Magnetic
nanoparticles are widely used in modern technologies, such as magnetic
recording of high density information, production of permanent magnets and
magnetic fluids, microelectronics, etc. [1, 2]. Recently, ensembles of magnetic
nanoparticles have been used with great success in biomedicine, magnetic
resonance imaging, targeted drug delivery, purification of biological media
from toxins, magnetic hyperthermia, to fight malignant neoplasms, and in many
other areas [3-6].
Meanwhile, physics of
ensembles of magnetic nanoparticles is very complex [7–11]. The physical
properties of an ensemble are determined by a large combination of geometric
and magnetic parameters, such as particle size and shape, saturation
magnetization, type of magnetic anisotropy, and values of
magnetic anisotropy constants. An ensemble density, viscosity, and temperature
of the medium in which magnetic nanoparticles are distributed are also of great
importance. For better understanding a complex behavior of an ensemble of
magnetic nanoparticles under various conditions, it is useful to first study an
evolution of a magnetic moment of an isolated single-domain magnetic
nanoparticle when an external magnetic field changes. This behavior
substantially depends on the type of effective magnetic anisotropy of the
nanoparticle [7–9].
Visualization of
complex physical processes is often used as one of the ways to study phenomena
occurring at the atomic level [12, 13]. For example visualization method is
used for studies of physical processes in nanostructures in [12]. Visualization
of complex, multidimensional numerical data allows us to understand the
processes taking place in the studied objects. Therefore, a graphic
representation of the various types of magnetic anisotropy that exist in
magnetic nanoparticles makes it easier for students and young researchers to
understand the essence of complex physical processes.
This article is
devoted to the development of a graphical representation of an energy density
of magnetic anisotropy of single-domain magnetic nanoparticles. A change in the
total energy of a magnetic nanoparticle under the influence of an external
uniform magnetic field is also considered. All images shown in this article
were obtained using Matplotlib data visualization package in Python programming
language.
As is known
[7–9], in a ferromagnetic sample, a modulus of a local magnetization vector
is
constant and equal to the saturation magnetization of a ferromagnet,
. The
direction of this vector is described by a unit magnetization vector,
. In most cases,
single-domain magnetic nanoparticles, which size is smaller than the diameter
of a single-domain particle
D
<
Dc,
are most interesting
for applications. A single-domain nanoparticle in the lowest energy state is
uniformly magnetized. Such a particle is a small permanent magnet, which
creates a sufficiently strong inhomogeneous magnetic field in the surrounding
space. The density of magnetic energy of a uniformly magnetized particle is
high, because it is proportional to [7–9] the square of the saturation
magnetization,
wm
~
Ms2.
However,
for a single-domain spherical nanoparticle, a magnetic energy density does not
depend on the direction of a single magnetization vector due to symmetry in a
distribution of magnetic charges on the surface of the sphere. At the same
time, experiment shows that in equilibrium, in absence of an external magnetic
field, the unit magnetization vector of a single-domain spherical nanoparticle
is oriented in strictly defined directions with respect to the symmetry axes of
the crystal lattice of the particle. These distinguished spatial directions are
called axes of easy magnetization of a nanoparticle (also, easy axes of
anisotropy of a particle). Since an exchange energy of a uniformly magnetized
particle is equal to zero, and a magnetic energy of a spherical nanoparticle
does not depend on the direction of a single magnetization vector, the
directions of easy anisotropy axes are determined by the form of the energy
density of magnetic crystallographic anisotropy,
[7-11].
For crystals
with a single axis of symmetry, it follows from general considerations [7–11]
that the energy density of magnetic crystallographic anisotropy can be
represented as an expansion in powers of a unit magnetization vector
,
|
(1)
|
where, for simplicity, it is assumed
that selected axis of symmetry of a crystal is parallel to the z axis of a
Cartesian coordinate system. In equation (1),
K1,
K2,
etc. there are uniaxial magnetic anisotropy constants, which usually decrease
in absolute value, |K1| > |K2|. From
equation (1) it follows that in the case
K1,K2
> 0 an energy density of uniaxial magnetic anisotropy has the smallest
value,
wa
= 0,
if
the unit magnetization vector is parallel to the z axis, i.e
or
. Thus,
for a particle with anisotropy energy density (1), these directions of the unit
vector are preferable.
On the other
hand, from symmetry considerations it follows [7–11] that for crystals with
cubic symmetry of the crystal lattice, an energy density of magnetic
crystallographic anisotropy can be specified as
,
|
(2)
|
When
K1c
> 0,
K2c
»
0 the
directions of easy magnetization of a particle with cubic anisotropy are
parallel to axes of a Cartesian coordinate system, since the magnetic
anisotropy energy (2) is minimal,
wa
= 0, when the unit
magnetization vector is parallel to the x, y, or z axes, i.e.
