Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Ferromagnetic semiconductors

A giant appears in spintronics

The unusual synergy between magnetism and semiconductivity in the ferromagnetic semiconductor GaMnAs makes it an attractive material for 'spintronics'. Reports of a new giant switching effect lend it an even greater allure.

The exciting prospect of semiconductor-based information processing (logic) and magnetic-based data storage (memory) operating together on the same device has created the fledgling interdisciplinary field of spintronics (spin electronics). Compared with existing charge-based microelectronics technology, the ability to control and manipulate the dynamics of both carrier charge and spin by external electric and magnetic fields (as well as light) is expected to lead to novel spintronics applications1,2. An important step in this direction is the observation3 by two research groups at Caltech and the University of California, Santa Barbara, of very large resistance jumps and associated 'giant' switching effects in epitaxial ferromagnetic semi-conductors (GaMnAs). The observations reported by Tang et al.3 in Physical Review Letters are intriguing and as yet unexplained. Much research will be needed to understand (and precisely control) these switching effects before any possible applications can be envisaged. But the size of the observed resistive jumps are so large — four orders of magnitude larger than the corresponding effect in metallic ferromagnets — that the results could turn out to be a key spintronics discovery.

Current information technology is based on a fundamental dichotomy: logic operations and information processing are carried out in semiconductor microprocessors, whereas data storage and memory reside in magnetic hard disks. An important spintronics goal is to create multifunctional novel materials that are both semiconducting and ferromagnetic, thus allowing in principle the possibility of combining logic and storage operations on the same chip. During the past five years, researchers have been actively studying and enhancing a new class of ferromagnetic semiconductor materials, the so-called diluted magnetic semiconductors, which are both semiconducting and (below a magnetic transition temperature, Tc) ferromagnetic4,5,6,7,8.

The most studied and (from technological and materials science perspectives) the most exciting of these ferromagnetic semiconductor materials is Ga1–xMnxAs (with x ≈ 0.05) where a small amount (about 5%) of Ga atoms in GaAs are carefully replaced, using low-temperature molecular beam epitaxy, by magnetically active Mn atoms leading to a ferromagnetic semiconductor with Tc 30–150 K. (The exact Tc of a particular sample depends on the Mn concentration and on many details of materials growth and preparation that are not yet completely understood, and is a matter of active research5,6.)

The exciting spintronics potential of GaMnAs arises not just from its simultaneous semiconducting and ferromagnetic nature, but also from the curious fact that ferromagnetism in GaMnAs results from a long-range coupling between the Mn atoms mediated by charge carriers (in this case, holes in the valence band of GaMnAs). These charge carriers in GaMnAs are also created or 'donated' by the Mn atoms, which act simultaneously as magnetic atoms and charge donors in this system4,5,6. This unusual synergy between magnetism and semiconductivity allows for the interesting (and potentially technologically useful) possibility of modulating magnetic behaviour by controlling the charge-carrier properties (for example, by using an external electric field or light) and vice versa. This interplay between electrical and magnetic properties of GaMnAs has already been reported in the literature7,8. Tang et al.3 offer an interesting observation on the apparent effect of the magnetic properties of GaMnAs on its semiconducting carrier transport behaviour.

In this general context, the observations by Tang and colleagues seem so surprising and serendipitous because their GaMnAs samples are quite ordinary. They use 150-nm-thick GaMnAs thin films on insulating GaAs substrates with a rather modest Tc 45 K (compared with the best-optimized GaMnAs samples having Tc as high as 120–150 K). As a semiconductor, the samples are also fairly ordinary, having mobilities of only about 4 cm2 V−1 s−1, compared with the best GaAs hole systems, which have mobilities easily a thousand times larger. The experimental methods used by Tang et al. are also fairly routine, involving the measurement of the 'planar' Hall resistance, in which a weak magnetic field is applied along the plane in the Hall bar geometry (Fig. 1a), and the transverse 'Hall' voltage (perpendicular to the direction of the current) is measured.

Figure 1: The giant planar Hall effect.
figure1

a, The Hall effect arises because a moving charge, q, in a magnetic field, H, experiences a force (the Lorentz force) proportional to v × H, and perpendicular to the direction of current flow. In a conductor (metal or semiconductor) this leads to the classical Hall resistance, RH, which is proportional to the carrier density and the external magnetic field, and can be measured from the transverse Hall voltage, VH. b, In the planar Hall effect, a weak magnetic field is applied along the plane of the sample (at an angle to the direction of the current flow) and the 'planar' RH is measured from the transverse VH. In their experiment, Tang et al.3 measure the planar RH in their GaMnAs samples in response to an applied in-plane magnetic field. The planar RH jumps are roughly the same size whether the sample is 1 mm wide (top) or 6 μm wide (bottom and SEM micrograph). Two distinct jumps (red and blue) are seen in the experiment.

