 3484. [14] A. Gigu"ere, N.H. Hai, N.M. Dempsey, D. Givord, J. Magn. Magn. Mater. 242–245 (2002) 581.

LETTER TO THE EDITOR

N.H. Hai et al. / Journal of Magnetism and Magnetic Materials 257 (2003) L139–L145 [15] A. Gigu"ere, N.M. Dempsey, M. Verdier, L. Ortega, D. Givord, IEEE Trans. Magn. (Proceedings Intermag 2002), to appear. [16] R.W.K. Honeycombe, The Plastic Deformation of Metals, Edward Arnold Ltd., London, 1968, 325pp. [17] B.E. Warren, X-ray Diffraction, Addison-Wesley Publishing Company, California, 1969, 206pp.

L145

[18] C.P. Luo, S.H. Shan, D.J. Sellmyer, J. Appl. Phys. 79 (1996) 4899. [19] O. Cugat, Micro-Actionneurs Electromagn!etiques— MAGMAS (MAGnetic Micro Actuators and Systems), Hermes-Lavoisier, Series EGEM, 2002, to appear. [20] N.H. Hai, N.M. Dempsey, D. Givord, in preparation.