This action might not be possible to undo. Are you sure you want to continue?
Spintronics: Fundamentals and applications
ˇ ´ Igor Zutic*
Condensed Matter Theory Center, Department of Physics, University of Maryland at College Park, College Park, Maryland 20742-4111, USA
¨ Institute for Theoretical Physics, Karl-Franzens University, Universitatsplatz 5, 8010 Graz, Austria
S. Das Sarma
Condensed Matter Theory Center, Department of Physics, University of Maryland at College Park, College Park, Maryland 20742-4111, USA
(Published 23 April 2004)
Spintronics, or spin electronics, involves the study of active control and manipulation of spin degrees of freedom in solid-state systems. This article reviews the current status of this subject, including both recent advances and well-established results. The primary focus is on the basic physical principles underlying the generation of carrier spin polarization, spin dynamics, and spin-polarized transport in semiconductors and metals. Spin transport differs from charge transport in that spin is a nonconserved quantity in solids due to spin-orbit and hyperﬁne coupling. The authors discuss in detail spin decoherence mechanisms in metals and semiconductors. Various theories of spin injection and spin-polarized transport are applied to hybrid structures relevant to spin-based devices and fundamental studies of materials properties. Experimental work is reviewed with the emphasis on projected applications, in which external electric and magnetic ﬁelds and illumination by light will be used to control spin and charge dynamics to create new functionalities not feasible or ineffective with conventional electronics.
I. Introduction A. Overview B. History and background 1. Spin-polarized transport and magnetoresistive effects 2. Spin injection and optical orientation II. Generation of Spin Polarization A. Introduction B. Optical spin orientation C. Theories of spin injection 1. F/N junction 2. F/N/F junction 3. Spin injection through the space-charge region D. Experiments on spin injection 1. Johnson-Silsbee spin injection 2. Spin injection into metals 3. All-semiconductor spin injection 4. Metallic ferromagnet/semiconductor junctions III. Spin Relaxation and Spin Dephasing A. Introduction 1. T 1 and T 2 2. Experimental probes B. Mechanisms of spin relaxation 1. Elliott-Yafet mechanism 323 323 325 325 328 329 329 331 333 333 337 338 340 340 341 342 345 346 346 347 348 349 349
D’yakonov-Perel’ mechanism a. Bulk III-V semiconductors b. Two-dimensional III-V semiconductor systems 3. Bir-Aronov-Pikus mechanism 4. Hyperﬁne-interaction mechanism C. Spin relaxation in metals D. Spin relaxation in semiconductors 1. Bulk semiconductors 2. Low-dimensional semiconductor structures 3. Example: Spin relaxation in GaAs a. Bulk n-GaAs b. GaAs-based quantum wells IV. Spintronic Devices and Applications A. Spin-polarized transport 1. F/I/S tunneling 2. F/I/F tunneling 3. Andreev reﬂection 4. Spin-polarized drift and diffusion B. Materials considerations C. Spin ﬁlters D. Spin diodes E. Spin transistors 1. Spin ﬁeld-effect transistors 2. Magnetic bipolar transistor 3. Hot-electron spin transistors F. Spin qubits in semiconductor nanostructures V. Outlook Acknowledgments References
351 353 354 356 357 358 360 360 361 363 363 365 366 366 366 368 371 372 373 376 377 380 380 381 382 384 385 386 386
*Present address: Center for Computational Materials Science, Naval Research Laboratory, Washington, D.C. 20735, USA. Electronic address: firstname.lastname@example.org † Electronic address: email@example.com
Spintronics is a multidisciplinary ﬁeld whose central theme is the active manipulation of spin degrees of free©2004 The American Physical Society
ˇ ´ Zutic, Fabian, and Das Sarma: Spintronics: Fundamentals and applications
dom in solid-state systems.1 In this article the term spin stands for either the spin of a single electron s, which can be detected by its magnetic moment g B s ( B is the Bohr magneton and g is the electron g factor, in a solid generally different from the free-electron value of g 0 2.0023), or the average spin of an ensemble of electrons, manifested by magnetization. The control of spin is then a control of either the population and the phase of the spin of an ensemble of particles, or a coherent spin manipulation of a single or a few-spin system. The goal of spintronics is to understand the interaction between the particle spin and its solid-state environments and to make useful devices using the acquired knowledge. Fundamental studies of spintronics include investigations of spin transport in electronic materials, as well as of spin dynamics and spin relaxation. Typical questions that are posed are (a) what is an effective way to polarize a spin system? (b) how long is the system able to remember its spin orientation? and (c) how can spin be detected? Generation of spin polarization usually means creating a nonequilibrium spin population. This can be achieved in several ways. While traditionally spin has been oriented using optical techniques in which circularly polarized photons transfer their angular momenta to electrons, for device applications electrical spin injection is more desirable. In electrical spin injection a magnetic electrode is connected to the sample. When the current drives spin-polarized electrons from the electrode to the sample, nonequilibrium spin accumulates there. The rate of spin accumulation depends on spin relaxation, the process of bringing the accumulated spin population back to equilibrium. There are several mechanisms of spin relaxation, most involving spin-orbit coupling to provide the spin-dependent potential, in combination with momentum scattering to provide a randomizing force. Typical time scales for spin relaxation in electronic systems are measured in nanoseconds, while the range is from picoseconds to microseconds. Spin detection, also part of a generic spintronic scheme, typically relies on sensing the changes in the signals caused by the presence of nonequilibrium spin in the system. The common goal in many spintronic devices is to maximize the spin detection sensitivity to the point that it detects not the spin itself, but changes in the spin states. Let us illustrate the generic spintronic scheme on a prototypical device, the Datta-Das spin ﬁeld-effect transistor (SFET; Datta and Das, 1990), depicted in Fig. 1. The scheme shows the structure of the usual FET, with a drain, a source, a narrow channel, and a gate for controlling the current. The gate either allows the current to ﬂow (ON) or does not (OFF). The spin transistor is similar in that the result is also a control of the charge cur-
FIG. 1. (Color in online edition) Scheme of the Datta-Das spin ﬁeld-effect transistor (SFET). The source (spin injector) and the drain (spin detector) are ferromagnetic metals or semiconductors, with parallel magnetic moments. The injected spinpolarized electrons with wave vector k move ballistically along a quasi-one-dimensional channel formed by, for example, an InGaAs/InAlAs heterojunction in a plane normal to n. Electron spins precess about the precession vector , which arises from spin-orbit coupling and which is deﬁned by the structure and the materials properties of the channel. The magnitude of is tunable by the gate voltage V G at the top of the channel. The current is large if the electron spin at the drain points in the initial direction (top row)—for example, if the precession period is much larger than the time of ﬂight—and small if the direction is reversed (bottom).
rent through the narrow channel. The difference, however, is in the physical realization of the current control. In the Datta-Das SFET the source and the drain are ferromagnets acting as the injector and detector of the electron spin. The drain injects electrons with spins parallel to the transport direction. The electrons are transported ballistically through the channel. When they arrive at the drain, their spin is detected. In a simpliﬁed picture, the electron can enter the drain (ON) if its spin points in the same direction as the spin of the drain. Otherwise it is scattered away (OFF). The role of the gate is to generate an effective magnetic ﬁeld (in the direction of in Fig. 1), arising from the spin-orbit coupling in the substrate material, from the conﬁnement geometry of the transport channel, and the electrostatic potential of the gate. This effective magnetic ﬁeld causes the electron spins to precess. By modifying the voltage, one can cause the precession to lead to either parallel or antiparallel (or anything between) electron spin at the drain, effectively controlling the current. Even though the name spintronics is rather novel,2 contemporary research in spintronics relies closely on a long tradition of results obtained in diverse areas of physics (for example, magnetism, semiconductor physics, superconductivity, optics, and mesoscopic physics) and establishes new connections between its different ˇ ´ subﬁelds (Rashba, 2002c; Zutic, 2002a). We review here both well-established results and the physical principles
1 While there are proposals for spintronic devices based on deoxyribonucleic acid (DNA) molecules (Zwolak and Di Ventra, 2002), the whole device, which includes electrodes, voltage/current source, etc., is still a solid-state system.
The term was coined by S. A. Wolf in 1996, as a name for a DARPA initiative for novel magnetic materials and devices.
Rev. Mod. Phys., Vol. 76, No. 2, April 2004
ˇ ´ Zutic, Fabian, and Das Sarma: Spintronics: Fundamentals and applications
relevant to present and future applications. Our strategy is to give a comprehensive view of what has been accomplished, focusing in detail on a few selected topics that we believe are representative for the broader subject within which they appear. For example, while discussing the generation of spin polarization, we survey many experimental and theoretical studies of both optical orientation and electrical spin injection and present a detailed and self-contained formalism of electrical spin injection. Similarly, when we discuss spin relaxation, we give a catalog of important work, while studying spin relaxation in the cases of Al and GaAs as representative of the whole ﬁeld. Finally, in the section on spin devices we give detailed physical principles of several selected devices, such as, for example, the above-mentioned DattaDas SFET. There have been many other reviews written on spintronics, most focusing on a particular aspect of the ﬁeld. We divide them here, for an easier orientation, into two groups, those that cover the emerging applications3 and those covering already well-established schemes and materials.4 The latter group, often described as magnetoelectronics typically covers paramagnetic and ferromagnetic metals and insulators, which utilize magnetoresistive effects, realized, for example, as magnetic read heads in computer hard drives, nonvolatile magnetic random access memory (MRAM), and circuit isolators (Wang et al., 2002). These more established aspects of spintronics have also been addressed in several books5 and will be discussed in another review6 complementary to ours. Spintronics also beneﬁts from a large class of emerging materials, such as ferromagnetic semiconductors (Ohno, 1998; Pearton et al., 2003), organic semiconductors (Dediu et al., 2002), organic ferromagnets (Pejak´ ovic et al., 2002; Epstein, 2003), high-temperature superconductors (Goldman et al., 1999), and carbon nanotubes (Tsukagoshi et al., 1999; Zhao et al., 2002), which can bring novel functionalities to the traditional devices. There is a continuing need for fundamental studies before the potential of spintronic applications can be fully realized. After an overview, Sec. I covers some basic historical and background material, part of which has already been extensively covered in the context of magnetoelec-
tronics and will not be discussed further in this review. Techniques for generating spin polarization, focusing on optical spin orientation and electrical spin injection, are described in Sec. II. The underlying mechanisms responsible for the loss of spin orientation and coherence, which impose fundamental limits on the length and time scales in spintronic devices, are addressed in Sec. III. Spintronic applications and devices, with the emphasis on those based on semiconductors, are discussed in Sec. IV. The review concludes with a look at future prospects in Sec. V.
B. History and background
1. Spin-polarized transport and magnetoresistive effects
Reviews on emerging applications include those of Das Sarma et al. (2000a, 2000b, 2000c, 2001); Wolf and Treger (2000); Das Sarma (2001); Wolf et al. (2001); Oestreich et al. ˇ ´ (2002); Rashba (2002c); Zutic (2002a, 2002b). 4 Established schemes and materials are reviewed by Tedrow and Meservey (1994); Prinz (1995, 1998); Gijs and Bauer (1997); Gregg et al. (1997); Ansermet (1998); Bass and Pratt, Jr. (1999); Daughton et al. (1999); Stiles (2004). 5 See, for example, the books of Hartman (2000); Ziese and Thornton (2001); Hirota et al. (2002); Levy and Mertig (2002); Maekawa et al. (2002); Parkin (2002); Shinjo (2002); and Chtchelkanova et al. (2003). 6 In preparation by S. S. P. Parkin for Reviews of Modern Physics.
Rev. Mod. Phys., Vol. 76, No. 2, April 2004
In a pioneering work, Mott (1936a, 1936b) provided a basis for our understanding of spin-polarized transport. Mott sought an explanation for an unusual behavior of resistance in ferromagnetic metals. He realized that at sufﬁciently low temperatures, where magnon scattering becomes vanishingly small, electrons of majority and minority spin, with magnetic moment parallel and antiparallel to the magnetization of a ferromagnet, respectively, do not mix in the scattering processes. The conductivity can then be expressed as the sum of two independent and unequal parts for two different spin projections— the current in ferromagnets is spin polarized. This is also known as the two-current model and has been extended by Campbell et al. (1967) and Fert and Campbell (1968). It continues, in its modiﬁcations, to provide an explanation for various magnetoresistive phenomena (Valet and Fert, 1993). Tunneling measurements played a key role in early experimental work on spin-polarized transport. Studying N/F/N junctions, where N was a nonmagnetic7 metal and F was an Eu-based ferromagnetic semiconductor (Kasuya and Yanase, 1968; Nagaev, 1983), revealed that I-V curves could be modiﬁed by an applied magnetic ﬁeld (Esaki et al., 1967) and now show potential for developing a solid-state spin ﬁlter. When unpolarized current is passed across a ferromagnetic semiconductor, the current becomes spin-polarized (Moodera et al., 1988; Hao et al., 1990). A series of experiments (Tedrow and Meservey, 1971b, 1973, 1994) in ferromagnet/insulator/ superconductor (F/I/S) junctions has unambiguously proved that the tunneling current remains spin polarized even outside of the ferromagnetic region.8 The Zeeman-
Unless explicitly speciﬁed, we shall use the terms ‘‘nonmagnetic’’ and ‘‘paramagnetic’’ interchangeably, i.e., assume that they both refer to a material with no long-range ferromagnetic order and with Zeeman-split carrier spin subbands in an applied magnetic ﬁeld. 8 It has been shown that electrons photoemitted from ferromagnetic gadolinium remain spin polarized (Busch et al., 1969).
ˇ ´ Zutic, Fabian, and Das Sarma: Spintronics: Fundamentals and applications
FIG. 2. (Color in online edition) Schematic illustration of electron tunneling in ferromagnet / insulator / ferromagnet (F/I/F) tunnel junctions: (a) Parallel and (b) antiparallel orientation of magnetizations with the corresponding spinresolved density of the d states in ferromagnetic metals that have exchange spin splitting ex . Arrows in the two ferromagnetic regions are determined by the majority-spin subband. Dashed lines depict spinconserved tunneling.
split quasiparticle density of states in a superconductor (Tedrow et al., 1970; Fulde, 1973) was used as a detector of spin polarization of conduction electrons in various ` magnetic materials. Julliere (1975) measured tunneling conductance of F/I/F junctions, where I was an amorphous Ge. By adopting the Tedrow and Meservey (1971b, 1973) analysis of the tunneling conductance ` from F/I/S to the F/I/F junctions, Julliere (1975) formulated a model for a change of conductance between the parallel (↑↑) and antiparallel (↑↓) magnetization in the two ferromagnetic regions F1 and F2, as depicted in Fig. 2. The corresponding tunneling magnetoresistance9 (TMR) in an F/I/F magnetic tunnel junction (MTJ) is deﬁned as TMR R R ↑↑ R ↑↓ R ↑↑ R ↑↑ G ↑↑ G ↑↓ , G ↑↓ (1)
charge current (for example, parallel or perpendicular) with respect to the direction of magnetization. ` Within Julliere’s model, which assumes constant tunneling matrix elements and that electrons tunnel without spin ﬂip, Eq. (1) yields 2P 1 P 2 , 1 P 1P 2
where conductance G and resistance R 1/G are labeled by the relative orientations of the magnetizations in F1 and F2 (it is possible to change the relative orientations, between ↑↑ and ↑↓, even at small applied magnetic ﬁelds 10 G). TMR is a particular manifestation of a magnetoresistance that yields a change of electrical resistance in the presence of an external magnetic ﬁeld.10 Historically, the anisotropic magnetoresistance in bulk ferromagnets such as Fe and Ni was discovered ﬁrst, dating back to the experiments of Lord Kelvin (Thomson, 1857). Due to spin-orbit interaction, electrical resistivity changes with the relative direction of the
where the polarization P i (NMi Nmi )/(NMi Nmi ) is expressed in terms of the spin-resolved density of states NMi and Nmi , for majority and minority spin in Fi , respectively. Conductance in Eq. (1) can then be expressed ¨ as (Maekawa and Gafvert, 1982) G ↑↑ NM1 NM2 Nm1 Nm2 and G ↑↓ NM1 Nm2 Nm1 NM2 to give Eq. ` (2).11 While the early results of Julliere (1975) were not conﬁrmed, TMR at 4.2 K was observed using NiO as a ¨ tunnel barrier by Maekawa and Gafvert (1982). ` The prediction of Julliere’s model illustrates the spinvalve effect: the resistance of a device can be changed by manipulating the relative orientation of the magnetizations M1 and M2 , in F1 and F2, respectively. Such orientation can be preserved even in the absence of a power supply, and the spin-valve effect,12 later discovered in multilayer structures displaying the giant magnetoresistance13 (GMR) effect (Baibich et al., 1988; Binasch et al., 1989) can be used for nonvolatile memory applications (Hartman, 2000; Hirota et al., 2002; Parkin,
` Starting with Julliere (1975) an equivalent expression (G ↑↑ G ↑↓ )/G ↑↑ has also been used by different authors and is often referred to as junction magnetoresistance (Moodera and Mathon, 1999). 10 The concept of TMR was proposed independently by R. C. Barker in 1975 [see Meservey et al. (1983)] and by Slonczewski (1976), who envisioned its use for magnetic bubble memory (Parkin, 2002).
Rev. Mod. Phys., Vol. 76, No. 2, April 2004
` In Sec. IV we address some limitations of the Julliere model and its potential ambiguities to identify precisely which spin polarization is actually measured. 12 The term was coined by Dieny et al. (1991) in the context of GMR, by invoking an analogy with the physics of the TMR. 13 The term ‘‘giant’’ reﬂected the magnitude of the effect (more than 10%), as compared to the better known anisotropic magnetoresistance ( 1%).
the discovery of large room-temperature TMR (Miyazaki and Tezuka. known as spin-transfer torque. Boukari et al. 16 Realization of the early magnetic random-access memory proposals used the effect of anisotropic magnetoresistance (Pohn et al. 2000). while the CPP version.. Magnetoresistance effects of similar magnitude have also been found in hybrid metal/ semiconductor granular ﬁlms (Akinaga. One of the keys to the success of magnetoresistancebased applications is their ability to control14 the relative orientation of M1 and M2 .. Park et al. are being investigated both as the nonferromagnetic region in MTJ’s and as the basis for an all-semiconductor junction that would demonstrate large TMR at low temperatures (Tanaka and Higo. The ﬂow of spin-polarized current can transfer angular momentum from carriers to ferromagnet and alter the orientation of the corresponding magnetization. Oiwa et al.. et al. 1998. Roche. and Roche. such as CrO2 (Soulen et al. 2003).... current in plane) or perpendicular (CPP. and Mueller (1983). Bhadra. and current-induced magnetization reversal has been demonstrated at room temperature (Katine et al. 2001. such a structure could be thought of as the For example. ` Julliere’s model also justiﬁes the continued quest for highly spin-polarized materials—they would provide large magnetoresistive effects. 2000. Parker et al. 1999. Falicov et al. Another approach to obtaining large room-temperature magnetoresistance ( 100% at B 100 G) is to fabricate ferromagnetic regions separated by a nanosize contact. 1987. 1995. (1991). et al. and Smith. Hirota et al. 2000. which could lead to improved magnetic read heads (Solin et al.. even in the absence of an applied magnetic ﬁeld. Myers et al. Tsoi et al. Fabian. 2002. desirable for device applications. (1). 2002) and electrically. In addition to ferromagnets. semiconductors. 2000).. which are now the basis for the several magnetic random-access memory prototypes16 (Parkin. Future generations of magnetic read heads are expected to use MTJ’s instead of CIP giant magnetoresonance. the density of states vanishes completely for the other spin. Mills. 2000). 2002) are based on the GMR effects. Moussa et al... and Das Sarma: Spintronics: Fundamentals and applications 327 FIG.. and these materials were reviewed by Pickett and Moodera (2001). 2000. 1999. 76. 2003). 3.. 2. reaching 750 000% at a magnetic ﬁeld of 4 T (Solin et al. An interesting realization of such control was proposed independently by Berger (1996) and Slonczewski (1996)... Nadgorny. Waintal et al.. Moodera et al. It was also shown that the magnetic ﬁeld generated by passing the current through a CPP giant magnetoresonance device could produce roomtemperature magnetization reversal (Bussmann et al. Boukari et al. 1998.. current perpendicular to the plane) to the interfaces between the different layers. Realization of a large roomtemperature GMR (Parkin. Levy and Mertig.. 2000. While in GMR or TMR structures the relative orientation of magnetizations will affect the ﬂow of spin-polarized current. 2002) and relates to the physics of the tunneling magnetoresistance effect (Mathon and Umerski.. 2002). 2002) and manganite perovskites (Park et al. 1967. there is evidence for high spin polarization in III-V ferromagnetic semiconductors like (Ga. 2003).. GMR structures are often classiﬁed according to whether the current ﬂows parallel (CIP. Tehrani et al. This phenomenon. with small magnetic ﬁeld (Parkin.. 1997). Sun. 2002. Tanaka.. 2002) or at high switching speeds (Schumacher et al. Stiles and Zangwill.. Vol. 1995) has renewed interest in the study of magnetic tunnel junctions. Most of the GMR applications use the CIP geometry. 1988). 1997. Mod. 2003a. where parallel and antiparallel orientations of the magnetizations in the two ferromagnetic regions are often denoted by ‘‘P’’ and ‘‘AP.’’ respectively (instead of ↑↑ and ↑↓). which would provide a lower tunneling barrier than the usually employed oxides. The challenge remains to preserve such spin polarization above room temperature and in junctions with other materials. No. 15 . since the surface (interface) and bulk magnetic properties can be signiﬁcantly different (Fisher. 1991) enabled a quick transition from basic physics to commercial applications in magnetic recording (Parkin. as depicted in Fig. 2002). et al. ﬁrst realized by Pratt et al. Another desirable property of semiconductors has been demonstrated by the extraordinary large room-temperature magnetoresistance in hybrid structures with metals. has since been extensively studied both theoretically and experimentally (Bazaliy et al. Janner. 2003.. In the context of ferromagnetic semiconductors additional control of magnetization was demonstrated optically. which determines the RC time constant of the MTJ cell. April 2004 14 Near the Fermi level they behave as metals only for one spin. by shining light (Koshihara et al.. Consequently. While many existing spintronic applications (Hartman. Numerical support for the existence of such materials—the socalled half-metallic ferromagnets15—was provided by de Groot. they predicted a reverse effect. by applying gate voltage (Ohno..ˇ ´ Zutic. Parkin. 2002). 1998a). Li.. Jiang. 1990). et al. Rev. 2002). 1999). 1997.. To improve the switching performance of related devices it is important to reduce the junction resistance. spins would be completely polarized even in the absence of magnetic ﬁeld. In an extreme case. 2003b). (Color in online edition) Schematic illustration of (a) the current in plane (CIP)... 2002. Phys. For simplicity. 3. 1998. (b) the current perpendicular to the plane (CPP) giant magnetoresistance geometry.Mn)As (Braden et al. The size of magnetoresistance in the GMR structures can be expressed analogously to Eq. 1991. Panguluri. is easier to analyze theoretically (Gijs and Bauer. Chiba. 1971. 2002) to perform switching between the ferromagnetic and paramagnetic states.