In the Hall effect, a voltage develops transverse to the direction of current flow in a conductor (either a metal or a semiconductor) in the presence of an applied magnetic field9 (Fig. 1a). It is a classical phenomenon, arising from the fact that a moving charge in a magnetic field H experiences a force (the Lorentz force) proportional to v × H, perpendicular to the direction of the charge velocity, v (that is, transverse to the current direction). In a non-magnetic system this leads to the classical Hall resistance, RH, which is proportional to the carrier density and the external magnetic field. Thus, the Hall measurement can be used either for a quantitative estimate of the carrier density or as a sensitive magnetic field sensor. (In a two-dimensional semiconductor there can also be a quantization of RH at very low temperatures leading to the quantized Hall effect, which, however, has no relevance to the situation of interest here.)

When Tang et al. measured the planar Hall resistance in GaMnAs at liquid helium temperature (4.2 K) they saw very large RH jumps (80 Ω) and associated switching effects. The effect appears to be independent of the device length over a 6 μm − 1 mm range (Fig. 1b). This observed 'giant' Hall effect is entirely unexpected because the corresponding RH jumps in metallic ferromagnets are only in the mΩ range (that is, more than 10,000 times weaker). In principle, this large an effect should enable a very sensitive measurement of the external magnetic field; it is also robust enough to be of potential interest in memory and storage technology. In addition, because GaMnAs is a semiconductor, the prospect of creating a device with logic and memory functionalities on the same GaMnAs chip is now, in principle, closer to reality. It should, however, be emphasized that this is still a low-temperature effect (4.2 K), and any widespread applications must await the development of materials in which the large planar RH jumps can be observed at room temperature or above.

The physical mechanism underlying these intriguing observations is unknown at this stage. The Hall effect in ferromagnetic materials is a highly complex subject, and to date it lacks a complete theoretical understanding. Complications arise because a ferromagnetic material has intrinsic internal magnetic fields associated with its spontaneous magnetization. Therefore, one must consider not only the regular Hall effect, but also the Hall effect associated with the magnetization. The invariable presence of the spin-orbit coupling in GaMnAs, which makes the orbital carrier dynamics (associated with electric current) intrinsically connected to the carrier spin dynamics (associated with ferromagnetism), further complicates the situation, not least because it leads to large anisotropies in the system that affect the Hall resistance. In addition, the detailed scattering mechanisms and magnetic domain structures affecting transport properties in GaMnAs are poorly understood and need to be further studied.

All these complications will undoubtedly keep researchers busy for a while trying to figure out how and why such a large planar Hall resistance switching effect occurs in GaMnAs microdevices (and how to control and enhance the effect, eventually, if possible, leading to a room-temperature effect). Nonetheless, by clearly showing that the giant planar Hall effect is also correlated with the system magnetization, the results of Tang et al. certainly open up an unexpected and important new direction for spintronics research in ferromagnetic semiconductors.

References

  1. 1

    Das Sarma, S. Am. Sci. 89, 516–523 (2001).

    Article  Google Scholar 

  2. 2

    Wolf, S.A. et al. Science 294, 1488–1495 (2001).

    CAS  Article  Google Scholar 

  3. 3

    Tang, H.X., Kawakami, R.K., Awschalom, D.D. & Roukes, M.L. Phys. Rev. Lett. 90, 107201 (2003).

    CAS  Article  Google Scholar 

  4. 4

    Ohno, H. Science 281, 951–956 (1998).

    CAS  Article  Google Scholar 

  5. 5

    Dietl, T., Ohno, H., Matsukura, F., Cibert, J. & Ferrand, D. Science 287, 1019–1022 (2000).

    CAS  Article  Google Scholar 

  6. 6

    Kaminski, A. & Das Sarma, S. Phys. Rev. Lett. 88, 247202 (2002).

    CAS  Article  Google Scholar 

  7. 7

    Ohno, H. et al. Nature 408, 944–946 (2000).

    CAS  Article  Google Scholar 

  8. 8

    Ohno, Y. et al. Nature 402, 790–792 (1999).

    CAS  Article  Google Scholar 

  9. 9

    Hall, E. Amer. J. Math. 2, 287–292 (1879).

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Das Sarma, S. A giant appears in spintronics. Nature Mater 2, 292–294 (2003). https://doi.org/10.1038/nmat883

Download citation

Further reading

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing