- Email: [email protected]
Effect of magnetoelastic anisotropy on properties of Finemet-type microwires
Journal of Alloys and Compounds 536S (2012) S291–S295
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jallcom
Effect of magnetoelastic anisotropy on properties of Finemet-type microwires M. Churyukanova a,∗ , V. Zhukova b , S. Kaloshkin a , A. Zhukov b a b
National University of Science and Technology «MISIS», Leninsky Prosp., 4, Moscow 119049, Russia Dpto. Fisica de Materiales, Fac. Quimicas, UPV/EHU, 20009 San Sebastian, Spain
a r t i c l e
i n f o
Article history: Received 24 June 2011 Received in revised form 10 October 2011 Accepted 19 October 2011 Available online 22 November 2011 Keywords: Finemet Microwire Amorphous alloy Curie temperature Heat capacity Relaxation Nanocrystallization
a b s t r a c t Magnetic properties and DSC peak near Curie temperature, TC , of amorphous and nanocrystalline microwires with different ratios = d/D were studied. The investigated compositions were close to Finemet-type: Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 , Fe71.8 Cu1 Nb3.1 Si15 B9.1 and Fe73.8 Cu1 Nb3.1 Si13 B9.1 . The effects of magnetoelastic energy, stored during the Finemet-type microwires fabrication, on hysteresis loops, TC and heat capacity of Finemet-type microwires were investigated. Hysteresis loops of all as-prepared microwires showed rectangular shape, typical for Fe-rich microwires. As expected, coercivity, Hc , increases with the decrease of the ratio . On the other hand, the change of heat capacity at TC , Cp , exhibits linear increase with the ratio . This relationship holds for microwires in the initial state as well as after annealing. Glass removal results in considerable change of both Hc and Cp , which reveals the effect of internal stresses. Structural relaxation of microwires results in a shift of TC calorimetric peak of amorphous phase to higher temperatures, while crystallization leads to peak disappearance. This effect was attributed to the dependency of TC calorimetric peak on the value of magnetostriction of magnetic phase, which declines to zero with Finemet alloys crystallization. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Amorphous and nanocrystalline magnetically soft glass coated microwires (typically of 10–30 m in diameter) attracted growing attention within the last few years owing to their outstanding hysteretic magnetic properties (magnetic bistability, enhanced magnetic softness, fast domain wall propagation) and GMI effect suitable for technological applications [1,2]. Particularly, recent studies have demonstrated that considerable improvement of soft magnetic properties and GMI effect of glass coated microwires is possible selecting appropriate chemical composition of metallic nucleus and adequate annealing conditions [1]. In some cases, nanocrystallization allows achieving good magnetic softness and enhanced GMI effect in ferromagnetic microwires [3,4]. Consequently, it is quite important to study the effect of magnetoelastic anisotropy on magnetic properties and Curie temperature of Finemet-type glass-coated microwires. It is worth mentioning that the simultaneous solidification of composite microwire consisting of ferromagnetic nucleus surrounded by glass coating introduces considerable internal stresses inside the ferromagnetic nucleus during simultaneous solidification of composite microwire due to the difference in the thermal expansion coefficients of the glass and the metal [1,5–7].
∗ Corresponding author. Tel.: +7 495 6384413; fax: +7 495 6384413. E-mail address: [email protected] (M. Churyukanova). 0925-8388/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2011.10.068
Generally magnetic properties and overall shape of hysteresis loops of amorphous microwires depend on the composition of the metallic nucleus as well as on the composition and thickness of the glass coating. As discovered before in [1], shape of hysteresis loops changes from rectangular, typical for amorphous Fe-rich compositions, to inclined, typical for Co-rich compositions. Microwires with vanishing magnetostriction exhibit quite soft magnetic properties. Such strong dependence of the hysteresis loops on these parameters should be attributed to the magnetoelastic energy given by:
Kme ≈
3 s i , 2
(1)
where s is the saturation magnetostriction and i is the internal stress. The magnetostriction constant depends mostly on the chemical composition and is vanishing in amorphous Fe–Co based alloys with Co/Fe ≈ 70/5 [1,8]. On the other hand, the estimated values of the internal stresses in these glass coated microwires arising from the difference in the thermal expansion coefficients of metallic nucleus and glass coating are of the order of 100–1000 MPa, depending strongly on the -ratio ( = d/D – ratio between the metallic core diameter, d, and total microwire diameter, D) [5–7]. The internal stresses increase with rising of the glass coating thickness (i.e. with -ratio decreasing). Such large internal stresses give rise to a drastic change of the magnetoelastic energy, Kme , even for small changes of the glass-coating thickness at fixed metallic core diameter. Additionally, such a change of the -ratio should be
