Accepted Manuscript

Magnetic properties and magnetization reversal in Co nanowires with different

morphology

Hongjian Li, Qiong Wu, Yi Peng, Huanhuan Xu, Jiuxing Zhang, Ming Yue

PII: S0304-8853(18)30902-8
DOI: https://doi.org/10.1016/j.jmmm.2018.08.063
Reference: MAGMA 64260

To appear in: Journal of Magnetism and Magnetic Materials

Received Date: 26 March 2018

Revised Date: 14 August 2018
Accepted Date: 22 August 2018

Please cite this article as: H. Li, Q. Wu, Y. Peng, H. Xu, J. Zhang, M. Yue, Magnetic properties and magnetization
reversal in Co nanowires with different morphology, Journal of Magnetism and Magnetic Materials (2018), doi:
https://doi.org/10.1016/j.jmmm.2018.08.063

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 Magnetic properties and magnetization reversal in Co nanowires

with different morphology

Hongjian Li, Qiong Wu, Yi Peng, Huanhuan Xu, Jiuxing Zhang, Ming Yue*
College of Materials Science and Engineering, Beijing University of Technology, Beijing, 100124, China

Abstract
In the present study, magnetic properties and demagnetization processes of Co
nanowires with various morphologies (ellipsoids, capped cylinders, and cylinders) are
investigated by means of Object Oriented Micromagnetic Framework (OOMMF)
software package with finite difference micromagnetic simulation. The results show
that the aspect ratio, morphology, and dipolar interactions of the nanowires play a key
role in determining the switching modes and hence the magnetic properties. For the
Co nanowires with the same morphology and diameter, the coercivity increases
substantially with increasing aspect ratio of the nanowires due to the increased shape
anisotropy when the aspect ratio is less than 3. For a single nanowire with a fixed
diameter of 15 nm and a length of 200 nm, the coercivity of ellipsoids of 14.45 kOe is
noticeably higher than those of capped cylinders of 12.55 kOe and cylinders of 11.15
kOe, indicating the considerable influence of the morphology on the coercivity of the
wires. The magnetization reversal mechanism of a single Co nanowire is mainly
described by a nucleation propagation process, which starts at the ends of the
nanowires. For the nanowires assembly, the magnetic moment reversal of some
nanowires occurs first, and the rest reverse gradually with increasing applied field and
ultimately the reversal of the whole assembly is achieved. This process could be
theoretically interpreted based on the nucleation of closure domains at the ends of the
wires as well as the subsequent depinning and propagation of domain walls.
Meanwhile, the squareness of the demagnetization curve and coercivity of the
nanowires assembly deteriorate faster than those of the single nanowires due to the
dipole interactions.
Keywords: Micromagnetic simulations; Co nanowires; packing density; morphology
Corresponding author: Ming Yue, E-mail: [email protected]