, etc. Thus, in
this case, a particle has 6 free easy magnetization. If a constant
K1c
< 0, then directions of easy magnetization are parallel to diagonals of a
cube, and the magnetic anisotropy energy (2) in this case has a minimum for the
vector form
.
It should be
noted, however, that expressions (1), (2) are only the simplest energy
contributions that determine a direction of the axes of easy magnetization of a
nanoparticle in space. Another important contribution to the effective magnetic
anisotropy is related to the difference between the shape of the particle and
the spherical one. This contribution is called the shape anisotropy energy
[7–9]. It is especially important for magnetically soft type nanoparticles with
a sufficiently high saturation magnetization. A famous Braun – Morrish theorem
[12] claims that a magnetostatic energy of a uniformly magnetized particle of
arbitrary shape, to a first approximation, coincides with a magnetostatic
energy of some equivalent ellipsoid. If axes of a Cartesian coordinate system
are chosen along the symmetry axes of the equivalent ellipsoid, then a magnetic
energy density of a nanoparticle will take the form [14]
,
|
(3)
|
where
Nx,
Ny,Nz
are demagnetizing factors of an equivalent ellipsoid in an
indicated coordinate system.
Obviously,
demagnetizing factors of a cube-shaped particle are equal to each other due to
symmetry,
Nx
=
Ny
=
Nz
=
4
p
/3.
Given that
Nx
=
Ny
=
Nz
=
4
p
/3, it
is easy to see that the magnetic energy of a uniformly magnetized cube, like
the magnetic energy of a spherical nanoparticle, does not depend on the
direction of a single magnetization vector and does not contribute to the
magnetic anisotropy of the particle. On the other hand, for a typical case of
an elongated spheroid, for which the transverse demagnetizing factors are
equal,
Nx
=
Ny
>
Nz,
the magnetic energy of the particle (3) can be written as
,
|
(4)
|
We note here that an addition of an
arbitrary constant to an energy density of magnetic anisotropy obviously does
not affect the direction of the light axes of the particle anisotropy, and does
not affect the dynamics of the unit magnetization vector [7–9]. This situation
is similar to a case of classical mechanics, where the potential energy of a
particle is given up to an arbitrary constant, because a force acting on the
particle is determined by a gradient of potential energy.
Comparing
equations (1) and (4), we can conclude that an elongated equivalent spheroid
can be characterized by an effective shape anisotropy constant,
.
Moreover, both the direction of the easy axis of anisotropy and the values
of the effective demagnetizing factors
Nx
and
Nz
are determined [14] by the real shape of the non-ellipsoidal nanoparticle. A
total energy of a magnetic anisotropy of a particle in a general case is the
sum of an energy of magnetocrystallographic anisotropy and a shape anisotropy
energy. In this case, it takes place [15] an effective combined magnetic
anisotropy of a nanoparticle,
.
In a presence of
a uniform external magnetic field
H, a single-domain nanoparticle
remains uniformly magnetized. In this case, stationary, time-independent,
directions of a unit magnetization vector are determined by a minima of a total
energy of a particle in an external magnetic field
,
|
(5)
|
Naturally, a position of minima and
maxima of a total energy of a single-domain nanoparticle and their number
changes with a change in a magnitude and direction of the external magnetic
field with respect to direction of easy axes of anisotropy of a particle.
In order to
clearly demonstrate a location of minima and maxima of an energy density of magnetic
anisotropy, it is convenient to set this energy in a spherical coordinate
system [7],
, by
expressing in equations (1) - (5) components of a unit magnetization vector
through spherical angles (q, j), so
that
,
,
. Then
in a spherical coordinate system it is possible to construct a certain surface
of a form
.
Fig.
1. An energy density of magnetic anisotropy for uniaxial and cubic types of magnetic
anisotropy of a nanoparticle.
Figure 1a shows a reduced energy
density of uniaxial magnetic anisotropy,
, equation (1),
for a case
K1
> 0,
K2
= 0. Obviously, an
energy density of magnetic anisotropy has deep minima for directions of a unit
magnetization vector close to the angles
q
= 0,
p
. In this case, it is said that a
particle has two deep potential wells separated by a high symmetric potential
barrier, which maximum corresponds to the angle
q
=
p
/2.
