Physica B 476 (2015) 118–121
Contents lists available at ScienceDirect
Physica B journal homepage: www.elsevier.com/locate/physb
Magnetic and structural properties of (Ru1 xCox)2FeSi alloys Bhargab Deka, A. Srinivasan n Department of Physics, Indian Institute of Technology Guwahati, Guwahati 781039, India
art ic l e i nf o
a b s t r a c t
Article history: Received 20 November 2014 Received in revised form 16 February 2015 Accepted 5 March 2015 Available online 6 March 2015
In this work, we report a systemetic study of Co substitution for Ru in Ru2FeSi. Our previous studies showed that Ru2FeSi and Co2FeSi are antiferromagnetic and ferromagnetic, respectively. In order to understand the influence of Co substitution for Ru, (Ru1 xCox)2FeSi alloys were prepared by arc melting, followed by annealing at 1273 K for 3 days. Structural and magnetic studies were carried out on the quaternary alloys by powder X-ray diffraction and vibrating sample magnetometry, respectively. At room temperature, Ru rich compositions i.e., x ¼0 and x ¼ 0.25 exhibited disordered B2 structure, but with increase in Co concentration L21 ordering appeared in the alloys. From magnetization measurements, it is seen that with increasing x, the ferromagnetic state becomes dominant with the appearance of spontaneous magnetization and increase in the value of magnetization of the alloys. Saturation magnetization measured at 15 kOe increased from 1.02 emu/g to 149.96 emu/g as x was increased from 0 to 1. & 2015 Elsevier B.V. All rights reserved.
Keywords: Heusler alloys Antiferromagnet Ferromagnet Arc melting
1. Introduction Heusler alloys, a large family of inter-metallic compounds, have attracted considerable attention since the pioneering work by Heusler in 1903 [1]. They are categorized into two distinct groups: half Heusler alloys with chemical formula of XYZ exhibiting C1b structure and full Heusler alloys with chemical formula of X2YZ exhibiting L21 structure (where X and Y are transition metals and Z is an element with sp valence electrons) [2–4]. The stable L21 structure of full Heusler alloys consists of four interpenetrating 1 1 1 FCC sub-lattices with Wyckoff positions of (0,0,0), ( 4 , 4 , 4 ), 1
1
1
3
3
3
( 2 , 2 , 2 ) and ( 4 , 4 , 4 ). This family of alloys offers the unique possibility of studying diverse magnetic phenomena like ferromagnetism, ferrimagnetism, antiferromagnetism, Pauli paramagnetism etc., in the same family of alloys [4]. Prediction of halfmetallic behavior first in Ni–Mn–Sb [3] and subsequently in Co2MnSi [5,6] have intensified the interest in these alloys. Halfmetallic ferromagnets are characterized by coexistence of metallic behavior for one spin direction and semiconducting behavior with energy band gap for other spin direction. This novel electronic structure results in 100% spin polarization at the Fermi level. The total spin magnetic moment per formula unit for X2YZ alloys is given by the Slater–Pauling (S–P) relation,
Mt = (Zt − 24) μ B
n
Corresponding author. Fax: þ91 361 2582749. E-mail addresses:
[email protected] (B. Deka),
[email protected] (A. Srinivasan). http://dx.doi.org/10.1016/j.physb.2015.03.006 0921-4526/& 2015 Elsevier B.V. All rights reserved.
(1)
where Zt is the total number of valence electrons in the unit cell [7]. This relationship provides a way for identifying new halfmetallic alloys suited for fabricating spintronic devices. In this context, Co2YZ and Fe2YZ based alloys have received a lot of attention because of their high Curie temperature and high magnetic moment as well as theoretically predicted half-metallic nature [8– 17]. Among these systems, Co2FeSi alloy has been shown to be ferromagnetic with high (1015 K) Curie temperature (TC) [17]. Discovery of half-metallicity in Heusler alloys has induced interest in the development of spin valve based giant magneto-resistance devices using Heusler alloy based electrode layers. However, such spin valves invariably use, an antiferromagnetic (AFM) Ir–Mn layer to pin one of the ferromagnetic layers [18,19]. AFM Ru2FeSi with Neel temperature (TN) of 280 K [20] provides hope to obtain Heusler alloys with high TN for fabricating an all Heusler spin valve with superior interfacial properties. Substitution of Co for Ru offers a means of obtaining FM and AFM materials in the same quaternary system. In this paper, we report the crystallographic and magnetic properties of bulk quaternary (Ru1 xCox)2FeSi alloys with 0 rx r1.