1
Introduction
Rare earth permanent magnetic materials are at the heart of modern technology from
electric motors to wind turbines[1-4]. Present high-performance permanent magnets
are mainly neodymium-based and samarium-based alloys[5, 6], but the shortage of
resource supplies and high cost of rare earths limits their applications[7]. Therefore,
development of new permanent magnetic materials with less or without rare-earth
elements is urgently needed. In this regard, various rare earth-free materials have been
explored in recent years, especially the polyol process has been applied to prepare
mono-dispersed magnetic nanowires and nanorods of 3d transition metals and their
alloys, which have some advantages such as high saturation magnetization and high
Curie temperatures compared to the rare earth compounds[8-10]. For 3d metals and
alloys with weak magnetocrystalline anisotropy, development of shape anisotropy to
achieve high coercivities has long been considered as a feasible but a challenging
route. Through calculations, the maximum values of coercivity resulting from the
shape anisotropy of the particles are 10.7 kOe and 8.9 kOe for iron and cobalt,
respectively[11]. And on this basis a lot of studies have shown the static magnetic
properties and magnetization reversal processes in a single and arrays of
nanowires[12-14]. In practice, however, magnets with magnetic hardening via shape
anisotropy always exhibited far less coercivities than the predicted values. For
example, elongated iron and iron-cobalt particles prepared by the electroplating
method had optimum coercivities of less than 1 kOe[15]. Recently, several novel
techniques have been developed to fabricate Fe-based and Co-based nanowires and
nanorods[10, 16, 17]. Among them, the cobalt nanowires with high aspect ratios are
the best candidates to develop a new generation of permanent magnets based on their
shape anisotropy. Owing to the superposition of the shape anisotropy and
magnetocrystalline anisotropy, Co nanowires with the easy magnetization direction
parallel to their geometric long axis are expected to exhibit both high coercivity and
good squareness in their demagnetization curves. In the future the application of these
suitable properties of Co nanowires might be realized so that the high energy product
magnets can be competitive with SmCo or NdFeB permanent magnets. Gandha et al.
2
prepared Co nanowires via a solvothermal chemical process with high room
temperature coercivity of 12.5 kOe, which even went far beyond the
magnetocrystalline anisotropy field of 8.9 kOe for infinitely long Co nanowires [18].
This new development demonstrated the feasibility of developing high coercivities via
a combination of shape anisotropy and crystal anisotropy and opened up the window
for a rare earth free magnet system with strong energy density. Nevertheless, how
much improvement is left regarding to the coercivity of Co nanowire is yet unknown.
In addition, the Co nanowires with various sizes and aspect ratios may exhibit a
different mode of magnetization reversal.
The micromagnetic simulation is one useful path to find the desired properties in
materials, which allow a relatively easy way to understand the properties of materials
compared with experimental work[19]. Previous theoretical studies have shown that
the tips play a key role on the magnetic properties of nanowires[20, 14, 21]. The
experimental results also show that the morphology has a great effect on the magnetic
performance[22-25]. However, the understanding of the magnetization reversal
mechanism and magnetic properties of nanowires with various morphologies still
remains ambiguous.
Meanwhile, characterization and understanding of the magnetic properties of
nanowire assemblies have been researched for decades[26, 27]. These studies shows
that the magnetostatic interactions among nanowires play an important role in the
magnetic properties of the nanowire assemblies. However, some problems remain
unclear. For example, the understanding of coercivity in terms of the dipolar
interactions has remained poorly quantitative. In addition, there are few reports on
investigating the morphology dependence on the magnetic properties and
magnetization reversal mechanisms of the nanowire assembly.
In the present study, by means of micromagnetic simulation, the magnetic
properties of Co nanowires (single wire and wire assembly) with various
morphologies were investigated quantitatively, and the magnetization reversal process
of Co nanowires with various morphologies was studied systematically. The results
show that the aspect ratio, morphology, and dipolar interactions of the nanowires play
3
key roles in determining the switching modes and the magnetic properties of the wires.
Optimized structural and magnetic properties are illustrated based on calculations.