Figure 1b shows a
reduced energy density of uniaxial magnetic anisotropy (1) for a case
K1
< 0,
K2
= 0. To exclude negative radius values
in a dependence
, a corresponding positive constant is
added to equation (1). As Fig. 1b, in a case under consideration, a minimum
energy corresponds to directions lying in a plane
q
=
p
/2. This type of
anisotropy is called a “easy plane” because an anisotropy energy remains
minimal, equal to zero, for all directions lying in this plane. Real particles
can have additional weak contributions to an energy of magnetic anisotropy,
which emit some directions in an indicated plane, separated, generally
speaking, by small energy barriers.
In Fig. 1c shows
a reduced energy density of cubic magnetic anisotropy,
, equation
(2), for a case
K1c
> 0,
K2c
= 0. In
this case, an anisotropy energy minima correspond to directions parallel to
axes of a Cartesian coordinate system. In total, there are thus 6 energy
minima, in other words, deep potential wells separated by energy barriers. In an
ideal case of a spherical nanoparticle, all these potential wells are
completely equivalent, so that in equilibrium a magnetic moment of a particle
can be directed in any of an indicated directions of easy magnetization. At the
same time, for a case
K1c
< 0,
K2c
= 0, shown
in Fig. 1d, a minimum energy corresponds to directions parallel to diagonals of
a cube. Therefore, in this case, there are 8 equivalent directions for which a
magnetic anisotropy energy of a particle has a minimum.
Energy
surfaces shown in Fig. 1a - 1d correspond to an ideal case of a spherical
magnetic nanoparticle. If a shape of a particle deviates from spherical, then an
energy of magnetic anisotropy, equations (1) - (3), is added to an energy of a
shape anisotropy, equation (4), which can make a significant contribution to a
total anisotropy energy for particles with high saturation magnetization.
Fig.
2. An energy density for a combined type of magnetic anisotropy of a particle
(cubic anisotropy + uniaxial shape anisotropy).
As an example,
Figure 2 shows the case of combined magnetic anisotropy. Here to the energy
density of cubic anisotropy,
, shown in Fig. 1c, a reduced energy
density of the shape anisotropy,
, is added. A
unit vector
n
indicates a direction of the easy axis of a shape
anisotropy.
First, we consider
a simple case when the axis of a shape anisotropy energy coincides with one of
the axes of cubic anisotropy (see Fig. 2a, 2b). In Fig. 2a, a ratio of
anisotropy constants was chosen small,
Kef/K1c
= 0.1. However, as shown in Fig. 2a, due to an influence of a shape anisotropy
energy, a surface of a combined anisotropy energy becomes, in fact, two-pit already
for small values of the
Kef/K1c
ratio, and two main, deep potential wells are located in directions parallel to
the vector
n. In this case, lateral energy minima are easily
distinguishable in Fig. 2a, they disappear only at
Kef/K1c
³
1.0.
So, Figure 2b shows a case of
Kef/K1c
= 2.0, when only two energy minima remain in directions parallel to the vector
n.
If a vector
n
is oriented arbitrarily with respect to cubic axes of anisotropy (see Fig. 2c,
2d), then a corresponding energy surface becomes very complex for sufficiently
small values of a ratio
Kef/K1c.
However, with an increase in this ratio,
Kef/K1c
= 0.1, a global energy surface also becomes two-fold, as Fig. 2c, 2d for the
directions of the vector
n
given by angles (
θ
=
p
/4,
φ
=
p
/
4)
è
(
θ
= -
p
/4,
φ
=
p
/
4),
respectively. Note that a barrier separating potential wells has a very
complex shape. A shape of an energy barrier between potential wells is
extremely important because it determines a position of an energy saddles, or
passes that connect adjacent potential wells. At finite temperatures, due to
thermal fluctuations, a magnetic moment of a nanoparticle can jump between
different potential wells, and such jumps in a magnetic moment are most likely
to occur through the saddle regions of the energy barrier.
As discussed
above, taking into account a magnetic anisotropy energy, a vector of a magnetic
moment of a nanoparticle has several stable directions in space for which a
total energy of a nanoparticle has a local minimum. In an absence of an
external magnetic field, neglecting thermal fluctuations, a magnetic moment of
a single-domain nanoparticle is in one of the energy minima. In an applied
external magnetic field, the energy minima of potential wells shift with respect
to each other, some minima disappear, others appear. Therefore, when an
external magnetic field changes, a magnetic moment of a particle can jump from
one potential well to another. In Fig. 3a-3d shows an example of a change in a
total energy density of a uniaxial nanoparticle,
,
|
(6)
|
under an influence of an external
magnetic field, where the magnitude of a magnetic field is normalized to the
particle anisotropy field,
,
Ha=2K1/Ms.