2000. discussed in Sec. when a charge current ﬂowed across the F/N junction (Fig. 4. 1999. However. Versluijs et al. 2. depicted in Fig. 2. 1989). Electrons ﬂow from F to N: (a) schematic device geometry.. Tsuei and Kirtley. spin-polarized carriers in a ferromagnet would contribute to the net current of magnetization entering the nonmagnetic region and would lead to nonequilibrium magnetization M. 2000.18 in which the hyperﬁne coupling between the electron and nuclear spins. Feher (1959) suggested several other methods. and Aronov and Pikus (1976) established several key concepts in electrical spin injection from ferromagnets into metals. for example. April 2004 .. it has been demonstrated that short spin relaxation times are useful in the implementation of fast switching (Nishikawa et al. Nonequilibrium spin is the result of some source of pumping arising from transport. 2004) and the creation of spin-polarized beams to measure parity violation in high-energy physics (Pierce and Celotta. 20 In an earlier work.. are not transport currents which could be employed for transporting spins and spin injection. it can also be used as a sensitive spectroscopic tool to study a wide variety of fundamental properties ranging from spin-orbit and hyperﬁne interactions (Meier and Zakharchenya. II. 1999. Spin injection and optical orientation Many materials in their ferromagnetic state can have a substantial degree of equilibrium carrier spin polarization. 18 The importance and possible applications of the Feher effect (Feher. sponsible for the dynamical nuclear polarization (Slichter. 4(b). Strand et al. No. which typically require current ﬂow and/or manipulation of the nonequilibrium spin (polarization). 1984). While for most applications it is desirable to have long spin relaxation times. is re- FIG.. 1984). 3(b). in which the use of a resonant microwave excitation causes the spin relaxation of the nonequilibrium electron population through hyperﬁne coupling to lead to the spin polarization of nuclei. 1988d). 1976b). Vol. Chung et al. 2003). who drove a direct current through a sample of InSb in the presence of a constant applied magnetic ﬁeld. it has been suggested that the spin-orbit interaction could lead to dissipationless spin currents in hole-doped semiconductors (Murakami et al. as illustrated in Fig. Phys. 1997. Aronov (1976b) predicted that. or resonance methods... Ngai et al. The principle was based on the Feher effect. 1976b. Aronov calculated that the electrical spin injection would polarize nuclei and lead to a measurable effect in the electron spin resonance.. Wei et al.17 The importance of generating nonequilibrium spin is not limited to device applications.. together with different temperatures representing electron velocity and electron spin populations. 1973). due to spin-orbit interaction. 4). semiconductors.. 2002). Garcia et al. (c) contribution of different spin-resolved densities of states to both charge and spin transport across the F/N interface. Such an effect can be thought of as a generalization of the Overhauser effect (Overhauser. Using an analogy with the quantum Hall effect.19 Motivated by the work of Clark and Feher (1963) and Tedrow and Meservey (1971b. Rashba (2003b) has pointed out that similar dissipationless spin currents in thermodynamic equilibrium. 19 Rev. and Das Sarma: Spintronics: Fundamentals and applications limiting case of the CPP giant magnetoresonance scheme in Fig. which is also equivalent to a nonequilibrium spin accumulation. that could produce a nonequilibrium electron population and yield a dynamical polarization of nuclei (see also Weger.. Electrical spin injection. 2001. 1999. 1. has already been realized experimentally by Clark and Feher (1963). 1959) to polarize electrons was discussed by Das Sarma et al. an example of a transport method for generating nonequilibrium spin. accompanied by dissipationless spin currents. Measuring polarization of electroluminescence was suggested as a technique for detecting injection of polarized carriers in a semiconductor (Scifres et al. Several decades later related experiments on spin injection are also examining other implications of dynamical nuclear polarization (Johnson.A. Imamura et al. Equilibrium polarization and the current ﬂow can be potentially realized. Tatara et al. this alone is usually not sufﬁcient for spintronic applications. 2001). 2003). 1999. with the spatial extent given by the spin diffusion length (Aronov. Once the pumping is turned off the spin will return to its equilibrium value. instead of microwave excitation. optical. 21 Supporting the ﬁndings of Clark and Feher (1963). 1984) to the pairing symmetry of high-temperature superconductors (Vas’ko et al.20 and superconductors. In the steady state M is re- 17 Important exceptions are tunneling devices operating at low bias and near equilibrium spin.. Aronov (1976a. Aronov and Pikus. (Color in online edition) Pedagogical illustration of the concept of electrical spin injection from a ferromagnet (F) into a normal metal (N). 76. 1976). 1953b). 1963). Fabian. spin injection of minority carriers was proposed in a ferromagnet/insulator/p-type semiconductor structure.. was ﬁrst measured in metals by Johnson and Silsbee (1985. This behavior. Mod. has already been studied in a large number of materials and geometries (Bruno. 1995). in spin-triplet superconductors and thin-ﬁlm ferromagnets ¨ (Konig et al. 1973) and the principle of optical orientation (Meier and Zakharchenya. (2000c) and Suhl (2002). also known as ballistic magnetoresistance. Unequal ﬁlled levels in the density of states depict spin-resolved electrochemical potentials different from the equilibrium value 0 .. 2000..21 Such a M.328 ˇ ´ Zutic. (b) magnetization M as a function of position—nonequilibrium magnetization M (spin accumulation) is injected into a normal metal. It is also instructive to compare several earlier proposals that use spin-orbit coupling to generate spin currents.
is taken to be ↑ or (numerical value 1) for spin up. 1985).22 Generation of nonequilibrium spin polarization and spin accumulation is also possible by optical methods known as optical orientation or optical pumping. and the sum of the spin-resolved components for a X X X particular quantity X. While electrical spin injection and optical orientation will be discussed in more detail later in this section.. Introduction Spin polarization not only of electrons. II. both the choice of the spin-resolved components and the relevant physical quantity X need to be speciﬁed.. It is important to emphasize that both the magnitude and the sign of the spin polarization in Eq. also referred to as spin-charge coupling. optical orientation in semiconductors has important differences related to the strong coupling between the electron and nuclear spin and the macroscopic number of particles (Paget et al. Even in the same homogeneous material the measured P X can vary for different X. No. Combining these properties of optical orientation with semiconductors tailored to have a negative electron afﬁnity allows photoemission of spinpolarized electrons to be used as a powerful detection technique in high-energy physics and for investigating surface magnetism (Pierce and Celotta. relevant to the detection technique employed. 1966. otherwise the information on the orientation of the magnetization in F1 will not be transferred to the F2 region. Conventionally. 1984). Vol. Rev. Polarized nuclei can exert large magnetic ﬁelds ( 5 T) on electrons. . we also survey here several other techniques for polarizing carriers. and resonance methods (as well as their combination) have all been used to create nonequilibrium spin. (3) depend on the choice of X. such as GaAs. 24 A similar geometry was also proposed independently by de Groot. Silsbee (1980) suggested that the nonequilibrium density polarization in the N region. (3) Transport. Jonker et al. 5) separated by a nonmagnetic region. 3 should not exceed the spin diffusion length.. Janner. In optical orientation. and it is crucial to identify which physical quantity—charge current. The spin polarization of electrical current or carrier density. 1999. To avoid ambiguity as to what precisely is meant by spin polarization. 1972). 1977. but also of holes. Fabian. along the spin angular momentum. This could be called the polarizer-analyzer method. or direction of light propagation. After introducing the concept of spin polarization in solid-state systems we give a pedagogical picture of electrical spin injection and detection of polarized carriers. While there are similarities with previous studies of free atoms (Cohen-Tannoudji and Kostler. generated in a nonmagnetic region. nuclei. Mod. with respect to the chosen axis of quantization. which could be further enhanced in quantum structures of reduced dimensionality or by applying a strain. 1973). April 2004 22 23 For example. In bulk III-V semiconductors. Silsbee (1980) proposed a detection technique consisting of two ferromagnets F1 and F2 (see Fig. carrier density. the optical counterpart of the transmission of light through two optical linear polarizers. 2003). Electron orbital momenta are directly oriented by light and through spin-orbit interaction electron spins become polarized. In a pioneering work Lampel (1968) demonstrated that spins in silicon can be optically oriented (polarized). conductivity. GENERATION OF SPIN POLARIZATION A.B we focus on optical orientation in semiconductors. and Das Sarma: Spintronics: Fundamentals and applications 329 alized as the balance between spins added by the magnetization current and spins removed by spin relaxation. and Mueller (1983). Hermann et al. From Fig. the ratio of the difference X s X X . This technique is derived from the optical pumping proposed by Kastler (1950) in which optical irradiation changes the relative populations within the Zeeman and hyperﬁne levels of the ground states of atoms.ˇ ´ Zutic. II. Happer. In semiconductors the terms majority and minority usually refer to relative populations of the carriers while ↑ or and ↓ or correspond to the quantum numbers m j with respect to the z axis taken along the direction of the light propagation or along the applied magnetic ﬁeld (Meier and Zakharchenya. magnetization. A simple reversal in the polarization of the illuminating light (from positive to negative helicity) also reverses the sign of the electron density polarization. 1984. is typically used to describe the efﬁciency of electrical spin injection. 5 it follows that the reversal of the magnetization direction in one of the ferromagnets The spin diffusion length is an important quantity for CPP giant magnetoresonance. equivalently. 2. Phys. In Sec.23 In ferromagnetic metals it is customary to refer to ↑ (↓) as carriers with magnetic moment parallel (antiparallel) to the magnetization or. The thickness of the N region in Fig. applied magnetic ﬁeld. 1984. as carriers with majority or minority spin (Tedrow and Meservey.24 F1 serves as the spin injector (spin aligner) and F2 as the spin detector. with the goal of implementing spinbased devices to amplify and/or switch current. say optical vs transport and bulk vs surface measurements (Mazin. and excitations can be deﬁned as P X X s /X. and ↓ or (numerical value 1) for spin down.. 1984). the angular momentum of absorbed circularly polarized light is transferred to the medium. 2003). acts as the source of spin electromotive force (emf) and produces a measurable ‘‘spin-coupled’’ voltage V s M. optical. Jonker et al. a well-established technique (Meier and Zakharchenya. where F1 and F2 were two half-metallic ferromagnets. or equivalently the nonequilibrium magnetization. optical orientation can lead to 50% polarization of electron density. 76. Using this concept. Meier and Zakharchenya.. or the density of states—is being measured experimentally.
An alternative scheme for producing pure spin currents was proposed by Bhat and Sipe (2000). 5. 1976) and in materials that lack inversion symmetry (Levitov et al. transport For example.C we focus on related theoretical work motivated by potential applications. Slichter. April 2004 25 methods to generate carrier spin polarization are not limited to electrical spin injection. and Das Sarma: Spintronics: Fundamentals and applications FIG. Using the principles of optical orientation (see Sec. A comprehensive survey of the related effects. in which spin photocurrent is driven. Watson et al. 1965. as discussed in the following sections. 1987. 2003) and nonadiabatic quantum spin pumping (Zheng et al. Since the experiments demonstrating the spin accumulation of conduction electrons in metals (Johnson and Silsbee. 2002a). A related effect. 2000). This could also be called a spin corkscrew effect. (2003). II. (b) density-of-states diagrams for spin injection from F1 into N. 2002. 1984). 1999..25 The generation of nonequilibrium spin polarization has a long tradition in magnetic resonance methods (Abragam.B. (c) spin accumulation in a device in which a superconductor (with the superconducting gap ) is occupying the region between F1 and F2. this would most likely work only as a transient effect (Fabian and Das Sarma. they also include scattering of unpolarized electrons in the presence of spin-orbit coupling (Mott and Massey. A quantum-mechanical interference between one. 1988a).B) in semiconductors of reduced dimensionality or lower symmetry. accompanied by the spin accumulation-generation of nonequilibrium magnetization M. 1998). or to the reversal of charge current j→ j. The current here is caused by the difference in spin-ﬂip scattering rates for electrons with different spin states in some systems with broken inversion symmetry. Mod. At F2 in the limit of low impedance (Z 0) electrical spin is detected by measuring the spin-polarized current across the N/F2 interface. spin injection could be detected through the spin accumulation signal as either a voltage or a resistance change when the magnetizations in F1 and F2 are changed from parallel to antiparallel alignment.2 and further discussion in Sec. (2003) and Hubner et al. I.. and spin detection: (a) two idealized completely polarized ferromagnets F1 and F2 (the spin-down density of states N↓ is zero at the electrochemical potential energy E 0 ) with parallel magnetizations are separated by the nonmagnetic region N. Only a spin current can ﬂow without a charge current. Si (1997. 76. would lead either to V s → V s . Phys. 1989).. 2001). has yet to be demonstrated. both the direction and the magnitude of a generated charge current can be controlled by circular polarization of the light. with only a spin current (j ↑ j ↓ ) and no charge current (j ↑ j ↓ 0)? While it is tempting to recall the SternGerlach experiment and try to set up magnetic drift through inhomogeneous magnets (Kessler. 2002).. ´ 1996. 1997).. No. which can be viewed as a transfer of the angular momentum of photons to directed motion of electrons. 2.. who were able to achieve coherent control of the spin current direction and magnitude by the polarization and relative phase of two exciting laser light ﬁelds. 1979) to recent demonstrations in semiconductor quantum wells .. It was proposed by D’yakonov and Perel’ (1971a. For example. for an instructive description of parametric pumping see Brouwer. in a short circuit (at small Z). 2003. Vol. Zhang. 2003. Charge current also can be driven by circularly polarized light (Ivchenko and Pikus.and two-photon absorptions of orthogonal linear polarizations creates an opposite ballistic ﬂow of spin-up and spin-down electrons in a semiconductor. 2003). Correspondingly. Spin and charge are carried by separate excitations and can lead to spatially separated spin and charge currents (Kivelson and Rokhsar. Zhao and Hershﬁeld (1995). as ¨ demonstrated by Stevens et al. 1998). is called the spin-galvanic effect (Ganichev and Prettl. studies probing the spin-charge separation in the non-Fermi liquids have been proposed by Kivelson and Rokhsar (1990). This interesting effect. 2003). 1980. also called the spin Hall effect (Hirsch. Sharma and Chamon. Fabian.. It would be interesting to know what the limits are on the magnitude of various spin polarizations. experiments on spin injection have also stimulated proposals for examining the fundamental properties of electronic systems. 1999). adiabatic (Mucciolo et al. since a nice mechanical analog is a corkscrew whose rotation generates linear directed motion. 1997) and carrier population (Fraser et al.. spin injection has been realized in a wide range of materials. (3)]. spin accumulation. In the limit of high impedance (Z ) spin is detected by measuring the voltage V s M developed across the N/F2 interface. and proximity effects (Ciuti et al. a consequence of Silsbee-Johnson spin-charge coupling (Silsbee. 1985). While in Sec. Johnson and Silsbee. This is called the circular photovoltaic effect (Ganichev and Prettl. 1961. Balents and Egger (2000. Kessler. Could we have a completely polarized current P j → . Rev.. II. (Color in online edition) Spin injection. 1971c) that a transverse spin current (and transverse spin polarization in a closed sample) would form as a result of spin-orbit coupling-induced skew scattering in the presence of a longitudinal electric ﬁeld. in an open circuit (in the limit of large impedance Z). Hache et al. motivated by the experimental demonstration of phasecoherent control of charge currents (Atanasov et al. from the circular photogalvanic effect (Asnin et al. However. 1990). 1976).330 ˇ ´ Zutic. see Eq.
valence heavy hole. this does not directly imply that it would be dissipationless. we deﬁne the spin polarization as (see Sec. Rev. 2001). Expressing the wave functions with the symmetry of s. respectively. 1/2.. that is. 2003. it follows from Table I that FIG. where Eg is the band gap and so the spin-orbit splitting. they lose spin orientation very fast. April 2004 26 Although holes are initially polarized too. or batteries. 2003. The circular polarization of the luminescence is deﬁned as In Si this condition is not fulﬁlled. X . 2001a. 2003).52 eV at T 0 K. 6(b). Lampel (1968) has used NMR to detect optical spin orientation. and p z orbitals as S . In the context of superconductors.D. 76. Let represent the helicity of the exciting light. Ganichev. Brataas et al.7. 2003. 2002. 2002). more precisely. 2. 2001. Pn 1 3 / 3 1 1/2 (6) is the spin polarization at the moment of photoexcitation. p x . conduction band. consider photons arriving in the z direction. II. Mod.. Y . see also Kittel. Ting and Cartoixa. The spin polarization of the excited electrons28 de. Instead of measuring the luminescence polarization. Long et al. 6(a). The band structure is depicted in Fig. often referred to as spin(-polarized) pumps. or the spin relaxation time of the carriers (Parsons. where s is the spin relaxation time (see Sec. 2003). even when it is feasible to generate pure spin current. 2002) and to extract such useful quantities as the spin orientation. However.1/2 Y1 1 3/2.. The spin is oriented against the direction of light propagation. could be useful for ultrafast spintronic applications (Dargys. SO. Ekimov and Safarov. There is a wide range of recent theoretical proposals for devices that would give rise to a spin electromotive ˇ ´ force (Zutic et al. 6. the luminescence (recombination radiation) will be partially polarized.A) Pn n n / n n . 1970. only the light and heavy hole subbands so contribute. Interband transitions in GaAs: (a) schematic band structure of GaAs near the center of the Brillouin zone ( point). Denoting by n and n the density of electrons polarized parallel (m j 1/2) and antiparallel (m j 1/2) to the direction of light propagation. 1976. et al. Mal’shukov ` et al. Meier and Zakharchenya. Garbuzov et al. 1963).. light hole. which is representative of a large class of III-V and II-VI zinc-blende semiconductors. B. 2001b.34 eV. We illustrate the basic principles of optical orientation by the example of GaAs. it was suggested that manipulating hole spin by short electric ﬁeld pulses. 1/2 Y 1 3/2. Other transitions are analogous.. We denote the Bloch states according to the total angular momentum J and its projection onto the positive z axis m j : J.. LH. and Z . 1970). the band wave functions can be written as listed in Table I (Pierce and Meier. is given by Ganichev and Prettl (2003). cells. CB. Ganichev. No..m j .1). Optical spin orientation In a semiconductor the photoexcited spin-polarized electrons and holes exist for a time before they recombine. If a fraction of the carriers’ initial orientation survives longer than the recombination time. and Das Sarma: Spintronics: Fundamentals and applications 331 (Ganichev et al. By measuring the circular polarization of the luminescence it is possible to study the spin dynamics of the nonequilibrium carriers in semiconductors (Oestreich et al. 1969. Fabian. Ivchenko.. it has been shown that Joule heating can arise from pure spin current ﬂowing through a Josephson junction (Takahashi et al. p y ... 27 For an outgoing light in the z direction the helicities are reversed.. the irreducible representations of the tetrahedron group T d (Ivchenko and Pikus. (5) For our example of the zinc-blende structure. spin-orbit split-off subbands. The circled numbers denote the relative transition intensities that apply for both excitations (depicted by the arrows) and radiative recombinations. Danilov. (b) selection rules for interband transitions between the m j sublevels for circularly polarized light and (positive and negative helicity). 1969. Governale et al. The same selection rules apply to the optical orientation of shallow impurities (Parsons. Ganichev et al. while the spin splitoff band is separated from the light and heavy hole bands by so 0. III). on the time scale of the momentum relaxation time (see Sec. Phys. where Y m is the l spherical harmonic. the recombination time. When we repopresent the dipole operator corresponding to the tical transitions as27 (X iY) Y 1 1 .ˇ ´ Zutic. with minor typos removed. if 26 s . 1971. Vol. 1/2 2 2 3 (4) transition between for the relative intensity of the the heavy ( m j 3/2 ) and the light ( m j 1/2 ) hole subbands and the conduction band. Ekimov and Safarov. However. To obtain the excitation (or recombination) probabilities.8 are the corresponding symmetries at the k 0 point. or. The relative transition rates are indicated in Fig. since there are more transitions from the heavy hole than from the light hole subbands. III. 1997). 28 . 1984).. 2002. For between E g pends on the photon energy and E g . et al.. 3/2 1 1/2. 2002. HH. The band gap is E g 1. 6. between momentum scattering events.