2. Experimental (Ru1 xCox)2FeSi (0 rx r1) alloy ingots were prepared by arc melting appropriate amounts of high purity (Z99.99%) elements in argon atmosphere. The ingots were then taken in sealed fused silica ampoules evacuated to 10 3 Pa and annealed at 1273 K for 3 days and quenched in ice water. Crystal structure of the alloys was determined with a powder X-ray diffractometer (Rigaku
B. Deka, A. Srinivasan / Physica B 476 (2015) 118–121
119
TTRAX III) emitting Cu Kα X-rays (λ ¼0.15406 nm). Magnetic properties were measured using a vibrating sample magnetometer (VSM, Lakeshore 7410). The overall composition of the alloys was verified using an energy dispersive spectrometer (EDS, Oxford) attached to a scanning electron microscope (SEM, Leo 1430 VP) and was found to be within 1% of the nominal composition.
3. Results and discussion Room temperature X-ray diffraction (XRD) patterns of (Ru1 xCox)2FeSi alloys are shown in Fig 1. A fully ordered Heusler (X2YZ) alloy with L21 structure gives Bragg reflections with nonzero structure factor when all indices are either even or odd [21,22]. The structure factor of the first three reflections are given by the relations
(
) (
F (111) = 4 f y − fz , ⎡ ⎤ F (200) = 4 ⎣2fx − f y + fz ⎦,
)
F (220) = 4 ⎡⎣2fx + f y + fz ⎤⎦
(2)
where fx, fy and fz are the average scattering amplitudes for respective sub-lattices. Absence of peaks with all odd indices in the XRD patterns indicates a transition from L21 to B2 structure due to intermixing of Y and Z elements. Degree of atomic ordering in the aloys can be estimated from the ratio of the intensity (Ihkl) of the super-lattice reflections from (111) and (200) planes using the relations
S = ((I200/220 )E /(I200/I220 )T )1/2
and
(1 − 2α) S = ((I111/220 )E /(I111/I220 )T )1/2
(3)
where the suffixes E and T denote data obtained experimentally (as depicted in Fig. 1) and data theoretically generated XRD patterns for the L21 unit cell. According to Webster and Ziebeck [23], S ¼1 and α ¼ 0 signify perfect L21 ordering in the Heusler alloy. It is obvious from Fig. 1 that (111) reflection is not present in the XRD patterns of alloys with x ¼0 and x ¼0.25. The estimated value of α is 0.5 for these two alloys, which suggests complete B2 structure with 50% of the Y (or Z) site is occupied by Z (or Y) atom. With further increase in Co substitution, (111) reflection appears in the XRD patterns of the alloys signifying appearance of L21 order in the alloys. α estimated for the alloys with x ¼0.5, 0.75 and 1.00 are 0.08, 0.06 and 0.03 respectively, which indicates 8%, 6% and 3% B2 disorder in the respective alloys. As x is increased, the peaks shift towards higher 2θ values. XRD patterns were further refined by Rietveld method using FullProf 2.00 software. The lattice
Fig. 1. Room temperature XRD patterns of (Ru1 xCox)2FeSi alloys with 0r xr 1.
