Field-induced Magnetic Transition in Cobalt-Ferrite

0.14 MB·English

by  Martin Kriegisch

MMM2011/CU-10
Field-induced Magnetic Transition in Cobalt-Ferrite
Martin Kriegisch, Reiko Sato-Turtelli, Herbert Mu¨ller, and Roland Gr¨ossinger
∗
Institute of Solid State Physics, Vienna University of Technology
Wiedner Hauptstrasse 8-10/E138, A-1040 Vienna, Austria
Weijun Ren and Zhidong Zhang
Shenyang National Laboratory for Materials Science,
Institute of Metal Research, and International Center for Materials Physics,
Chinese Academy of Sciences, Shenyang 110016, People’s Republic of China
2
(Dated: January 13, 2012)
1
0
We present magnetostriction and magnetization measurements of a cobalt ferrite (Co0.8Fe2.2O4)
2
single crystal. We observe unusual behaviour in the magnetic hard axis of the single crystal which

n manifests in a jump of the magnetization curve at a critical field. This first order magnetization
a process (FOMP) which is explained as an anisotropy driven transition is visible at temperatures
J lowerthan150K.Byanalyzingtheanisotropyconstantswefoundthatthehigherorderanisotropy
2 constantK2 dominatestheanisotropy energy. Inthemagnetostriction measurementstheFOMPis
1 clearlyvisibleasahugejumpinthe[111]direction,whichcanbeexplainedbymeansofageometric
model.