Oskotskij et al. There are plenty of unpolarized electrons in the conduction band available for recombination. The steadywhere state polarization is independent of the illumination intensity. underlining the vital role of spin-orbit coupling for spin orientation. The effective carrier lifetime is J n 0 /G.. has indeed been demonstrated (Vasilev et al. Angular and spin part of the wave function at . (11) gives Pn Pn t 0 / 1 s (12) 1/rp 0 is the electron lifetime. (9) and (11) give (D’yakonov and Perel’. (13) In contrast to the previous case. No. 1971).4) in a homogeneously doped semiconductor. where J represents the intensity of the illuminating light. 1969) and P n 0.3As (Ekimov and Safarov. photoluminescence is then The polarization of the P circ n n 3n 3n 3n 3n n n Pn 2 1 .332 ˇ ´ Zutic. with index zero denoting the equilibrium values.3/2 3/2.06 in Ga0. 1970) showed an effective spin orientation close to the maximum value of P n (t 0) 1/2 for a bulk unstrained zinc-blende structure. (10) 8 P circ I I / I I .m j 1/2. 1969). consider the simple model of carrier recombination and spin relaxation (see Sec. P n depends on the photoexcitation intensity G.42 0. Similarly.A. the electron and hole densities are n and p. 1971. and Eq. the balance between spin relaxation and spin generation is expressed by rsp s/ s P n t 0 G. that is. Phys. Vol.7Al0. being reduced from the initial spin polarization P n (t 0). To derive the steady-state expressions for the spin polarization due to optical orientation. 1984) between the terms optical spin orientation and optical spin pumping. After the illumination is switched off. 4 (8) n is the electron spin density and P n (t where s n 0) is the spin polarization at the moment of photoexcitation. representing spin polarization of the ‘‘ground’’ state. spin pumping is very effective. Unless speciﬁed otherwise.9.1/2 1/2. Zakharchenya et al. and Das Sarma: Spintronics: Fundamentals and applications TABLE I. 1/2 3/2. Sometimes a distinction is made (Pierce and Meier. the carrier (now hole) lifetime 1/rn 0 has no effect on P n . The ﬁrst term in Eq. 76. The steady-state degree of spin polarization depends on the balance between the spin excitation and decay. 1971). 6 suggests that a removal of the heavy/light hole degeneracy can substantially increase P n (D’yakonov and Perel’. 1 1/ s rp / . From Eqs.1/2 1/2. 3/2 S↑ S↓ Wave function 7 (1/3) 1/2 (X iY)↓ Z↑ (1/3) 1/2 (X iY)↑ Z↓ (1/2) (X iY)↑ (1/6) 1/2 (X iY)↓ 2Z↑ (1/6) 1/2 (X iY)↑ 2Z↓ (1/2) 1/2(X iY)↓ 1/2 where r measures the recombination rate. 1969). we shall use the term optical orientation to describe both spin orientation and spin pumping. or equivalently the nonequilibrium magnetization. It varies with frequency roughly as ( ) ( E g ) 1/2 (Pankove. where n n 0 and p p 0 . (5). 1970.. 1976. If it is comparable to or shorter than s . The balance between direct electron-hole recombination and optical pair creation can be written as Rev.46 0. 29 (11) If the excitation involves transitions from the spin Eg split-off band. 1971b) P n P n t 0 / 1 n 0 /G s . n n 0 . (9) J. On the other hand. indicating that / s 1. up to the limit of complete spin polarization.3As (Ekimov and Safarov. Spin pumping works because the photoexcited spin-polarized electrons do not need to recombine with holes. the electron spin density. 30 The polarization of the photoluminescence is P circ P n (t 0)P n (Parsons. 2. (9) and (10) we obtain the steadyˇ ´ state electron polarization as (Zutic et al. (10) describes the disappearance of the spin density due to carrier recombination. Holes are assumed to lose their spin orientation very fast.1/2 3/2. While photoexcitation with circularly polarized light creates spin-polarized electrons. Fig. Symmetry 6 r np n 0 p 0 G. 1969) and p. 1984). Fabian.. However. or due to conﬁnement in quantum well heterostructures. 1/2 3/2. The spin is thus pumped in to the electron system. will decrease exponentially with the inverse time constant 1/T s 1/ 1/ s (Parsons. and G is the electron-hole photoexcitation rate.and n-Ga0.08 in GaSb (Parsons.7Al0. while the second term describes the intrinsic spin relaxation. An increase in P n and P circ in GaAs strained due to a lattice mismatch with a substrate. Both spin orientation and spin pumping were demonstrated in the early investigations on p-GaSb (Parsons. For spin pumping in an n-doped sample. given by Eq. 30 The effect of a ﬁnite length for the light absorption on P n is discussed by Pierce and Celotta (1984). Mod. 1/2 1/2. Eqs. detecting P n greater than 0. 1997). 1993. IV. if so . April 2004 In a p-doped sample p p 0 . Meier and Zakharchenya. Early measurements of P n 0. (7) where I is the radiation intensity for the helicity . the nonequilibrium spin decays due to both carrier recombination and spin relaxation. The latter term is reserved for the majority carriers (electrons in n-doped samples). the electrons will not be spin polarized (P n P circ 0).. 2001b) Pn Pn t 0 1 n 0 p 0 /np . so they are treated as unpolarized.. The former term is used in relation to the minority carriers (such as electrons in p-doped samples) and represents the orientation of the excited carriers. as expected for a pumping process. 29 . The absorption length 1 is typically a micron for GaAs.
Aronov and Pikus. C. See. denoted with solid lines) and the average electrochemical potential ( F . Johnson (2001. and Wei (2002). 2. et al. Eqs. 34 Even some dc spin injection experiments are actually performed at low (audio-frequency) bias. g 0 2. 7. At the interface x 0 both the spinresolved electrochemical potentials ( . Early work also included spin injection into a semiconductor (Sm. form a contact at x 0. The spin diffusion lengths L sF and L sN characterize the decay of s ↑ ↓ (or equivalently the decay of spin accumulation and the nonequilibrium magnetization) away from the interface and into the bulk F and N regions. Mod. consistent with the experiments of Vekua et al. respectively.8Al0. 1969). we also give results for nondegenerate semiconductors in which the violation of local charge neutrality.. Tang et al. Theories of spin injection FIG. (1999. see footnote 29. (2002).0023 (Kittel.2As. April 2004 . the decrease of the steady-state magnetization can have two sources: (a) spins which were excited at random time and (b) random diffusion of spins towards a detection region. (Color in online edition) Spatial variation of the electrochemical potentials near a spin-selective resistive interface at an F/N junction. 1979) and a superconductor (S. showing a Hanle curve different from the usual P n (B) 1/B 2 Lorentzian behavior (D’yakonov and Perel’. Here we follow the approach of Rashba (2000. Aronov 1976a). 33 Johnson and Silsbee base their approach on irreversible thermodynamics and consider also the effects of a temperature gradient on spin-polarized transport. Jedema. for example. it will induce spin precession with the Larmor frequency L B gB/ . the projection of the electron spin along the exciting beam will decrease with the increase of transverse magnetic ﬁeld. (1971) to separately determine both and s at various temperatures. ↑. III. Masterov and Makovskii. The steady-state spin polarization of the precessing electron spin can be calculated by solving the Bloch-Torrey equations (Bloch. Vol. F and N.ˇ ´ Zutic. Nijboer. Spin injection in F/N junctions was subsequently studied in detail by Johnson and Silsbee (1987. omitted in this section.1). with a harmonic time dependence. The Hanle effect was ﬁrst measured in semiconductors by Parsons (1969). resulting in a net magnetization. (1976) in Ga0. spins precess along the applied ﬁeld acquiring random phases relative to those which were excited or have arrived at different times. (1987). At high ﬁelds P n (B) 1/B 1/2. (52)–(54) describing the spin dynamics of diffusing carriers. (2002). 1996). Consequently. and Das Sarma: Spintronics: Fundamentals and applications 333 When magnetic ﬁeld B is applied perpendicular to the axis of spin orientation (transverse magnetic ﬁeld). This is also known as the Hanle effect (Hanle. (2003). In p-doped semiconductors the Hanle curve shows a Lorentzian decrease of the polarization (Parsons. N . Goldman et al.A. 1. The theory of the Hanle effect in n-doped semiconductors was developed by D’yakonov and Perel’ (1976). The relative magnitudes of three char- Reviews on spin injection have covered materials ranging from semiconductors to high-temperature superconductors and have addressed the implications for device operation as well as for fundamental studies in 31 In our convention the g factor of free electrons is positive. 2 P n (B) P n (B 0)/(1 L T s ) . Maekawa et al. In a steady state a balance between nonequilibrium spin generation and spin relaxation is maintained. 7. Schmidt and Molenkamp (2002). 1924). 1984a). leads to the spin dephasing (see Sec. dashed lines) are discontinuous. who showed the nonLorentzian decay of the luminescence for the regimes of both low ( J / s 1) and high ( J / s 1) intensity of the exciting light. 76. 1976. 2002b) and consider a steady-state34 ﬂow of electrons along the x direction in a three-dimensional (3D) geometry consisting of a metallic ferromagnet (region x 0) and a paramagnetic metal or a degenerate semiconductor (region x 0). Hershﬁeld and Zhao (1997). (2001). The notation introduced here emphasizes the importance of different (and inequivalent) spin polarizations arising in spin injection. 32 Rev. electric ﬁelds.32 In addition to degenerate conductors. Fabian. Torrey. leading to depolarization of the luminescence. Measurements of the Hanle curve in GaAlAs were used by Garbuzov et al. 1988a).↓. If a transverse magnetic ﬁeld is applied. 1956). The two regions. No. As a result. and others.33 van Son et al. Valet and Fert (1993). as depicted in Fig. Osofsky (2000). where B is the Bohr magneton and g is the electron g factor. 2002a). 2001). Consider spins excited by circularly polarized light (or by any means of spin injection) at a steady rate.31 The spin precession. where P n (B 0) is the polarization at B 0 from Eq. and carrier band bending require solving the Poisson equation. Recent ﬁndings on the Hanle effect in nonuniformly doped GaAs and reanalysis of some earlier studies are given by Dzhioev et al. Generalization to ac spin injection. was studied by Rashba (2002a). 1946. solid-state systems. in analogy to the depolarization of the resonance ﬂuorescence of gases. (12) and T s 1 is the effective spin lifetime given by 1/T s 1/ 1/ s . together with the random character of carrier generation or diffusion. Phys. examined in these works. F/N junction A theory of spin injection across a ferromagnet/ normal metal (F/N) interface was ﬁrst offered by Aronov (1976b).
1991. each given by the ratio of the spin diffusion length and the effective bulk conductivity in the corresponding region. where r c →0. generally different from the density polarization P n (n ↑ n ↓ )/n. Hartman. 1988a. due to spin-orbit coupling or magnetic impurities). Hershﬁeld and Zhao. 2003). Using a detailed balance N↑ / ↑↓ N↓ / ↓↑ (Hershﬁeld and Zhao. This is no longer possible for a more complicated geometry (Takahashi and Maekawa. Quantities pertaining to a particular region are denoted by the index F or N... Spinresolved charge current (density) in a diffusive regime can be expressed as j where tial is . and D D . 1997. 40 The effects of nonconserving interfacial scattering on spin injection were considered by Valet and Fert (1993). the current polarization38 P j j s /j. Kravchenko. (15) and (18) it follows that s satisﬁes the diffusion equation (van Son et al. (20) From Eqs. (19) where s ↑ ↓ . 36 More generally. 1 or ↓ for spin down along the chosen quantization axis. Valet and Fert. n n n 0 the change of electron density from the equilibrium value for spin . Gijs and Bauer. 39 Spin accumulation is also relevant to a number of physical phenomena outside the scope of this article. Fabian. Margulis and Margulis. s 0. (17) where ↑ ↓ and N N↑ N↓ is the density of states. and Rashba (2002b). 2003). 2000. for a noncollinear magnetization. Vol. 1993.’’ X s . In terms of the average electro( ↑ chemical potential ↓ )/2. (15) j j ↑ j ↓ const.r F . MacDonald. and the electric potential. Fert and Lee (1996). Two limiting cases correspond to the transparent limit.36 In the steady state the continuity equation is j q n n . Mod. to the tunneling rates in the quantum Hall regime (Chan et al. and the low-transmission limit.. while the spin counterpart. Equation (21) has also been used to study the diffusive spin-polarized transport and spin accumulation in ferromagnet/superconductor structures (Jedema et al. Aronov and Pikus..334 ˇ ´ Zutic. Hirota et al. For example. Equation (18) implies the conservation of charge current Correspondingly. the difference of the spin-polarized currents j s j ↑ j ↓ is position dependent. 2. 1999). 2002b). s is often referred to as the (nonequilibrium) spin accumulation39 and is used to explain the GMR effect in CPP structures (Johnson. 2002). Stiles and Zangwill. In the absence of spin-ﬂip scattering40 at the F/N interface (which can arise. which can be factored out. s (18) where s ↑↓ ↓↑ /( ↑↓ ↓↑ ) is the spin relaxation time. the Einstein relation implies that the ratio of the diffusion coefﬁcient and the mobility is k B T/q. 1997). j becomes a second-rank tensor (Johnson and Silsbee. Other ‘‘spin quantities. (15) and ↑ n ↓ 0 shows that the local charge quasineutrality n ↑ is proportional to the nonequilibrium spin density s s n↑ n ↓ (s s 0 s n ↑ n ↓ ). and superscripts in other quantities). 2000. 76. spin angular momentum and magnetic moment are in opposite directions. for example. (16) 1 N↑ N↓ s. 1997. 1987. 38 Rev. 2q N↑ N↓ (22) is the average time for ﬂipping a spin to a and spin. spin up refers to carriers with majority spin. April 2004 . Phys. Gijs and Bauer. Using Eq. 1997. s with q proton charge. Valet and Fert. Spin-resolved quantities are labeled by 1 or ↑ for spin up. 37 In the nondegenerate case of Boltzmann statistics.. in metallic systems (Tedrow and Meservey. 1973. (14) is conductivity and the electrochemical potenqD / n . These boundary conditions were used by Aronov (1976b. 1999. (15) and (17). 35 For this simple geometry various resistances have a common factor of the cross-sectional area. 2002). 1993. For a free electron. Schmidt et al. These are the contact resistance r c and the two characteristic resistances r N and r F . (21) ¯ where the spin diffusion length is L s (D s ) 1/2 with ¯ the spin-averaged diffusion coefﬁcient D ( ↓ D ↑ D ↓ )/ N(N↓ /D ↑ N↑ /D ↓ ) 1 . the spin current is continuous and thus P jF (0 ) P jN (0 ) P j (omitting x 0 for brevity.. This is also referred to as a spin injection coefﬁcient (Rashba. and what precisely is denoted by ‘‘spin up’’ varies in the literature (Jonker et al. unless explicitly deﬁned. Conventionally. The preceding equations are simpliﬁed for the N re¯ gion by noting that /2. 2000) 2 s 2 s /L s . the continuity equation can be expressed as j q2 N↑ N↓ N↑ N↓ . Some care is needed to establish the appropriate boundary conditions at the F/N interface. This means that the spin (angular momentum) of such carriers is antiparallel to the magnetization. is related to the conductivity polarization P as Pj 2 ↑ ↓/ s /j P . No. where r c r N . For a degenerate conductor37 the Einstein relation is q 2N D . P further satisﬁes P s /2 j/ . for example. 2002) together with Eqs.. are analogously expressed with the corresponding (spin) polarization given by P X X s /X. 1999). and Das Sarma: Spintronics: Fundamentals and applications acteristic resistances per unit area35 determine the degree of current polarization injected into a nonmagnetic material. D diffusion coefﬁcient. 1994.
Stiles and Penn. to the F and N regions. (23) (24) is the contact conductivity. where it was ﬁrst derived. Hershﬁeld and Zhao. 1994). sN sN 0 e x/L sN . r c → G( ) 1 yields Eq. n . Yeh et al. Unless the F/N contact is highly transparent. Rashba. where →G. typically measured by optical techniques. sF derivation of P j from Eq.. April 2004 41 Carrier density will also be inﬂuenced by the effect of screening. 1992) that employing resistive (tunneling) contacts could lead to an efﬁcient spin injection. (30) a half-metallic ferromagnet implies a large rF . and the boundary condition is j 0 where ↑ ↓ N 0 F 0 . By examining Eq. (31). Hershﬁeld and Zhao.43 An equivalent form for P j in Eq. (A11) in Johnson and Silsbee (1987) is recovered. 2003. 1997. i n. 1999). and different scattering mechanisms and is usually obtained from model calculations (Schep et al. can be mapped to Eq. (27).. For a free-electron model ↑ ↓ can be simply inferred from the effect of the exchange energy. 1985. and Das Sarma: Spintronics: Fundamentals and applications 335 1976) without relating P j to the effect of the F/N contact or material parameters in the F region. (25) (26) (27) 0 ↓. From Eqs... 2002). and f are of the order of unity. (31) we can both infer some possible limitations and deduce several experimental strategies for effective spin injection into semiconductors. separately. 1987. 1988d) and superconductors. (27)] is complicated by the inﬂuence of disorder. Smith and Silver. 2000).. does not itself imply a large spin accumulation or a large density polarization.45 a systematic understanding was provided by Rashba (2000) and supported by subsequent experimental and theoretical studies (Fert and Jaffres. (31) was obtained by Hershﬁeld and Zhao (1997) and for r c 0 results from van Son et al.F → n. For a perfect Ohmic contact r c 0. By applying Eq.. and high-temperature superconductors (Hass et al. 43 The substitutions are P j → * . (A11) from Johnson and Silsbee (1987).ˇ ´ Zutic. 45 The inﬂuence of the resistive contacts on spin injection can also be inferred by explicitly considering resistive contacts (Johnson and Silsbee. 1993. Rashba. Dong et al.41 It is useful to note42 that Eq. 1997. Takahashi and Maekawa. 1997. L sN. (23) and (24) it follows that sN N 0 0 /4 sF F 0 2r c P j P r c 1 P P j j. A microscopic determination of the corresponding contact resistance [see Eq. creating a substantial current polarization P j by direct electrical spin injection from a metallic ferromagnet into a semiconductor proved to be more difﬁcult (Hammar et al. r F → f / f ( f p f ). 1997. . a detailed correspondence between Eq. (18) in Rashba (2000). 2003. where magnetic susceptibility). written as Eq. was proposed by Aronov and Pikus (1976) and ﬁrst measured in metals by Johnson and Silsbee (1985). for which injection has been reported in both conventional (Johnson. 1997). one can obtain the amplitude of spin accumulation in terms of the current and density-of-states spin polarization and the effective resistances r F and r N . 2000).. (19). Such a M. 2001. 2002b. 1994.44 While there was early experimental evidence (Alvarado and Renaud. is discontinuous across the interface (Johnson and Silsbee. determining P n requires using Poisson’s equation or a condition of the local charge quasineutrality. one must 2 additionally have ( ) 0 and ( i p i2 ) 0. . f label N and F regions. j. the typical resistance mismatch r F r N (where F is a metallic ferromagnet) implies inefﬁcient spin injection with P j r F /r N 1. Phys. 2 For example. In particular. No. (31). Mod. As can 0 2r F P j P N. 2. as a result of electrical spin injection. 76. (28) A nonzero value for sN (0) implies the existence of nonequilibrium magnetization M in the N region (for is the noninteracting electrons q s B M/ . Johnson and Byers. . Rashba (2002d). Fabian. 2001). 42 E. representing an efﬁcient spin injection. P → . To ensure that resistances and the spin diffusion lengths in Johnson and Silsbee (1987) are positive. 1988c. F j. f (for normal and ferromagnetic regions. away from the interface. In contrast to normal metals (Johnson and Silsbee. (2000). 2002) is needed for a more detailed understanding of spin injection. is characterized by the corresponding spin diffusion length sF sF 0 e x/L sF . Even in the absence of the resistive contacts. referred to as the conductivity mismatch problem by Schmidt et al. In contrast to the Rev. Vol. 2000. (1987) are recovered. (29) and (25) the current polarization can be obtained as Pj r cP r FP F /r FN . Erwin et al. 1999. and n. r N → n / n n . respectively. It is important to emphasize that a measured highly polarized current. Johnson. (29) (30) where r N L sN / r F L sF 4 ↑F ↓F From Eqs. P →p. 1996. 1999. 2003). Filip et al. 2001. effective spin injection into a semiconductor can be achieved if the resistance mismatch is reduced by using for spin injectors either a magnetic semiconductor or a highly spin-polarized ferromagnet. Zhu et al. 44 From Eq. (31) where r FN r F r c r N is the effective equilibrium resistance of the F/N junction. Vas’ko et al. assuming n f 1... F/ sN 0 2r N P j j.. Monzon and Roukes. (31) and Eq. r c → G( 2 2 ) 1 . Valet and Fert. surface roughness. respectively). which would yield spin-dependent Fermi wave vectors and transmission coefﬁcients. where the effective contact resistance is rc ↑ The decay of s . which changes with the dimensionality of the spin injection geometry (Korenblum and Rashba. Continued work on the ﬁrst-principles calculation of F/N interfaces (Stiles.F . I.
