The substitution of Cu for Ni clearly effects on magnetic transitions, magnetocaloric effects and magnetic orders of these alloy ribbons.. That leads to great interest in researching mag
Trang 1Magnetic, magnetocaloric and critical properties of Ni 50x Cu x Mn 37 Sn 13
rapidly quenched ribbons
The-Long Phanc, Seong Cho Yuc,⇑, Nguyen Huy Dana,⇑
a
Institute of Materials Science, Vietnam Academy of Science and Technology, 18 Hoang Quoc Viet, Hanoi, Viet Nam
b Faculty of Physics, VNU University of Science, 334 Nguyen Trai, Hanoi, Viet Nam
c
Department of Physics, Chungbuk National University, Cheongju 361-763, South Korea
a r t i c l e i n f o
Article history:
Received 23 September 2014
Received in revised form 12 October 2014
Accepted 24 October 2014
Available online 1 November 2014
Keywords:
Magnetically ordered materials
Rapid-solidification
Phase transitions
Magnetocaloric
a b s t r a c t
Magnetic, magnetocaloric and critical properties of Ni50xCuxMn37Sn13 (x = 0, 1, 2, 4 and 8) rapidly quenched ribbons have been studied The substitution of Cu for Ni clearly effects on magnetic transitions, magnetocaloric effects and magnetic orders of these alloy ribbons With increasing the Cu-concentration, the martensitic–austenitic phase transition shifts to lower temperature, from 265 K (for x = 0) to 180 K (for x = 4), and disappears with x = 8, while the Curie temperature, TC, is almost unchanged Both the con-ventional and inverse magnetocaloric effects are observed The obtained values for the maximum inverse magnetic entropy changes, |DSm|max, of the ribbons are relatively large, 5.6 J kg1K1(for x = 0) and 5.4 J kg1K1(for x = 4) with the external magnetic field changeDH = 12 kOe The critical parameters (TC, b,cand d) of the ribbons are determined from the static magnetic data at the second order ferromag-netic–paramagnetic transition by using both the Arrott–Noakes and Kouvel–Fisher methods The results reveal that the long-range ferromagnetic order in the alloy is tendentiously dominated by increasing the Cu-concentration
Ó 2014 Elsevier B.V All rights reserved
1 Introduction
Magnetocaloric effect (MCE) is defined as the heating up or
cooling down a magnetic material by an applied magnetic field
Magnitude of MCE can be determined by direct measurement of
adiabatic temperature change (DTad) or indirect measurement of
magnetic entropy change (DSm)[1] Historically, MCE was first
dis-covered by Warburg[2]in 1881, basing on the temperature change
of iron in an applied magnetic field After that, the first MCE theory
and device were established by Bitter, Giauque and Mac Dougall
[3,4], who used the MCE of paramagnetic Gd2(SO4)38H2O salts to
achieve the temperature less than 1 K In 1997, the achievement
of the giant magnetocaloric effect (GMCE) in Gd–Si–Ge alloys
around 300 K [5] manifested application potential of magnetic
refrigeration technology at room temperature, which promised
for a new generation of solid refrigerant, energy-saving and
envi-ronmental protection chillers That leads to great interest in
researching magnetic materials possessing large magnetic phase
transitions around room temperature because of the closely
relationship of magnetic transitions with GMCEs Such kinds of the material include Gd-containing compounds (Gd–Ge–Si)[5,6], As-containing alloys (Mn–As)[7], La-containing alloys (La–Fe–Si)[8], Heusler alloys (Ni–Mn–Sn, Ni–Mn–Ga, Ni–Mn–In)[9–12]and fer-romagnetic perovskite manganites (La–Ca–Mn–O) [13,14] There are some material families which can exhibit both the first- and second-order magnetic phase transitions and thus both the inverse and conventional GMCEs can be respectively obtained Due to the coexistence of both the first-order and second-order magnetic transitions around room temperature, Ni–Mn–X (X = Sn, Sb, Ga and In) Heusler alloys have been attractive to research for GMCEs Besides that, so many theoretically and experimentally investiga-tions on these alloys have been carried out for other application potentials such as shape memory effect, giant magnetoresistance, and high spin polarization [15–18] Particularly, the interesting magnetic properties of nonstoichiometric Ni–Mn–Sn have been concentratedly researched because of exhibiting large value of adi-abatic temperature (DTad) and magnetic entropy (DSm) changes
[13,20] In recent years, the substituting some other elements such
as Ag, and Co for Ni or Mn on ferromagnetic Ni–Mn–Sn based Heusler alloys have been extensively studied to change magnetic interaction, Curie temperature and martensitic–austenitic phase transition[19–21], and to enhance magnetocaloric effect[21,22]
http://dx.doi.org/10.1016/j.jallcom.2014.10.126
0925-8388/Ó 2014 Elsevier B.V All rights reserved.