Method
In order to understand the magnetization reversal process, it is important to
estimate the static and dynamic magnetization characteristics of the nanowires. The
Object Oriented Micromagnetic Framework (OOMMF) software package with finite
difference micromagnetic simulation was used for theoretical cognizance. The details
of this simulation method have been described in previous papers[28, 29].
The effect of the morphology on magnetic properties and magnetization reversal
processes in single and assemblies of Co nanowires were investigated by means of
OOMMF micromagnetic simulations[30]. The key issue for micromagnetic
simulation is based on the experimental results. Depending on the chemical synthesis
conditions, the kinetic control of the growth allows to modify the morphologies of the
nanowires[31, 32]. In a more realistic view, these morphologies will be the basis for
our numerical modeling. For micromagnetic simulations, the modes of nanowires
morphology were changed by varying the morphology of nanowire tips while the
middle part exhibited the regular diameter along their length. The schematics of the
nanowires with different morphologies which were used for micormagnetic
simulations, are shown in Scheme 1. For the single Co nanowire model the diameter
(D) is set as 15 nm and the maximum length (L) is 200 nm. For the hexagonal Co
nanowires assembly model, the diameter (D) and the length (L) are the same with that
of the single Co nanowire model, with the distance spacings (V) of 20 nm for 16
nanowires.
In the experiment, each wire is a single crystal with the c-axis (002) along the
long axis of the wire[31]. Thus, a large external field is applied along the long axis of
wires and the magnetic field sweeps from +20 kOe to -20 kOe in steps of 0.1 kOe.
However, in the practical experiment, the reversal mechanism will also influenced by
the unavoidable deviation of applied magnetic field. The coercivity and the
remanence decrease monotonously with the increase of easy axis deviation, indicating
4
the large impact of the easy axis orientation on the magnetic performance. The
magnetization reversal processes with various applied field deviations from easy axis
is also quite different from that with the applied field parallel to the easy axis. More
details and discussions for the magnetization reversal processes in magnetic
nanowires depend upon the direction of applied magnetic field can be found in Ref 33.
The parameters L, D, and AR = L/D are the length, diameter, and the aspect ratio of
the nanowire, respectively. Based on the finite difference method of OOMMF, to
obtain accurate results with appropriate calculation times, the single nanowire has
been divided into many 0.5 nm×0.5 nm×1 nm cells, while cell sizes of 0.5 nm×0.5
nm×0.5 nm as the aspect ratio of single nanowire is less than 3, meanwhile the
nanowire assemblies have been divided into 1nm×1 nm×2 nm cells. The material
parameters are adjusted as saturation magnetization Ms = 17.86 kGs; crystalline
anisotropy constant K1 = 0.54 × 107 erg·cm-3; exchange constant A = 20 × 10-7
erg·cm-1, following ref. 34 and 35.

Scheme1 The schematics of the nanowires with different morphologies which were used for

micormagnetic simulations, (a) cylinders, (b) capped cylinders and (c) ellipsoids.

Results and Discussion

1. Dependence of aspect ratio on the magnetic properties of single cylinders Co
nanowires
5
Fig.1. (a) The simulated second quadrant of hysteresis loops of single cylinder Co nanowires

with different aspect ratios (AR) where D = 15 nm. Inset shows the amplified

demagnetization curves near nucleation position; (b) The coercivity of Co nanowires with

different aspect ratio (AR) where D = 15 nm. Inset shows the corresponding maximum

energy product. The analytical coercivity, calculated based on Stoner-Wohlfarth (SW) theory,

is shown together for comparison.

Fig.1 (a) shows the calculated second quadrant of hysteresis loops of the single
cylinders cobalt nanowires with various aspect ratios (AR), where the diameters of
nanowires (D) are kept at a constant value of 15 nm which is close to the typical size
of the cobalt nanowires prepared by chemical processes. The hysteresis loops for
different aspect ratios are nearly rectangular with a remanence magnetization of 17.86
kGs which is equal to the saturation magnetization, causing a higher energy product.
The maximum energy product was calculated from the M (H) curve with different
aspect ratios as shown in the Fig.1 (b) inset. The obtained maximum energy products
(BH)max, with aspect ratio AR = 1, is 76.97 MGOe. As the aspect ratio is increased to
10, the energy product increases to 79.66 MGOe which is close to the highest
theoretical value of 79.74 MGOe. However, the packing density and dipolar
interactions of the bulk magnet reduce the energy product potentially. Therefore, in
the latter section, we performed micromagnetic simulations of Co nanowire
assemblies. In addition, fig.1 (b) analyzes the dependence of the coercivity on the
aspect ratio of the nanowires. The coercivity is strongly affected by the aspect ratio of
the Co nanowires with the ratios less than 3, the coercivity rises quickly with

6
increasing aspect ratio due to the increasing shape anisotropy[25]. While no
significant increase of the coercivity has been observed for aspect ratio higher than 3
because the dependence on the aspect ratio is suppressed, which shows good
agreement with previous investigations[22]. Thus, the coercivity values are
substantially smaller than expected from the SW model. But the dependence of
coercivie field on the aspect ratio is similar to the SW model.