S292
M. Churyukanova et al. / Journal of Alloys and Compounds 536S (2012) S291–S295
related to the change of the magnetostriction constant with applied stress [1]: s =
M dH o s k 3
d
,
(2)
where o Ms is the saturation magnetization. It is worth mentioning, that residual stresses of glass-coated microwires arising during simultaneous solidification of metallic nucleus and glass coating, mostly have been estimated from the simulations of the process of simultaneous solidification of metallic nucleus inside the glass tube [5–7] and experimental determination of such residual stresses is rather complex. One of the experimental evidence of existence of such stresses is the dependence of hysteresis loops and particularly magnetic properties (coercivity, remanent magnetization) on -ratio [1,9]. Consequently, tailoring of the magnetoelastic energy, Kme , is essentially important for achieving optimal magnetic properties of glass-coated microwires [1,10–12] Accordingly, any method allowing estimation of internal stresses in glass-coated microwires is quite suitable for soft magnetic properties optimization. Lately a few publications on utilisation of differential scanning calorimetry, DSC, method for the studies of the properties of amorphous alloys in vicinity of Curie temperature have been released [13,14]. This method allows determination of TC and activation energy of relaxation and crystallization processes. Besides, DSC method allows precise determination of TC (with reproducibility ±0.2 K). In this paper we studied effect of magnetoelastic energy stored during the Finemet-type microwires fabrication on hysteresis loops, Curie temperature and heat capacity. 2. Experimental Four Finemet-type compositions were investigated: Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 (A), Fe71.8 Cu1 Nb3.1 Si15 B9.1 (B), Fe73.8 Cu1 Nb3.1 Si13 B9.1 (C) and Fe70.8 Cu1 Nb3.1 Si16 B9.1 (D). Glass-coated microwires with different metallic nucleus diameter, d, and total microwire diameter, D, were produced by modified Taylor–Ulitovsky method [1]. Hysteresis loops have been determined by flux metric method, as described elsewhere [1,3]. DSC measurements were performed using DSC 204 F1 Netzsch calorimeter in Ar atmosphere at a heating rate of 10 K/min. In order to increase the sensibility of DSC measurements the mass of the samples was increased up to 30–50 mg (details see in [8]) that was possible due to special high-volume aluminum containers. The Curie temperature, the peak area and the change of the specific heat near the Curie point were estimated from DSC curves using standard IT application. X-ray diffraction (XRD) analysis was carried out on Rigaku ULTIMA-4 diffractometer in CuK␣ radiation.
Fig. 1. Hysteresis loops of as-prepared Fe73.8 Cu1 Nb3.1 Si13 B9.1 microwires with different -ratios: (a) = 0.87, (b) = 0.38.
Curie temperature is very sensitive to the composition of the Finemet amorphous alloys: increase of Fe and B content results in the rise of TC , while Cu and Nb reduce its value [16]. This is also true for the studied microwires: increase of Fe content from 70.8% (alloy A) to 73.8% (B) leads to TC decrease from 310.4 ◦ C to 299.3 ◦ C. Heat treatment of microwires leads to a change of the position and the shape of DSC peak in the vicinity of the Curie point (Fig. 4). The shape of TC calorimetric peak was estimated as the difference of heat capacities Cp at TC and at 20 ◦ C above this point.
3. Results Hysteresis loops of as-prepared microwires showed rectangular shape, typical for Fe-rich amorphous microwires (Fig. 1) [1]. As expected, coercivity, Hc , of as-prepared microwires depends on ratio = d/D (Fig. 2). Conventional (without magnetic field or applied stress) annealing at temperatures below the nanocsrystallization did not affected both coercivity value and overall shape of hysteresis loops, as previously observed for amorphous Fe-rich microwires [13]. On the other hand, glass removal leads to decrease in coercivity of microwires for as-prepared samples and after annealing. DSC curve reflects the thermal properties of microwires and depends on many parameters of production and subsequent heat treatment [14]. Typical DSC curves of as-prepared Finemet-type microwires are shown in Fig. 3. Magnetic transition at the temperature near 300 ◦ C (TC ) and two peaks of crystallization (T1 and T2 ) can be seen. T1 corresponds to the release of ␣-Fe nanocrystals, and T2 – boride phase [15].
Fig. 2. Coercivity dependence on -ratio for as-prepared Fe73.8 Cu1 Nb3.1 Si13 B9.1 microwires.
M. Churyukanova et al. / Journal of Alloys and Compounds 536S (2012) S291–S295
S293
Fig. 3. DSC curves of as-prepared Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 microwire with = 0.5.
Fig. 5. X-ray diffraction of Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 microwire after annealing at 460 ◦ C: (a) as-prepared; (b) 1 min; (c) 8 h.