. For such a resistor scheme.C. Accumulated spin near the F/N interface. (31) and (32) could be obtained by considering an equivalent circuit ˜ ˜ scheme with two resistors R ↑ . 1990. 2. initially spin-unpolarized carriers (say. II. which is carried by quasiparticles (Takahashi and Maekawa. as discussed in Sec. 2002). 2004. the spin diffusion length. the sign of the spin accumulation (nonequilibrium magnetization) is reversed in the F and N regions. IV. 76. Using Eqs. 1991). while their product. 1999). and Das Sarma: Spintronics: Fundamentals and applications be seen from Eq. The increase in the total resistance with spin injection can be most dramatic if the N region is taken to be a superconductor (S).. In a spindiscriminating transport process the resulting degree of spin polarization is changed.. impedes the ﬂow of spins and acts as a ‘‘spin bottleneck’’ (Johnson.. R in Eq. .. 1987) and illustrated by considering the simpliﬁed schemes in Figs. combined with Eqs. Vas’ko et al.N 0. (15) and (17). Johnson and Silsbee. Vol.3).. (31). Yeh et al. (22). see Fig. L sF / F L sN / N . Mod. A rise of sN must be accompanied by the rise of sF [their precise alignment at the interface is given in Eq. similar to spin injection. 1999.. (25)] or there will be a backﬂow of the nonequilibrium spin back into the F region. (32) is then obtained as the ˜ ˜ R↓ ↑ ↓ difference between the total resistance of the nonequilibrium spin-accumulation region of the length L sF ˜ ˜ ˜ L sN [given by the equivalent resistance R ↑ R ↓ /(R ↑ ˜ )] and the equilibrium resistance for the same reR↓ gion. Rev. homogeneous. (32) where R 0 is the nonequilibrium resistance. 1997. at low temperature EuS and EuSe. 1997.336 ˇ ´ Zutic. IV.C. discussed in Sec. Filip et al. appropriate for bulk. in contrast to the behavior sketched in Fig. 2002). Takahashi et al. it follows that Rj N (0) F (0) P F sF (0)/2. The spin injection process alters the potential drop across the F/N interface because differences of spindependent electrochemical potentials on either side of the interface generate an effective resistance R. In the preceding analysis. In the S region the quasiparticle 2 2 1/2 energy is E k ( k ) . Opening of a superconducting gap implies that a superconductor is a low carrier system for spin. Such a dispersion relation results in a smaller diffusion coefﬁcient and a longer spin-ﬂip time than in the N region. (30). Poisson’s equation was not invoked and the local charge neutrality n ↑ n ↓ was used only to derive Eq. and R r N r FP 2 F r cP 2 r Fr c P F P 2 /r FN . Petukhov has shown (Jonker et al.. where R is the junction resistance.48 Focusing on bulk samples in which both the size of the F and N regions and the corresponding spin diffusion lengths are much larger than the Debye screening length. 1980. see Sec... yields A similar result was stated previously by Johnson and Silsbee (1988a).r N (such as a tunnel or Schottky contact) would contribute to effective spin injection with P j P being dominated by the effect r c and not the ratio r F /r N . 7. 1967. Eq. Wei et al. No. 5(c)]. (25)]. ˜ ˜ ˜ 2L sN / N and R ↑ R ↓ where R L sF / F 1/ 4r FN ... there are also proposals for generating highly polarized currents in a CIP-like geometry [Fig. 2003a) that Eqs. Consequently the effect of spin ﬁltering. one can ﬁnd that the quasineutrality condition. where k is the single-particle excitation energy corresponding to the wave vector k and is the superconducting gap [see Fig. Hao et al. Spin injection depletes the superconducting condensate and can result in switching to a normal state of much higher resistance (Dong et al.1). (26).. by noting that ˜ ˜ ˜ j ↑ R ↑ j ↓ R ↓ . By integrating Eq. 47 While most of the schemes resemble a CPP geometry [Fig. 1985. 2003). at far right in the N region. (31) is obtained as P j PR (R ↑ ˜ ˜ )/(R R ). three-dimensional N and F regions and degenerate (semi)conductors. In the extreme case. remains the same (Yamashita et al. the backﬂow of spin driven by diffusion creates an additional resistance for the charge ﬂow across the F/N interface. April 2004 46 48 For spin injection in nondegenerate semiconductors (with the carriers obeying the Boltzmann statistics) there can be large effects due to built-in ﬁelds and deviation from local charge neutrality. Eq.. where sF. far to the left of the F/N interface...3. 1999. in the ab↓ sence of spin injection. 1988. 2002). Consequently. 5(c). where R 0 1/ (R 0 r c if ↑ ) is the equilibrium resistance. 46 This limit is also instructive to illustrate the principle of spin ﬁltering (Esaki et al. (20) for N and F regions separately. A critical review of possible spurious effects in reported experiments (Gim et al. 5 and 7. 3(a)] (Gurzhi et al. near the interface [recall Eq.. R ↓ connected in parallel. 2001. Fabian. Moodera et al. 3(b)]. can act as spin-selective barriers. Because both spin and charge are carried by electrons in spin-charge coupling. injected from a nonmagnetic material) via spin ﬁltering could attain a complete polarization. 2001) has also stimulated the development of a novel detection technique which uses scanning tunneling spectroscopy combined with pulsed quasiparticle spin injection to minimize Joule heating (Ngai et al. Yamashita et al. Based on an analogy with the charge transport across a clean N/superconductor (S) interface (see Sec. Phys. For a strong spinﬁltering contact P P F .A. together with a ﬁnite spin relaxation and a ﬁnite spin diffusion. leads to the generation of (nonequilibrium) spin polarization.A. and (31) allows us to express R R 0 R. (1987) explained R by invoking the consequences of current conversion from spin-polarized. The concept of the excess resistance R can also be explained as a consequence of the Silsbee-Johnson spincharge coupling (Silsbee. to completely unpolarized. van Son et al. (21) also applies to the diffusive spin-polarized transport and spin accumulation in ferromagnet/superconductor structures (Jedema et al. IV. 2003).47 For example. 1999. a spin-selective resistive contact r c r F .
1985) was also found to be proportional to current polarization and the spin accumulation ( s s M. and r N are deﬁned as in the case of an c F/N junction and The above analysis of the F/N bilayer can be readily extended to the geometry in which two inﬁnite F regions D L R rF rL rR rF c c R r F coth d/L sN . which can be evaluated from Eqs. (42) 0 x d. (26) and (33) as N 0 F 0 r c 1 P P j j P NF 0 sF 0 /2. F/N/F junction at the two conwhere the current spin polarization tacts in the F/N/F geometry can be expressed (Rashba. is L sF L 2r F P L P LF je x/L sF . x/L sF ) R R sF R 2r F P R P RF je (d j . that the difference of the spinresolved electrochemical potential. It follows from Eq. (19). (31) for the left and right contact (with the appropriate parameters for the F/N/F junction) as if it were surrounded by the inﬁnite F and N regions. P L cosh d x /L sN j (41) x d. From Eq. (34). 2002b): R R ↑↓ R ↑↑ 4P jL P jR L R r FN r FN r N P L. as depicted in Fig. 0 sN e x/L sN .3). Within a simpliﬁed semi-inﬁnite geometry for the F and N regions. 1988b). Johnson and Silsbee. 5 and high impedance measurements at the N/F2 junction. April 2004 49 Up to a factor of 2. r L. (43) L. and Das Sarma: Spintronics: Fundamentals and applications 337 PN s /2. (39) where P j is given in Eq. No. j L The creation of nonequilibrium spin in the N region results in the spin emf in the F/N structure which can be used to detect electrical spin injection. 2001. II. 1980. the consequence of the spin injection is an increase in the resistance R R 0 R. (36) while the discontinuity of sF in Eq.ˇ ´ Zutic. by assuming the continuity of the spin current at L. (28). (40) sN 2r N P R cosh x/L sN j j/sinh d/L sN . 2002b) in terms of the P j calculated for F/N junction with the inﬁnite F and N regions in Eq. Vol. 5(b)] or by optical pumping (see Sec. d/L sN →0. 2. The resulting potential drop can be calculated by modifying sN in Eq. To calculate the open circuit voltage (j 0) the continuity of spin current at x 0 should be combined with the fact that P j j j s . (25) yields . 76. In particular.D.. see Sec. Eq.. sN /2r FN . At the contact x 0 there is a potential drop. The nonequilibrium resistance R is also always positive for spin-conserving contacts (Rashba.R j D sinh d/L sN . By P jL. (19) and the appropriate effective resistances. (33) where the density-of-states spin polarization of P N vanishes in the N region. 2002) and the allmetallic spin injection of Johnson and Silsbee (1985). Parkin. (34) are separated by an N region of thickness d. The quantities pertaining to the two ferromagnets are deﬁned as in the case of an F/N junction and labeled by the superscripts L and R (left and right regions. (19) it follows that js 0 2 ↑ F ↓ sF 0 L sF 1 2 s sN N 0 L sN sN . (43) has also been obtained by Hershﬁeld and Zhao (1997) using Onsager relations. Many applications based on magnetic multilayers rely on the spin-valve effect. we can evaluate the contact potential drop as N 0 F 0 r F P NF r c P sN /2r FN . Johnson and Silsbee. 2. Rev. Analogously to the F/N junction.. in which the resistance changes due to the relative orientations of the magnetization in the two F regions. R remains ﬁnite.B). sN sN sN x 0. In the context of the spindetection scheme from Fig. Bass and Pratt. 2002b). L r2 rN rF rL rR N c c (44) A missprint in sF (0) from Rashba (2002b) has been corrected. 2000.R we denote the P j calculated in Eq. (31). we consider an effect of spin pumping in the N region. 5. II. the resistance change between antiparallel and parallel magnetization orientations in the two ferromagnets can be expressed using current polarization of an inﬁnite F/N junction P jL. (37) sN r c r F /r FN By substituting this solution into Eq. 2003). The geometry considered here is relevant for CPP giant magnetoresistance (Gijs and Bauer. realized either by electrical spin injection from another F region [as shown in Fig. 2002b) FN N F Pj sN /2. (38) The total potential drop (recall j 0) at the F/N junction50 is (Rashba. respectively).R (Rashba. the spin-coupled voltage V s (Silsbee. over the equilibrium value R 0 ( L ) 1 ( R ) 1 . R. 1999. In the limit of a thin N region.R where r F . even when r c 0. sN 49 0 0 r F /r FN sN js 0 . In the opposite limit. 50 A similar potential drop was also calculated across a ferromagnetic domain wall (Dzero et al. in agreement with experiments on allsemicondcutor trilayer structures (Schmidt et al. Phys. . (35) where sN ( ) represents the effect of homogeneous spin pumping in the N region. for d L sN . responsible for the spin accumulation. Mod. 1997. Fabian.R .
Zwierzycki et al. 2001. While phenomenological models. 53 Equilibrium magnetization can be a consequence of doping with magnetic impurities. For r c which would exceed the resistance of the N and F bulk regions. 2003). The p-n junction has a magnetic n region53 with a net equilibrium electron R spin P n0 .F . 76. Mod. the spin injection efﬁciency can attain P j (r c↑ r c↓ )/r c (Kravchenko and Rashba. Nitta. 2003). We introduce the Sharvin resistance R Sharvin (Sharvin. Following Kravchenko and Rashba (2003) the effective contact resistance r c r c↑ r c↓ (recall that it is deﬁned per unit area) is obtained as rc 4R Sharvin /A 1 t L t R /t L . both affecting transport.. (46) Considering the spin injection from F into a ballistic N region in the presence of diffusive interfacial scattering. 2003) by identifying r c 1/4 . In a 3D geometry the resistance is R Sharvin 4 l 3 a2 e 2 k 2A h 2 1 examined ballistic spin injection in the F/N junction.. Let us now consider spin injection through the depletion layer in magnetic p-n junctions (Fabian et al. The transport equations are solved selfconsistently with Poisson’s equation.. Hu.4. Considering spin-orbit coupling and the potential scattering at the F/N interface modeled by the function. 2001. of diffusive transport through the contact with a l.1. II. First-principles calculations were also used for ballistic spin injection from a ferromagnetic metal into a semiconductor (Mavropoulos et al.D. Typical examples are the Schottky contact and the depletion layer in p-n junctions. 3. (45) L R For a symmetric F/N/F junction. Matsuyama et al.338 ˇ ´ Zutic. 1999).3. 2002. 2001. and Das Sarma: Spintronics: Fundamentals and applications R P jL P jR exp d/L sN ... ´ 1966. Mireles and Kirczenow. arising in ballistic transport between two inﬁnite regions connected by a contact (an oriﬁce or a narrow and short constriction) of radius much smaller than the mean free path. April 2004 Interfaces making up a semiconductor often develop a space-charge region—a region of local macroscopic charges. 2003). Phys. (47) where h/e 2 25. The system is depicted in Fig. 1980. Zutic et al. IV. discussed in Sec. (1995). 2003) and Prins et al. de Jong. In the limit of coherent (specular) scattering51 and high interfacial quality it was shown that different band structure in the F and the N regions would contribute to a signiﬁcant contact resistance and an efﬁcient spin injection (Zwierzycki et al. a l. They show that even a spin-independent barrier can be used to enhance the spin injection and lead to an increase in the conductance polarization. Liu. Johnston-Halperin et al. 2001. Microscopic studies of spin-polarized transport and spin-resolved tunneling through space-charge regions are still limited in scope. The opposite limit. IV.A. Hu and Matsuyama (2001) have Rev.. Van Dorpe. corresponds to the Maxwell or Drude resistance R Maxwell /2a. Jansen et al. there are many theoretical studies treating the ballistic case and phasecoherent transport in both F/N and F/N/F junctions (Hu and Matsuyama. 8. 2. while Osipov and Bratkovsky (2003) proposed an efﬁcient spin injection method using a -doped Schottky contact. Fabian. A is the contact area. 2003). Spin injection through the space-charge region .C. The physics is based on drift and diffusion52 limited by carrier recombination and spin relaxation. it follows that R 4r N r c P r FP D sinh d/L sN F . 2002. et al. 52 Tunneling or ﬁeld emission becomes important. such as the one introduced in Sec. where R stands for the right (here n) region. Simple models in which the N region is a degenerate semiconductor often adopt an approach developed ﬁrst for charge transport in junctions involving superconductors.. 1994. or of using a ferromagnetic semiconductor (Ohno... as described in more detail in Sec. for example. 2002.. it is instructive to reconsider the effect of contact resistance (Kravchenko and Rashba. The studies of intermediate cases provide an interpolation scheme between the R Maxwell and R Sharvin for various ratios of a/l (Wexler.81 k is the quantum of resistance per spin. where r c. in thin Schottky barriers or in p-n junctions and heterostructures at large reverse biases (Kohda et al. 2003). While most of the experimental results on spin injection are feasible in the diffusive regime.R represent the transmission coefﬁcients for electrons reaching the contact from the left and from the right and satisfy t L t R 1. capture a remarkable wealth of spin injection physics. Nikolic and Allen. the importance of the resistive contacts to efﬁcient spin injection. 1998. ˇ ´ 2002a. Connection with the results in the diffusive regime can be obtained (Kravchenko and Rashba. Holes are assumed to be unpolarized. 2003).. IV. yielding large carrier g factors. 2003). Additional examples of magnetic p-n junctions are discussed in Sec. . and applying magnetic ﬁeld. where the phase coherence is lost and the Boltzmann equation can be applied.. An important issue to be resolved is whether there will be spin accumu- 51 The wave-vector component along the interface is conserved during scattering. et al. they carry little information about spin-dependent processes right at the interfaces. Pearton et al. 2002. Wunnicke et al.. Non-self-consistent analyses of a Schottky-barrier spin injection were performed by Albrecht and Smith (2002. 1965).A. (48) where t L. No.F r c.. similarly to the diffusive regime. (24). 2002). taking full account of electric ﬁeld due to accumulated charges. Vol. The difﬁculty lies in the need to consider selfconsistently simultaneous charge accumulation and electric-ﬁeld generation (through Poisson’s equation). where the contact conductivity was introduced in Eq. and k is the Fermi wave vector. showing.
P n polarization is the same across the depletion layer. in which the injected carrier density through the depletion region is still smaller than the equilibrium carrier density. not the spin polarization P n . The geometry is depicted in the inset. On the n side there are more spin-up than spin-down electrons.. Only with bias increasing to the highinjection limit (typically above 1 V) is spin injected. 2002). however. (2001b).54 To be important. 2002a): L Pn L P n0 1 R P n0 R P n0 2 2 R L R P n 1 P n0 P n0 R L R P n P n0 P n0 1 . Under a large bias. Under applied forward bias V the charge current ﬂows to the right.. Spin injection of nonequilibrium spin is also very effective if it proceeds from ˇ ´ the p region (Zutic et al.. Adapted from Z ´ ˇ ´ tions with a magnetic p region (Zutic et al. which shows a junction with a spin-split conduction band in the n region with spin-polarized electrons (solid circles) and unpolarized holes (empty circles). At room temperature the thermal energy is 25 meV. can be transˇ ´ ferred through the depletion layer at small biases (Zutic et al. 8. 2002. and Das Sarma: Spintronics: Fundamentals and applications 339 FIG. Fabian. This is the normal limit of diode operation. (2001). which is.5. 2003a. magnetic drift. can also enhance spin injection (Fabian et al. which is the case for a ˇ ´ spin-polarized solar cell (Zutic et al.. Flensberg ˇ ´ et al.. If 2q is the spin splitting of the conduction 0) exp(q /kBT). 2002a. (49) lation in the p region if a forward bias is applied to the junction. No.ˇ ´ Zutic. Since current is proportional to both the carrier density and the transmission rate. See. but the result shown in Fig. P n represents the nonequilibrium electron polarization. electrons ﬂow to the p region. Note. spin is extracted (depleted) from the nonmagnetic n region. 2002a). 54 . The resulting spin accumulation in the n region extends the spin diffusion range. and Fabian et al. but spin diffusion as well. In addition to spin injection.. April 2004 where L (left) and R (right) label the edges of the space-charge (depletion) region of a p-n junction. The equilibrium density polarization in the n region is about 0. where it is driven by electric drift leading to nonequilibrium spin population already in the n region (Fabian et al. Vol. There is no spin accumulation. 2002a).. greater by q . Martin (2003). The case discussed in Fig. These results were also conﬁrmed in the junctions with two differently doped n regions (Pershin and Privman. While such large ﬁelds are usually present inside the space-charge regions. the nonequilibrium spin tion (P n0 P n0 ). 2001b. At small biases there is no spin injection. Similarly for spin down. For a homogeneous equilibrium magnetizaL R L R P n . while the spin diffusion length can be several microns. (2002a). for example. Spin injection appears only at large biases. ˇ ´ Zutic et al. In other words. n ↑ ( )/n ↑ ( bias. Under a forward band. Spin ˇ ´ polarization of the current is discussed by Zutic et al. that the term ‘‘spin polarization density’’ used in Pershin and Privman (2003a. CorR respondingly. spin extraction has also been predicted in magnetic p-n juncRev. (2001a) ˇ utic et al.. the spin-resolved current is unaffected by 2q and there is no spin current ﬂowing through the depletion layer. 2002). in magnetically inhomogeneous semiconductors. (Color in online edition) Spin injection through the space-charge region of a magnetic p-n junction. in accord with the result of no spin injection. 76. by q and by the spin diffusion length. Fabian et al. manifested by the increase of P n in the p region. For Boltzmann statistics. Phys.55 more than 100 V/cm at room temperature. The explanation for the absence of spin injection at small biases and for nondegenerate doping levels (Boltzmann statistics is applicable) is as follows. Bratkovsky and Osipov (2003). will spin be injected across the depletion layer? Naively the answer is yes. Equation (49) demonstrates that only nonequilibrium spin. n ↑ n ↓ . since spin is carried by electrons.. Yu and ´ Flatte (2002a). such as the thermal or Fermi energy. Electric ﬁeld in the bulk regions next to the space charge is important only at large biases. they exist in the adjacent bulk regions only at the high injection limit and affect transport and spin injection. ˇ ´ where it is driven by electric drift (Zutic et al. leading to spin ampliﬁcation—increase of the spin population away from the spin source. Then P n 0. 8 suggests a more complicated answer. It affects not only spin density. Zutic et al. the electric ﬁeld needs to be very large.. The ﬂow is limited by thermal activation over the barrier (given by the built-in electrostatic potential minus bias). 2. As a result. (2002a).8 V) there is no spin injection. 2001a). In addition to electric drift. 2001b). for the spin-up electrons. arising from a spin L R Pn source. That spin injection efﬁciency can increase in the presence of large electric ﬁelds due to an increase in the spin diffusion length (spin drag) was ﬁrst shown by Aronov and Pikus (1976) and was later revisited by other authors.2 and 1. The curves. labeled by V.. Fabian et al. 55 The critical magnitude is obtained by dividing a typical energy.. 2003b) is actually the spin density s n ↑ n ↓ . 8 is for P n0 L 0. already present in the bulk region. 2003b). and Vignale and D’Amico (2003). 2002). appears only at large biases (1. the rate of transmission of spin-up electrons over the barrier is exp( q /kBT). show the electron density polarization proﬁles P n (x) for the depicted geometry and GaAs materials parameters. evaluated at R.5 V). The following formula was obtained for spin injection at small biases (Fabian et al. Spin injection. the two exponential factors cancel out. At low bias (0. Margulis and Margulis (1994). Mod.