Fig. 2 (a-d) The magnetic moment distribution of the top, middle and bottom layer of the Co

nanowire and (e-h) the average angle between the magnetic moment and the long axis of the

nanowires (AR = 10, D = 15 nm) when H = 0 kOe (remanence state), -10.6 kOe (nucleation
7
point) , -11.1 kOe (near the coercivity) and -11.2 kOe (after the coercivity). The simulated

second quadrant of hysteresis loops of the Co nanowires as showed in Fig.1 (a).

Fig. 2 shows the magnetic moment distribution and average angular distributions
in the Co nanowire with D = 15 nm and AR = 10, where θ is the angle between the
magnetic moment and the applied field. We define the nucleation field as the applied
field at which the magnetic moment reversal starts at the two ends of the nanowires,
and define the coercive field as the field at which the magnetization is equal to zero.
All the magnetic moments are along the length direction of the nanowire in a large
positive applied field, corresponding to a coherent state with θ = 0°. Such a coherent
state does not change until H = 0 kOe, as shown in Fig. 2(a) and Fig. 2(d), when the
magnetic moment at the end of the nanowires deviates slightly about 5.28°.
Meanwhile the magnetic moment in the middle part (along length direction of the
nanowire) maintains the initial state. Therefore, the remanence is close to the
saturation magnetization. After this zero point, the ends gradually reverse with
increasing reverse field, which leads to an obvious deviation of the magnetic moments.
Once the applied field H reaches to -10.6 kOe [Fig. 2(b) and Fig. 2(e)], the average θ
in the end increases to 8.71°, while the magnetic moments of the rest of the nanowire
retain their previous state. For example, the magnetic moment in the middle point of
the wires deviates less than 1° compared with its remanence state. At H = -11.1 kOe
[Fig. 2(c) and Fig. 2(f)], a considerable deviation occurs in the end of the nanowire.
At this point near the coercivity, the end and the middle angles are 13.77o and 0.76°,
respectively. Further increases of the applied field results in the gradually complete
reversal of the two ends. The magnetic moment in the central region of the nanowire
reverses drastically and coherently at about H = -11.2 kOe.
The above analysis show that the magnetic moment deviation starts from the
ends of the nanowires and forms two nucleation centers in the corners. Because of the
nearest neighbor exchange interaction effects, the pioneering nucleation points
propagate into their neighborhoods. As the applied field transfers to a negative
direction, more magnetic moments deviate from the previous state. Near the
coercivity point, the deviation of the magnetic moments reaches the most remarkable
8
state. A slight increase of the applied field leads to a drastic magnetic moment reversal
and helps the system to pass the coercivity point. After the magnetization reversal, all
the magnetic moments are oriented at the current applied field direction with θ = 180°.
Therefore, the magnetization reversal mechanism of single a Co nanowire is mainly
described by a nucleation propagation process, which starts at the two ends of the
nanowire[36, 13, 37].
2. Dependence of morphology on the magnetic properties of single Co nanowire

Fig.3. Hysteresis loops (a), B-H curves, and energy products (b) of the experimental [38] and

simulated different morphology Co nanowires with the field applied parallel to the long axis.