In general the behaviour of the DSC curves is the same as was observed for amorphous ribbons of similar compositions [17]. During the annealing below 500 ◦ C TC increased with annealing temperature without any change in the shape of the peak. Short term annealing below crystallization temperature resulted in relaxation of the amorphous phase, there were no traces of crystalline phase on XRD patterns (Fig. 5a and b). When the annealing temperature was approaching the temperature of crystallization onset, the curves started to change significantly. The peak became smoother and completely disappeared after annealing at 580 ◦ C. This was linked to partial crystallization of the amorphous phase. The same character of the changes of DSC curves was observed for longer annealing times at lower temperatures. X-ray study revealed the emergence of 17 vol.% of ␣-Fe with crystallite size of 12 nm after annealing at 460 ◦ C during 8 h (Fig. 5c). In spite of predicted pressure Curie temperature dependence [18], our results did not show clear evidence of such dependence. For a given composition TC remains almost constant for all -ratios of microwires (Fig. 6). TC values depend only on the degree of structural relaxation achieved by annealing. However this statement is true only for relatively low time–temperature heat treatment procedures. On the other hand, at higher annealing temperatures the internal
stress dependence of Curie temperature may appear: the internal stresses in microwires start to affect the crystallization process of the amorphous nucleus. This can results in a significant change of TC calorimetric peak up to its complete disappearance (Fig. 4). While TC value is not sensible to the -ratio, the shape of the TC calorimetric peak reveals a strong dependence on this parameter for the studied microwires (Fig. 7). The dependencies of Cp on ratio for different treatments have similar shape: Cp reduces with decreasing -ratio. The strongest dependence is observed for the as-prepared microwires (Fig. 7a). Annealing results in the reduction of the slope of the dependences, however it remains considerable, showing influence of internal stresses on the heat capacity peak in Curie temperature (Fig. 7b and c). Cp falls down to zero for low values, corresponding to the highest magnitudes of internal stresses, which means that TC calorimetric peak gradually disappears. In other words, DSC cannot detect magnetic transformation in case of strong tensile stress. At the same time, the disappearance of the TC calorimetric peak corresponds to abrupt increase of coercivity of the samples at < 0.4 (Fig. 2). This effect needs special consideration. The result on decreasing of magnetic permeability of the amorphous core of microwire at stress increase seems more or less understandable (see the discussion of Fig. 2). However influence
Fig. 4. DSC curves of Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 microwire with = 0.5 for annealing during 4 min at different temperatures.
Fig. 6. Curie temperature dependences on -ratio for Fe73.8 Cu1 Nb3.1 Si13 B9.1 microwires: (a) as-prepared; (b) after annealing at 400 ◦ C during 1 min; (c) after annealing at 480 ◦ C during 1 min.
S294
M. Churyukanova et al. / Journal of Alloys and Compounds 536S (2012) S291–S295
structure relaxation of amorphous phase at annealing. Precipitation of ␣-Fe nanocrystals in amorphous phase at further annealing is accompanied by gradual degradation of corresponding to TC heat capacity peak. The -shaped time dependencies, however, were not characterized by definite final value of TC : long-term annealing of amorphous samples resulted in continuation of slow increase of TC at each annealing temperature. TC measurements may detect the very first stages of amorphous phase crystallization. The fact that TC data correlate with structural alterations reflects the redistribution of components in amorphous phase during annealing. 4. Conclusions
Fig. 7. The change of heat capacity depending on -ratio for Fe73.8 Cu1 Nb3.1 Si13 B9.1 microwires: (a) as-prepared; (b) 400 ◦ C 1 min; (c) 480 ◦ C 1 min.
of tensile stress on heat capacity is rather unexpected. As-prepared amorphous microwires have positive magnetostriction constant of around 20 ppm. In principle by measuring change of heat capacity at the Curie point we study alterations of oscillation spectra of metallic atoms in the transition point from ferromagnetic to paramagnetic state. This follows from the definition of heat capacity. We cannot observe any difference in heat capacities of the amorphous phase at temperatures lower and higher TC . This means that internal stresses are high enough, that the elongation of the metallic nucleus under these stresses is not less than magnetostrictive elongation in ferromagnetic state. This elongation does not change at the Curie point and as a result DSC does not register any alterations in heat capacity. This means that internal stresses are high enough and consequently the metallic nucleus subjected to high enough deformation. Considering, this case we can assume, that change of magnetic ordering does not affect significantly the metallic nucleus strain. Accordingly, strain related with magnetic ordering does not changed at the Curie point and as a result DSC does not register any alterations in heat capacity. As-prepared amorphous microwires have positive value of magnetostriction of around 20 ppm. In principle by measuring change of heat capacity in the Curie point we study alterations of oscillation spectra of metallic atoms in the transition point from ferromagnetic to paramagnetic state. This follows from the definition of heat capacity. We cannot observe any difference in heat capacities of the amorphous phase at temperatures lower and higher TC . This means that outer stresses are so high, that the elongation of the metallic nucleus under these stresses is not less than magnetostrictive elongation in ferromagnetic state. This elongation does not change at the Curie point and as a result DSC does not register any alterations in heat capacity. Glass removal results in considerable change of heat capacity Cp (Fig. 7). This confirms that internal stresses affect the value of the peak of heat capacity in the Curie point. Removal of the glass coating releases the stresses induced by the coating on the magnetic nucleus, and the sample undergoes higher deformation under magnetization due to magnetostriction. This is manifested in larger change in heat capacity on DSC curve. In any case DSC measurements can detect changes of the heat capacity in the vicinity of TC only in the case of changing average interatomic distances. In the case of zero magnetostriction composition and/or high stresses hindering the sample deformation during ferromagnetic–paramagnetic transition at Curie point, DSC cannot detect this transition. As we have shown earlier for Finemet-type amorphous ribbonshaped alloys [17], the increase of TC takes place due to the atomic
Increasing of internal stresses in the amorphous metallic nucleus caused by the reduction of the -ratio, results in the rise of coercivity, Hc . DSC study in the vicinity of magnetic transformation around the Curie point, determined as a heat capacity peak, did not exhibit any dependence of TC on -ratio. However, for the highest internal stresses values, corresponding to the lowest , heat capacity peak in TC could not be registered. On the other hand, the change of heat capacity at TC , Cp , decreases with decreasing the -ratio. Higher internal stresses correspond to lower heat capacity peak. This effect has been attributed to the dependence of TC calorimetric peak on the magnetostriction value of the magnetic phase as well as on the change of interatomic distances under applied outer stresses. Structural relaxation of microwires results in a shift of TC calorimetric peak of amorphous phase to higher temperatures, while crystallization leads to peak disappearance. Thus, precise DSC measurement can be a sensitive method for compositional and structural control of amorphous and nanocrystalline soft magnetic microwires. Acknowledgements This work was supported by the EU under project EM-safety, by ERA-NET program under Project “SoMaMicSens” (MANUNET2010), by Spanish Ministry of Science and Innovation, MICINN under Project MAT2010-18914 and by the Basque Government under Saiotek 09 Mic Magn project. A. Zh. and V. Zh. wish to acknowledge the support of the Basque Government under Program of Mobility of the Investigating Personnel of the Department of Education, Universities and Investigation under Grants MV2010-2-31 and MV-2009-2-24 respectively. References [1] A. Zhukov, V. Zhukova, Magnetic properties and applications of ferromagnetic microwires with amorphous and nanocrystalline structure, Nova Science Publishers, Inc., 400 Oser Avenue, Suite 1600, Hauppauge, NY 11788, 2009, p. 162. [2] D.C. Jiles, Acta Mater. 51 (2003) 5907. [3] J. Arcas, C. Gómez-Polo, A. Zhukov, M. Vázquez, V. Larin, A. Hernando, Nanostruct. Mater. 7 (8) (1996) 823–834. [4] C. Dudek, A.L. Adenot-Engelvin, F. Bertin, O. Acher, J. Non-Cryst. Solids 353 (2007) 925. [5] H. Chiriac, T.-A. Ova!ri, A. Zhukov, J. Magn. Magn. Mater. 254–255 (2003) 469–471. [6] J. Velázquez, M. Vazquez, A. Zhukov, J. Mater. Res. 11 (1996) 2499. [7] A.S. Antonov, V.T. Borisov, O.V. Borisov, A.F. Prokoshin, N.A. Usov, J. Phys. D: Appl. Phys. 33 (2000) 1161. [8] H. Fujimori, K.I. Arai, H. Shirae, H. Saito, T. Masumoto, N. Tsuya, Jpn. J. Appl. Phys. 15 (4) (1976) 705. [9] A. Zhukov, J. Magn. Magn. Mater. 242–245 (2002) 216–223. [10] V. Zhukova, A. Chizhik, A. Zhukov, A. Torcunov, V. Larin, J. Gonzalez, IEEE Trans. Magn. 38 (2002) 3090. [11] V. Zhukova, M. Ipatov, A. Zhukov, Sensors 9 (2009) 9216. [12] A. Zhukov, M. Ipatov, J. Gonzalez, J.M. Blanco, V. Zhukova, J. Magn. Magn. Mater. 321 (2009) 822–825. [13] A. Zhukov, Adv. Funct. Mater. 16 (2006) 675.
M. Churyukanova et al. / Journal of Alloys and Compounds 536S (2012) S291–S295 [14] V. Zhukova, S. Kaloshkin, A. Zhukov, J. Gonzalez, J. Magn. Magn. Mater. 249 (2002) 108–112. [15] S.D. Kaloshkin, I.A. Tomilin, Thermochim. Acta 280 (1996) 303–317. [16] E.V. Kaevitser, S.D. Kaloshkin, V.V. Tcherdyntsev, et al., J. Magn. Magn. Mater. 258 (2003) 539–542.
S295
[17] S. Kaloshkin, M. Churyukanova, V. Zadorozhnyi, I. Shchetinin, R.K. Roy, J. Alloys Compd. 509 (S1) (2011) 400–403. [18] J.M. Barandiarán, P. Gorria, I. Orúe, M.L. Fdez-Gubieda, F. Plazaola, A. Hernando, Phys. Rev. B 54 (1996) 3026–3029.