From Johnson and Silsbee. 2002a. thermionic emission) remains to be investigated. and the measurement of voltage will give a spin-coupled signal V s related to the relative orientation of magnetizations in F1 and F2. B 0 100 G. For the Hanle effect B must have a component perpendicular to the orientation axes of the injected spins. there is another reversal of magnetization. After averaging over several cycles. Spin injection data from bulk Al wire sample. Torrey. only M(0)cos . 1956. diffusing population of nonequilibrium spin-polarized electrons injected at x 0. Particularly relevant is the question of whether spin phase is conserved during spin injection. FIG. also discussed in Secs. Experiments on spin injection 1. 1985. Malajovich et al. Inset: amplitude of the observed Hanle signal as a function of orientation angle of magnetic ﬁeld. Negative magnetic ﬁeld is applied parallel to the magnetization ( z axis) in the two ferromagnetic regions. Fabian.57 The results. at approximately B 02 . 2. (39) that the spincoupled signal is proportional to M. center to center (Johnson and Silsbee. equipotentials characterizing electrical current ﬂow. applied at an angle to the z axis in the y-z plane. April 2004 56 57 This method for detecting the effects of spin injection is also referred to as a potentiometric method. 10 summarizes results from a series of Hanle experiments on a single sample. Schematic top view of nonlocal. Johnson and Silsbee. 1988a. 10. 1980. are given in Fig. from negative to positive values. . where the separation between the spin injector and detector L x is variable. Johnson and Silsbee (1985) point out that in the depicted geometry there is no ﬂow of the charge current for x 0 and that in the absence of nonequilibrium spins a voltage measurement between x L x and x b gives zero. In one detection scheme a single ferromagnetic pad was used as a spin injector while the distance to the spin detector was altered by selecting different Py pads to detect V s and through the spatial decay of this spin-coupled voltage infer L sN . 1985.1 magnetization in one of the ferromagnetic regions is reversed. No. quasi-one-dimensional geometry used by Johnson and Silsbee (1985): F1 and F2. dotted lines. 10. ˆ injected nonequilibrium magnetization is M(0)z then at ﬁnite ﬁeld M precesses about B with a cone of angle 2 . of a magnitude several times larger ˆ than a typical ﬁeld for magnetization reversal. the component B. (52)–(54)].2. at B 01 there is a reversal of magnetization in one of the ferromagnetic ﬁlms accompanied by a sign change in the spincoupled signal. 1946. 76. Rev. 1985) was demonstrated in a ‘‘bulk wire’’ of aluminum on which an array of thin ﬁlm permalloy (Py) pads (with 70% nickel and 30% iron) was deposited spaced in multiples of 50 m. at B 0. D. by studying spin evolution in transport through a n-GaAs/n-ZnSe heterostructure. at B 0. An in-plane ﬁeld (B z). in the analysis of the experimental data. 1988d). 58 Recall from the discussion leading to Eq. (2001) showed. The voltage detector58 senses the remaining part of the magnetization projected on the axis of the detector M(0)cos cos (Johnson and Silsbee. 56 This procedure is illustrated in Fig. is applied to deﬁne the direction of magnetization in the injector and detector. From Johnson. and at B 0. resulting in parallel orientation of F1 and F2 and a V s of magnitude similar to that for the previous parallel orientation when B z B 01 . The problem of spin evolution in various transport modes (diffusion. Only projection of B perpendicular to the spin axis applies a torque and dephases spins. As the ﬁeld is increased. that the phase can indeed be preserved. gray shading. and Das Sarma: Spintronics: Fundamentals and applications FIG. As the ﬁeld sweep is performed.B and III. and described by Bloch-Torrey equations [Bloch. is small enough that the magnetizations in ferromagnetic thin ﬁlms remain in the x-z plane (see Fig. The The spin relaxation time in a ferromagnet is often assumed to be very short. II. 9. A more effective detection of the spin injection is realized through measurements of the Hanle effect. tunneling. The inset of Fig. As B z is further increased. 9. both the spin diffusion length and M are taken to vanish in the F region (Silsbee. If. Correspondingly. Theoretical understanding of spin injection has focused largely on spin density while neglecting spin phase. Phys. see Eqs. Injected spin-polarized electrons will diffuse symmetrically (at low current density the effect of electric ﬁelds can be neglected). corresponding to the polarizer-analyzer detection and the geometry of Fig.. 9). 9.2 the magnetization in the other region is also reversed (both are along z axis).340 ˇ ´ Zutic. 1988d) to serve as spin injectors and detectors. Mod. Johnson-Silsbee spin injection The ﬁrst spin polarization of electrons by electrical spin injection (Johnson and Silsbee.A. Vol. the two metallic ferromagnets having magnetizations in the x-z plane. will survive. 1988a). which is important for some proposed spintronic applications. with darker shades corresponding to higher density of polarized electrons. The magnitude of B.
is corrected from the one originally given in Fig. as in the bulk-wire geometry of Johnson and Silsbee (1985).. When polarizer-analyzer detection was used.3) by spin polarization.. 61 d 100 nm was much smaller than the separation between F1 and F2 in bulk Al wires (Johnson and Silsbee. Imamura. 60 The ﬁtting parameters are s . 2000). 2002). where theoretical estimates of V s did not lead to P 1. The saturation of this increase can be inferred from Eqs.6) K. The measured spin-coupled signals61 in Au ﬁlms were larger than the values obtained in bulk Al wires (Johnson and Silsbee. to the spin-coupled voltage. Model calculations (Chui. and the penetration depth of the quasiparticle in the Nb ﬁlms was measured to be 16 nm. 1988d). as compared to 2 nm in Johnson (1994). The studies of spin injection were extended to the thin-ﬁlm geometry. gave the expected P 1 values (Johnson. 76. realizing the theoretical assertion of Aronov (1976a) that nonequilibrium spin could be injected into a superconductor. 2002). 2001. between the magnetization in F1 and F2 (Johnson and Silsbee. 1993b. 2002a). as compared to the usual spinunpolarized quasiparticle injection. The spin injection technique of Johnson and Silsbee was also applied to semiconductors. (2002). Two insulating Al2 O3 ﬁlms were inserted between F1 and F2 (both made of Py) and a Nb ﬁlm [see Fig.1). Mod. Rev. which treat the F/N/F junction as a whole. and R is the Ohmic resistance (Johnson. Vol. (1). V s 1/d.C.. stimulated by the proposal of Rashba (2000) to employ spinselective diffusive contacts (Sec. in the inset.62 A modiﬁcation of the bipolar spin-switch structure was used to demonstrate spin injection into a niobium ﬁlm (Johnson. show that the magnetic susceptibility in N can be much smaller than the freeelectron value and can increase the predicted V s 1/ . Monzon et al. 1995). These corrections to the free-electron picture of an F/N/F junction are smaller for larger d. P . April 2004 59 Theoretical estimates for V s from which P 1 was inferred are modiﬁed when one considers the Coulomb interaction and proximity effects—near the N/F interface the spin splitting of the carrier bands in the N region will be ﬁnite even at equilibrium. The Hanle effect was also studied theoretically by solving the Bloch-Torrey equations for an arbitrary orientation.60 This spin injection technique using the few-pV resolution of a superconducting quantum interference device (SQUID) and with an esti0. characterized by the angle . Analysis from Sec. In interpreting the results..59 Results conﬁrm the ﬁrst application of the Hanle effect to dc spin injection. This topic is reviewed by Tang et al. However. was already anticipated by Silsbee (1980).. the relative change at low temperatures and for L x L s was R/R 5%. 2002a). The results were interpreted as support for enhanced depletion of the superconducting condensate (and correspondingly the reduction of the critical current I c ) by spin-polarized quasiparticles. determined by the relative orientation of the magnetization in F1 and F2. A similar trend.C. 10. 1997. 1993b) similar to the one depicted in Fig. 1985). such as Al. van Wees. 2000. Fabian. 2. While in a good conductor. Chui and Cullen. II. 2000). The origin of this discrepancy remains to be fully resolved (Johnson. 1996. Hershﬁeld and Zhao.3 (36. and since the diffusion coefﬁcient is obtained from Einstein’s relation L s is known. 2. above and below T c .. 2002a). Fert and Lee.07 provided an accuracy able to detect mated P P n 5 10 12. Phys. and the amplitude of the Hanle effect was about 104 larger (Johnson. 2000. 2002a. the parameters L s 450 (180) m and P 0. 2001). Results obtained from the Hanle effect. where R is deﬁned as in Eq. Related measurements were recently performed in a CPP geometry (Gu et al.ˇ ´ Zutic.08) were extracted. The geometry employed is depicted in Fig. 1988d). 1999) for which various explanations were offered (Hammar et al. The measurements were performed near the superconducting transition temperature T c with the data qualitatively similar. Initial experiments on using a metallic ferromagnet to inject spin into a twodimensional electron gas (2DEG) showed only a very low ( 1%) efﬁciency (Hammar et al. et al. 1985. 1993a. and (Johnson and Silsbee. the spin-orbit coupling and the energyindependent density of states at the Fermi level were taken into account (Silsbee. 1988d). 62 . 5(a). causing speculation that a single-spin sensitivity might be possible in smaller samples (Johnson and Silsbee. 9. 5(b)]. From the Hanle curve V s (B ) measured at T 4. 3 of Johnson and Silsbee (1985). 1985. potentially important for applications. II. 1994). and Das Sarma: Spintronics: Fundamentals and applications 341 predicted angular dependence for the amplitude of the Hanle signal (proportional to the depolarization of M M(0)cos2 is plotted in the in a ﬁnite ﬁeld) M(0) inset together with the measured data. 1988a). The corresponding temperature dependence of CPP giant magnetoresistance is well explained by the theory of Yamashita. Geux et al. the subsequent measurements have showed substantially more efﬁcient spin injection into a 2DEG after an insulating layer was inserted (Hammar and Johnson. Spin injection into metals An important part of the operation of CPP giant magnetoresistance structures is the presence of nonequilib- The range of the angle .A. (43) and (44) for d L sN and been discussed by Fert and Lee (1996) and Hershﬁeld and Zhao (1997). one of the ﬁtting parameters from the measured data P sometimes exceeded 1—which corresponds to complete interfacial polarization. IV.06 (0. 1995. No. as obtained in the magnetic-ﬁeld sweep from Fig. also known as the ‘‘bipolar spin switch’’ or ‘‘Johnson spin transistor’’ (Johnson. (2003) and the modiﬁcation of Andreev reﬂection (see Sec. on simi- lar samples. the observed resistance change R was small ( n ).2 shows that measurement of R could be used to determine the product of injected current polarizations in the two F/N junctions.
1998. it would be desirable to be able to inject spin into silicon at room temperature. 2002. (2001) were given by Jedema. No. Nijboer. Takahashi and Maekawa.342 ˇ ´ Zutic. 76. et al. 1988d). or Ni) were chosen to be of different sizes to provide different coercive ﬁelds. This is different from the usual (quasi-)one-dimensional analysis in which spin and charge currents ﬂow along the same paths. 1984.C. the presence of interfacial spin-selective resistance can substantially change the spin injection efﬁciency and inﬂuence the resistance mismatch between the F and N regions [see Eq. 2002). In the second type of experiment. there were few other experimental studies directly concerned with spin injection into metals. 1997). p. et al. anisotropic magnetoresistance and the Hall signal) from effects intrinsic to spin injection. 2003. used a Cr. (19)]. except for the work of Johnson and Silsbee. 1993b.D.. for a wide range of applications and for compatibility with complementary metal-oxide semiconductors (CMOS..5 m was extracted at room temperature. 1998) was used to infer that there is also a large contact resistance in all-metal spin injection experiments (Johnson. Costache.. have the desirable properties of providing injection of spinpolarized electrons (with spin lifetimes typically much longer than for holes) and large spin splitting 0. which have since largely been ignored. deﬁned analogously to Eq.5 eV at 4. (31). Estimates of how these considerations would affect the results of Jedema et al. In the ﬁrst type of experiment the cross-shaped region was deposited directly over the F region (Fig. II.02 at 4. 1993b) were interpreted to be three to four orders of magnitude smaller.63 for a conclusive understanding it will be crucial to have in situ measurements. 1999). 1991.1 and II.. II. through the diffusion of nonequilibrium spin. Heersche. Nonlocal measurements. 1983) as the spin injector.. twodimensional modeling might be necessary (Johnson and Byers. 2001) scaled to the size of the samples used by Johnson (1993a. A theoretical analysis (Jedema et al. Heersche.. 2002b). 11). 2003. By applying a transverse ﬁeld B z (see Fig. Costache. Mod.. 3. while more difﬁcult to fabricate than the subsequent class of III-V ferromagnetic semiconductors. 5 and 9 (discussed by Johnson. The results for R (Jedema et al. was measured as a function of an in-plane magnetic ﬁeld. Wong et al. 1990. Johnson and Silsbee. 2002. rium spin polarization in nonmagnetic metallic regions.. In analyzing data for the van der Pauw cross. From Co/Al2 O3 /Al/Al2 O3 /Co structures L s 0. Viglin et al. 1983)] with nearly complete spin polarization and a Curie temperature T C of up to 130 K (HgCr2 Se4 ) (Osipov et al. Jedema. (Color in online edition) Schematic representation of (a) local and (b) nonlocal geometry used to measure the effects of spin injection and spin accumulation. a twodimensional geometry has an important effect—while the electric current is following the paths depicted in Fig. This proved to be equivalent to Johnson and Silsbee’s (1988d) solution to the Bloch-Torrey equations. with r F r N ] even without using resistive contacts. the maximum current polarization obtained was P j 0.2 K.2. 11. 63 . 2002) at both low (4. 2001. Heersche.64 The experiments were motivated by the theoretical For example. 11. The cross-shaped nonmagnetic region was made of Al or Cu (Jedema. Jedema. et al. Jedema. et al.. For a spin injection from Py into Cu.. the measured resistance of clean F/N contacts in CPP giant magnetoresistance (Bussmann et al.. and the spin-coupled resistance R. Takashi and Maekawa. 2002a).1).C. were performed using the van der Pauw geometry depicted in Fig. All-semiconductor spin injection If a magnetic semiconductor could be used as a robust spin injector (spin aligner) into a nonmagnetic semiconductor it would facilitate the integration of spintronics and semiconductor-based electronics. allowing an independent reversal of magnetization in F1 and F2. 2002a) as outlined in Sec. Early studies (Osipov et al. 2001..2 K) and room temperature. 11. Jedema. 1998). 40). et al. (2002b) as well as by others (Johnson and Byers. similar to the approach shown in Figs. (1). 2002) was performed assuming no interfacial resistance (r c 0) and the continuity of the electrochemical potentials at the F/N interface (see Sec. were shown to simplify the extraction of spurious effects (for example. April 2004 of the contact resistance obtained on different samples. Studies of spin-injection parameters in such systems have been reviewed by Bass and Pratt (1999) and Gijs and Bauer (1997). Comparable resistivities of magnetic and nonmagnetic semiconductors could provide efﬁcient spin injection [see Eq.C. Phys.1. The analysis of the Hanle signal was performed by averaging contributions of different lifetimes (D’yakonov and Perel’. 2002a. until recently. tunneling contacts were fabricated by inserting Al2 O3 as an insulator into the regions where F1 and F2 overlapped with the cross. Co. 11) the precession of the injected nonequilibrium spin was controlled and the amplitude of the Hanle effect was measured (Jedema. In various structures (Jedema. 2003). II. who analyzed the importance of multidimensional geometry. Ultimately. Nijboer. 2002) the two ferromagnetic regions (made of Py. as compared to the local or conventional spin-valve geometry. In addition to comparing characteristic values Rev. 64 These materials. Jedema. and Das Sarma: Spintronics: Fundamentals and applications FIG.. As discussed in Secs. 2003). For a full understanding of the van der Pauw cross geometry. However. et al. 2. et al.. A series of experiments (Jedema et al. the spin current. would have similar ﬂow in all four arms (Johnson. Fabian.and Eu-based chalcogenide ferromagnetic semiconductor (FSm) (Nagaev.2 K for n-doped HgCr2 Se4 (Nagaev. Vol..
respectively. Figure 12 depicts a scheme for realization of allsemiconductor electrical spin injection and optical detection (Fiederling et al. EDSR is the spin-ﬂip resonance absorption for conduction electrons at Zeeman frequency.Mn)VI spin aligner and unpolarized holes injected from the p-doped GaAs. Vol. generated microwave emission. 1988) S J sp d S /V 0 . 1976).05Se has g * 500 (Dietl.. 12. 1994). and V 0 is the volume of an elementary cell... and is the exchange integral for s-like 6 electrons (see Table I in Sec. Displayed is a spin light-emitting diode (LED.Mn)As (Ohno. In (II.Mn)VI materials can be incorporated in highquality heterostructures with different optically active III-V nonmagnetic semiconductors which. generally different from the free-electron value. Jonker et al. In the second approach ferromagnetic semiconductors like (Ga. Rev. April 2004 65 where J sp d is the electron-ion exchange coupling. 76. Similar analysis also applies to g factors of holes. in a spin LED. 2002) quantum dots. (b) conduction and valence bands of a spin aligner in an external magnetic ﬁeld. 2003).. (c) sketch of the corresponding band edges and band offsets in the device geometry. Adapted from Fiederling et al. (50) and (51) it follows that g * g * (H) can even change its sign. The most recent experiments using semiconductor spin injectors can be grouped into two different classes.. due to its large negative ( 50) g factor (McCombe and Wagner. Ferromagnetic semiconductor spin injectors formed p-n and n-n heterostructures with a nonmagnetic semiconductor InSb. 1988) of the conduction band. Such splitting arises due to sp-d exchange between the spins of conduction electrons and the S 5/2 spins of the localized Mn2 ions. in the presence of a highly inhomogeneous magnetic ﬁeld could lead to the charge ´ carrier localization (Berciu and Janko. 1999). 1981). proportional to the steady-state magnetization. 2000). 2.. II. Young. 2000). which is excited by the electric-ﬁeld vector of an incident electromagnetic wave. at low Mn concentration and at low temperatures.. From Eqs. In this case carriers are excited by electrical means and we speak of electroluminescence rather than photoluminescence.. 1988.ˇ ´ Zutic. is extensively reviewed by Rashba and Sheka (1991). electrons and holes recombine in a quantum well or a p-n junction and produce electroluminescence. Mattana et al. spin-down electrons and unpolarized holes are depicted by solid and empty circles. The selection rules for the recombination light are the same as discussed in Sec. Phys.. Chun et al. developed by Rashba and Sheka (1961). Similar to an ordinary LED (Sze.. there is a giant Zeeman splitting E g * B H (Furdyna. In the quantum well.B). Fabian. while injection to the higher Zeeman level. given by (Furdyna. Both approaches were also employed to inject spins into CdSe/ZnSe (Seufert et al. 1998) was tuned by an applied magnetic ﬁeld (from 35 GHz at 400 G up to 1. Jonker et al. can also serve as spin detectors. Aronov and Pikus. in the intrinsic GaAs quantum well. The g * factor for H 0 can exceed65 100 and is given by (Brandt and Moshchalkov. 1984. 2004) and InAs (Chye et al. (1990) and Viglin et al.Mn)As g* 100 at 30 K (Zudov et al. No. (1991) prompted a related prediction (Margulis and Margulis.. Furdyna. The measurements of Osipov et al.4 THz at 20 kG) and only seen when electrons ﬂowed from FSm into an Sm region. 2002). These included CdMnTe (Oestreich et al. 2003) were used. 1971). (50) where g is the H 0 II-VI ‘‘band’’ value g. and ZnMnSe (Jonker et al. II. in which g * is the effective electron g factor. and Das Sarma: Spintronics: Fundamentals and applications 343 work of Aronov (1976a. with the Zeeman splitting of a valence band being typically several times larger than that of a conduction band (Brandt and Moshchalkov.95Mn0.. et al. 1999. function (Ashcroft and Mermin. B s is the Brillouin At low temperatures ( 1 K) Cd0. 1984). The theory of EDSR. by providing circularly polarized luminescence. 1988) g* g M/ g Mn 2 BH FIG. 1999. (II. magnetization M S z B s (g Mn B SH)/(k B T) . (51) .Mn)VI paramagnetic semiconductors were employed as the spin aligners. 1994) that spin injection could be detected through changes in electric dipole spin resonance (EDSR). producing circularly polarized light. Mod. would be changed by spin injection. 1999. leading to population inversion. The choice of InSb is very suitable. The injection to the lower Zeeman level increased the absorption of electron spin resonance. In experiments of . The observed absorption and emission of microwave power (Osipov et al. 1999). Such large g factors.. 2002.. Gaj. as a consequence of radiative recombination of spin-polarized carriers. respectively. 1999) in a Faraday geometry where both the applied B ﬁeld and the direction of propagation of the emitted light lie along the growth direction.Mn)VI materials. (Color in online edition) Schematic device geometry and band diagram of a spin LED: (a) Recombination of spinpolarized electrons injected from the (II. In one approach (II. while in n-doped (In. 1976) predicting that the electron spin resonance signal.. However. BeMnZnSe (Fiederling et al. for detecting the effects of spin injection through electron spin resonance. the emitted light is circularly polarized.B.
