Size dependence of magnetic characteristics measured on separate nickel particles
Interference electron microscopy was applied to measure the coercive force, the magnetic saturation and the residual magnetization of separated nickel particles. Nickel particles with perfect sphericity and radius from 10 to 100 nm were produced directly in the interference electron microscope by me...
Збережено в:
| Дата: | 1999 |
|---|---|
| Автори: | , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут фізики напівпровідників імені В.Є. Лашкарьова НАН України
1999
|
| Назва видання: | Semiconductor Physics Quantum Electronics & Optoelectronics |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/119880 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Size dependence of magnetic characteristics measured on separate nickel particles / S.A. Nepijko, R. Wiesendanger // Semiconductor Physics Quantum Electronics & Optoelectronics. — 1999. — Т. 2, № 3. — С. 5-9. — Бібліогр.: 24 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-119880 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1198802017-06-11T03:02:30Z Size dependence of magnetic characteristics measured on separate nickel particles Nepijko, S.A. Wiesendanger, R. Interference electron microscopy was applied to measure the coercive force, the magnetic saturation and the residual magnetization of separated nickel particles. Nickel particles with perfect sphericity and radius from 10 to 100 nm were produced directly in the interference electron microscope by means of wire explosion caused by the passage of an electric current pulse through it. We find a decrease of the magnetic saturation and an increase of the coercive force with decreasing size of the separate particles. If there are neighbouring particles, with decreasing the distance to them coercive force is characterized by more smooth size dependence and has less absolute value. This observation shows the contribution of the interparticle interaction. 1999 Article Size dependence of magnetic characteristics measured on separate nickel particles / S.A. Nepijko, R. Wiesendanger // Semiconductor Physics Quantum Electronics & Optoelectronics. — 1999. — Т. 2, № 3. — С. 5-9. — Бібліогр.: 24 назв. — англ. 1560-8034 PACS: 75.50.K http://dspace.nbuv.gov.ua/handle/123456789/119880 en Semiconductor Physics Quantum Electronics & Optoelectronics Інститут фізики напівпровідників імені В.Є. Лашкарьова НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| description |
Interference electron microscopy was applied to measure the coercive force, the magnetic saturation and the residual magnetization of separated nickel particles. Nickel particles with perfect sphericity and radius from 10 to 100 nm were produced directly in the interference electron microscope by means of wire explosion caused by the passage of an electric current pulse through it. We find a decrease of the magnetic saturation and an increase of the coercive force with decreasing size of the separate particles. If there are neighbouring particles, with decreasing the distance to them coercive force is characterized by more smooth size dependence and has less absolute value. This observation shows the contribution of the interparticle interaction. |
| format |
Article |
| author |
Nepijko, S.A. Wiesendanger, R. |
| spellingShingle |
Nepijko, S.A. Wiesendanger, R. Size dependence of magnetic characteristics measured on separate nickel particles Semiconductor Physics Quantum Electronics & Optoelectronics |
| author_facet |
Nepijko, S.A. Wiesendanger, R. |
| author_sort |
Nepijko, S.A. |
| title |
Size dependence of magnetic characteristics measured on separate nickel particles |
| title_short |
Size dependence of magnetic characteristics measured on separate nickel particles |
| title_full |
Size dependence of magnetic characteristics measured on separate nickel particles |
| title_fullStr |
Size dependence of magnetic characteristics measured on separate nickel particles |
| title_full_unstemmed |
Size dependence of magnetic characteristics measured on separate nickel particles |
| title_sort |
size dependence of magnetic characteristics measured on separate nickel particles |
| publisher |
Інститут фізики напівпровідників імені В.Є. Лашкарьова НАН України |
| publishDate |
1999 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/119880 |
| citation_txt |
Size dependence of magnetic characteristics measured on separate nickel particles / S.A. Nepijko, R. Wiesendanger // Semiconductor Physics Quantum Electronics & Optoelectronics. — 1999. — Т. 2, № 3. — С. 5-9. — Бібліогр.: 24 назв. — англ. |
| series |
Semiconductor Physics Quantum Electronics & Optoelectronics |
| work_keys_str_mv |
AT nepijkosa sizedependenceofmagneticcharacteristicsmeasuredonseparatenickelparticles AT wiesendangerr sizedependenceofmagneticcharacteristicsmeasuredonseparatenickelparticles |
| first_indexed |
2025-07-08T16:50:12Z |
| last_indexed |
2025-07-08T16:50:12Z |
| _version_ |
1837098273387577344 |
| fulltext |
5© 1999, Institute of Semiconductor Physics, National Academy of Sciences of Ukraine
Semiconductor Physics, Quantum Electronics & Optoelectronics. 1999. V. 2, N 3. P. 5-9.