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jallcom
Effect of magnetoelastic anisotropy on properties of Finemet-type microwires M. Churyukanova a,∗ , V. Zhukova b , S. Kaloshkin a , A. Zhukov b a b
National University of Science and Technology «MISIS», Leninsky Prosp., 4, Moscow 119049, Russia Dpto. Fisica de Materiales, Fac. Quimicas, UPV/EHU, 20009 San Sebastian, Spain
a r t i c l e
i n f o
Article history: Received 24 June 2011 Received in revised form 10 October 2011 Accepted 19 October 2011 Available online 22 November 2011 Keywords: Finemet Microwire Amorphous alloy Curie temperature Heat capacity Relaxation Nanocrystallization
a b s t r a c t Magnetic properties and DSC peak near Curie temperature, TC , of amorphous and nanocrystalline microwires with different ratios = d/D were studied. The investigated compositions were close to Finemet-type: Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 , Fe71.8 Cu1 Nb3.1 Si15 B9.1 and Fe73.8 Cu1 Nb3.1 Si13 B9.1 . The effects of magnetoelastic energy, stored during the Finemet-type microwires fabrication, on hysteresis loops, TC and heat capacity of Finemet-type microwires were investigated. Hysteresis loops of all as-prepared microwires showed rectangular shape, typical for Fe-rich microwires. As expected, coercivity, Hc , increases with the decrease of the ratio . On the other hand, the change of heat capacity at TC , Cp , exhibits linear increase with the ratio . This relationship holds for microwires in the initial state as well as after annealing. Glass removal results in considerable change of both Hc and Cp , which reveals the effect of internal stresses. Structural relaxation of microwires results in a shift of TC calorimetric peak of amorphous phase to higher temperatures, while crystallization leads to peak disappearance. This effect was attributed to the dependency of TC calorimetric peak on the value of magnetostriction of magnetic phase, which declines to zero with Finemet alloys crystallization. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Amorphous and nanocrystalline magnetically soft glass coated microwires (typically of 10–30 m in diameter) attracted growing attention within the last few years owing to their outstanding hysteretic magnetic properties (magnetic bistability, enhanced magnetic softness, fast domain wall propagation) and GMI effect suitable for technological applications [1,2]. Particularly, recent studies have demonstrated that considerable improvement of soft magnetic properties and GMI effect of glass coated microwires is possible selecting appropriate chemical composition of metallic nucleus and adequate annealing conditions [1]. In some cases, nanocrystallization allows achieving good magnetic softness and enhanced GMI effect in ferromagnetic microwires [3,4]. Consequently, it is quite important to study the effect of magnetoelastic anisotropy on magnetic properties and Curie temperature of Finemet-type glass-coated microwires. It is worth mentioning that the simultaneous solidification of composite microwire consisting of ferromagnetic nucleus surrounded by glass coating introduces considerable internal stresses inside the ferromagnetic nucleus during simultaneous solidification of composite microwire due to the difference in the thermal expansion coefficients of the glass and the metal [1,5–7].
∗ Corresponding author. Tel.: +7 495 6384413; fax: +7 495 6384413. E-mail address: [email protected] (M. Churyukanova). 0925-8388/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2011.10.068
Generally magnetic properties and overall shape of hysteresis loops of amorphous microwires depend on the composition of the metallic nucleus as well as on the composition and thickness of the glass coating. As discovered before in [1], shape of hysteresis loops changes from rectangular, typical for amorphous Fe-rich compositions, to inclined, typical for Co-rich compositions. Microwires with vanishing magnetostriction exhibit quite soft magnetic properties. Such strong dependence of the hysteresis loops on these parameters should be attributed to the magnetoelastic energy given by:
Kme ≈
3 s i , 2
(1)
where s is the saturation magnetostriction and i is the internal stress. The magnetostriction constant depends mostly on the chemical composition and is vanishing in amorphous Fe–Co based alloys with Co/Fe ≈ 70/5 [1,8]. On the other hand, the estimated values of the internal stresses in these glass coated microwires arising from the difference in the thermal expansion coefficients of metallic nucleus and glass coating are of the order of 100–1000 MPa, depending strongly on the -ratio ( = d/D – ratio between the metallic core diameter, d, and total microwire diameter, D) [5–7]. The internal stresses increase with rising of the glass coating thickness (i.e. with -ratio decreasing). Such large internal stresses give rise to a drastic change of the magnetoelastic energy, Kme , even for small changes of the glass-coating thickness at fixed metallic core diameter. Additionally, such a change of the -ratio should be