2003) using magnetic resonance force microscopy (Sidles et al. which were injected from the p-doped GaAs. Both results clearly demonstrated a robust low-temperature spin injection using spin LED’s.94Mn0. see Fig. an issue not limited to semiconductor heterostructures. At small angles to the axis of growth (see Fig..B) of electroluminescence. lose their polarization very quickly after being injected into a nonmagnetic semiconductor. 1988. is that the most (III. the probability of the spin ﬂip of an injected electron is very high. Similar values of P circ were also measured in a resonant tunneling diode based on ZnMnSe (Gruber et al. T 4 K. Vol. leading to the selection rules for a bulk GaAs (recall P circ P n /2. The results of Jonker et al.. and forward bias. 2000) similar to the actual fabrication conditions employed in conventional electronics. II. 12 the injected spin-down electrons are majority electrons with their magnetic moments parallel to the applied magnetic ﬁeld. and the reports of high T C in some compounds suggest that all-semiconductor spin LED’s could operate at room temperature. 6. as a function of B and the thickenss of the magnetic spin aligner (0 nm.. and Das Sarma: Spintronics: Fundamentals and applications Fiederling et al.. Thurber and Smith. 2002). 2001. from stacking faults at the ZnMnSe/AlGaAs interface (Stroud et al. (2000). (1999) and Jonker et al.. The principles of optical orientation discussed in Sec. The behavior of P circ(B). P circ increased with the thickness of the magnetic layer. Gaj. the magﬁeld. Spin injection can be achieved even with no external ﬁeld. respectively). Stroud et al. Subsequent studies (Park et al. 2001). described above. Waag et al. (2000). (Color in online edition) Electroluminescence in a spin light-emitting diode (LED): (a) Electroluminescence (EL) spectra from a surface-emitting spin LED with a Zn0. Fiederling et al. analyzed for netic ﬁeld is applied along the surface normal (Faraday geometry) and the spectra are dominated by the heavy hole exciton. 6 are used to infer P n in a quantum well. (b) magnetic ﬁeld dependence of the EL circular polarization. 2002) have supported the lifting of degeneracy between the light and heavy hole bands. expected for the (II. due to spin-orbit coupling. 1999) using P circ (deﬁned in Sec. results of the spin injection show . Fabian. from Sec. 1988).B and the selection rules for GaAs sketched in Fig. 1995). For quantum wells of approximately the same width (150 nm) the conversion of P circ to P n used by Fiederling et al. 1991). as a consequence of quantum well conﬁnement and lifting of the degeneracy between light and heavy hole states in the valence band ( 5 – 6 MeV). Phys. 12). (1999) differed by a factor of 2 from that used by Jonker et al. 13....B. suggesting the ﬁnite spin relaxation time needed for initially unpolarized electrons to relax into the lower (spin-down) Zeeman level. Rev.. from P circ P n (Weisbuch and Vinter. These ﬁndings illustrate the importance of interface quality and the effect of defects on the spin injection efﬁciency. and 300 nm. 2000. Adapted from Jonker et al. (2000) were similar to those of Fiederling et al.06Se contact for selected values of applied magnetic (positive and negative helicity). No. 13. (1999) assumed that conﬁnement effects were negligible. (1999) for the thickest magnetic region.1) at the interface. An approximate linear decrease of P circ with the density of stacking faults was shown to be consistent with the inﬂuence of spin-orbit interaction as modeled by Elliot-Yafet scattering (see Sec.Mn)V ferromagnetic semiconductors are also used to inject spin in spin-LED structures as depicted in Fig. II. The corresponding data are shown in Fig. In Fig.Mn)VI/AlGaAs interface was studied (Fiederling et al. It would be valuable to perform realistic calculations of a spinpolarized transport and spin injection which would treat the whole spin LED as a single entity. The electrons further traveled (by drift and diffusion) to an intrinsic GaAs quantum well.Mn)V’s have spin-polarized holes (rather than electrons) as the main carriers which. The robustness of measured P circ was studied by intentionally changing the density of linear defects...Mn)VI semiconductors. The maximum P circ 43% was interpreted as implying nearly 90% polarized injected electrons. Mod. 2. Consequently. Related information is currently being sought by spatial imaging of the spin polarization in spin LED’s (Thurber et al. III. April 2004 66 ing air-exposed interfaces (Park et al.344 ˇ ´ Zutic. could be well explained by the magnetization described with the Brillouin function (Furdyna. electrons entering from the n contact were almost completely polarized in the spin-down state as they left the spin aligner and were injected across the (II. at B 1 T. 2001. however. 2001. Jonker et al. Spin injection using the spin LED’s.66 The efﬁciency of electrical spin injection across the (II. The spatial separation and spin relaxation between the injection and detection points (in a quantum well) make a fully quantitative analysis of the injected polarization more difﬁcult. (III. It is also feasible us- FIG. Jonker et al. up to the saturation value (B 3 T). 76. leading to a small spin polarization. 12.. The nonspherically symmetric defect potential (entering the spin-orbit interaction) causes a highly anisotropic loss of spin polarization. is not limited to structures grown by molecular-beam epitaxy. (2000) inferred P n 50%. 3 nm.B). II. where they recombined with the unpolarized holes. 2002.Mn)VI/AlGaAs interface. The drawback.
. offer the possibility of spin injection at room temperature. some of them highly spin polarized. A small measured signal (P circ 1% at 5 K). Liu. an intrinsic GaAs spacer of thickness d was introduced between the spin aligner (Ga. this result supports the importance of the tunneling contact for efﬁcient spin injection.Mn)VI/II-VI/(II..C. 2000. for a repeated experiment (Young et al. Prins et al. All-electrical spin injection studies of trilayer structures (II. Young. 2002). At 100 K nearly fully spin-polarized injection of electrons was reported. see Sec..D). Isak´ ovic et al. Spin injection via Schottky contacts at room temperature was demonstrated in a Fe/GaAs junction by Zhu et al. Tang et al.3. Spin injection into (110) GaAs at room temperature has already been demonstrated using vacuum tunneling from a polycrystalline Ni STM tip and optical detection via circularly polarized luminescence (Alvarado and Renaud. which selfconsistently incorporates the effects of band bending and deviation from local charge neutrality. In the experiment of Ohno. 2000. 1999. 1997). or effects within the noise (Filip et al.5%. A strong suppression of this signal at applied bias of 10 mV was attributed to the nonlinear regime of spin injection. Metallic ferromagnet/semiconductor junctions A large family of metallic ferromagnets.. 2002). 12). with R/R 1%. 1999. 1995).. Alvarado.. It was shown that the minority spin electrons (spin ↓ in the context of metals. (2001). 2001).. Mod. 4.3 and IV. 2002).A) in Ni produced the dominant contribution to the tunneling current. and Kreuzer et al. II. the analysis was performed only on the electroluminescence [for possible difﬁculties see Fiederling et al. 2002). 2. 1992. 2002) as they are transfered across the nonmagnetic semiconductor. 1999).Mn)VI have displayed up to 25% magnetoresistance at B 5 T and T 4 K (Schmidt et al. 2001). II.. 2002) using a Faraday geometry (as in Fig. A highly efﬁcient spin injection of P n 80% in GaAs has been realized using (Ga.. discussed in Secs. the same change of thickness P circ was reduced from 7% to 0. was also obtained in an additional experiment (Young et al. Similar studies of spatially resolved spin injection. The electroluminescence in a quantum well was measured perpendicular to the growth direction [the easy magnetization axis of (Ga. 1997) as well as after the initial experiments (Monzon et al. (2003)]. Vol.Mn)As as a spin injector in a Zener diode structure (Van Dorpe. 2001) which can be signiﬁcantly affected by the interface (midgap) states at the Schottky barrier (Jonker et al. discussed in the next section. but the experimental focus has since shifted to other approaches.. and the resulting polarization was inferred to be P n ( 31 5. Control measurements have been performed to address these issues (Hammar et al. (1999). Hammar et al. 2002) by using molecular-beam epitaxy to grow MnAs.. in which the effects of quantum conﬁnement and quenching spin-orbit coupling lead to longer s . further analyses of the measurements of P circ have substantially reduced these estimates to 25% (Egelhoff et al.C.67 In contrast. (1982).. in which the effects of band bending and charge accumulation at the (II. initially showed only small effects (Gardelis et al. Direct spin injection from a ferromagnet into a 2D electron gas. Lee et al.Mn)As and the (In.Mn)As and the applied magnetic ﬁeld were both perpendicular to the growth direction].D.. and Das Sarma: Spintronics: Fundamentals and applications 345 only a small degree of hole polarization.1 and II. 2000). The measured I-V curves display a complicated behavior (Hirohata et al. who reported detection of P circ 2% using spin LED structures and optical detection as described in Sec. There was no preferential behavior for spin injection using different azimuthal orientations of the epitaxial MnAs layer. II. II. see Ando et al.. 68 . It would be instructive to analyze these measurements by adopting the approach discussed in the context of magnetic p-n junctions (Secs. (2002). as discussed in Sec. LaBella et al.6)% (Alvarado and Renaud. 2001. The tunneling properties of a Schottky barrier were discussed by Meservey et al. (1995). However. The detection employed the technique of an oblique Hanle effect.. a behavior that remains to be understood considering that the hole spins should relax fast (Hilton and Tang.. A theoretical explanation for this behavior is still lacking. van Wees. Phys. These studies were extended (Ramsteiner et al. et al. (1982).C. II. have employed a single-crystal Ni (100) tip (LaBella et al. April 2004 67 For a comprehensive review of the 2D electron gas. The possibility of spurious effects arising from the Hall and anisotropic magnetoresistance signals in similar structures was suggested earlier (Monzon et al. 2003). This debate about the presence/absence of spin injection effects via Ohmic contacts stimulated further studies.. with both measured electroluminescence and B along the growth direction. A possible exception is the quantum well.Mn)VI/II-VI interface were important (Schmidt et al.1. even in the absence of applied magnetic ﬁeld. Hammar and Johnson.. Gibson and Meservey (1985).. 2002). 2000. 1992). 2000).68 motivated by the proposal of Datta and Das (1990). sensitive to the topography of the GaAs surface. consistent with the expectation for holes as the injected spin-polarized carriers.2. Rev. Such inefﬁciency could be attributed to the resistance mismatch in the F and N regions. a ferromagnetic metal. Fabian. on top of the GaAs to provide high-quality interfaces (Tanaka.ˇ ´ Zutic. No. 2002. which could have been expected from the symmetry between the conduction-band wave functions in MnAs and GaAs. The corresponding relation between the P circ and hole density polarization P p is not straightforward.C.. P circ was approximately independent of the GaAs thickness (d 20– 420 nm).Ga)As quantum well. 76. Even though the spin injection in future spintronic devices will likely be implemented by some means other than vacuum tunneling. et al..
An oblique magnetic ﬁeld is applied to give a net out-of-plane component of injected spin which could contribute to the emission of circularly polarized light. Fabian. Furthermore. 2002. transistorlike devices. Typical operating parameters are 1 mA and 2 V.. This approach allows one to apply a magnetic ﬁeld several times smaller than would be needed to pull the magnetization out of plane [for Fe it is 2 T (Hanbicki et al.Ga)As/GaAs spin LED. From Hanbicki et al. Different coercive ﬁelds in the two ferromagnetic electrodes allowed them to measure a spin-valve effect with R/R 40% at 11 K.. at which the Fe magnetization is entirely along the surface normal. An oblique Hanle effect (D’yakonov et al. as used in related previous experiments (Hanbicki and Jonker. as discussed in Sec. used in the context of semiconductors and ferromagnetic metals. 2002. allowing for spin-encoded information to travel macroscopic distances. Optical detection in spin LED structures. 2003).D. II.. (Color in online edition) Electroluminescence (EL) and polarization due to spin injection from Fe Schottky contact: (a) EL spectra from a surface-emitting spin LED with an Fe/AlGaAs Schottky tunnel contact for selected values of applied magnetic ﬁeld. et al.1) was used (Motsnyi et al. 2000). As discussed in Sec. (2004) have demonstrated that MgO can be a suitable choice of an insulator for highly efﬁcient spin injection into GaAs. 2002). see also Secs. The spectra are dominated by the heavy hole exciton.D. Jiang et al. The geometry used is similar to that sketched in Fig. Hanbicki et al. a consequence of weak spin-orbit coupling (Davis and Bussmann. Experimental results are given in Fig. Motivated by these ﬁndings Xiong et al. 12(a).. (2004) have replaced one of the LSMO electrodes by Co.2 T.. 69 . tunnel contacts formed between a metallic ferromagnet and a semiconductor can provide effective spin injections. Mod.B and II.. Jansen. with an insulating layer (AlOx ) separating the ferromagnetic spin injector and the (Al.E. 2002). Direct electrical spin injection has also been demonstrated in organic semiconductors (Dediu et al. analyzed for circular polarization.C. 2. Related theoretical studies of the ferromagnetic metal/conjugated polymer interfaces were reported by Xie et al. No. Phys. IV. 1974.. using standard measurements of the Hanle curve. II.346 ˇ ´ Zutic.1. (b) Magnetic-ﬁeld dependence of the EL circular polarization of the heavy hole exciton. 2002. The decrease of magnetoresistance with increasing thickness of the N region was used to infer L sN 100 nm at room temperature. as discussed in Sec. as discussed in Sec. Motsnyi. 69 as compared to the usual III-V inorganic semiconductors. normalized to the polarization of a ferromagnet. Hot-electron spin injection above the Schottky barrier is another method for using a high polarization of metallic ferromagnets to create a nonequilibrium spin in a semiconductor even at room temperatures. III.3MnO3 (LSMO). Vol. While quantum well emission efﬁciency limits detection at higher temperatures ( 100 K). Furthermore. 2003. and Das Sarma: Spintronics: Fundamentals and applications FIG. April 2004 injection. giving up to P n 16%. Typically such injection is performed in three-terminal.A. The magnetization easy axis lies in the plane of the ferromagnet. 2003). s.3. the same tunnel injector should also be suitable for efﬁcient spin injection even at room temperature. II. at room temperature. there are often different conventions for deﬁning the sign of P n (Hanbicki et al. is what makes spintronics a viable option for technology. Van Dorpe. 2003). (2003). Spin LED with GaAs/ AlGaAs quantum well was used to detect P n 50% at 100 K injected from CoFe/MgO (100) tunnel injector. The fact that nonequilibrium electronic spin in metals and semiconductors lives relatively long (typically a nanosecond). The resulting spin diffusion length is a combination of low mobility. and long spin relaxation times. II. The large difference in intensity between these components indicates successful spin injection from the Fe into the GaAs quantum well and reveals an electron spin polarization in the quantum well of 32%. After introducing the concepts of spin relaxation and spin This is also a typical value for other organic semiconductors (Krinichnyi. 2003. and N is sexithienyl (T6 ). 2003) to detect spin Rev. was used to show carrier polarization of P n 30% using Fe as a spin injector (Hanbicki et al. 76. 10 4 cm2 V 1 s 1 (about 107 times smaller than for the bulk GaAs). Introduction Spin relaxation and spin dephasing are processes that lead to spin equilibration and are thus of great importance for spintronics.7Sr0. where F is La0.. 2002)] and to measure polarized luminescence in a Faraday geometry. The polarization tracks the hard axis magnetization of the Fe contact and saturates at an applied magnetic-ﬁeld value 4 M 2. a -conjugated rigid-rod oligomer organic semiconductor. Some care has to be taken in deﬁning the efﬁciency of the spin injection. SPIN RELAXATION AND SPIN DEPHASING A.3. The magnetic ﬁeld is applied along the surface normal (Faraday geometry).. 14. Magnetoresistance measurements were performed in an F/N/F junction.. one can extract separately the spin lifetime and carrier recombination time. a colossal magnetoresistive manganite. 14.
70 . Spatial inhomogeneities of B 0 . decay. while the symbol T 2 is reserved for irreversible loss of the ensemble spin phase. and diffusion of electronic magnetization M in an applied magnetic ﬁeld B(t) B 0 z B1 (t). gives a contribution to 1/T 2 of the order of 2 p .. Torrey. Dyakonov-Perel’. spin relaxation time T 1 (often called longitudinal or spin-lattice time) and spin dephasing time T 2 (also called transverse or decoherence time) are deﬁned via the equations for the spin precession. 1956) for magnetization dynamics. but the rootmean-square phase change increases with time as ( 2 ) 1/2 ( 0 c )(t/ c ) 1/2. April 2004 Time T 2 is classically the time it takes for an ensemble of transverse electron spins. The phase relaxation time t is deﬁned as the time over which the phase 2 ﬂuctuations reach unity: 1/t 0 c. T 1 and T 2 Spin relaxation and spin dephasing of a spin ensemble are traditionally deﬁned within the framework of the Bloch-Torrey equations (Bloch. Rev. Time T 1 is the time it takes for the longitudinal magnetization to reach equilibrium. like those coming from hyperﬁne ﬁelds. the g-factor-induced broadening is inhibited by motional narrowing70 and need not be generally considered as contributing to T 2 [see. Vol. Kaplan. As shown below. and Das Sarma: Spintronics: Fundamentals and applications 347 dephasing times T 1 and T 2 . or the average time of spin precession in one direction. 1956. (54) y 2 Mz M0 z z is the electron gyromagnetic ratio ( B Here B g/ is the Bohr magneton and g is the electronic g factor). This is analogous to precession frequency ﬂuctuations of localized spins due to inhomogeneities in the static ﬁeld B 0 . too. g. but also slightly different g factors and thus different precession frequencies. 76. 1959): Mx t My t Mz t M B M B M B x Mx T2 My T2 T1 D D 2 Mx . 2 In isotropic and cubic solids T 1 T 2 if B 0 1/ c . For mobile electrons. In T 1 processes an energy has to be taken from the spin system.. spatial inhomogeneities play an important role and are often observed to affect T 2 . Mod. although for conduction electrons to a very good approximation T * T 2 . For an ensemble of mobile electrons the measured T 1 and T 2 come about by averaging spin over the thermal distribution of electron momenta. the ‘‘inhomogeneous broadening’’ is a microsecond. which can potentially be eliminated by spin-echo experi* ments. However. 2. de Sousa. Although relaxation and decoherence processes in a manyspin system are generally too complex to be fully described by only two parameters. the average spin phase does not change. and Das Sarma. T 1 and T 2 are nevertheless an extremely robust and convenient measure for quantifying such processes in many cases of interest. derives the magnetization dynamics. it is the time of thermal equilibration of the spin population with the lattice. respectively. which is much more than the observed values for T 2 . 0 c 1. where c is the correlation time of the random force. one starts with a microscopic description of the spin system (typically using the densitymatrix approach). and g as large as 0. My . For localized electrons (e. No. sometimes the symbol T 2 (Hu. describing quantitatively very well the dynamics of mobile electron spins (more properly. usually by phonons. To obtain microscopic expressions for spin relaxation and dephasing times. for donor states in semiconductors). Dupree and Holland (1967)]. a ˆ ˆ transverse oscillating part B1 perpendicular to z (Torrey. If the spin is subject to a random force that makes spin precession equally likely clockwise and anticlockwise. which are commonly called s throughout this paper. 2001) is used to describe spin dephasing of ensemble spins. In general. Consider a spin rotating with frequency 0 . D 2 (52) (53) Mz . Those times Motional (dynamical) narrowing is an inhibition of phase change by random ﬂuctuations (Slichter. The Bloch equations are phenomenological. D is the diffusion coefﬁcient (for simplicity we assume an isotropic or a cubic solid with scalar D). Electrons in different momentum states have not only different spinﬂip characteristics. scattering times of 1 ps. We then survey recent works on electronic spin relaxation in nonmagnetic metals and semiconductors. 1. since momentum scattering (analogous to intersite hopping or exchange interaction of localized spins) typically proceeds much faster than spin-ﬂip scattering. with a static longitudinal component B 0 ˆ (conventionally in the z direction) and. however. and hyperﬁneinteraction processes. where c is the so-called correlation or interaction time: 1/ c is the rate of change of the effective dephasing magnetic ﬁeld (see footnote 70). Equivalently. For B 0 ﬁelds of the order of 1 T. Phys. To describe such reversible phase losses. using the important examples of Al and GaAs for illustration. frequently. Phase losses occur during time intervals of c . where the B 0 -dependent precession frequency ( g/g) B 0 and p is the momentum scatspread is tering time. due to the itinerant nature of electrons. * T 2 T 2 . and M 0 B 0 is z the thermal equilibrium magnetization with denoting the system’s static susceptibility. This is valid for rapid ﬂuctuations. Bir-Aronov-Pikus. magnetization) in experiments such as conductionelectron spin resonance and optical orientation. and compares it with the Bloch equations to extract T 1 and T 2 . Indeed. to the lattice. are inhibited by motional narrowing. 1989).g. motional narrowing of the g-factor ﬂuctuations. to lose their phase due to spatial and temporal ﬂuctuations of the precessing frequencies. in electronic systems c is given either by p or by the time of the interaction of electrons with phonons and holes. The spin phase changes by 0 t over time t. Fabian. 1946. we discuss four major physical mechanisms responsible for spin equilibration in nonmagnetic electronic systems: Elliott-Yafet.01. initially precessing in phase about the longitudinal ﬁeld.ˇ ´ Zutic.