PACS: 75.50.K
Size dependence of magnetic characteristics
measured on separate nickel particles
S. A. Nepijko1, R. Wiesendanger2
1 Institute of Physics, National Academy of Siences of Ukraine, 46, Prospect Nauki,252022 Kiev, Ukraine
2 Institute of Applied Physics and Microstructure Research Center, University of Hamburg, Jungiusstraße 11, D-20355 Hamburg, Germany
Abstract. Interference electron microscopy was applied to measure the coercive force, the magnetic
saturation and the residual magnetization of separated nickel particles. Nickel particles with perfect
sphericity and radius from 10 to 100 nm were produced directly in the interference electron microscope
by means of wire explosion caused by the passage of an electric current pulse through it. We find a
decrease of the magnetic saturation and an increase of the coercive force with decreasing size of the
separate particles. If there are neighbouring particles, with decreasing the distance to them coercive
force is characterized by more smooth size dependence and has less absolute value. This observation
shows the contribution of the interparticle interaction.
Keywords: ferromagnetic particles, size dependence, coercive force, magnetic saturation, residual mag-
netization
Paper received 25.05.99; revised manuscript received 01.10.99; accepted for publication 12.07.99.
1. Introduction
Measurements of magnetic properties of separate small par-
ticles is of particular interest. These allow one to distinguish
contribution of size-dependent properties from contribution
of effects caused by interaction between particles. To do
this, an interference electron microscope (IEM) can be ap-
plied [1]. Experiments on observation of magnetic structure
of Ni particles [2-4] as well as measurement of the tempera-
ture dependence of the Curie temperature of Ni particles [5]
were performed by means of this technique.
In the present paper, size dependence of a several mag-
netic characteristics, such as the coercivity, the magnetization
of saturation and the remanent magnetization, were measured
by IEM. This study was carried out on small Ni particles.
We do not contrast the interference electron microscopy
with the Lorentz microscopy [6] and the scanning magnetic-
force microscopy (MFM) [7] which are also informative
when the magnetic properties of separate particles are un-
der investigation. On the contrary, comparison of results
obtained by these methods is of interest in the field of over-
lapping their potentialities.
2. Experimental
The study was conducted in a 100 kV IEM with a thermi-
onic cathode. To raise the spatial coherence of the electron
beam, it was extended by means of the condensing lenses.
Then it was strongly narrowed by a diaphragm, so that used
area of the cathode comprised 100 nm. As a result, the beam
intensity was weak, and the exposure time under shooting
was long and accounted for 10-20 s. The length of coherence
was sufficiently large, it was equal near 0,5 m. An electro-
static analogue of the Fresnel biprism [8,9] was applied in
our IEM. For this purpose, a filament 0,7 µm in diameter
was installed under the lower focal plane of the microscope
objective lens normal to its optical axis. Since the filament
diameter is less than one of the coherent electron beam, the
filament divides this beam into two parts. The part of beam,
passing on the one side of the filament, serves as the probe
beam. The part of beam on the other side of the filament
was used as the reference beam. Trajectories of electrons,
passing on the both sides of the filament, are bent towards
each other if a positive potential is applied to the filament.
The width of the shadow from the filament at the IEM screen
decreases as Ub increases. In our case the beams overlap at
Ub ≥ 3 V. The interference fringes were observed in the over-
lap region. The higher is the voltage applied to the filament,
the more is the overlap region. When the amount and spatial
density of the fringes grow, distance between them reduces
[10]. A settled phase shift 2π corresponds to this distance.