S292
M. Churyukanova et al. / Journal of Alloys and Compounds 536S (2012) S291–S295
related to the change of the magnetostriction constant with applied stress [1]: s =
M dH o s k 3
d
,
(2)
where o Ms is the saturation magnetization. It is worth mentioning, that residual stresses of glass-coated microwires arising during simultaneous solidification of metallic nucleus and glass coating, mostly have been estimated from the simulations of the process of simultaneous solidification of metallic nucleus inside the glass tube [5–7] and experimental determination of such residual stresses is rather complex. One of the experimental evidence of existence of such stresses is the dependence of hysteresis loops and particularly magnetic properties (coercivity, remanent magnetization) on -ratio [1,9]. Consequently, tailoring of the magnetoelastic energy, Kme , is essentially important for achieving optimal magnetic properties of glass-coated microwires [1,10–12] Accordingly, any method allowing estimation of internal stresses in glass-coated microwires is quite suitable for soft magnetic properties optimization. Lately a few publications on utilisation of differential scanning calorimetry, DSC, method for the studies of the properties of amorphous alloys in vicinity of Curie temperature have been released [13,14]. This method allows determination of TC and activation energy of relaxation and crystallization processes. Besides, DSC method allows precise determination of TC (with reproducibility ±0.2 K). In this paper we studied effect of magnetoelastic energy stored during the Finemet-type microwires fabrication on hysteresis loops, Curie temperature and heat capacity. 2. Experimental Four Finemet-type compositions were investigated: Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 (A), Fe71.8 Cu1 Nb3.1 Si15 B9.1 (B), Fe73.8 Cu1 Nb3.1 Si13 B9.1 (C) and Fe70.8 Cu1 Nb3.1 Si16 B9.1 (D). Glass-coated microwires with different metallic nucleus diameter, d, and total microwire diameter, D, were produced by modified Taylor–Ulitovsky method [1]. Hysteresis loops have been determined by flux metric method, as described elsewhere [1,3]. DSC measurements were performed using DSC 204 F1 Netzsch calorimeter in Ar atmosphere at a heating rate of 10 K/min. In order to increase the sensibility of DSC measurements the mass of the samples was increased up to 30–50 mg (details see in [8]) that was possible due to special high-volume aluminum containers. The Curie temperature, the peak area and the change of the specific heat near the Curie point were estimated from DSC curves using standard IT application. X-ray diffraction (XRD) analysis was carried out on Rigaku ULTIMA-4 diffractometer in CuK␣ radiation.
Fig. 1. Hysteresis loops of as-prepared Fe73.8 Cu1 Nb3.1 Si13 B9.1 microwires with different -ratios: (a) = 0.87, (b) = 0.38.
Curie temperature is very sensitive to the composition of the Finemet amorphous alloys: increase of Fe and B content results in the rise of TC , while Cu and Nb reduce its value [16]. This is also true for the studied microwires: increase of Fe content from 70.8% (alloy A) to 73.8% (B) leads to TC decrease from 310.4 ◦ C to 299.3 ◦ C. Heat treatment of microwires leads to a change of the position and the shape of DSC peak in the vicinity of the Curie point (Fig. 4). The shape of TC calorimetric peak was estimated as the difference of heat capacities Cp at TC and at 20 ◦ C above this point.
3. Results Hysteresis loops of as-prepared microwires showed rectangular shape, typical for Fe-rich amorphous microwires (Fig. 1) [1]. As expected, coercivity, Hc , of as-prepared microwires depends on ratio = d/D (Fig. 2). Conventional (without magnetic field or applied stress) annealing at temperatures below the nanocsrystallization did not affected both coercivity value and overall shape of hysteresis loops, as previously observed for amorphous Fe-rich microwires [13]. On the other hand, glass removal leads to decrease in coercivity of microwires for as-prepared samples and after annealing. DSC curve reflects the thermal properties of microwires and depends on many parameters of production and subsequent heat treatment [14]. Typical DSC curves of as-prepared Finemet-type microwires are shown in Fig. 3. Magnetic transition at the temperature near 300 ◦ C (TC ) and two peaks of crystallization (T1 and T2 ) can be seen. T1 corresponds to the release of ␣-Fe nanocrystals, and T2 – boride phase [15].
Fig. 2. Coercivity dependence on -ratio for as-prepared Fe73.8 Cu1 Nb3.1 Si13 B9.1 microwires.
M. Churyukanova et al. / Journal of Alloys and Compounds 536S (2012) S291–S295
S293
Fig. 3. DSC curves of as-prepared Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 microwire with = 0.5.
Fig. 5. X-ray diffraction of Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 microwire after annealing at 460 ◦ C: (a) as-prepared; (b) 1 min; (c) 8 h.