We note that it is s that is relevant for spintronics (spin transport) applications.B). Using simple qualitative analysis Yafet (1963) showed that. As Dzhioev. is employed to deduce s unambiguously. Speciﬁc examples of this will be discussed in studying spin relaxation in twodimensional semiconductor heterostructures (Sec. while theoretically it is often more convenient to calculate T 1 . or magnetoresistance (Dobers et al. see also Sec. In the motionally narrowed case this difference between T 1 and T 2 can be considered as arising from the tensorial nature of the spin relaxation rate. Zakharchenya. In many cases a single symbol s is used for spin relaxation and dephasing (and called indiscriminately either of these terms). A momentum scattering time obtained from conductivity (analogous to s ) would thus be very different from a single-state momentum time (analogous to sc ). Wangsness and Bloch. and that T 2 changes with the direction by at most a factor of 2. spin that is oriented along the direction of the magnetic ﬁeld can precess a full period about the perpendicular ﬂuctuating ﬁeld. we discuss the connection between s and the single-spin decoherence time72 sc . Optical spin orientation (Meier and Zakharchenya. Since in a steady state the excited spin polarization depends not only on s but also on the carrier recombination time. with little justiﬁcation. As the external ﬁeld increases. polarization of luminescence by a transverse magnetic ﬁeld (see Sec. Feher and Kip. or if one is working at small magnetic ﬁelds. spin resonance techniques are aided by other measurements. although it is often. feeling the same dephasing ﬁelds as transverse components. The equality between the relaxation and dephasing times was noticed ﬁrst in the context of NMR (Bloch. et al. An analogy with momentum scattering may be helpful. so T 1 T 2 is fulﬁlled for magnetic ﬁelds up to several tesla. Finally. Vol. In GaAs. or a few. 76. the precession angle of the longitudinal component is reduced. In classical terms. where spin coherence needs to last for at least 104 106 gate operations for the computation to be fault tolerant (Preskill. assumed that they are similar. 1952. April 2004 . 2000). while sc is relevant for solid-state quantum computing. the inequality T 2 2T 1 holds. optical detection (Weisbuch and Hermann. the equality T 1 T 2 no longer holds. II. in which each spin ﬂip adds to the equilibration of both spin-up and spin-down states. Fabian. 1946. while there is no general relation between the two times.. inhibiting dephasing.71 Throughout this paper we adopt this notation to avoid unnecessary confusion. Typically changes in surface impedance and in the transmission coefﬁcient of the sample are observed. Kaplan. 1977). Phys. the energy uncertainty of the spin levels determining the longitudinal spin is greater than the Zeeman splitting B 0 of the levels. 2.. if it does not matter what experimental situation is involved. The splitting then does not play a role in dephasing. 1998. 1959).D. spins. sc can be smaller than s by several orders of magnitude in n-GaAs at low doping densities where electrons are localized in donor states. The equality of the two times is very convenient for comparing experiment and theory. which is the singlespin correlation time. 1955. since measurements usually yield T 2 . 1953) and later extended to electronic spin systems (Pines and Slichter. Electron-electron collisions lead to individual momentum equilibration while conserving the total momentum and hence do not contribute to charge current (unless umklapp processes are taken into account). 72 To distinguish ensemble and individual spin dephasing. Experiments of type (a) are exempliﬁed by conduction-electron spin resonance (CESR) and optical orientation in combination with the Hanle effect. the Hanle effect. e. Often sc s . 1955. Rev. Time sc becomes important for applications in spin-based quantum computing (Loss and DiVincenzo. we use the term decoherence in relation to single. Much remains to be learned about sc . 1958). It is not clear at present how much s and sc differ for different materials under different conditions. 1983). Andreev and Gerasimenko. No.g. 2. 1984) is a method of exciting spin-polarized carriers (electrons and holes) in direct-gap semiconductors like GaAs by absorption of circularly polarized light (see Sec.a.D. 1955) and donor states in semiconductors like Si (Feher and Gere. even in the case of full motional narrowing of the spin-spin interactions and g-factor broadening. Hu and Das Sarma. CESR was the ﬁrst technique used to detect the spin of conduction electrons in metals (Griswold et al. II. as is the case for a direct exchange interaction causing single-spin decoherence.2).. 1959) one can obtain both the carrier g factor and T 2 . Experiments detecting spin relaxation and decoherence of conduction electrons can be grouped into two broad categories: (a) those measuring spectral characteristics of magnetization depolarization and (b) those measuring time or space correlations of magnetization. The technique measures signatures of resonant absorption of microwaves by a Zeeman-split spin system of conduction electrons. Korenev. The relative magnitudes of s and sc depend on many factors. 1988). By comparing the absorption resonance curve with theory (Dyson. while contributing to ensemble dephasing only as a dynamical averaging factor (the exchange interaction preserves total spin). While this is correct for just one spin state. Mod. III.1). doubling the rate for magnetization relaxation. (2002) recently suggested.B. Experimental probes 71 Sometimes one ﬁnds a spin relaxation time 2 s . and Das Sarma: Spintronics: Fundamentals and applications are typically smaller than a picosecond.. A qualitative reason for T 1 T 2 is that if the phase acquires a random contribution during a short time interval c . and the dephasing ﬁeld will act equally on the longitudinal and transverse spins. photoconductivity (Stein et al. 1998)..348 ˇ ´ Zutic. If the system is anisotropic. III. conventionally by spin relaxation we mean magnetization relaxation. The injected spin polarization can be detected by observing circularly polarized luminescence resulting from recombination of the spin-polarized electrons and holes.3.
D’yakonov-Perel’. 1980). BirAronov-Pikus. timeresolved (pump-probe) photoluminescence (in which the probe is used on the same principle as optical orientation). Time-resolved photoluminescence detects. The most important examples are the Johnson-Silsbee spin injection experiment. 73 .. circular polarization of the recombination light. but rather a We do not consider magnetic scattering. In addition.ˇ ´ Zutic. seen as an oscillation of the transmitted radiation amplitude with changing static magnetic ﬁeld (Janossy. No. which changes in random directions every time the electron scatters to a different momentum state. The Hanle effect. April 2004 lasers for spin excitation has the great advantage of not only detecting. resulting from the lack of inversion symmetry and from the spin-orbit interaction. as shown using Faraday rotation on bulk GaAs and GaAs/ ZnSe heterostructures (Awschalom and Kikkawa. In the presence of the spin-orbit interaction V so 4m 2 c 2 V sc p• ˆ . Vol. II. Fabian. to investigate the spin dynamics of bulk CdTe (Kimel et al. electrons ballistically propagating through the sample experience Larmor precession. 2000). 1980). scattering due to an exchange interaction between conduction electrons and magnetic impurities. but also manipulating spin dephasing. electrons are ﬁrst injected by electrical spin injection from a ferromagnetic electrode into a metal or semiconductor. and ˆ are the Pauli matrices.D. Pump experiments. zero-ﬁeld (or zero-g-factor material) data can be measured. for example. and time-resolved Faraday and (magneto-optic) Kerr rotation. In the Johnson-Silsbee scheme.. The detection is by means of spin-charge coupling. 1996). ˆ (55) where m is the free-electron mass. and Faraday rotation was used to study spin coherence in nanocrystals of CdSe (Gupta et al. The advantage of optical orientation over CESR is that. it is the hyperﬁne interaction of the electron spins and nuclear moments which dominates spin dephasing of localized or conﬁned electron spins. 2. The Bir-Aronov-Pikus mechanism is important for p-doped semiconductors. p ˆ i is the linear momentum operator. in semiconductor heterostructures (quantum wells and quantum dots) based on semiconductors with a nuclear magnetic moment. 1994). Kikkawa et al. Type (b) experiments measure magnetization in a time or space domain. emitting microwave radiation. Using Rev. Mechanisms of spin relaxation Four mechanisms for spin relaxation of conduction electrons have been found relevant for metals and semiconductors: the Elliott-Yafet.. that is.. resulting in Larmor waves. The D’yakonov-Perel’ mechanism explains spin dephasing in solids without a center of symmetry. too. 76. can be used in combination with JohnsonSilsbee spin injection. This technique was used. 1. one can extract the spin diffusion length L s and 2 thus the spin relaxation time T 1 L s /D. another ferromagnetic electrode detects the amount of spin as a position-dependent quantity. As the spin diffuses throughout the sample. which agrees with T 1 determined from the measurement of L s . B. An informal review of spin relaxation of conduction electrons can be found in Fabian and Das Sarma (1999c). The Kerr effect was used. after creation of spin-polarized carriers. 2001). if carrier lifetime is known. to detect a 500-ps spin coherence time T 2 of free excitons in a GaAs quantum well (Heberle et al. single-electron (Bloch) wave functions in a solid are no longer the eigenstates of ˆ z . 1999. Finally. The last two methods can follow coherent (in the ensemble sense) dynamics of electron spin precession. using a circularly polarized laser pulse. in connection with electrical spin injection. smaller s values can be detected. due to spin-orbit coupling induced by ions.1). and Das Sarma: Spintronics: Fundamentals and applications 349 The Hanle effect has also been a great tool for investigating T 2 in metals. Spin injection experiments (Johnson and Silsbee. whereby an emf appears across the paramagnet/ ferromagnet interface in proportion to the nonequilibrium magnetization (Silsbee. In very clean samples and at low temperatures. Spin dephasing occurs because electrons feel an effective magnetic ﬁeld. Mod. yielding directly T 1 T 2 . A precursor to the Hanle effect in spin injection was transmission-electron spin resonance (TESR). Elliott-Yafet mechanism Elliott (1954) was the ﬁrst to realize that conductionelectron spins can relax via ordinary momentum scattering (such as by phonons or impurities) if the lattice ions induce spin-orbit coupling in the electron wave function. 2002) and coherent control of spin dynamics of excitons in GaAs quantum wells (Heberle et al. transmitted through (Faraday) or reﬂected by (Kerr) a magnetized sample. The Faraday and (magneto-optic) Kerr effects are the rotation of the polarization plane of a linearly polarized light. for example. in which nonequilibrium electron spin excited in the skin layer of a metallic sample is transported to the other side. V sc is the scalar (spin-independent) periodic lattice potential. the evolution of magnetization as it dephases in a transverse magnetic ﬁeld.73 In the Elliott-Yafet mechanism electron spins relax because the electron wave functions normally associated with a given spin have an admixture of the opposite-spin states. 1985) detect the amount of nonequilibrium magnetization by observing charge response (see Sec. The angle of rotation is proportional to the amount of magnetization in the direction of the incident light. and hyperﬁne-interaction mechanisms. and probe experiments employing magneto-optic effects can now follow.. in which the electron-hole exchange interaction gives rise to ﬂuctuating local magnetic ﬁelds ﬂipping electron spins.. with 100-fs resolution. Phys. The Kerr effect is more useful for thicker and nontransparent samples or for thin ﬁlms fabricated on thick substrates. By ﬁtting the spatial dependence of magnetization to the exponential decay formula.
consider a band structure in the absence of spin-orbit coupling. 1963). III. (61) E 1. s 10 g 2/ p. we ﬁnd by perturbation theory g b . phonon-modiﬁed V so is not important for polyvalent metals. Spin hot spots are discussed in more detail in Sec. whose spin relaxation is dominated by spin hot spots—states with anomalously large b —as shown by the explicit calculation of Fabian and Das Sarma (1999b). 1954). in combination with momentum scattering. Indeed. q is the frequency of phonons with momentum q k k and branch index . Without the spin-orbit interaction this expectation value is zero. They must be combined with the Elliott processes discussed above to form a consistent picture of phonon-induced spin relaxation. and the spin-ﬂip matrix element is g kn↑. The ratio is different for scat- . ˆ z kn↓ ). No. The brackets ¯ kn in Eq. Considering the spin-orbit interaction to be a small parameter. Turning V so on couples electron states of opposite Pauli spins which are of the same k (because V so has the period of the lattice). There is another important spin-ﬂip scattering mechanism that involves phonons. g is determined by the expectation value of ˆ z . and so is the amplitude of the matrix element of V so between the two states. 1954) kn↑ kn↓ pressed through the spin-ﬂip Eliashberg function 2 S F( ) as (Fabian and Das Sarma. We now give a recipe.C. The Bloch wave functions for this calculation are chosen to satisfy ( kn↓ . The spin-orbit coupling alone does not lead to spin relaxation. (60) denote Fermi-surface averaging. Vol. (63) to give the Elliott relation (62). 1954): 1/ s g 2/ p. The two Bloch states are degenerate: they are connected by spatial inversion and time reversal (Elliott. Because the spin-orbit interaction is normally much smaller than the distance between the bands. This relation follows from the fact that for a momentum scattering interaction V i the spin-ﬂip scattering probability in the Born approximation is proportional to ( kn↑ . The spin relaxation rate due to phonon scattering according to the Elliott-Yafet mechanism can be exRev. (56)] and spin-down [Eq. 1999b) 1/ s 8 T 0 d 2 sF N / T. V k n ↓ 2 . (64) The Elliott relation is only a very rough estimate of The experimentally relevant ratio p / s depends on the scattering mechanism. If the solid possesses a center of symmetry. (56) (57) b * kn r ↑ ]e ik"r. spin-up and spin-down states follows from the fact that in most cases the typical values of a are close to unity. The Elliott relation relates s to the shift g of the electronic g factor from the free-electron value of g 0 2. where the two processes have similar magnitudes. where N( ) exp( /kBT) 1 1 is the phonon distribution function. M is the ion mass. Fabian. 1978) gives the revised estimate 1/ s. 1997) and for bandstructure systems by Yafet (1963). However. (63) On the other hand.350 ˇ ´ Zutic.k n ↓ uq • kn↑ .k n ↓ q kn k n .V i k n ↓ ) 2 b 2 ( kn↑ . A periodic. Mod. perturbation theory gives b so / gS 2M g kn↑. and Das Sarma: Spintronics: Fundamentals and applications mixture of the Pauli spin-up ↑ and spin-down ↓ states. (62) where p is the momentum relaxation time. lattice-ion-induced spin-orbit interaction is modiﬁed by phonons and can directly couple the (Pauli) spin-up and spin-down states. which combines with Eq. ˆ z kn↑ ) lattice-ion interaction V contains both scalar and spinorbit parts: V V sc V so . Momentum scattering is typically caused by impurities (at low T) and phonons (at high T). respectively. (59) r r a kn r ↑ a * kn r ↓ b kn r ↓ ]e ik"r . 76. (57)] states can couple and lead to spin relaxation. That it makes sense to call kn↑ and kn↓ . There are two important relations giving an order-ofmagnitude estimate of s . An empirical test of the Elliott relation for simple metals in which spin relaxation is due to thermal phonons (Beuneu and Monod. after a Fermi-surface averaging we get the estimate 1/ s b2 / p. April 2004 where uq is the polarization of the phonon with momentum q and in branch . (58) where E is the energy distance between the band state in question and the state (of the same momentum) in the nearest band. the Bloch states of ‘‘spin-up’’ and ‘‘spin-down’’ electrons with the lattice momentum k and band index n can be written as (Elliott. while b 1. (60) Here g S is the number of electron states per spin and atom at the Fermi level. see also Grimaldi and Fulde.. the case of elemental metals which Elliott considered. Such processes were ﬁrst considered for a jellium model by Overhauser (1953a. Phys. The corresponding theory was systematically developed by Yafet (1963). for calculating phonon-induced 1/ s from the known band and phonon structure. The spin-ﬂip Eliashberg function gives the contribution of the phonons with frequency to the spin-ﬂip electron-phonon interaction. useful for elemental metals and semiconductors. Realizing that the spin-conserving scattering probability gives the momentum relaxation rate. At larger T. especially at low temperatures (Yafet. the z component of the orbital momenl tum in a Bloch state. the spin-up [Eq.0023 (Elliott. The periodic ( kn↑ . 2.V i k n ↑ ) 2 . as well as its temperature dependence: the Elliott and the Yafet relations. but different n. 2 sF where we write the explicit dependence of the complex lattice-periodic coefﬁcients a and b on the radius vector r.
Equation (66) shows that the Elliott-Yafet mechanism is important for small-gap semiconductors with large spin-orbit splitting (the prototypical example is InSb). Without inversion symmetry the momentum states of the spin-up and spin-down electrons are not degenerate: E k↑ E k↓ . 6). 1984). In heterostructures the symmetry is broken by the presence of asymmetric conﬁning potentials. to an increase in 1/T 1 . The ratio p / s for impurity and phonon scattering in Al and Cu is compared by Jedema et al. especially in systems where spin hot spots are important (see Sec. ˜ is deﬁned later. as most important states are usually around high-symmetry points. E g is the energy gap. (69). The corresponding Hamiltonian term describing the precession of electrons in the conduction band is H k 1 2 • k . boundaries.C). D’yakonov-Perel’ mechanism s Eg Ek Eg 2 1 . At low temperatures. We expect that such averaging leads. The magnetic-ﬁeld dependence of the Elliott-Yafet spin relaxation has not been systematically investigated. Vol. 2. By careful symmetry considerations 1/ p (T) Yafet (1963) proved that 1/T 1 T 5 at low temperatures. For example. 1984): 1 Ek A so 2 so s from Eq. in general. III.k n↓ (k k ) 4 as k →k (Yafet. The intrinsic ﬁeld derives from the spin-orbit coupling in the band structure. Spin splittings induced by inversion asymmetry can be described by introducing an intrinsic k-dependent magnetic ﬁeld Bi (k) around which electron spins precess with Larmor frequency (k) (e/m)Bi (k). Yafet’s T 5 law stems from the nontrivial fact that for spin-ﬂip electron-phonon scattering g kn↑. (66) with E k E F . This work led to a deeper understanding of the spin relaxation processes in polyvalent metals (Silsbee and Beuneu. (65) as follows directly from Eq. 2. The Yafet relation is only qualitative. which is of order 1. Momentum-dependent spin precession described by H. and Das Sarma: Spintronics: Fundamentals and applications 351 tering off impurities. and so is the spin-orbit splitting of the valence band (see Fig. although one would expect it to be within an order of magnitude. 74 . scattering by heavy impurities induces an additional spin relaxation channel where spin ﬂip is due to the spin-orbit interaction induced by the impurities. the ratio is comparable for scattering by light impurities and by boundaries. phonon. p Ek (66) An efﬁcient mechanism of spin relaxation due to spinorbit coupling in systems lacking inversion symmetry was found by D’yakonov and Perel’ (1971e). For degenerate electron densities the spin relaxation time is given by Eq. 76. 1998). However. The Yafet relation was tested experimentally by Monod and Beuneu (1979). 1975. 1963). (67) where p (E k) is the momentum scattering time at energy E k . where cyclotron effects become important. connecting the temperature dependence of 1/T 1 with that of the resistivity : 1/T 1 T b2 T . 1/ s (T) T 2 / p (T).ˇ ´ Zutic. In metals this means a constant at low T and a linear increase at large T. 1983. Fabian and Das Sarma. 74 leads to spin dephasing. The temperature dependence of 1/ s for metals and degenerate semiconductors follows the dependence of 1/ p . or phonons. Fabian. Here we give a formula for the spin relaxation of conduction electrons with energy E k in the frequently studied case of III-V semiconductors (Chazalviel. together with momentum scattering characterized by momentum relaxation time p . Analytic formulas for the ElliottYafet mechanism due to electron-electron scattering are given by Boguslawski (1980). The numerical factor A. For nondegenerate semiconductors. Scattering by phonons is too complex to be simply equated with the ratio p / s for impurity or boundary scattering. one needs to average over cyclotron trajectories on the Fermi surface to obtain 1/ s . (65) over a large temperature range. Realistic calculations of the Elliott-Yafet s in semiconductors can be performed analytically using approximations of the band and phonon structures. It is In the qualitative reasonings below we use p instead of the effective correlation time ˜ for during momentum scattering. (66) one needs to know p . justifying Eq. Pikus and Titkov. so the D’yakonov-Perel’ mechanism does not apply to them. Elemental semiconductors like Si possess inversion symmetry in the bulk. electron-hole). Phys. treating the effects of inversion asymmetry by introducing intrinsic precession helps to give a qualitative understanding of spin dephasing. (63) after realizing that (T). (2003). while only a quadratic dependence holds for spin-conserving scattering.. Kramer’s theorem still dictates that E k↑ E k↓ . To estimate Rev. No. ¨ similarly to the Bloch-Gruneisen law for resistivity. Equation (63) then does not hold. April 2004 where are the Pauli matrices. In the important case of scattering by charged impurities ( p T 3/2) 1/ s T 1/2. depends on the dominant scattering mechanism (charge or neutral impurity. It often sufﬁces to know the doping or temperature dependence of p to decide on the relevance of the Elliott-Yafet mechanism (Pikus and Titkov. While the microscopic expression for (k) needs to be obtained from the band structure. Most prominent examples of materials without inversion symmetry come from groups III-V (such as GaAs) and II-VI (ZnSe) semiconductors. while for nondegenerate densities the thermal averaging leads to a substitution of thermal energy k B T for E k and thermalaveraged momentum relaxation time for p . This corresponds to 2 4 the long-wavelength behavior S F( ) of the spinﬂip Eliashberg function. in Eq. where inversion symmetry is broken by the presence of two distinct atoms in the Bravais lattice. Mod.