Thus, as a voltage applied to the filament grows, modulus of
bending of the interference fringes, caused by the phase shift
∆Φ of the probe beam relative to the reference one, dimin-
ishes. Because of this, the filament voltage was also limited
from above, and in our case it was equal to 3 V < Ub < 10 V.
S. A. Nepijko, R. Wiesendanger: Size dependence of magnetic characteristics measured...
6 SQO, 2(3), 1999
Ni particles were produced direct by in IEM by means
of wire explosion caused by the passage of an electric current
pulse through it [11]. A source of this kind can be made
rather small in size in order to place it in limited free space
under the IEM objective lens. Moreover, this source pro-
duces particles of near ideal spherical shape that is of great
importance because magnetic properties of small particles
depend strongly on their shape. Sizes of particles were var-
ied by the pulse duration, the value of current passing
through the filament and its diameter. In films prepared in
such way, the areas can be found which contain particles
well separated from each other. In that case, it’s possible to
neglect the probable interparticle interaction. At last,
employment of the explosion method makes it possible to
hope that prepared nickel particles are clean and perfect
enough in spite of the fact that they were prepared not in ul-
trahigh vacuum. The vacuum in the sample area was 10-7 mbar.
A 10-nm thick carbon film on the copper grid served as
a substrate. The substrate electrical conductivity excluded
from charging of the particles concerned under the electron
beam [12]. The carbon film rolled up in places where there
were its ruptures. Particles lying on these rolled regions
(Fig.1) were chosen for investigation.
A specimen was somewhat lifted relative to its standard
position in the objective lens. This lowered resolution by
an order of magnitude, and it amounted to 4 - 5 nm. How-
ever, this enabled the specimen to be placed between two
diminutive solenoidal coils, the axis of which intersects
normally to the IEM optical axis. Value of the local mag-
netic field, created by the coils, varied in the range 0 ≤ H ≤
103 Gs (from 0 up to 1,5.103 Gs), but it reach 104 Gs in
pulse. The latter exceeded the magnetic field of the objec-
tive lens close to the specimen and was used in order to
orientate the magnetic moment of the particles under study
in the plane normal to the IEM optical axis.
3. Experimental results and their discussion
A spherical Ni particle obtained by the Ni wire explosion
was chosen on the rolled section of the carbon surface. Its
magnetic moment was settled in the plane normal to the
IEM optical axis by means of a pulse field produced by the
solenoids. The following measurements were performed in
the constant external uniform magnetic field oriented, as in
the case of the pulse field, along the solenoid axis in the
direct or opposite direction.
The uniform magnetic field does not result in bending
of the interference fringes because the probe and reference
coherent electron beams get the same phase in this field.
However, in the uniform magnetic field a ferromagnetic
particle obtains the magnetic moment that leads to
nonuniformity of the magnetic field in its vicinity. Then the
coherent electron beams pass through regions with differ-
ent magnetic fields, and the phase difference arises between
them that causes the interference fringes to bend. It is
necessary that the magnetic moment of a particle has
component normal to the direction of electron motion in
the beam [1]. If the magnetic moment of a particle is paral-
lel to the electron beam, then the positive phase shift under
movement of electron toward this particle is completely
compensated by the phase shift of opposite sign under elec-
tron moving away from the particle (in this case the picture
is entirely symmetrical). A character of bending of the in-
terference fringes is analyzed in ref.[1]. When the external
magnetic field enhances, the magnetic moment of the fer-
romagnetic particle grows as well, and this growth has satu-
ration. Accordingly, the bend of these interference fringes
increases with saturation, too. In experiment the external
magnetic field enhanced slowly till changes in bending the
interference fringes were still observed. We are interested
not in the value of external magnetic field itself, but in value
and shape of bending the interference fringes out of the
particle involved. From these data using Eqs.(15), (4) and
(3) from ref.[1] the magnetic moment of the particle in satu-
ration can be calculated as well as the magnetic moment
per unit of the volume of particle, i.e. the magnetization of
saturation Is.
After it the current through the solenoids as well as the
external magnetic field went down to zero. The magnetic
Fig.1. A schematic of the spherical particle under study (1), the carbon
substrate with roll (2), the probing (3) and reference (4) coherent electron
beams and the IEM elements: illuminating system consisting of G - source
of electrons, D - diaphragm, K - condencing lens, B - filament with a posi-
tive potential applied to it (an electrostatic analog of the Fresnel biprism),
S - screen. Objective, magnifying and projective lenses are not shown.