In general the behaviour of the DSC curves is the same as was observed for amorphous ribbons of similar compositions [17]. During the annealing below 500 ◦ C TC increased with annealing temperature without any change in the shape of the peak. Short term annealing below crystallization temperature resulted in relaxation of the amorphous phase, there were no traces of crystalline phase on XRD patterns (Fig. 5a and b). When the annealing temperature was approaching the temperature of crystallization onset, the curves started to change significantly. The peak became smoother and completely disappeared after annealing at 580 ◦ C. This was linked to partial crystallization of the amorphous phase. The same character of the changes of DSC curves was observed for longer annealing times at lower temperatures. X-ray study revealed the emergence of 17 vol.% of ␣-Fe with crystallite size of 12 nm after annealing at 460 ◦ C during 8 h (Fig. 5c). In spite of predicted pressure Curie temperature dependence [18], our results did not show clear evidence of such dependence. For a given composition TC remains almost constant for all -ratios of microwires (Fig. 6). TC values depend only on the degree of structural relaxation achieved by annealing. However this statement is true only for relatively low time–temperature heat treatment procedures. On the other hand, at higher annealing temperatures the internal
stress dependence of Curie temperature may appear: the internal stresses in microwires start to affect the crystallization process of the amorphous nucleus. This can results in a significant change of TC calorimetric peak up to its complete disappearance (Fig. 4). While TC value is not sensible to the -ratio, the shape of the TC calorimetric peak reveals a strong dependence on this parameter for the studied microwires (Fig. 7). The dependencies of Cp on ratio for different treatments have similar shape: Cp reduces with decreasing -ratio. The strongest dependence is observed for the as-prepared microwires (Fig. 7a). Annealing results in the reduction of the slope of the dependences, however it remains considerable, showing influence of internal stresses on the heat capacity peak in Curie temperature (Fig. 7b and c). Cp falls down to zero for low values, corresponding to the highest magnitudes of internal stresses, which means that TC calorimetric peak gradually disappears. In other words, DSC cannot detect magnetic transformation in case of strong tensile stress. At the same time, the disappearance of the TC calorimetric peak corresponds to abrupt increase of coercivity of the samples at < 0.4 (Fig. 2). This effect needs special consideration. The result on decreasing of magnetic permeability of the amorphous core of microwire at stress increase seems more or less understandable (see the discussion of Fig. 2). However influence
Fig. 4. DSC curves of Fe70.8 Cu1 Nb3.1 Si14.5 B10.6 microwire with = 0.5 for annealing during 4 min at different temperatures.
Fig. 6. Curie temperature dependences on -ratio for Fe73.8 Cu1 Nb3.1 Si13 B9.1 microwires: (a) as-prepared; (b) after annealing at 400 ◦ C during 1 min; (c) after annealing at 480 ◦ C during 1 min.
S294
M. Churyukanova et al. / Journal of Alloys and Compounds 536S (2012) S291–S295
structure relaxation of amorphous phase at annealing. Precipitation of ␣-Fe nanocrystals in amorphous phase at further annealing is accompanied by gradual degradation of corresponding to TC heat capacity peak. The -shaped time dependencies, however, were not characterized by definite final value of TC : long-term annealing of amorphous samples resulted in continuation of slow increase of TC at each annealing temperature. TC measurements may detect the very first stages of amorphous phase crystallization. The fact that TC data correlate with structural alterations reflects the redistribution of components in amorphous phase during annealing. 4. Conclusions
Fig. 7. The change of heat capacity depending on -ratio for Fe73.8 Cu1 Nb3.1 Si13 B9.1 microwires: (a) as-prepared; (b) 400 ◦ C 1 min; (c) 480 ◦ C 1 min.
of tensile stress on heat capacity is rather unexpected. As-prepared amorphous microwires have positive magnetostriction constant of around 20 ppm. In principle by measuring change of heat capacity at the Curie point we study alterations of oscillation spectra of metallic atoms in the transition point from ferromagnetic to paramagnetic state. This follows from the definition of heat capacity. We cannot observe any difference in heat capacities of the amorphous phase at temperatures lower and higher TC . This means that internal stresses are high enough, that the elongation of the metallic nucleus under these stresses is not less than magnetostrictive elongation in ferromagnetic state. This elongation does not change at the Curie point and as a result DSC does not register any alterations in heat capacity. This means that internal stresses are high enough and consequently the metallic nucleus subjected to high enough deformation. Considering, this case we can assume, that change of magnetic ordering does not affect significantly the metallic nucleus strain. Accordingly, strain related with magnetic ordering does not changed at the Curie point and as a result