]
i PACSnumbers: 75.80.+q,75.47.Lx,75.30.Kz,75.50.-y
c
Keywords: Magnetostriction, Cobaltferrite,FOMP,Anisotropy
s
-
l
r
t INTRODUCTION holder. The error from transferring the single crystal
m
from one sample holder to the other was usually smaller
.
t Cobalt Ferrite is under examination since more than than 1.5◦.
a
m sixty years, but still there is quite a lot of open prob- The magnetostriction was measured with a miniature
lems. In1988Guillotobservedajumpinthefielddepen- capacitive dilatometer described in [3] using a cryostat
-
d dence of the magnetization in pure Cobalt Ferrite with with a variable temperature insert (VTI) and a super-
n the composition Co1.04Fe1.96O4 and also in Cd substi- conducting 9 T coil.
o
tuted Cobalt Ferrite [1]. This jump was explained as
c
a spin-flip. Because of the rather low critical field this
[
explanation was not conclusive. Therefore within the
2 present work this transition was studied in more detail. RESULTS AND DISCUSSION
v
7
2 Magnetization
0 SAMPLE PREPARATION
4
Thedegreeofinversionofthecationdistributionofthe
1. The single crystal with the composition Co0.8Fe2.2O4 inverse spinel (A2+)[B32+]O4 was calculated to i = 0.625
1 was grown by flux method. The starting materials are with (Co20+.8 i Fe30+.2+i)[Co2i+Fe32+i]O4 and the magnetic
11 18 g Na2B4O7 · 10H2O (Borax), 2.3 g CoO (99.99%), momentofµ−=µB Sites−µA S−ites =4.1µB atT =5K.
− −
: and 6.7 g Fe2O3 (99.99%). After sufficiently mixing, the The value of saturation magnetization at 5 and 10 K is
v materials were put in a tightly closed Pt crucible and practically the same due to the the very high ordering
i
X heated from room temperature to 1370 ◦C at a rate of temperature.
100 C/h,andthenheldforaperiodof6h;slowlycooled The magnetization measurements revealed that the
r ◦
a from 1370 to 990 C at 2 C/h, followed by a furnace [100] axis is the easy axis of magnetization of Co-ferrite
◦ ◦
cooling by switching off the power supply [2]. The com- over the whole temperature range. Below 150 K a jump
position of the single crystal was checked by XRD and inthe magnetizationatfields occurs asit wasalsofound
SEM investigation. in a similar material (Co1.04Fe1.96O4) in Ref. [1]. In
figure 1 the normalized magnetization at T = 10 K is
plotted versus the external magnetic field. The jump
EXPERIMENTAL PROCEDURES is clearly visible at roughly 7 T in the [111] axis. The
criticalfield differs fromthatpublished by Guillotwhich
Magnetization was measured from 5 to 400 K in a can be understood regarding the different sample com-
vibrating sample magnetometer with a superconducting position. A linear fit between 1 and 6 T was performed
9 T coil. The single crystal was oriented by XRD in showingthatthe extrapolationto 0 T leads to avalue of
a Laue setup and then transfered to the VSM sample 1 , which indicates that the magnetization vector lies in
√3
2
FIG. 1. Normalised Magnetization of Co0.8Fe2.2O4 single FIG. 2. Offset of the linear fits of normalized magnetization
crystal at T =10 K of Co0.8Fe2.2O4 as a function of temperature. The straight
lines indicate the geometrical fit
Magnetic Anisotropy
The magnetic anisotropy was determined with the in-
the[100]axisandonlytheprojection(cos(α))ofthe[100] tegralmethodEA =R0MSHdM alongthemeasuredcrys-
magnetizationvectorinto the [111]axis is measured. By tallographic axis of the single crystal. Applying the for-
increasing the magnetic field the vector starts to rotate mulaEA =K0+K1(α21α22+α22α23+α23α21)+K2(α21α22α23),
towardsthe [111]axis. Howeverassoonasthe measured where the αi are the direction cosines in polar coordi-
moment in the [111] axis achieves the value of √12, the nates, the anisotropy constants K0, K1 and K2 were
measuredmomentbecomesequaltothe [110]valuemea- determined. For using the integral method it is pre-
suredat0T.Atthispointthesystemneedsmoreenergy requisite to saturate the sample fully. It is reportedthat
(field)toovercomethemagneticanisotropyenergycorre- the saturation along the intermediate and hard axis is
sponding to the [110]axis. As a consequencethe magne- reached at around 18 T at T = 4 K [1]. Our measure-
tization vector rotates in the [110] axis to a local energy ments wereonly performedupto 9T,thereforeextrapo-
minimuminordertoreduceenergy. Infigure2theoffset latingtheM(H)curvestosaturationcausesaratherlarge
of the linear fits assuming such a geometrical model be- error (± 20% for K1 and ± 30% K2), but they are still
low the transitionare shown. The errordue to misalign- in the range of reported values of K1 in literature.
ment is found to be 1.5◦ for the [111] axis and 0.3◦ for In figure 3 the anisotropy constants K0, K1 and K2 of
the[110]axis. Obviouslyourgeometricmodelworkswell the single crystal and K1 of a polycrystalline sample are
from5Kupto250K.Above250Ktheanisotropyalong plotted against the temperature. For obtaining K1 for
the [110] and [111] axis decreases and the magnetization the polycrystalline sample we used the law of approach
vectorcanrotatesmoothly (withoutadiscontinuity)out to saturation. Measuring the polycrystalline sample the
of the easy axis. maximum magnetic field was only 9 T and the sample
By this geometricalconsiderations one can see that the not fully saturated. Accordingly all reported values in
origin of the magnetization jump is not a spin-flip, but literature using the law of approach to saturation are
a rotation in the magnetization due to the higher or- notgivingthecorrectvalueofK1,becausetheyestimate
deranisotropyconstantswhichcauseamorecomplicated only K1 and not higher order anisotropy constants.
shape of the anisotropy energy EA. ItisinterestingtonotethatnovaluesforK2arereported
Such anisotropy driven magnetization jumps are gener- in literature. Above T = 150 K the value of K2 is ∼ 6
ally called FOMP (First Order Magnetization Process) times higher than K1 and below the FOMP the factor
andwerealsofound inother complexsystems andalloys is even much higher, yielding in an increased anisotropy
(as e.g. PrCo5 - see [4], RE2Fe17C [5], REFe11Ti [6] or along the [111] and [110] axes.
Nd2Fe14B [7]).
3
(remanence) the magnetostriction measurements at
low temperatures are very difficult and need a special
measuringprocedureaswillbepublishedelsewhere. But
when increasing the magnetic field above the critical
field, a huge jump in the magnetostriction occurs. At
higher fields due to rotational magnetization process no
hysteresis effects are observed.
CONCLUSIONS
We have proved that the jump in the magnetization
can be explained as an anisotropy driven transition
which is called “FOMP”. The transition is caused by a
rotationofthemagnetizationvectorjumpingoveranen-
ergybarrier. At low temperatures the secondanisotropy
constant K2 is increasing which strengthens the [100]
FIG. 3. Anisotropy constants of Co0.8Fe2.2O4 single crystal axisaseasyaxisandunderlinesthe[110]asintermediate
andonepolycrystallinespecimenasafunctionoftemperature
and the [111] as hard axis. These assumptions are also
supported by geometric considerations explaining the
values of the M(H) curves as measured in the different
crystallographic directions. Additionally we present an
accurate measurement of λ111 at a temperature of 4.2 K
showingalsothiscriticaltransitioninthemagnetoelastic
behaviour.
The financial support by the FWF under the NFN-
project numbers S 10406 and S 10403 is gratefully ac-
knowledged.
∗
[email protected]
[1] M. Guillot, J. Ostorero and A. Marchand, Zeitschrift fu¨r
Physik B Condensed Matter, 71, p.193, 1988
[2] W. Wang and X. Ren, Journal of Crystal Growth, 289,
p.605, 2006
FIG. 4. Magnetostriction measurements of Co0.8Fe2.2O4 sin- [3] M.Rotter,H.Mu¨ller,E.Gratz,M.DoerrandM.Loewen-
gle crystal at T =4.2K haupt,Review of Scientific Instruments,69, p.2742, 1998
[4] G. Asti and F. Bolzoni, Journal of Applied Physics, 50,
p.7725, 1979)
Magnetostriction [5] R. Gr¨ossinger, X. C. Kou, T. H. Jacobs and
K.H.J.Buschow,JournalofAppliedPhysics,69,p.5596,
1991
We are presenting the first magnetostriction mea-
[6] X.Kou,T.Zhao,R.Gr¨ossinger, H.Kirchmayr,X.Liand
surement at T = 4.2 K demonstrating also the effect F. de Boer, Physical Review B, 47, p.3231, 1993
of such a field induced transition in the magnetoelastic [7] X.C.Kou,M.Dahlgren,R.Gr¨ossingerandG.Wiesinger,
behaviour as shown in figure 4. Due to hysteresis effects Journal of Applied Physics, 81, p.4428, 1997