S. A. Nepijko, R. Wiesendanger: Size dependence of magnetic characteristics measured...
7SQO, 2(3), 1999
moment, calculated from bending of the interference fringes
at disconnected external magnetic field, associates with the
remanent magnetization Ir.
To measure one more characteristic of magnetic reversal,
the current was passed through the solenoids in the oppo-
site direction and increased. When the opposing external
magnetic field is applied, and the interference fringes bend
in the opposite direction, the value of this field is the
coercivity Hc.
Measurements of three characteristics of the magneti-
zation curve (hysteresis loops) were carried out on Ni par-
ticles of different sizes, results are presented in Fig.2a-c.
The coercivity Hc rises when the radius R of Ni particles
decreases, i.e. the coercivity is size-dependent. The
coercivity tends to diminish with increasing distance be-
tween particles. The experimental points in Fig.1 are in the
hatched region for this case. Measurements of value of the
coercivity do not require any recalculation of form of the
experimentally observed interference fringes, i.e. they are
characterized by high precision. It falls as the concentration
of particles grows. Really, if there are adjacent particles,
the effective magnetic field, switching the magnetic moment
of particles, is already poorly determined because it is a
superposition of the external field and the magnetic fields
of adjacent particles.
Measured dependence Hc(R) is understandable qualita-
tively. Indeed, decrease of sizes of ferromagnetic particles
is accompanied by reconstruction of their magnetic struc-
ture. In this case, when transition from multi-domain to
mono-domain state takes place, possibility of nucleation of
magnetic reversal centers decreases that leads to rise of the
coercivity.
The value of demagnetizing factor changes with increas-
ing concentration of the ferromagnetic particles because of
their interaction, which is mainly magnetostatic. As a result,
the critical size of the mono-domain state R0 [13] grows.
Therefore, a maximum of the dependence Hc(R) shifts to-
wards greater sizes. This results in the fact that at a settled
particle size (R > R0) the coercivity should rise as the bulk
concentration of particles increases. It is also clear that the
magnetic reversal of particles in ensemble has some pecu-
liar features. Fig.3a-d illustrates schematically possible types
of the magnetic reversal in a separate particle and their
ensembles that is discussed in literature. When distance
between particles reduces, type of the magnetic reverse
changes from twisting (Fig.3a) to parallel rotation of the
elementary magnetic moments in particles (Fig.3b) [14,15].
In this case the simplest consideration of character of mag-
netic reversal of mono-domain neighboring particles sup-
poses that all their magnetic moments turn simultaneously
and coherently (Fig.3c). It proved to be reasonable that the
10 30 50 70 R , nm
H , G
I , G
I , G
300
600
900
200
300
300
400
a
b
c
c
s
r
Fig.2. Size dependencies of (a) the coercivity Hc(R); (b) the magnetiza-
tion of saturation Is(R) and (c) the remanent magnetization Ir(R). They
were measured on separate Ni particles of radius R. In the hatched region
there are experimental points for the case when there are neighboring par-
ticles close to the particle under study.
a b c d
Fig.3. Possible types of the magnetic reversal of mono-domain particles isolated by twisting (a) and parallel rotation (b) as well as in the case of their
ensemble. The magnetic reversal can be both coherent (c) and incoherent (d).
S. A. Nepijko, R. Wiesendanger: Size dependence of magnetic characteristics measured...
8 SQO, 2(3), 1999
coercivity of a chain of spherical particles reversing the
magnetism according to this scheme was much more than
in the case of a separate particle [16]. However, under the
action of external field the vectors of magnetic moments
can also turn to opposite direction in this way, and that is
shown in Fig. 3d. This process is preferable in terms of
energetics [16]. The real situation is certainly more
complicated because of multidomain state of particles, size
dispersion of particles and dispersion of distances between
them. Our experimental measurements show a rise of the
coercivity with decreasing distance between particles
(Fig.2a, hatched region). This strongly supports the model
of coherent magnetic reversal.