Fig. 2. Variation of lattice constant (a) and effective anisotropy constant (Keff) as a function of Co (x) substitution in (Ru1 xCox)2FeSi
constant a obtained from the Rietveld analysis of the cubic samples with x¼0, 0.25, 0.50, 0.75 and 1.00 are 5.88 Å, 5.83 Å, 5.76 Å, 5.70 Å and 5.63 Å, respectively. Linear variation in lattice constant with x indicates that Co atoms effectively substitute for Ru atoms in the lattice (Fig. 2). A continuous contraction of the lattice occurs with an increase in x because the atomic radius of Co (¼1.67 Å) is smaller than that of Ru (¼1.89 Å). Lattice constant of the end members of this alloy system i.e. Ru2FeSi and Co2FeSi are 5.88 Å and 5.63 Å, respectively, which is in agreement with earlier experimental observations [13,20]. Fig. 3 shows the thermo-magnetization (M–T) curves for the alloy with x¼0 recorded under an applied field of 1.5 kOe. The alloy shows AFM behavior with a peak in magnetization data at 270 K which can be associated with its Neel temperature (TN). Measured TN is in close agreement with earlier reported value of 280 K [20] for this alloy. As Co concentration is increased from x¼ 0.25 to x¼ 1.00, magnetization value in the M–T measurement at the same applied field starts increasing due to increased ferromagnetic interaction (not shown). TC of the alloys is well above the ambient temperature. Fig. 4 shows the field dependence of magnetization (M–H curves) at room temperature for all the samples. Since TN for Ru2FeSi is 270 K, M–H loop recorded at 260 K is shown in the inset to illustrate the weak magnetism exhibited by this alloy in the AFM state. Magnetization shows linear field dependence for the alloy with x¼ 0, confirming its AFM nature. With increase in x, spontaneous magnetization appears in the alloys. Magnetization at an applied field of 15 kOe increases from 1.02 emu/g to 149.96 emu/g as x is increased from 0 to 1.00. In Ru2FeSi with B2 structure, the spacing between the (111) layers which contain both Fe and Si atoms is 1.70 Å (¼a/2√3). Ru atoms occupy (000) and 111 ( 2 2 2 ) positions which are placed in between the (111) layers. Fe atoms occupying lattice points in (111) layers are coupled ferromagnetically within the layer. However, the alloy behaves as
Fig. 3. Thermo-magnetization (M–T) curve of Ru2FeSi alloy recorded at 1.5 kOe.
120
B. Deka, A. Srinivasan / Physica B 476 (2015) 118–121
eventually results in a decrease in the total magnetic moment of the alloys [25]. A theoretical estimate [25] showed that 12.5% DO3 type disorder in the alloy with x ¼1 (Co2FeSi) can reduce its Ms value of the predicted value of 6.0 μB to 5.5 μB. Hence, Ms of 5.42 μB obtained for the current sample suggests that about 13% of DO3 type disorder can be present in this alloy. Increase in the deviation of Ms from S–P rule with increasing x indicates an increase of DO3 type disorder in the alloys with increasing x. Field dependence of magnetization of a ferromagnetic material can be expressed in terms of the law of approach to saturation as [26,27]
⎛ c ⎞ ⎟ M (H) = Ms ⎜1 − ⎝ H2 ⎠ Fig. 4. M–H curves of (Ru1 xCox)2FeSi alloys with 0 rx r 1.
antiferromagnetic due to the antiferromagnetic coupling between the adjacent (111) layers [20,21]. For Co substitution for Ru up to 25%, B2 structure is still preserved in this alloy system. But the smaller atomic radius of Co as compared to Ru results in reducing this (111) layer spacing up to 1.68 Å. This leads to a collapse of the antiferromagnetic coupling between the (111) layers and induces ferromagnetic interaction in the alloy. With further increase in Co concentration, L21 order develops in the alloys. In the L21 structure, (111) layers containing only Fe atoms are sandwiched between a (111) layer of Si atoms and the Fe layers are coupled ferromagnetically. The change in the interlayer spacing with increased substitution of Co for Ru thus results in strengthening the ferromagnetic interaction between the (111) Fe layers. It can be seen from Fig. 4 that the M–H curves for the alloys with x¼0.50, 0.75 and 1.00 are completely saturated. Room temperature initial magnetization (M–H) curves obtained for these three alloy compositions