⇑ Corresponding authors.
E-mail addresses: scyu@chungbuk.ac.kr (S.C Yu), dannh@ims.vast.ac.vn
(N.H Dan).
Contents lists available atScienceDirect
Journal of Alloys and Compounds
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 / j a l c o m
Trang 2By using melt-spinning process to fabricate Ni–Mn–Sn based
Heus-ler alloys, the magnetization and magnetocaloric effect of samples
can be considerably improved[23–26]
In this work, we investigated magnetic, magnetocaloric and
critical properties of Ni50xCuxMn37Sn13(x = 0, 1, 2, 4 and 8)
rib-bons prepared by melt-spinning method
2 Experiment
Cu-doped Ni 50x Cu x Mn 37 Sn 13 pre-alloys (x = 0, 1, 2, 4 and 8) were initially
fab-ricated by arc-melting technique from pure elements (99.9%) of Ni, Cu, Mn and Sn in
argon environment The pre-alloys were turned over and arc-melted five times to
ensure their homogeneity These ingots were then melt-spun on a single roller
sys-tem with a velocity of the copper roller of 40 m/s to obtain alloy ribbons Thickness
of the ribbons is about 30lm Structure of the Ni 50x Cu x Mn 37 Sn 13 alloy ribbons was
examined by powder X-ray diffraction (XRD) technique using Cu K a radiation with
measuring step of 0.02° at room temperature Magnetic properties of the ribbons
were investigated by magnetization measurements on a vibrating sample
magne-tometer (VSM).
3 Result and discussion
X-ray diffraction patterns of Ni50xCuxMn37Sn13 (x = 0, 1, 2, 4
and 8) alloy ribbons are presented in Fig 1(a) All of the
samples exhibit two main (2 2 0) and (4 0 0) diffraction peaks of
(Ni,Cu)2MnSn phase corresponding to basic index of L21-austenitic
crystalline structure with Fm3m space group [22] In order to investigate the influence of Cu-concentration on the structure of the alloy, the change of lattice constant (a) and average size (d)
of crystallites was calculated by using Scherrer–Debye’s formula:
d ¼ kk
where d – size of the crystallites, k – wavelength of the X-ray radi-ation, h – Bragg angle, k – shape factor of 0.9 and b – peak width measured at half of maximum intensity
It is easily seen inFig 1(a) that the full width at half maximum of diffraction peak of all the Cu-doped (x = 1, 2, 4 and 8) samples is larger than that of the undoped (x = 0) one The results inFig 1(b) show that by substituting Cu for Ni, the lattice constant of
Ni50xCuxMn37Sn13 alloy is slightly raised up This probably is due to the larger lattice constant of Cu2MnSn crystalline phase (a = 6.17 Å) in comparison with that of Ni2MnSn phase (a = 6.05 Å)[27] The average crystallite size, which is calculated basing on the (2 2 0) diffraction peaks, decreases from 22.8 to 7.6 nm with increasing Cu-concentration from 0 to 8 at% The change of the lattice constant and average crystallite size might affect on magnetic properties of the alloy as presented below
Fig 2(a) exhibits hysteresis loops of the Ni50xCuxMn37Sn13
(x = 0, 1, 2, 4 and 8) ribbons at room temperature All the samples
Fig 1 (a) XRD patterns, (b) lattice constant and average crystallite size of
Ni 50x Cu x Mn 37 Sn 13 (x = 0, 1, 2, 4 and 8) alloy ribbons (the solid lines are to guide
Fig 2 (a) Hysteresis loops of Ni 50x Cu x Mn 37 Sn 13 (x = 0, 1, 2, 4 and 8) ribbons (the inset enlarges the typical loops at low magnetic field); (b) thermomagnetization curves of Ni 50x Cu x Mn 37 Sn 13 (x = 0, 1, 2, 4 and 8) ribbons measured in magnetic
Trang 3behave as the soft magnetic material with a low coercive force less
than 10 Oe (see inset ofFig 2(a)) InFig 2(b), the temperature
dependence of magnetization of the Ni50xCuxMn37Sn13 ribbons
with an applied magnetic field H = 12 kOe in the temperature
range from 100 K to 380 K is showed The thermomagnetization
curves of these samples reveal that there is an appearance of
mag-neto-structural phase transition, or martensitic–austenitic (M–A)
phase transition[10,25,28] The M–A phase transition temperature
is decreased from 265 K (x = 0) to 180 K (x = 4) with increasing
Cu-concentration As presented in [25], when increasing
Sn-concentration by 1%, the TM–Aof the Ni50Mn37xSn13+xalloy ribbon
is rapidly decreased from 265 to 165 K The substitution of Cu for
Ni in the Ni50xCuxMn37Sn13ribbon alloys makes the TM–Adecrease
more slowly from 265 to 180 with increasing Cu-concentration
from 0 to 4 at% Our obtained results can be promised to produce
Ni50xCuxMn37Sn13multi-layer refrigerant for applying to the
mag-netic refrigeration technology While the TM–A of the alloy is
strongly influenced by Cu-concentration, the Curie temperature
of the austenitic phase is almost unchanged (Fig 2(b)) This can
be explained by the dependence of the Curie temperature on the
exchange interaction in the materials In the Ni50xCuxMn37Sn13
alloy ribbons, the ferromagnetic exchange interaction of the
transi-tion metal atoms of Ni and Mn essentially decides value of their
Curie temperature The substitution of Cu for Ni almost does not
influence on the exchange interaction of Ni and Mn atoms resulting
in no large change of the Curie temperature of the austenitic phase
In order to calculate the magnetic entropy change (DSm) of the
alloy ribbons, the series of M(H) curves were determined at various
temperature around the first-order and second-order magnetic
phase transition temperatures (Fig 3(a)) TheDSmvalues of the
Ni50xCuxMn37Sn13 alloy ribbons were indirectly calculated from
M(H) data by using the following Maxwell’s relation:
DSm¼
Z H 2
H 1
@M
@T
H
Fig 4 shows DSm(T) curves with magnetic field change
DH = 12 kOe of the Ni50xCuxMn37Sn13(x = 0 and 4) ribbons Both
the inverse (IMCE) and conventional (CMCE) magnetocaloric
effects are revealed on the DSm(T) curves The peak of IMCE is
shifted to lower temperature from 264 K to 183 K with increasing
the Cu-concentration from 0 to 4 at%, and the maximum values of
the inverse magnetic entropy change, |DSm|max, are 5.6 J kg1K1
(for x = 0) and 5.4 J kg1K1(for x = 4) In accordance with Biswas
et al.[34], theDSmquantity follows the power lawDSm Hnwith
n 1 for Heusler alloys, our estimated |DSm|max values for the
Ni50xCuxMn37Sn13 (x = 0 and 4) alloy ribbons are greater than
20 J kg1K1 with magnetic field changeDH = 50 kOe Thus, our
materials have exhibited GMCE in comparison with that of the
Ni50Mn37Sn13 ingot (|DSm|max= 18 J kg1K1) [9]and the Gd5Si
2-Ge2 alloy (|DSm|max= 19 J kg1K1) [5] and had a potential for
application of magnetic refrigeration
As for CMCE, the peak ofDSm(T) is happened near room
temper-ature and just slightly changed with the variation of
Cu-concentra-tion Although the conventional |DSm|maxvalues are smaller, about
1.4 J kg1K1(for x = 0) and 1.2 J kg1K1(for x = 4), than those of
the inverse ones, the values of full width at half maximum (FWHM)
of the conventional entropy change peak are relatively large, about
39 K (for x = 0) and 30 K (for x = 4), in comparison with those of the
inverse ones The refrigerant capacity, RC, of the material can be
estimated by using the following relation:
The calculated values of the conventional and inverse RC (with
DH = 12 kOe) of the Ni50xCuxMn37Sn13alloy ribbons are 55 J kg1
and 28 J kg1 (for x = 0), 36 J kg1 and 22 J kg1 (for x = 4),
respectively
To clearly understand the magnetic orders at the second order phase transition (SOPT), the Arrott plots or M2–H/M plots are con-structed from M(H) data (Fig 3(b)) Because the ferromagnetic– paramagnetic transition at Curie temperature is a continuous
Fig 3 (a) M(H) curves of Ni 46 Cu 4 Mn 37 Sn 13 (x = 4) alloy ribbons around the first-order (dash lines) and second-first-order (solid lines) transitions; (b) Arrott plots for the
Ni 46 Cu 4 Mn 37 Sn 13 (x = 4) alloy ribbons.
Fig 4 DS m (T) curves of Ni 50x Cu x Mn 37 Sn 13 (x = 0 and 4) ribbons with magnetic field change DH = 12 kOe.
Trang 4phase transition, the power law dependence of spontaneous
mag-netization Ms(T) and inverse initial susceptibilityv1
0 (T) on reduced temperaturee= (T TC)/TCwith the set of critical exponents of b,c,
d, etc., can be determined by using the following Arrott–Noakes
rela-tions[29,30]:
MSðTÞ ¼ M0ðeÞb e<0; ð4Þ
v1
At Curie temperature TC, the exponent d is determined by the
relation of magnetization and applied magnetic field:
where M0, H0/M0and D are the critical amplitudes
The spontaneous magnetization Ms(T) and initial inverse
suscep-tibilityv1
0 (T) of the material can be obtained from constructing and
linearly fitting of Arrott plot of M2versus H/M at high magnetic field
The values of Ms(T) andv1
0 (T) as functions of temperature T are plot-ted for the Ni50xCuxMn37Sn13samples with x = 1 and 4 In accordance
with Eqs.(4) and (5)for Ms(T) andv1
0 (T), the power law fits are used to extract b,cand TC(Fig 5(a and b)) With the Ni49Cu1Mn37Sn13alloy
ribbon, the continuous power law fittings for Ms(T) and v1
0 (T) give the critical values of b = 0.442 ± 0.005, TC= 305.03 ± 0.48
and c= 1.183 ± 0.08, TC= 304.83 ± 0.07, respectively (Fig 5(a))
Similarly, for the Ni46Cu4Mn37Sn13 alloy ribbon, those values are b = 0.480 ± 0.011, TC= 314.19 ± 0.04 and c= 0.876 ± 0.011, TC= 314.22 ± 0.31 (Fig 5(b)) In comparison with some standard models such as mean field theory (b = 0.5,c= 1 and d = 3.0), 3D-Heisenberg model (b = 0.365, c= 1.336 and d = 4.8) and 3D-Ising model (b = 0.325, c= 1.241 and d = 4.82) [32], our critical parameters attained in this method fall between those of mean field and 3D-Heisenberg models However, the critical parameters of the alloy with x = 4 are closer to those of the mean field theory of long-range ferromagnetic orders
By using Kouvel–Fisher (KF) method[31], the critical exponents can be obtained more accurately From the critical exponents deter-mined from Arrott–Noakes relations, the plots of M1/bversus [H/ M]1/ccan be constructed at various temperature Again, the values
of Ms(T) and v1
0 (T) are determined from interception of linear extrapolation of these plots with M1/band [H/M]1/caxes, respectively The slopes of 1/b and 1/care determined by using these equations:
MsðTÞ½dMsðTÞ=dT1¼ ðT TCÞ=b ð7Þ
v1
0 ðTÞ½dv1
0 ðTÞ=dT1¼ ðT TCÞ=c ð8Þ
Fig 6shows the KF plots with the critical parameters obtained from fittings, b = 0.449 ± 0.033, TC= 305.67 ± 0.14 K (to Eq (7)) and c= 1.192 ± 0.025, TC= 305.46 ± 0.20 K (to Eq (8)) for x = 1,
Fig 5 Temperature dependence of spontaneous magnetization M s (T) and inverse
initial susceptibility v1
0 (T) along with fittings to Arrott–Noakes relations for
Ni Cu Mn Sn ribbons with (a) x = 1 and (b) x = 4.
Fig 6 Kouvel–Fisher plots for spontaneous magnetization M s (T) and inverse initial susceptibilityv1 (T) of the samples with (a) x = 1 and (b) x = 4.
Trang 5and b= 0.489 ± 0.018, TC= 314.07 ± 0.24 K (to Eq (7)) and
c= 0.881 ± 0.021, TC= 314.06 ± 0.13 K (to Eq (8)) for x = 4 We
can realize that these critical parameters are in good agreement
with those obtained from Arrott–Noakes method The value of b
parameter of the Ni49Cu1Mn37Sn13sample is still smaller than that
of Ni46Cu4Mn37Sn13sample
The value of d parameter can be obtained by fitting isothermal
magnetization at T TCto Eq.(3) From linear fitting the log–log
plots of M(H) at T = 305 K for x = 1 and T = 314 K for x = 4, the d
val-ues are respectively determined as 2.857 and 3.584 The Wildom
scaling Eq.(9)can be also used to evaluate the accuracy of fitting
process[33]:
From the values of b andcof our above obtained results, values
of d are calculated to be 2.802 and 3.660 for x = 1 and 4,
respec-tively The values of d obtained from the two methods are quite
in good agreement
The reliability of the obtained critical parameters can be verified
by using the static-scaling theory, which predicts that the
isother-mal magnetization is a universal function of e and H:
Mjejb¼ fðHjejðbþcÞÞ ð10Þ
Here, f+for T > TCand ffor T < TCare regular analytical
func-tions From the values of b andc, which we obtained above, the
static-scaling plots of M/ebversus H/eb+cin log scale are constructed
inFig 7(a) and (b) for x = 1 and 4, respectively The two parts of the plots at T > TCand T < TCare separate The falling of data on the two separated branches exhibits characteristic of continuous phase transitions and proves the reliability of our achieved critical parameters
The obtained results can be used to explain the variation of the M–A phase transition on the Cu-doped samples As presented
in Ref [25], the Ni50Mn37Sn13 ribbon shows short-range ferromagnetic orders (b = 0.385 ± 0.035) of the ferromagnetic exchange interaction By substituting Cu for Ni in alloy, the TM–Aof the Ni50xCuxMn37Sn13 alloy shifts to lower temperature with increasing x That means the austenitic phase with long-range ferromagnetic orders is enhanced by Cu-concentration This might
be due to the stronger covalent hybridization between the d-states
of Cu and Mn atoms in comparison with that of Ni and Mn atoms[27] Therefore, ferromagnetic exchange interaction of the
Ni50xCuxMn37Sn13 alloy ribbons is changed from short-range to long-range by substituting Cu for Ni As a brief summary of the influence of Cu-concentration on the Ni50xCuxMn37Sn13 rapidly quenched ribbons, the obtained parameters of the structure and properties are listed inTable 1
4 Conclusion
By substituting Cu for Ni of Ni50xCuxMn37Sn13(x = 0, 1, 2, 4 and 8) ribbons, the lattice constant is slightly increased with increasing Cu-concentration, while the average crystalline size are strongly decreased All the alloy ribbons exhibit soft magnetic materials with coercive force less than 10 Oe With increasing Cu-concentra-tion, the Curie temperature, TC, of the alloy almost unchanges, whereas the temperature of martensitic–austenitic transition,
TM–A, considerably decreases The magnetocaloric effects are relatively large at both the TCand TM–A The austenitic phase with long-range ferromagnetic order is enhanced by Cu-concentration
Acknowledgements
This work was supported by Vietnam Academy of Science and Technology under Grant number of VAST03.04/14-15 and the Con-verging Research Center Program through the Ministry of Science, ICT and Future Planning, Korea (2014048835) A part of the work was done in the Key Laboratory for Electronic Materials and Devices, and Laboratory of Magnetism and Superconductivity, Institute of Materials Science, Vietnam
References
[1] V.K Pecharsky, K.A Gschneidner, A.O Pechasky, A.M Tishin, Phys Rev B 64 (2001) 144406
[2] K.A Gschneidner, V.K Pecharsky, Annu Rev Mater Sci 30 (2000) 387 [3] W.F Giauque, D.P MacDougall, Phys Rev 43 (1933) 768
Fig 7 The static-scaling performances of Ni 50x Cu x Mn 37 Sn 13 ribbons with (a) x = 1
Table 1 Influence of Cu-concentration (x) on lattice constant (a), average crystallite size (d) coercive force (H c ), martensitic–austenitic transition temperature (T M–A ), maximum inverse magnetic entropy change (|DS m | max ) and critical parameters (T C , b,c, d) of the
Ni 50x Cu x Mn 37 Sn 13 rapidly quenched ribbons.
a (Å) 5.994 6.000 6.000 6.015 6.005
d (nm) 22.8 10.2 7.7 7.5 7.6
T M–A (K) 265 231 204 180 –
|DS m | max (J kg 1
K 1
Trang 6[5] V.K Pecharsky, K.A Gschneidner, Phys Rev Lett 78 (1997) 4494
[6] V Provenzano, A.J Shapiro, R.D Shull, Nature 429 (2004) 853
[7] S Gama, A.A Coelho, A Campos, A.M.G Carvalho, F.C.G Gandra, P.J Ranke, N.A.