Measurement of the magnetization of saturation Is is
characterized by the maximum sensitivity when the abso-
lute value of the bend of above interference fringes is maxi-
mum. As illustrated in Fig.1b, the magnetization of satura-
tion Is tends to diminish with decreasing radius of particles
R. In refs.[17-21] the suggestion has been made that similar
behavior of the magnetization of saturation of the small fer-
romagnetic particles, i.e. of the magnetic moment of vol-
ume unit, can be caused by oxidation of their surface. In
our case, Ni particles were produced directly in the IEM
column, i.e. they were not transferred via atmosphere. How-
ever, since vacuum in the region close to specimen
comprised 10-7 mbar, the particles could partly oxidize. One
should take into consideration that the size dependence of
the magnetization of saturation Is = Is(∞)(1- a/R)3 describes
results of the experimental measurements only under a rather
strange condition of an increasing depth of the oxidized layer
with increasing sizes of particles. Hence it follows that oxi-
dation is not the only reason of lowering the magnetization
of saturation with decreasing sizes of particles. The same
objections are also true when an idea about chemisorption
of, for example, hydrogen is proposed [22] or existence is
assumed of «dead» atom layers on the particle surface which
do not take part in ferromagnetism [23,24]. Relying on these
reasons, a conclusion can be drawn that dependence of the
magnetization of saturation on the particles’ sizes is not a
result of their oxidation. It seems reasonable to say that as
the particle size diminishes, more and more significant part
of spins adjoining the surface has a stable direction which
does not coincide with the direction along which spins are
lined up in the bulk of the particle. As a result, two spin
subsystems arise in the small ferromagnetic systems. One
of them is collinear in the bulk, and in near-surface layer
the second subsystem has a component of magnetization
normal to the spin orientation of the first one (Fig.4).
Results of measuring the remanent magnetization Ir are
presented in Fig.1c. Because of small accuracy and sensi-
tivity of these measurements, it is possible to say only about
a tendency of Ir growth with decreasing radius of particles
R. This tendency appears to be correct. Indeed, as the par-
ticles’ sizes reduce, Is diminishes as well, i.e., the height of
hysteresis loop is lowered. However, since its area remains
constant, it follows that Ir rises.
In measurements of the magnetic characteristics of sepa-
rate particles by means of the interference electron micros-
copy the value of the bend of the interference fringes does
not exceed several percent from the distance between adja-
cent fringes. In order to register so small bending, the fol-
lowing steps were taken:
- spherical particles on the substrate roll were chosen for
measurements. If the probing beam and, especially, if both
the probing and reference coherent beams intersect the sub-
strate, then, because of its heterogeneity in thickness and
density, fluctuations of phase comprise already several per-
cent from 2π (the phase shift between two adjacent inter-
ference fringes), i.e., they are comparable with the measured
signal;
- bending the interference fringes was analyzed out of a
particle, but not on its image. Out of Ni particle bending the
interference fringes is caused by its magnetic moment,
whereas on the particle image bending is stimulated by the
magnetic moment as well as by the inner potential. The sec-
ond effect is considerably stronger [1]. In this case measure-
ments are not limited by arbitrary large sizes of particles;
- effect of bending the interference fringes caused by char-
ging of a particle under the electron beam [12] was excluded
by using a substrate with good conductivity;
- the value of interference fringes’ bending was determined
from conditions that their shape is described by the expres-
sion (15) from ref.[1] and sum of least squares. It should be
noted here that it does not matter how the orthogonal axes
Ox and Oy are lined in the interferogram plane. Mutual
orientation of the direction of external magnetic field and
of the biprism filament is of no importance, too.
Conclusion
Measurements of the size dependence of the coercivity
Hc(R), the magnetization of saturation Is(R) and the remanent
magnetization Ir(R) were performed on separate Ni particles.
This allows judging about the value and character of defor-
mation of the hysteresis loop as sizes of Ni particles de-
crease. Hc and Ir increase and Is decreases with diminishing
sizes of Ni particles. Influence of interaction of particles
with each other on the value of their magnetic characteristics
has been shown. In particular, Hc goes down as concentration
of Ni particles rises.