Fig.3 (a) shows the experimental and simulated different morphology Co

nanowires with the field applied parallel to the long axis. In the experiment, Co
nanowires with mean diameters of 15 nm and mean length of 200 nm were obtained,
which provides an ultrahigh coercivity of 12.5 kOe at room temperature [31], which is
the highest value reported so far for Co nanowires. The morphology of Co nanowires
is cylindrical with rounded tip (capped cylinders) and ellipsoidal tip (ellipsoids)
prepared via a solvothemal chemical process. Such a high coercivity is comparable to
that of rare earth magnets, demonstrating the application potential of 3d metallic
nanowires as hard magnetic materials. In particular, the saturation magnetization (Ms)
is 16.5 kGs, which beats most rare-earth permanent magnets. The remanent
magnetization ratio Mr/Ms is 0.92 and the loop squareness is 0.78, leading to a
maximum energy product [(BH)max] of 44.0 MGOe. In order to understand the effects
of the different morphologies on magnetic properties and magnetization processes of

9
Co nanowires, the OOMMF micromagnetic simulation was used. For these Co
nanowires with different morphologies (cylinders, capped cylinders and ellipsoids),
we set the maximum diameter of 15 nm and the length of 200 nm with the magnetic
field applied parallel to the nanowire axis. The magnetization curves of these objects
exhibited square hysteresis loops when the magnetic field was applied parallel to the
nanowires with remanence close to saturation. Impressively, nanowires with the same
aspect ratio and diameter show very different coercivities. Our results strongly
confirm that once the aspect ratio of the nanowire is larger than 10, the morphology of
the nanowires, instead of the length, plays a key role [20]. The coercivity of ellipsoids
is the highest, followed by capped cylinders and cylinders [39]. The difference of
coercivity confirms the importance of the edge morphology and suggests that the
reversal mechanism of these objects might be different. Meanwhile, the remanence is
not well matched, owing to the less orientation and uniformity morphology of the
magnetic measurement sample. In addition, the hysteresis loop of the single nanowire
exhibits a typical square loop whereas the hysteresis loop of the experimental data
exhibits a worse squareness. The B (H) curves were calculated from the M (H) curve,
and the maximum of the energy product for different morphology nanowires was
taken into account. As shown in the Fig.3 (b), for all these objects, the maximum
energy products are 79.66 MGOe, close to the theoretical upper limit of energy
product in Co nanowires. The experimentally obtained maximum energy product[38]
is 44.0 MGOe which is estimated based on a 100 percent volume fraction of closely
packed Co nanowires using the theoretical density of 8.92 g/cm3. The (BH)max value
is dependent on the shape of the magnetization loop, which is directly related to the
alignment and the magnetic volume fraction of the nanowires. The difference in
maximum energy product between a single Co nanowire and the experimental data
also demonstrates that the dipolar interactions between the nanowires cannot be
neglected.

10
Fig. 4 The magnetic moment distribution in the different states of the demagnetization

process of (a) single cylinders, (b) capped cylinders and (c) ellipsoids of Co nanowires for

magnetic field applied parallel to the nanowire axis with maximum diameter D = 15 nm and

length L = 200 nm. The color scale represents the value of each component of the normalized

magnetization.

In order to further understand the mechanism of magnetization reversal in these

different objects, the demagnetization processes of Co nanowires with different
morphology were investigated. Fig. 4 shows the magnetic moment distribution in the
different states of demagnetization process of Co nanowires with various
morphologies. The external field is applied along the direction of the wires’ axes. For

11
ellipsoidal and capped cylindrical nanowires, the magnetic moment distribution
variation tendency is in good agreement with the cylindrical nanowires. The magnetic
moment deviation starts from the ends of the nanowires to form two nucleation
centers, and the rest of the wires remain mostly homogeneously magnetized. This
incoherent magnetization reversal is mainly due to the non-uniform distribution of the
demagnetizing field. However, the reversal of magnetic moment of ellipsoidal Co
nanowires requires a larger external field than cylinders, which explains why
ellipsoidal nanowires have the highest coercivities.

3. Dependence of dipolar interactions on the magnetic properties of Co nanowire

assemblies

Fig.5. Hysteresis loops of nanowires assembly with different morphologies (D =15 nm, L =

200 nm, V = 20 nm and N = 16). The inset shows the coercivity of single nanowire and

nanowire assemblies with the same diameter and length (D = 15 nm and L = 200 nm).