DSC does not register any alterations in heat capacity. As-prepared amorphous microwires have positive value of magnetostriction of around 20 ppm. In principle by measuring change of heat capacity in the Curie point we study alterations of oscillation spectra of metallic atoms in the transition point from ferromagnetic to paramagnetic state. This follows from the definition of heat capacity. We cannot observe any difference in heat capacities of the amorphous phase at temperatures lower and higher TC . This means that outer stresses are so high, that the elongation of the metallic nucleus under these stresses is not less than magnetostrictive elongation in ferromagnetic state. This elongation does not change at the Curie point and as a result DSC does not register any alterations in heat capacity. Glass removal results in considerable change of heat capacity Cp (Fig. 7). This confirms that internal stresses affect the value of the peak of heat capacity in the Curie point. Removal of the glass coating releases the stresses induced by the coating on the magnetic nucleus, and the sample undergoes higher deformation under magnetization due to magnetostriction. This is manifested in larger change in heat capacity on DSC curve. In any case DSC measurements can detect changes of the heat capacity in the vicinity of TC only in the case of changing average interatomic distances. In the case of zero magnetostriction composition and/or high stresses hindering the sample deformation during ferromagnetic–paramagnetic transition at Curie point, DSC cannot detect this transition. As we have shown earlier for Finemet-type amorphous ribbonshaped alloys [17], the increase of TC takes place due to the atomic
Increasing of internal stresses in the amorphous metallic nucleus caused by the reduction of the -ratio, results in the rise of coercivity, Hc . DSC study in the vicinity of magnetic transformation around the Curie point, determined as a heat capacity peak, did not exhibit any dependence of TC on -ratio. However, for the highest internal stresses values, corresponding to the lowest , heat capacity peak in TC could not be registered. On the other hand, the change of heat capacity at TC , Cp , decreases with decreasing the -ratio. Higher internal stresses correspond to lower heat capacity peak. This effect has been attributed to the dependence of TC calorimetric peak on the magnetostriction value of the magnetic phase as well as on the change of interatomic distances under applied outer stresses. Structural relaxation of microwires results in a shift of TC calorimetric peak of amorphous phase to higher temperatures, while crystallization leads to peak disappearance. Thus, precise DSC measurement can be a sensitive method for compositional and structural control of amorphous and nanocrystalline soft magnetic microwires. Acknowledgements This work was supported by the EU under project EM-safety, by ERA-NET program under Project “SoMaMicSens” (MANUNET2010), by Spanish Ministry of Science and Innovation, MICINN under Project MAT2010-18914 and by the Basque Government under Saiotek 09 Mic Magn project. A. Zh. and V. Zh. wish to acknowledge the support of the Basque Government under Program of Mobility of the Investigating Personnel of the Department of Education, Universities and Investigation under Grants MV2010-2-31 and MV-2009-2-24 respectively. References [1] A. Zhukov, V. Zhukova, Magnetic properties and applications of ferromagnetic microwires with amorphous and nanocrystalline structure, Nova Science Publishers, Inc., 400 Oser Avenue, Suite 1600, Hauppauge, NY 11788, 2009, p. 162. [2] D.C. Jiles, Acta Mater. 51 (2003) 5907. [3] J. Arcas, C. Gómez-Polo, A. Zhukov, M. Vázquez, V. Larin, A. Hernando, Nanostruct. Mater. 7 (8) (1996) 823–834. [4] C. Dudek, A.L. Adenot-Engelvin, F. Bertin, O. Acher, J. Non-Cryst. Solids 353 (2007) 925. [5] H. Chiriac, T.-A. Ova!ri, A. Zhukov, J. Magn. Magn. Mater. 254–255 (2003) 469–471. [6] J. Velázquez, M. Vazquez, A. Zhukov, J. Mater. Res. 11 (1996) 2499. [7] A.S. Antonov, V.T. Borisov, O.V. Borisov, A.F. Prokoshin, N.A. Usov, J. Phys. D: Appl. Phys. 33 (2000) 1161. [8] H. Fujimori, K.I. Arai, H. Shirae, H. Saito, T. Masumoto, N. Tsuya, Jpn. J. Appl. Phys. 15 (4) (1976) 705. [9] A. Zhukov, J. Magn. Magn. Mater. 242–245 (2002) 216–223. [10] V. Zhukova, A. Chizhik, A. Zhukov, A. Torcunov, V. Larin, J. Gonzalez, IEEE Trans. Magn. 38 (2002) 3090. [11] V. Zhukova, M. Ipatov, A. Zhukov, Sensors 9 (2009) 9216. [12] A. Zhukov, M. Ipatov, J. Gonzalez, J.M. Blanco, V. Zhukova, J. Magn. Magn. Mater. 321 (2009) 822–825. [13] A. Zhukov, Adv. Funct. Mater. 16 (2006) 675.
M. Churyukanova et al. / Journal of Alloys and Compounds 536S (2012) S291–S295 [14] V. Zhukova, S. Kaloshkin, A. Zhukov, J. Gonzalez, J. Magn. Magn. Mater. 249 (2002) 108–112. [15] S.D. Kaloshkin, I.A. Tomilin, Thermochim. Acta 280 (1996) 303–317. [16] E.V. Kaevitser, S.D. Kaloshkin, V.V. Tcherdyntsev, et al., J. Magn. Magn. Mater. 258 (2003) 539–542.
S295
[17] S. Kaloshkin, M. Churyukanova, V. Zadorozhnyi, I. Shchetinin, R.K. Roy, J. Alloys Compd. 509 (S1) (2011) 400–403. [18] J.M. Barandiarán, P. Gorria, I. Orúe, M.L. Fdez-Gubieda, F. Plazaola, A. Hernando, Phys. Rev. B 54 (1996) 3026–3029.