are depicted in Fig. 5. Saturation magnetization (Ms) for the alloys with x ¼0.50, 0.75 and 1.00 are 4.99 μB (5.00 μB), 5.45 μB (5.50 μB) and 5.42 μB (6.00 μB), respectively, where the values within brackets are the values predicted by the S–P rule (Eq. (1)). Thus, the measured Ms values are lower than the predicted values and this deviation from the S–P rule increases with increase in x. The probable reason for the lower Ms could be the presence of a small amount of DO3 type disorder in the alloy due to swapping of some Co (8c) and Fe (4b) atoms from their designated Wycoff sites in the lattice. Nearly equal scattering factors of Co and Fe makes this disorder hard to detect from XRD studies [24]. Due to this disorder, magnetic moment of Fe atom occupying 8c site decreases from 3.16 μB to 2.45 μB, which is only partly compensated by the moderate increase from 1.54 μB to 1.95 μB in the moment of Co atom occupying 4b site. This
(4) 2
where Ms is the saturation magnetization and the term c/H is attributed to the magneto-crystalline anisotropy. The effective anisotropy constant (Keff) for a cubic compound is described by the relation [28]
⎛ 105c ⎞1/2 ⎟ Keff = μ0 M s ⎜ ⎝ 8 ⎠
(5)
Keff values for the alloys with x¼0.50, 0.75 and 1.00 calculated from the M–H data are depicted in Fig. 5 and listed in Table 1. Keff increases with increase in Co concentration indicating larger magnetic anisotropy in Co rich alloys (Fig. 2). Magneto-crystalline anisotropy of Co (5.30 105 J m 3) is higher than that of Fe (0.48 105 J m 3) [29]. As discussed above, DO3 type disorder can be present in the alloys with x 40.25. Since this disorder is associated with the distribution of magnetic Fe and Co atoms in the alloy, an increase in DO3 disorder with increasing Co substitution can lead to magnetic anisotropy. Relatively larger percentage of DO3 disorder in the alloy with x ¼1 is probably responsible for the unusually large Keff exhibited by this alloy. When compared to other bulk Heusler alloys such as Ni50Mn30Ga20 (1.01 105 J m 3) [27] and Co43Ni22Ga29Fe6 (2.2 105 J m 3) [30], Keff value for this alloy system is higher. The high Keff observed in the bulk samples can increase further in nanogranular state and may find application in magnetic media.
4. Conclusions Bulk (Ru1 xCox)2FeSi alloys have been prepared by arc melting method. Alloys with x ¼0.50, 0.75 and 1.00 crystalize in L21 structure, whereas, alloys with x ¼0 and 0.25 exhibit B2 disorder. Antiferromagnetism of Ru2FeSi gets destroyed when Co is substituted for Ru atoms. Due to substitution of Ru with the smaller Co atom, the AFM interaction of nearest (111) layers of Fe collapses and ferromagnetism develops in these alloys. M–H measurements reveal that all the alloys exhibit low coercivity (18.5–22.0 Oe) and small hysteresis loop area (6.03 103–14.35 103 erg/g). Ms for Table 1 Lattice constant (a), Neel temperature, (TN), Curie temperature (TC), saturation magnetic moment (Ms), effective anisotropy constant (Keff) data obtained for (Ru1 xCox)2FeSi alloys.
Fig. 5. Initial magnetization curves of (Ru1 xCox)2FeSi alloys with 0.50 rx r 1.
x
a [Å]
TN [K], TC [K]
Ms
0 0.25 0.50 0.75 1.00
5.88 5.83 5.76 5.70 5.63
270 (TN) – – – 1015 (TC)a
– – 5.00 5.50 6.00
(S–P)
[μB]
Ms
(300 K)
[μB]
– – 4.99 7 0.002 5.45 7 0.002 5.42 7 0.002
Keff [ 105 J m 3] – – 4.02 7 0.07 6.107 0.05 11.177 0.06
a TC of Co2FeSi is beyond the measurement range of 300 K. Cited value is taken from Ref. [17].
B. Deka, A. Srinivasan / Physica B 476 (2015) 118–121
the alloys with x ¼0.50 and x ¼0.75 follows S–P rule but for x ¼1.00 deviates from it. These studies show that alloys with AFM as well as ferromagnetic interactions of different strengths could be prepared in this alloy system. Ferromagnetic alloys exhibited relatively high magnetic anisotropy as compared to other bulk Heusler alloys.
Acknowledgment Financial assistance from the Council of Scientific and Industrial Research, India vide Project no: 03(1236)/12/EMR-II and the Directorate of Extramural Research and Intellectual Property Rights, Defence Research and Development Organization, India vide Project no: ERIPR/ER/I00392/Ml/01/1381 are gratefully acknowledged.