It is well known that the dipolar interactions considerably affect the magnetic
behavior of the nanowires assembly[19]. Therefore, the dipolar interactions were
taken into account in this theoretical simulation process. For the present model, we set
16 Co nanowires which have 200 nm in length and 15 nm in diameter. According to
the experimental observation[38], the distance spacing of 20 nm is selected for this
simulation. Fig.5 shows the simulated magnetic properties of Co nanowires assembly

12
with different morphology with the field applied is parallel to the long axis. For the
cylindrical nanowire assembly, the effective dipolar interactions among the nanowires
have a strong impact on the macroscopic magnetic behavior of the system[40]. Clear
steps and plateaus can be found in the demagnetization curve of nanowire assembly,
which shows that the dipolar interactions among the wires have a significant influence
on the switching field. This process could be theoretically interpreted based on the
nucleation of closure domains at the ends of the wires as well as the subsequent
depinning and propagation of domain walls[41]. These dipolar interactions are the
same for the ellipsoidal and capped cylindrical nanowires assemblies. It is
straightforward to extract the coercivity from the calculated hysteresis loops. The
inset in Fig. 5 shows the comparison of the coercivity of a single Co nanowire with
that of an assembly of nanowires with the same morphology. Compared with a single
nanowire, the low coercivity and poor squareness of the assembly of nanowires are
the result of dipolar interactions[42]. Meanwhile, for cylindrical nanowire assembly,
the packing density is 35.33%, which means the magnetic volume fraction of the
nanowire assembly[26]. The (BH)max of the cylindrical nanowires assembly is 26.03
MGOe, which is close to the experimental value of SmCo permanent magnets[7],
indicating that these Co nanowires can potentially be ideal bulk materials for high
performance bonded magnets.

13
Fig. 6 Magnetic moment distribution of the Co nanowires assembly (D = 15 nm, L = 200 nm,

V = 20 nm and N = 16) with different morphologies (a) cylinders, (b) capped cylinders and (c)

ellipsoids for magnetic field applied parallel to the nanowire axis in remanence state, before

and after the coercivity. The color scale represents the value of each component of the

normalized magnetization.

14
 To get an understanding of the magnetization reversal mechanisms of Co
nanowires assembly, the magnetic moment distribution during the reversal process is
investigated. Fig.6 shows the magnetic moment distribution states of the different
morphology nanowires of the 4×4 hexagonal assembly in its remanence state, before
and after coercive field. The magnetization reversal mechanism is different from that
in a single wire. As shown in Fig.6 (a), some cylindrical nanowires switch in the
direction of the external field, while others remain magnetized antiparallel to the field.
This denotes that the external field is not strong enough to switch the whole nanowire
assembly. The reversal of some nanowires occurs because the stray field of
neighboring cylindrical nanowires adds to the external field and leads to a higher field
where the magnetic moments are effectively exposed. In this way, the moment
reversal of some nanowires occurs first. However, these reversed nanowires generate
a stray field opposite to the external field and reduce the local field. Therefore, they
need a higher external field to compensate the dipolar contribution. With increasing
external field, those non-reversed nanowires gradually reverse, and finally finish the
reversal of the whole assembly [27, 43-45]. For ellipsoidal and capped cylindrical
nanowires assembly, the variation in the magnetic moment distribution are in good
agreement with the cylindrical nanowires assembly. However, due to the effect of the
various morphologies of the nanowire assembly, the deflection degree of the magnetic
moment in the two ends of the cylindrical is much larger than that of the ellipsoidal,
when H = 0 kOe. In addition, for ellipsoidal nanowire assembly, a larger external field
is needed to achieve the same magnetization reversal state of the cylindrical nanowire
assembly.
Conclusions
The magnetic properties and demagnetization processes of Co nanowires with
various morphologies have been investigated by means of micromagnetic simulation.
For the Co nanowires with the same morphology and diameters, the coercivity
increases substantially with increasing aspect ratio of the nanowires due to the
increased shape anisotropy when the aspect ratio is less than 3. For a single nanowire
with fixed diameter and length, the coercivity of ellipsoids is noticeably higher than
15
those of capped cylinders and cylinders, indicating the considerable influence of the
morphology on the coercivity of the wires. The magnetization reversal mechanism of
single Co nanowire is mainly described by a nucleation propagation process, which
starts at the ends of the nanowires. For the nanowires assembly, the magnetization
reversal process could be theoretically interpreted based on the nucleation of closure
domains at the two ends of the wires as well as the subsequent depinning and
propagation of the domain wall. Meanwhile, the squareness of the demagnetization
curve and coercivity of the nanowires assembly deteriorates faster than those of the
single nanowires due to the dipole interactions. In the future, the application of these
properties of Co nanowires might be realized so that the high energy product magnets
can be competitive with SmCo or NdFeB permanent magnets. The challenges to get a
new generation of permanent magnets are how to synthesize ferromagnetic nanowires
with optimized structural and magnetic properties.

