Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (428.74 KB, 43 trang )
the electrons spend most of their time in between neighboring atoms when the
interatomic distance is small. This gives rise to antiparallel alignment and therefore
negative exchange. (antiferromagnetic), Fig. 12.
Fig. 12. Antiparallel alignment for small interatomic distances.
If the atoms are far apart the electrons spend their time away from each other
in order to minimize the electron-electron repulsion. This gives rise to parallel
alignment or positive exchange (ferromagnetism), Fig. 13.
Fig. 13. Parallel alignment for large interatomic distances.
For direct inter-atomic exchange j can be positive or negative depending on
the balance between the Coulomb and kinetic energies. The Bethe-Slater curve
represents the magnitude of direct exchange as a function of interatomic distance.
Cobalt is situated near the peak of this curve, while chromium and manganese are
on the side of negative exchange. Iron, with its sign depending on the crystal
structures probably around the zero-crossing point of the curve, Fig. 14.
•
Fig. 14. The Bethe-Slater curve.
Indirect exchange
Indirect exchange couples moments over relatively large distances. It is the
dominant exchange interaction in metals, where there is little or no direct overlap
37
between neighboring electrons. It therefore acts through an intermediary, which in
metals are the conduction electrons (itinerant electrons). This type of exchange is
better known as the RKKY interaction named after Ruderman, Kittel, Kasuya and
Yoshida. The RKKY exchange coefficient j oscillates from positive to negative as
the separation of the ion changes and has the damped oscillatory nature shown in
Fig. 15. Therefore depending on the separation between a pair of ions their
magnetic coupling can be ferromagnetic or antiferromagnetic. A magnetic ion
induces a spin polarization in the conduction electrons in its neighborhood. This
spin polarization in the itinerant electrons is felt by the moments of other magnetic
ions within the range leading to an indirect coupling.
In rare-earth metals, whose magnetic electrons in the 4f shell are shielded by
the 5s and 5p electrons, direct exchange is rather and indirect exchange via the
conduction electrons gives rise to magnetic order in these materials.
Fig. 15: The coefficient of indirect (RKKY) exchange versus the interatomic spacing
a.
MnO
• Superexchange in antiferromagnetic
38
Figure 16: Antiferromagnetic MnO [Reprinted from Blundell, Stephen;
1st
Magnetism in Condensed Matter,
Edition; Oxford Univ. Press]
• electrons interact with neighbouring electrons
⇒ direct exchange, no intermediary
•
BUT: direct exchange often not possible
⇒ insufficient direct overlap between orbitals
⇒ for example, 4f electrons in rare earths strongly localized, close to the nucleus,
so direct exchange interaction uneffective.
The exchange interaction described in the previous section stated a direct
exchange, so electrons would interact only with their next neighbours. But if the
electrons are strongly localised there is only a small probability for them to interact
with electrons an neighbouring atoms. This is for example the case for the 4f
electrons in the rare earths but MnO still exhibits the characteristics of an
antiferromagnet. The reason is that in the MnO-lattice (see Fig. 16) the Oxygen
O2
takes up the rôle of an intermediary transmitting the exchange forces. This is then
called superexchange.
4.2 Anisotropic
exchange: Crystalline anisotropy
By looking at the Heisenberg Hamiltonian:
39
ur uu
r
µ = −2
H
J
S
.
S
∑ i> j ij i j ,
It is obvious that the asscociated exchange energy and therefore the
magnetization should be totally isotropic, as
µ
H
is invariant with respect to choice
of coordinate systems. So the experimental fact, that the magnetization curve of
iron depends on the positioning of the lattice in respect to the external magnetic
field, is very surprising. How can the Heisenberg Hamiltonian, which is only spin
dependent, be affected by the structure of the lattice? The answer is simple: Even
though the spin part
χ
Φ
and the spatial part
of the wave function do not depend
on the same variables they are connected by the spin-orbit coupling! So
transmitts the structure of the environment (symmetry of the lattice) onto
χ
Φ
via
LS-coupling (→ [Bl])!
exchangeuu
depends on scalar
renergy
uu
r
S1.S 2
product
⇒ invariant with respect to choice of
coord. sys.
⇒ magnetization of ferromagnets
considered isotropic
BUT: experimentally was found:
⇒ magnetization lies along certain axes
Anisotropy is caused by spin-orbit
coupling
Figure 17: Magnetization curve of
iron [Reprinted from Heiko Lueken,
Magnetochemie, 1. Auflage;
Teubner Verlag]
40
•
spatial wave functions reflect symmetry of the lattice because of the
interactions between the lattice atoms that arise from the crystalline electric
•
fields and the overlap of the wave functions
via spin-orbit coupling, spins are made aware of this anisotropy
41
CONCLUSIONS
We have seen that magnetic properties of matter originate essentially from
the magnetic moment of electrons in incomplete shells in the atoms and from
unpaired electrons. The incomplete shell may be, for example, the 3d shell in the
case of the elements of the iron group, or the 4f shell in the rare earths (Sections 1
and 2). The cause for some solids, called (anti)ferromagnets, to exhibit
macroscopic magnetic characteristics was given by the exchange interaction, which
is quantum mechanical in nature. Dipolar interaction, which has a tendency to align
microscopic magnetic moments as well, does not present a cause for
(anti)ferromagnetism due to energetical reasons (Section 3). This text was then
concluded by qualitatively explaining why the exchange interaction, though short
range in nature, is able to cause ferromagnetic characteristics in compounds like
MnO where there is no direct overlap of the Mn-orbitals and how anisotropy can
occure in crystals even though the Heisenberg hamiltonian, describing
(anti)ferromagnetism, is totally isotropic (Section 4).
42
REFERENCES
[1]
[2]
Lueken, Heiko; Magnetochemie, 1. Auflage; Teubner Verlag
Blundell, Stephen; Magnetism in Condensed Matter,
1st
Edition; Oxford
Univ. Press
[3]
H. Haken, H.C. Wolf; Atom- und Quantenphysik, 8. Auflage; Springer
Verlag
[4] Demtröder, Wolfgang; Experimentalphysik Band 3, 3. Auflage; Springer
Verlag
1st
[5]
J. Stöhr, H.C. Siegmann; Magnetism,
Edition; Springer Verlag
[6]
Bringer, Andreas; Heisenberg Model; Institut für Festkörperforschung
and Institut for Advanced Simulation, Forschungszentrum Jülich, D52425 Jülich,
Germany
[7]
Aharoni, Amikam; Introduction to the Theory of Ferromagnetism,
2nd
Edition; Oxford Univ. Press
[8]
Guimarães, A. P.; Magnetism and Magnetic Resonance in Solids,
1st
John Wiley & Sons, Inc.
[9]
Simonds, J. L. (1995), “Magnetoelectronics Today and Tomorrow”,
Physics Today 48, 26-32.
43
Edition;