A magnitude of a
magnetic field in Fig. 3a-3d is equal to
H
= 0, 300, 700, 4000 Oe,
respectively. An uniaxial magnetic anisotropy constant and a saturation
magnetization are
K1
= 0.4·105erg/cm3
and
Ms
= 400 emu/cm3, respectively. A direction of A
magnetic field is determined by the angles (
θ
H
=
p
/2,
φ
H
=
0). In Fig. 3 this direction is shown by an arrow.
Fig.
3. Change in a total energy density of a uniaxial nanoparticle under an
influence of an external magnetic field directed perpendicular to an easy axis
of anisotropy.
As Fig. 3a - 3d show,
with an increase of an external magnetic field
H, potential wells
of a particle gradually merge, and in fields
H
³
2000 Oe, only
one potential well remains. Note that when constructing the energy surfaces
shown in Fig. 3b - 3d, we add the corresponding positive constant to equation
(6) to make the radius values positive. As indicated in the
introduction, addition of an arbitrary constant to a total energy of a particle
does not affect the physics of the processes, but leads to a corresponding
change in the relative scales of the figures.
A similar
picture of transformation of energy surfaces for nanoparticles with different
types of magnetic anisotropy. For simplicity, only symmetric directions of an
external magnetic field with respect to axes of a magnetic anisotropy of a
nanoparticle are considered. In Fig. 4–7 show a conversion of a total energy
density depending on a type of particle anisotropy and a magnitude of an
applied external magnetic field. In each case, a direction of a magnetic field
is indicated by an arrow.
Fig.
4. An influence of an external magnetic field on a surface shape of a total
energy density of a nanoparticle in a case of cubic magnetic anisotropy at
K1c
< 0,
K2c
= 0. (Field direction
θ
H
=
p
/2,
φ
H
= 0)
Figure
4 shows an influence of an external field on a surface shape of a total energy density
of a nanoparticle in a case of cubic anisotropy; an anisotropy energy density
of this type is described by formula (2). Here, magnetic anisotropy constants
and saturation magnetization are
K1c
= - 0.4·105erg/cm3,
K2c
= 0,
Ms
= 400 emu/cm3.
Fig.
5. An influence of an external magnetic field on a surface shape of a total
energy density of a nanoparticle in a case of combined anisotropy (cubic
anisotropy + uniaxial shape anisotropy) for
K1c
< 0,
K2c
= 0. (Field direction
θ
H
=
p
/2,
φ
H
= 0)
Figure
5 shows an influence of an external field on a surface shape of a nanoparticle
total energy density in the case of combined anisotropy. Here, combined
anisotropy is the sum of crystalline anisotropy,
, with shape
anisotropy,
. A
uniaxial magnetic anisotropy constant and saturation magnetization are
K1c
= -
0.4·105erg/cm3,
K2c
= 0,
Ms
= 400 emu/cm3, a ratio of a shape anisotropy constant to a
crystalline anisotropy constant is,
Kef/K1c
= 0.1. A direction of an easy shape anisotropy vector,
n, is
indicated by an arrow.
Fig.
6. An influence of an external magnetic field on a surface shape of a nanoparticle
total energy density in a case of cubic anisotropy for
K1c
> 0,
K2c
= 0. (Field direction
θ
H
=
p
/2,
φ
H
= 0)
Figure
6 shows a form of a reduced energy density of cubic magnetic anisotropy,
,
equation (2), depending on a magnitude of an external field, for a case
K1c
> 0,
K2c
= 0. Constants of magnetic anisotropy and saturation
magnetization are
K1c
= 0.4·105erg/cm3,
K2c
= 0,
Ms=400 emu/cm3.
Fig.
7. An influence of an external magnetic field on a surface shape of a total
energy density of a nanoparticle in a case of uniaxial anisotropy at
K1
< 0,
K2
= 0 (Field direction
θ
H
=
p
/2,
φ
H
= 0)
Figure 7 shows a reduced energy density of uniaxial
magnetic anisotropy (1) as a function of the external field for the case
K1
< 0,
K2
= 0. Constants of magnetic anisotropy and
saturation magnetization are
K1
= - 0.4·105erg/cm3,
K2
= 0,
Ms
= 400 emu/cm3.
Figures 4-7 show that
for large values of an applied external field,
H
> 4000 Oe, a
complex geometry of a nanoparticle energy density, caused by a contribution of
anisotropy, completely disappears. A single minimum of energy appears along the
direction of the magnetic field.