References [1] F. Heusler, Verh. Dtsch. Phys. Ges. 5 (1903) 219. [2] P.J. Webster, K.R.A. Ziabeck, J. Phys. Chem. Solids 34 (1973) 1647. [3] R.A. de Groot, F.M. Mueller, P.G. van Engen, K.H.J. Buschow, Phys. Rev. Lett. 50 (1983) 2024. [4] I. Galanakis, P.H. Dederichs, Half-Metallic Alloys: Fundamentals and Applications, Springer, Berlin, 2005. [5] S. Ishida, S. Fujii, S. Kashiwagi, S. Asano, J. Phys. Soc. Jpn. 64 (1995) 2152. [6] S.F. Cheng, B. Nadgorny, K. Bussmann, E.E. Carpenter, B.N. Das, G. Trotter, M. P. Raphael, V.G. Harris, IEEE Trans. Magn. 37 (2001) 2176. [7] I. Galanakis, P.H. Dederichs, N. Papanikalou, Phys. Rev. B 66 (2002) 174429.
121
[8] V. Niculescu, T.J. Burch, K. Rai, J.I. Budnick, J. Magn. Magn. Mater. 5 (1977) 60. [9] K.H.J. Buschow, P.G. van Engen, J. Magn. Magn. Mater. 25 (1981) 90. [10] K.H.J. Buschow, P.G. van Engen, R. Jongebreur, J. Magn. Magn. Mater. 38 (1983) 1. [11] M.Z.E. Bruck, F.R. de Boer, Z. Li, G. Wu, J. Phys. D: Appl. Phys. 37 (2004) 2049. [12] R.Y. Umetsu, K. Kobayashi, A. Fujita, K. Oikawa, R. Kainuma, K. Ishida, N. Endo, K. Fukamichi, A. Sakuma, Phys. Rev. B 72 (2005) 214412. [13] S. Wurmehl, G.H. Fecher, H.C. Kandpal, V. Ksenofontov, C. Felsera, H. Lin, Appl. Phys. Lett. 88 (2006) 032503. [14] T.M. Nakatani, A. Rajanikanth, Z. Gercsi, Y.K. Takahashi, K. Inomata, K. Hono, J. Appl. Phys. 102 (2007) 033916. [15] K.R. Kumar, K.K. Bharathi, J.A. Chelvane, S. Venkatesh, G. Markandeyulu, N. Harishkuma, IEEE Trans. Magn. 45 (2009) 10. [16] R.Y. Umetsu, A. Okubo, R. Kainuma, J. Appl. Phys. 111 (2012) 073909. [17] B. Deka, D. Chakraborty, A. Srinivasan, Physica B 448 (2014) 173. [18] K. Nikolaev, P. Kolbo, T. Pokhil, X. Peng, Y. Chen, T. Ambrose, O. Mryasov, Appl. Phys. Lett. 94 (2009) 222501. [19] T.M. Nakatani, N. Hase, H.S. Goripati, Y.K. Takahashi, T. Furubayashi, K. Hono, IEEE Trans. Magn. 48 (2012) 1751. [20] S.N. Mishra, D. Rambabu, A.K. Grover, P.N. Tandon, J. Magn. Magn. Mater. 51 (1985) 359. [21] A. Szytula, H. Ptasiewicz-Bak, J. Leciejewicz, J. Magn. Magn. Mater. 80 (1989) 195. [22] B. Deka, R. Das, A. Srinivasan, J. Magn. Magn. Mater. 347 (2013) 101. [23] P.J. Webster, K.R.A. Ziebeck, J Phys. Chem. Solids 34 (1973) 1647. [24] B. Balke, S. Wurmehl, G.H. Fecher, C. Felser, C. Maria, M. Alves, F. Bernardi, J. Morais, Appl. Phys. Lett. 90 (2007) 172501. [25] Z. Gercsi, K. Hono, J. Phys. Condens. Matter 19 (2007) 326216. [26] Z.Q. Jin, W. Tang, J.R. Zhang, H.X. Qin, Y.W. Du, Eur. Phys. J. B 3 (1998) 41. [27] C. Jiang, Y. Muhammad, L. Deng, W. Wu, H. Xu, Acta Mater. 52 (2004) 2779. [28] X.Y. Xiong, T.R. Finlayson, B.C. Muddle, J. Phys. D: Appl. Phys. 34 (2001) 2845. [29] H.P. Wijn, Magnetic Properties of Metals, Springer, Berlin, 1991. [30] S. Sarma, A. Srinivasan, AIP Conf. Proc. 1347 (2011) 111.