Oliveira, Phys Rev Lett 93 (2004) 237202
[8] J Lyubina, K Nenkov, L Schultz, O Gutfleisch, Phys Rev Lett 101 (2008)
177203
[9] T Krenke, E Duman, M Acet, E.F Wassermann, X Moya, L Manosa, A Planes,
Nat Mater 4 (2005) 450
[10] T Krenke, M Acet, E.F Wassermann, X Moya, L Mañosa, A Planes, Phys Rev.
B 72 (2005) 014412
[11] A Aliev, A Batdalov, S Bosko, V Buchelnikov, V Khovailo, I Dikshtein, V.
Koledov, R Levitin, V Shavrov, T Takagi, J Magn Magn Mater 272–276
(2004) 2040
[12] P.A Bhobe, K.R Priolkar, A.K Nigam, Appl Phys Lett 91 (2007) 242503
[13] M.H Phan, H.X Peng, S.C Yu, J Magn Magn Mater 316 (2007) 562
[14] X Moya, L.E Hueso, F Maccherozzi, A.I Tovstolytkin, D.I Podyalovskii, C.
Ducati, L.C Phillips, M Ghidini, O Hovorka, A Berger, M.E Vickers, E Defaÿ,
S.S Dhesi, N.D Mathur, Nat Mater 12 (2013) 52
[15] I Dubenko, M Khan, A.K Pathak, B.R Gautam, S Stadler, N Ali, J Magn Magn.
Mater 321 (2009) 754
[16] V.V Khovaylo, K.P Skokov, O Gutfleisch, H Miki, T Takagi, T Kanomata, V.V.
Koledov, V.G Shavrov, G Wang, E Palacios, J Bartolomé, R Burriel, Phys Rev.
B 81 (2010) 214406
[17] W.R Branford, S.K Clowes, M.H Syed, Y.V Bugoslavsky, S Gardelis, J.
Androulakis, J Giapintzakis, C.A.E Grigorescu, A.V Berenov, S.B Roy, L.F.
Cohen, Appl Phys Lett 84 (2004) 13
[18] R.A Groot, F.M Mueller, P.G Engen, K.H.J Buschow, Phys Rev Lett 50 (1983) 25
[19] S.K Bose, J Kudrnovsky, V Drchal, I Turek, Phys Rev B 82 (2010) 174402 [20] S Roy, E Blackburn, S.M Valvidares, M.R Fitzsimmons, S.C Vogel, M Khan, I Dubenko, S Stadler, N Ali, S.K Sinha, J.B Kortright, Phys Rev B 79 (2009)
235127 [21] V.K Sharma, M.K Chattopadhyay, A Khandelwal, S.B Roy, Phys Rev B 82 (2010) 172411
[22] D.H Wang, C.L Zhang, H.C Xuan, Z.D Han, J.R Zhang, S.L Tang, B.X Gu, Y.W.
Du, J Appl Phys 102 (2007) 013909 [23] B Hernando, J.L Sánchez Llamazares, J.D Santos, M.L Sanchez, L.L Escoda, J.J Sunol, R Varga, C Garcıa, J Gonzalez, J Magn Magn Mater 321 (2009) 763 [24] D Wua, S Xuea, J Frenzel, G Eggelerc, Q Zhai, H Zheng, Mater Sci Eng A 534 (2011) 568
[25] T.L Phan, P Zhang, N.H Dan, N.H Yen, P.T Thanh, T.D Thanh, M.H Phan, S.C.
Yu, Appl Phys Lett 101 (2012) 212403 [26] J.D Santos, T Sanchez, P Alvarez, M.L Sanchez, J.L Sánchez Llamazares, B Hernando, J Appl Phys 103 (2008) 07B326
[27] J Kubler, A.R Williams, C.B Sommers, Phys Rev B 28 (1983) 1745 [28] S Aksoy, M Acet, P.P Deen, L Mañosa, A Planes, Phys Rev B 79 (2009)
212401 [29] A Arrott, J.E Noakes, Phys Rev Lett 19 (1967) 786 [30] S Rößler, U.K Rößler, K Nenkov, D Eckert, S.M Yusuf, K Dörr, K.H Müller, Phys Rev B 70 (2004) 104417
[31] J.S Kouvel, M.E Fisher, Phys Rev 136 (1964) A1626 [32] H.E Stanley, Oxford University Press, 1971.
[33] B Widom, J Chem Phys 41 (1964) 1633 [34] Anis Biswas, T.L Phan, N.H Dan, P Zhang, S.C Yu, H Srikanth, M.H Phan, Appl Phys Lett 103 (2013) 162410