Acknowledgment
This work was supported by the National Natural Science Foundation of China
(51401001, 51701109, and 51331003) and International S&T Cooperation Program of
China (2015DFG52020).
References
[1] Rong C B, Wu Y Q, Wang D P, Zhang Y, Poudyal N, Kramer M J and Liu J P 2012 J. Appl. Phys. 111
07A717
[2] Rong C B, Wang D, Nguyen V V, Daniil M, Willard M A, Zhang Y, Kramer M J and Liu J P 2013 J. Phys.
D: Appl. Phys. 46 045001
[3] Li H L, Li X H, Guo D F, Lou L, Li W and Zhang X Y 2016 Nano Lett. 16 5631
[4] Li H L, Lou L, Hou F C, Guo D F, Li W, Li X H, Gunderov D V, Sato K and Zhang X Y 2013 Appl. Phys.
Lett. 103 142406
[5] Liu Z W, Huang H Y, Gao X X, Yu H Y, Zhong X C, Zhu J and Zeng D C 2011 J Phys D Appl Phys. 44
025003
[6] Poudyal N and Liu J P 2013 J Phys D Appl Phys. 46 043001
[7] Li X, Lou L, Song W, Huang G, Hou F, Zhang Q, Zhang H T, Xiao J, Wen B and Zhang X 2017 Adv.
Mater. 29
[8] Piraux L, Dubois S, Ferain E, Legras R, Ounadjela K, George J M, Maurice J L and Fert A 1997 J.
Magn. Magn. Mater. 165 352
[9] Han G C, Zong B Y and Wu Y H 2002 IEEE Trans. Magn. 38 2562
[10] Maurer T, Ott F, Chaboussant G, Soumare Y, Piquemal J Y and Viau G 2007 Appl. Phys. Lett. 91
16
172501
[11] Kittel C 1949 Rev. Mod. Phys. 21 541
[12] Dmytriiev O, Al-Jarah U A S, Gangmei P, Kruglyak V V, Hicken R J, Mahato B K, Rana B, Agrawal M,
Barman A, Matefi-Tempfli M, Piraux L and Matefi-Tempfli S 2013 Phys. Rev. B. 87 174429
[13] Pal S, Saha S, Polley D and Barman A 2011 Solid State Commun. 151 1994
[14] Bance S, Fischbacher J, Schrefl T, Zins I, Rieger G and Cassignol C 2014 J. Magn. Magn. Mater. 363
121
[15] Luborsky F E 1961 J. Appl. Phys. 32 S171
[16] Anagnostopoulou E, Grindi B, Lacroix L M, Ott F, Panagiotopoulos I and Viau G 2016 Nanoscale. 8
4020
[17] Cattaneo L, Franz S, Albertini F, Ranzieri P, Vicenzo A, Bestetti M and Cavallotti P L 2012
Electrochimica Acta. 85 57
[18] Gandha K, Elkins K, Poudyal N, Liu X and Liu J P 2014 Sci. Rep. 4 5345
[19] Bance S, Fischbacher J, Schrefl T, Zins I, Rieger G and Cassignol C 2014 J. Magn. Magn. Mater. 363
121
[20] Zighem F and Mercone S 2014 J. Appl. Phys. 116 193904
[21] Ott F, Maurer T, Chaboussant G, Soumare Y, Piquemal J Y and Viau G 2009 J. Appl. Phys. 105
013915
[22] Pousthomis M, Anagnostopoulou E, Panagiotopoulos I, Boubekri R, Fang W Q, Ott F, Atmane K A,
