Magnetic semiconductors are
semiconductor materials that exhibit both
ferromagnetism (or a similar response) and useful
semiconductor properties. If implemented in devices, these materials could provide a new type of control of conduction. Whereas traditional electronics are based on control of
charge carriers (
n- or
p-type), practical magnetic semiconductors would also allow control of quantum
spin state (up or down). This would theoretically provide near-total
spin polarization (as opposed to
iron and other metals, which provide only ~50% polarization), which is an important property for
spintronics applications, e.g.
spin transistors.
While many traditional magnetic materials, such as
magnetite, are also semiconductors (magnetite is a
semimetal semiconductor with
bandgap 0.14 eV), materials scientists generally predict that magnetic semiconductors will only find widespread use if they are similar to well-developed semiconductor materials. To that end, dilute magnetic semiconductors (DMS) have recently been a major focus of magnetic semiconductor research. These are based on traditional semiconductors, but are
doped with
transition metals instead of, or in addition to, electronically active elements. They are of interest because of their unique
spintronics properties with possible technological applications.[1][2] Doped
wide band-gap metal oxides such as
zinc oxide (ZnO) and
titanium oxide (TiO2) are among the best candidates for industrial DMS due to their multifunctionality in
opticomagnetic applications. In particular, ZnO-based DMS with properties such as transparency in visual region and
piezoelectricity have generated huge interest among the scientific community as a strong candidate for the fabrication of
spin transistors and
spin-polarizedlight-emitting diodes,[3] while
copper doped TiO2 in the
anatase phase of this material has further been predicted to exhibit favorable dilute magnetism.[4]
The pioneering work of Dietl et al. showed that a modified Zener model for magnetism[7]
well describes the carrier dependence, as well as anisotropic properties of
GaMnAs.
The same theory also
predicted that room-temperature
ferromagnetism should exist in heavily
p-typedoped ZnO and GaN doped by Co and Mn, respectively.
These predictions were followed of a flurry of theoretical and experimental studies of various oxide and nitride semiconductors,
which apparently seemed to confirm room temperature ferromagnetism in nearly any semiconductor or insulator material
heavily doped by
transition metal impurities.
However, early
Density functional theory (DFT) studies were clouded by band gap errors and overly delocalized defect levels,
and more advanced DFT studies refute most of the previous predictions of ferromagnetism.[8]
Likewise, it has been shown that for most of the oxide based materials studies for magnetic semiconductors
do not exhibit an intrinsic carrier-mediated ferromagnetism as postulated by Dietl et al.[9]
To date,
GaMnAs remains the only semiconductor material with robust coexistence of ferromagnetism persisting up to rather high Curie temperatures around 100–200 K.
The manufacturability of the materials depend on the thermal equilibrium
solubility of the
dopant in the base material. E.g., solubility of many dopants in
zinc oxide is high enough to prepare the materials in bulk, while some other materials have so low solubility of dopants that to prepare them with high enough dopant concentration thermal nonequilibrium preparation mechanisms have to be employed, e.g. growth of
thin films.
Permanent magnetization has been observed in a wide range of semiconductor based materials.
Some of them exhibit a clear correlation between
carrier density and magnetization,
including the work of
T. Story and co-workers where they demonstrated that the ferromagnetic Curie temperature of
Mn2+-doped
Pb1−xSnxTe can be controlled by the
carrier concentration.[10]
The theory proposed by Dietl required
charge carriers in the case of
holes to mediate the
magnetic coupling of manganese
dopants in the prototypical magnetic semiconductor, Mn2+-doped
GaAs. If there is an insufficient hole concentration in the magnetic semiconductor, then the
Curie temperature would be very low or would exhibit only
paramagnetism. However, if the hole concentration is high (>~1020 cm−3), then the
Curie temperature would be higher, between 100 and 200 K.
[7]
However, many of the semiconductor materials studied exhibit a permanent magnetization extrinsic
to the semiconductor host material.[9]
A lot of the elusive extrinsic ferromagnetism (or phantom ferromagnetism)
is observed in thin films or nanostructured materials.[11]
Several examples of proposed ferromagnetic semiconductor materials are listed below. Notice that many of the observations and/or predictions below remain heavily debated.
Manganese- and
iron-doped
indium oxide, ferromagnetic at room temperature. The ferromagnetism appears to be mediated by carrier-electrons,[14][15] in a similar way as the
GaMnAs ferromagnetism is mediated by carrier-holes.
^Philip, J.; Punnoose, A.; Kim, B. I.; Reddy, K. M.; Layne, S.; Holmes, J. O.; Satpati, B.; LeClair, P. R.; Santos, T. S. (April 2006). "Carrier-controlled ferromagnetism in transparent oxide semiconductors". Nature Materials. 5 (4): 298–304.
Bibcode:
2006NatMa...5..298P.
doi:
10.1038/nmat1613.
ISSN1476-1122.
PMID16547517.
S2CID30009354.
^Lany, Stephan; Raebiger, Hannes; Zunger, Alex (2008-06-03). "Magnetic interactions of Cr − Cr and Co − Co impurity pairs in ZnO within a band-gap corrected density functional approach". Physical Review B. 77 (24): 241201.
Bibcode:
2008PhRvB..77x1201L.
doi:
10.1103/PhysRevB.77.241201.
ISSN1098-0121.