Fig.4. A model representation of the domain structure of a small particle
having two spin subsystems - in bulk and in near-surface layer.
S. A. Nepijko, R. Wiesendanger: Size dependence of magnetic characteristics measured...
9SQO, 2(3), 1999
References
1. S.A.Nepijko, R.Wiesendanger: Appl. Phys. A 65, 361 (1997)
2. A.Tonomura, T.Matsuda, J.Endo: Phys. Rev. Lett. 44, 1430 (1980)
3. T.Matsuda, A.Tonomura, R.Suzuki, J.Endo, N.Osakabe, H.Umezaki,
H.Tanabe, Y.Siguta, H.Fujiwara: J. Appl. Phys. 53, 5444 (1982)
4. J.N.Chapman, R.P.Ferrier, U.J.Heyderman, S.McVitie,
W.A.P.Nicholson, B.Bormans: Inst. Phys. Conf. Ser. No.138: Section
1 (Proc. Electron Microscopy and Analysis Group Conf. EMAG93,
Liverpool, 1993) P.1-8
5. S.A.Nepijko, R.Wiesendanger: Europhys.Lett. 31, 567 (1995)
6. T.Arii, S.Yatsuya, N.Wada, K.Mihama: Jap. J. Appl. Phys. 17, 259
(1978)
7. A.A.Bukharaev, D.A.Ovchinnokov, N.I.Nurgazizov, E.F.Kukovitskii,
M.Kleiber, R.Wiesendanger: Phys. Solid State, 40, 1163 (1998)
8. G.Möllenstedt, H.Düker: Z. Physik 145, 377 (1956)
9. V.Dragos, A.Delong: Cesk. Cas. Fys. 13, 278 (1963); Slaboproudy
obsor 25, 523 (1964)
10. I.G.Stoyanova, I.F.Anaskin: Physical principles of transmission elec-
tron microscopy, Nauka, Moscow 1972. In Russian
11. V.N.Byckov, Yu.A.Kulyupin, A.A.Lisenko, S.A.Nepijko: Fiz. Met. &
Metalloved. (USSR) 52, 1176 (1981). In Russian. English translation
in: Phys. Met. & Metallogr. (GB)
12. V.N.Byckov, Yu.A.Kulyupin, S.A.Nepijko, E.A.Pashitskii:
Radiotekhnika & Elektronika, 26, 1055 (1981). In Russian
13. S.A.Nepijko: Physical properties of small metal particles, Naukova
Dumka, Kiev 1985. In Russian
14. A.Aharoni: phys. status solidi 16, 3 (1966)
15. A.Aharoni: J. Appl. Phys. 35, 347 (1964)
16. I.S.Jacobs, C.P.Bean: Phys. Rev. 100, 1060 (1955)
17. C.R.Abeledo, P.W.Selwood: J.Appl. Phys. 32, suppl. 229S (1961)
18. A.Tasaki, S.Tomiyama, S.Iida, R.Uyeda: Jap. J. Appl. Phys. 4, 707
(1965)
19. Yu.I.Petrov, Yu.I.Fedorov: Zh. Tekh. Fiz. (USSR) 37, 726 (1967). In
Russian. English translation in: Soviet Phys. Tech. Phys. (USA) 12,
(1967)
20. A.E.Ermakov, O.A.Ivanov, Ya.S.Shur, R.M.Grechishkin, G.V.Ivanova:
Fiz. Met. & Metalloved. (USSR) 33, 558 (1972). In Russian. English
translation in: Phys. Met. & Metallogr. (GB)
21. I.N.Shabanov, A.E.Ermakov, V.A.Trapeznikov, Ya.S.Shur: Fiz. Met.
& Metalloved. (USSR) 38, 314 (1974). In Russian. English transla-
tion in: Phys. Met. & Metallogr. (GB)
22. P.W.Selwood: Adsorption and collective paramagnetism, Acad. Press,
New York 1962
23. L.Lieberman, J.Clinton, D.M.Edwards, J.Mathon: Phys. Rev. Lett.
25, 232 (1970)
24. L.N.Liebermann, D.P.Fredkin, H.B.Chore: Phys. Rev. Lett. 22, 539
(1969)
|