Piquemal J Y, Lacroix L M and Viau G 2015 Nano Research. 8 2231
[23] Soumare Y, Garcia C, Maurer T, Chaboussant G, Ott F, Fievet F, Piquemal J Y and Viau G 2009 Adv.
Funct. Mater. 19 1971
[24] Mercone S, Zighem F, Leridon B, Gaul A, Schoenstein F and Jouini N 2015 J. Appl. Phys. 117
203905
[25] Shirato N, Sherrill S, Gangopadhyay A K and Kalyanaraman R 2016 J. Magn. Magn. Mater. 407 328
[26] Panagiotopoulos I, Fang W Q, Ott F, Boue F, Ait-Atmane K, Piquemal J Y and Viau G 2013 J. Appl.
Phys. 114 143902
[27] Vázquez M and Vivas L G 2011 physica status solidi (b). 248 2368
[28] Zhang W, Zhao G P, Yuan X H and Ye L N 2012 J. Magn. Magn. Mater. 324 4231
[29] Yuan X H, Zhao G P, Yue M, Ye L N, Xia J, Zhang X C and Chang J 2013 J. Magn. Magn. Mater. 343
245
[30] Donahue M J P, Donald Gene. OOMMF User's guide. US Department of Commerce, Technology
Administration, National Institute of Standards and Technology; 1999.
[31] Liu Q Y, Cao X F, Wang T J, Wang C G, Zhang Q and Ma L L 2015 Rsc Advances. 5 4861
[32] Li Y, Liu Q and Shen W 2011 Dalton Trans. 40 5811
[33] Rana B, Ganguly A, and Barman A 2011 IEEE Trans. Magn.47 2859
[34] Vila L, Darques M, Encinas A, Ebels U, George J M, Faini G, Thiaville A and Piraux L 2009 Phys. Rev.
B. 79 172410
[35] Zhang X C, Zhao G P, Xia J, Yue M, Yuan X H and Xie L H 2014 Chinese Physics B. 23 097504
[36] Suarez O J, Perez L M, Laroze D and Altbir D 2012 J. Magn. Magn. Mater. 324 1698
[37] Ivanov Y P, Vazquez M and Chubykalo-Fesenko O 2013 J Phys D Appl Phys. 46 485001
[38] Gandha K, Mohapatra J and Liu J P 2017 J. Magn. Magn. Mater. 438 41
[39] Ke L Q, Skomski R, Hoffmann T D, Zhou L, Tang W, Johnson D D, Kramer M J, Anderson I E and
Wang C Z 2017 Appl. Phys. Lett. 111 022403

17
[40] Hertel R 2001 J. Appl. Phys. 90 5752
[41] Sampaio L C, Sinnecker E H C P, Cernicchiaro G R C, Knobel M, Vazquez M and Velazquez J 2000
Phys. Rev. B. 61 8976
[42] Gao J H, Sun D L, Zhan Q F, He W and Cheng Z H 2007 Phys. Rev. B. 75 064421
[43] Zhong K H, Huang Z G, Che Z G, Feng Q and Yang Y M 2007 Trans. Nonferrous Met. Soc. China. 18
[44] Knobel M, Sampaio L C, Sinnecker E H C P, Vargas P and Altbir D 2002 J. Magn. Magn. Mater. 249
60
[45] Velázquez J, G , Vázquez M and Hernando A 1999 J. Appl. Phys. 85 2768

18