Thus, we see that
properties of magnetic nanoparticles are determined by a whole set of geometric
and magnetic parameters, namely, an external shape of particles, a distribution
of easy axes of magnetocrystalline anisotropy, a value of magnetic anisotropy
constants, and saturation magnetization. For applications, it is also very
important whether a particle is single-crystal or consists of individual
crystallites of different spatial orientations connected by an exchange
interaction. In real experimental ensembles, there is usually a significant
scatter of nanoparticles in size and shape. Particles are often
polycrystalline. In polycrystalline nanoparticles, anisotropic interactions are
significantly averaged, which affects a characteristic size of a single domain
and a coercive force of nanoparticles [16]. Therefore, magnetic properties of
such ensembles of nanoparticles are very difficult to control.
The
authors gratefully acknowledge the financial support of the Ministry of
Education and Science of the Russian Federation in the framework of Increase Competitiveness
Program of NUST “MISIS”, contract ¹K2A-2019-034.
1.
S. Sun, C. B. Murray, D. Weller, L. Folks, A. Moser, Monodisperse FePt
Nanoparticles and Ferromagnetic FePt Nanocrystal Superlattices // Science, Vol.
287, 2000, pp. 1989-1992.
2.
López-Ortega, A., Lottini, E., Fernández, C. D. J., Sangregorio,
C. Exploring the Magnetic Properties of Cobalt-Ferrite Nanoparticles for the
Development of a Rare-Earth-Free Permanent Magnet // Chemistry of Materials,
Vol. 27, 2015, pp. 4048-4056.
3.
Q.A. Pankhurst,
N.K.T. Thanh,
S.K. Jones,
J. Dobson, Progress
in
applications
of
magnetic
nanoparticles
in
biomedicine //
J.
Phys. D: Appl. Phys., Vol. 42, 2009, p. 224001
4.
Dutz S., Hergt R., Magnetic nanoparticle heating and heat transfer on a
microscale: Basic principles, realities and physical limitations of
hyperthermia for tumour therapy // Int. J. Hyperthermia, Vol. 29, 2013, pp. 790-800.
5.
Ortega D., Pankhurst Q.A., Magnetic hyperthermia // In O’Brien P editor Nanoscience:
Nanostructures through Chemistry Royal Society of Chemistry: Cambridge,
Vol. 1, 2013, pp. 60–88.
6.
Périgo E.A., Hemery G., Sandre O., Ortega D., Garaio E., Plazaola F.,
Teran F.J., Fundamentals and advances in magnetic hyperthermia //
Appl. Phys. Rev., Vol. 2,
2015, p. 041302.
7.
S. Chikazumi, Physics of Magnetism. New York: Wiley, 1964.
8.
W.F. Brown, Jr., Micromagnetics. New York – London: Wiley-Interscience, 1963.
9.
A. Aharoni, Introduction
to the Theory of Ferromagnetism. Oxford: Clarendon Press, 1996
10.
L.D. Landau, E.M. Lifshitz, Electrodynamics of Continuous Media. New York:
Pergamon, New York, 1984.
11.
A. I. Akhiezer, V. G. Baryakhtar, and S. V. Peletminskii, Spin Waves. Amsterdam:
North-Holland, 1968.
12.
V.V. Pilyugin, I. Millman. Visual Analytics and its Application in the
Activities of the Laboratory "Scientific Visualization" MEPhI (2019).
Scientific Visualization 11.5: 46 - 55, DOI: 10.26583/sv.11.5.05
13.
Yu. S. Lagutin, A. A. Lagutina, F. V. Grigoriev, V. B. Sulimov, A. V.
Tikhonravov (2018) Visualization of nanosize pores in the atomistic structure
of thin films.
Scientific
Visualization 10.1: 128 - 134, DOI: 10.26583/sv.10.1.10
Translated
with www.DeepL.com/Translator (free version)14. W. F. Brown, Jr. and A. H.
Morrish, Effect of a Cavity on a Single-Domain Magnetic Particle // Phys. Rev.,
Vol. 105, 1957, p. 1198.
15. N.A. Usov, J.M. Barandiarán,
Magnetic nanoparticles with combined anisotropy //
J. Appl. Phys., Vol. 112, 2012, p. 053915.
16.
V. A. Bautin, A. G. Seferyan, M. S. Nesmeyanov, N. A. Usov, Properties of
polycrystalline nanoparticles with uniaxial and cubic types of magnetic
anisotropy of individual grains // J. Magn. Magn. Mater.,
Vol. 460, 2018,
p. 278.