Analogous expressions for s x and s y can be written by ˙ ˙ index permutation. one deﬁnes77 l p /˜ l to measure the effectiveness of momentum scattering in randomizing Larmor frequencies. April 2004 . and Das Sarma: Spintronics: Fundamentals and applications 1 important to note. that is. the usual motional narrowing result is obtained: 1/ s 2 a v p (see footnote 70). As a result. ˜ l accounts for the relative angle between before and after scattering. Indeed. that the analogy with real Larmor precession is not complete. but later evaluate it. the spin phase follows a random walk: after time t. If the band structure is isotropic and scattering is both elastic and isotropic. for B 0 and no spin diffusion. A more rigorous expression for s in regime (ii) has been obtained by solving the kinetic rate equation for a spin-dependent density matrix (D’yakonov and Perel’. which amounts to t/ p steps of the random walk.ij (here i and j are the Cartesian coordinates) whose diagonal 1/ s.. On the other hand. however. though they do not contribute to the momentum relaxation time which appears in the measured mobility (Glazov and Ivchenko. Two limiting cases can be considered: (i) p a v 1 and (ii) p a v 1. While increased scattering intensity makes the Elliott-Yafet mechanism more effective. spin dephasing depends on the spin direction and on the dephasing rates of the perpendicular spin components.75 which depends on the distribution of values of (k). whose magnitude and direction change randomly with the average time step of p . Generally l 1 for l 1. The electron spin rotates about the intrinsic ﬁeld at an angle a v p .ij 1 l p i j. Deﬁning s as the time at which (t) 1. 2002.. The total spin in this regime initially dephases reversibly due to the anisotropy in (k). normal (not umklapp) electron-electron collisions should also be included in the effective spin randomization time ˜ . where 0 is the Larmor frequency of the external ﬁeld. 1/ s. while in case (ii) it is the distribution of the sums of the intrinsic Larmor frequencies (the total phase of a spin after many steps consists of a sum of randomly selected frequencies multiplied by p ). The faster the momentum relaxation. according to the central-limit theorem. Rev. the slower the spin dephasing. has a signiﬁcantly reduced variance. The effective momentum scattering time is introduced as 75 The reversible decay need not be exponential. This regime can be viewed from the point of view of individual electrons as a spin precession about ﬂuctuating magnetic ﬁelds. (68) 76 In fact. Fabian. 76. (70) In general. Equations (70) are valid for small magnetic ﬁelds. While these two mechanisms coexist in sx x z sy y z .ii and off-diagonal 1/ s. Phys. for i j. where a v is an average magnitude of the intrinsic Larmor frequency (k) over the actual momentum distribution. Case (i) corresponds to the situation in which individual electron spins precess a full cycle before being scattered to another momentum state. Case (ii) is what is usually meant by the D’yakonovPerel’ mechanism. Mod. when randomizing scattering takes place. Since it is useful to express the results in terms of the known momentum relaxation times76 p ˜ 1 . Spin-independent collisions with impurities or phonons do not lead to phase randomization during the collision itself. inconsistently. satisfying 0 p 1. we see that spin decay is described by the tensor 1/ s. (69) for the lth polar harmonic of (k) (Pikus and Pikus. spin phases are randomized between collisions. 77 Pikus and Titkov (1984) initially deﬁne l as here. (67) produces an equal number of spin-up and spin-down states. terms are 1/ s. is in general proportional to the width of the distribution: 1/ s . 1995). 2. evolution of the z component of spin s is (Pikus and Titkov. 1989). The spin is irreversibly lost after time p .ij . 2002). where the bar denotes averaging over directions of k. (68). while H of Eq. 1971d. The more collisions there are.352 ˇ ´ Zutic. in the Elliott-Yafet process the precession frequency is conserved between collisions and the loss of phase occurs only in the short time during collision. The difference between cases (i) and (ii) is that in case (i) the electron spins form an ensemble that directly samples the distribution of (k). Vol. 1984) sz ˙ ˜ l sz 2 2 z 1/˜ l W 1 1 P l cos d cos . Comparing with the Bloch-Torrey equations (52)– (54). The most important difference between the ElliottYafet and the D’yakonov-Perel’ mechanisms is their opposite dependence on p . in D’yakonov-Perel’ spin dephasing. leading to motional narrowing. An applied magnetic ﬁeld induces a macroscopic spin polarization and magnetization.] In two dimensions P l (cos ) is replaced by cos(l ) in Eq. 1971e). (69) where W( ) is the rate of momentum scattering through angle at energy E k . before experiencing another ﬁeld and starting to rotate with a different speed and in a different direction. the greater is the loss of phase memory. and P l is the Legendre polynomial whose order l is the power of k in (k). which. it decreases the effectiveness of the D’yakonov-Perel’ processes. No. In a sense the two mechanisms are similar to collision broadening and motional narrowing in NMR (Slichter.ii 1 l p 2 2 i . since electrons precess with different frequencies depending on their momenta. in analogy with collision broadening of spectral lines. Both limits (i) and (ii) and the transition between them have been experimentally demonstrated in n-GaAs/AlGaAs quantum wells by observing temporal spin oscillations over a large range of temperatures (and thus p ) (Brand et al. as the inverse l → l 1 . momentum scattering is more effective in randomizing spins than in randomizing momentum. but help to establish the random-walk-like evolution of the phase. 2003). The spin dephasing rate. [It is assumed that (k) k l in Eq. the phase progresses by (t) t/ p .
(71) is proportional to the cube of the lattice momentum. which is of order 1. 1984). The material-speciﬁc parameters are the band gap E g and the conduction electron mass m c .ˇ ´ Zutic. which is rarely the case. while for Since D p D’yakonov-Perel’ L s does not depend on the momentum scattering time and for a degenerate electron system should be a constant independent of T. Application of longitudinal (to the initial spin direction) magnetic ﬁeld suppresses the D’yakonov-Perel’ mechanism (Pikus and Titkov. (70) to the study of spin dephasing in bulk and two-dimensional III-V semiconductor heterostructures. For the important case of charged impurity scattering ( p T 3/2). of the order of v F / a v . 1984). the orbital motion induced by B is the cause for suppression of spin relaxation in semiconductors. Mod. 2 (75) The above expression describes D’yakonov-Perel’ spin dephasing of degenerate (E k E F ) or hot78 electrons in bulk III-V semiconductors.5. (71) Here k i are the lattice wave-vector components along the crystal principal axes. 2. Fabian. (66). the D’yakonov-Perel’ spin dephasing increases much faster with increasing electron energy and is expected to This is strictly true only if the spin relaxation of the hot electrons is faster than energy relaxation by optical-phonon emission. it is best to deduce p (E k) and p (T) from mobility measurements and use Eqs. where L is the Larmor precession due to B. when c p 1. is Q 16 35 1 3 Ek 32 105 1 3 p Ek 2 3 Ek 2 Eg . For GaAs 0. as was ﬁrst found by Dresselhaus (1955). 1984) 1/ s be dominant at large donor doping levels and at high temperatures.. Vol. Since the prefactors may vary with different scattering mechanisms. (73) 7 2 5 . 1984). The Elliott-Yafet mechanism can be dominant in small-band-gap and large-spin-orbitsplitting materials. we obtain the expected result that i j the off-diagonal elements of 1/ s. due to bulk inversion asymmetry.k z k 2 k 2 . their relative strength depends on many factors. . The temperature behavior of D’yakonovPerel’ spin dephasing in nondegenerate samples is 1/ s T 3 m (T). y z z x x y (72) 2 2m 3 E g c 1/2 . Perhaps the most robust trend is that the D’yakonov-Perel’ mechanism becomes more important with increasing band gap and increasing temperature. or the equations given below for 1/ s (T). The most frequently used ways to distinguish between various methods of spin relaxation are comparing the electron density (through the variation of the Fermi energy) and the temperature dependences of 1/ s with the theoretical estimates. and Das Sarma: Spintronics: Fundamentals and applications 353 systems lacking inversion symmetry. (70). The coefﬁcient Q. In the rest of the section we apply Eq. 1/ s T 3/2 . (71) for in Eq. 1/ s Q m 2 k BT 3 . We do not know of an experimental veriﬁcation of this distinction. Spin splitting of conduction electrons and heavy and light holes in GaAs quantum wells. Rev. For scattering by charged impurities 1/ s T 9/2. a. Compared to the Elliott-Yafet expression. for acoustic phonons 3 1. 1984) for two reasons: (i) The B ﬁeld suppresses precession along the transverse intrinsic ﬂuctuating ﬁelds when L p 1. 76. . (66) and (73).7 (Pikus and Titkov. it follows that c L . while for optical polar phonons 3 41/6.k y k 2 k 2 . Another interesting distinction between the two mechanisms is revealed by the dependence of their spin diffusion lengths L s D s on momentum scattering. which has a similar effect to averaging by random scattering. Contrary to the EY mechanism. is a dimensionless parameter specifying the strength of the spin-orbit interaction. If the electrons obey nondegenerate statistics. where c is the cyclotron frequency. if the opticalphonon emission is particularly fast. (ii) k is orbitally averaged. No. was calculated by Rashba and Sherman (1988). For impurity scattering 3 6..ij vanish for cubic systems and the diagonal elements are all equal to (Pikus and Titkov. greater impurity density will decrease the importance of the D’yakonov-Perel’ processes. thermal averaging over the Boltzmann distribution gives (Pikus and Titkov. For scattering by ionized impurities Q 1. Phys. The temperature dependence of 1/ s comes from the temperature dependence of p . while scattering by polar optical or piezoelectric phonons gives Q 0. for Elliott-Yafet L s p . and scattering by acoustic phonons (deformation potential) Q 2. Since for conduction electrons m c m e . One has to consider either the spin relaxation at different energy levels during the cascade process of optical-phonon emission or. The two mechanisms can also be easily distinguished by their opposite dependence on momentum relaxation. Eq. which is the usual case of p-doped materials.07 (Marushchak et al. Bulk III-V semiconductors In bulk III-V semiconductors the intrinsic Larmorfrequency vector of Eq. The spin splitting described by Eq. spin relaxation only during the ﬁnal stages of energy relaxation by acoustic-phonon emission (see Pikus and Titkov 1984). 1971d) k where k x k 2 k 2 . April 2004 78 where the power law p E k is assumed for momentum relaxation time. (67) due to the lack of inversion symmetry is (D’yakonov and Perel’. 2 Eg (74) where m p (E k)E k / E k . and using (4/105)k 6 ij . Substituting Eq.8.
k y . The tensor ˆ depends on the orientation of n with respect to the principal crystal axes (D’yakonov and Kachorovskii. k n 0. the directions follow a ‘‘breathing’’ pattern as shown in Fig. 1986). predict a different dependence of s on the quantum-well orientation relative to the principal axes.7794(2m e E/ 2 ) 2/3 [see. For . the vector pattern is also the pattern of the spin on the Fermi surface. one obtains 1/ s. The ﬁeld for BIA  lies perpendicular to the plane. 80 A trivial typo in xx is corrected.xx 1/ s. The spin dephasing tensor 1/ s. 79 This readily gives (D’yakonov and Kachorovskii. 1/2 s. All other cases have constant ﬁelds lying in the plane.. z (80) (81) n xn y 9 n 2 n 2 n 2 n 2 x z y z and analogously for other components. Treating wave ˆ vectors k in Eq.80 xx xy 4n 2 n 2 n 2 x y z n2 n2 y z 2 9n 2 1 . Figure 15 shows the vector ﬁeld patterns of the intrinsic magnetic ﬁelds for both bulk and spin inversion asymmetry.yy 1/2 s. (71) and (77).354 ˇ ´ Zutic. however. (79): 1/ s.ij ij Tr ˆ ij / 0 s Ek . 4/3 0 s . x 2 1 n2 . for example. 15. 76. (78) Ek 2 kn 2 Ek p 2 2m c E g 2 2 Ek . The bulk inversion asymmetry term comes from the bulk Dresselhaus spin splitting. We follow D’yakonov and Kachorovskii (1986) in discussing the three important cases of . No. Eq. April 2004 . (83) As expected for the case of the in-plane ﬁeld. 2. For a rectangular quantum well of width a.ii 16/9 s and 1/ s. (71) as operators k i . For  quantum wells. Quantum averaging of can be done using the formula ˆ ˆ ˆ k ik jk l 2 k n k in jn l k jn ln i k ln in j While the magnitude of (k) is constant over the Fermi surface.zz off-diagonal components vanish. . By diagonalizing 1/ ij we obtain 1/ s 1/2 s. Both H BIA and H SIA lead to spin splitting of the conduction band linear in k. Vol. having a constant magnitude (refer to Fig. In the following. deﬁned in Eq. Vector ﬁelds (k) (k) on the Fermi surface (here a circle) for the structure inversion asymmetry (SIA) and bulk inversion asymmetry (BIA). (79) FIG. (84) The intrinsic magnetic ﬁeld lies in the plane. de Sousa and Das Sarma (2003c)]. and  quantum wells. (70). Spin relaxation 0 0 rates are now 1/ s. (82) In two-dimensional III-V semiconductor systems (quantum wells and heterostructures) there are two distinct Hamiltonian terms that contribute to D’yakonovPerel’ spin dephasing: the bulk inversion asymmetry term H BIA and the structure inversion asymmetry term H SIA . a perpendicular spin dephases twice as fast as a parallel one. which appears only in asymmetric systems. leads to momentum quantization along the conﬁnement unit vector n of the quantum well. and Das Sarma: Spintronics: Fundamentals and applications and similarly for y and z by index permutation. 15. Terms cubic in k were omitted from the above equation. is readily evaluated using Eqs. as2 suming that for narrow quantum wells k 2 k n . The explicit knowledge of is useful in qualitative analysis of spin dephasing for particular orientations of quantum wells.yy 1/2 zz 1/ s . k denotes the wave vector for a Bloch state propagating in 2 ˆ the plane. The spin relaxation times follow from 0 Eq. The two terms. This is shown here only for the SIA case. 1986) x 2 k n 2n x n y k y n z k z kx n2 n2 . As the opposite spins have different energies. 1/ 0 s . Fabian. as spin dephasing rates of spins parallel and perpendicular to the plane. k 2 kn b. 15).i j 4/9 s . y z Rev. since (k) lies in the plane. Other which 1/ s. Since (k) is also the spin quantization axis.0 .79 1/ where 1 0 s s. the Fermi circle becomes two concentric circles with opposite spins. Diagonalizing the tensor gives k i k j k l . The most interesting case is the  orientation for 0 1/ s. with the magnitude varying along the Fermi surface. For a triangular well of conﬁning potential 2 eEz. (76) (77) Averaging over the directions of k in a plane perpendicular to n can be performed by using k i k j (k 2 /2)( ij n i n j ). (85) As for the  case. k 2 2/)k n k n . and evaluating as the expectation value in the conﬁned states. Two-dimensional III-V semiconductor systems k x .xx 1/ s. and k n (k•n) 2 denotes the expectation value of the square of the component of the wavenumber operator normal to the plane in the lowest subband state.ij . Mod. 2 k n ( /a) 2 . Deﬁning 1/ s and 1/ s. the lifetime of a spin parallel to the plane is twice that of the spin perpendicular to the plane.xy 1/8 s . Phys. but the analogy extends to all examples. (71).
Wissinger et al. This is due to the fact that . according to the direction of the spin pattern— parallel or antiparallel to (k). by diagonalizing 1/ s. Perhaps the most appealing fact about structure inversion asymmetry is that BR can be tuned electrostatically. (90) Since the intrinsic precession vector (k) for the structure inversion asymmetry always lies in the plane. 1/ s. with the precession vector k 2 BR k n . 1984b) occurring in asymmetric quantum wells or in deformed bulk systems. 15. (2003).E. April 2004 relative importance of the Dresselhaus and Rashba spin splittings in III-V heterostructure systems. Calculations of s based on the D’yakonov-Perel’ mechanism. As for the bulk inversion asymmetry case. 1960. were performed by Sherman (2003a). Band-structure k•p calculations of BR for quantum wells in GaAs/AlGaAs heterostructures can be found in Pfeffer and Zawadzki (1995). 76. k 2 k n k x /2 1. Phys. see Sec. 1.A and IV. Note that the magnitude of (k) changes along the Fermi surface. a perpendicular spin should dephase twice as fast as a spin in the plane. (1998). and in p-InAs metaloxide-semiconductor ﬁeld-effect transistor channel in Lamari (2003). Another interesting feature of bulk and structure inversion asymmetry ﬁelds is that injection of electrons along a quasi-one-dimensional channel can lead to large relaxation times for spins oriented along (k). (88) Here BR is a parameter depending on spin-orbit coupling and the asymmetry of the conﬁning electrostatic potentials arising from the growth process of the heterostructure. The corresponding Hamiltonian is that of Eq. such as the effective mass. Indeed. in InSb/ InAlSb asymmetric quantum well it can be found in Pfeffer and Zawadzki (2003). The relativistic transformation then gives rise to a magnetic ﬁeld (spin-orbit coupling) acting on the electron spin. 2003). and Das Sarma: Spintronics: Fundamentals and applications 355 1/ s. Spin dephasing in  quantum wells can still be due to the cubic terms in k left out of Eq. 0. or the asymmetry in the penetration of the electron wave function into the barriers (de Andrada e Silva et al. a calculation for InGaAs/InP quantum well is reported in En¨ gels et al. The splitting can also arise in nominally symmetric heterostructures with ﬂuctuations in doping density (Sherman. where k is the wave vector for the states in the channel (Hammar and Johnson. This interesting fact qualitatively distinguishes structure from bulk inversion asymmetry and can be used in assessing the Rev.1) in which BR is tailored by a gate. (67). 1/2 s. unlike the previous cases.ij one ﬁnds that 1/ s. Mod. Majewski and Vogl (2003) have recently calculated the structure inversion asymmetry by local density functional methods and .. Rashba and Efros. assisted by external bias. This is in general wrong. 2003. The Bychkov-Rashba ﬁeld always lies in the plane. 1997). and thus of the interpretation of experimental results on splitting in semiconductor heterostructures. Electrons moving along  experience little spin dephasing. This led to one of the pioneering spintronic proposals by Datta and Das (1990. 2002). Model spin dephasing calculations based on structure inversion asymmetry were carried out by Pareek and Bruno (2002). (86) The perpendicular spin does not dephase. (78) as 1/ and ij 0 s 4 2 BR mc 2 Ek p (89) 1 n in j . Vol. 15) and thus cannot affect the precession of the perpendicular spin. has been debated. Adding to the controversy. This tailoring. Research on spin inversion asymmetry is largely motivated by Datta-Das spin ﬁeld-effect transistor proposal (see Secs. I. For a clear qualitative explanation of the involved physics see Pfeffer and Zawadzki (1999) and Zawadzki and Pfeffer (2001). Bychkov and Rashba. The parameter BR is then assumed to be directly proportional to the conﬁning electric ﬁeld. and Kainz et al. as shown in Fig. IV. 2. Pfeffer (1997). since the average electric force acting on a conﬁned particle of uniform effective mass is zero. (1997). has been controversial and the microscopic origin of the Bychkov-Rashba Hamiltonian. (2003). Fabian. as was shown for  quantum wells by Averkiev and Golub (1999) and Kainz et al. with structural asymmetry due to doping ﬂuctuations in the heterostructure interface. This plane anisotropy can be easily seen by adding the corresponding vector ﬁelds in Fig. Using the same procedure as for bulk inversion asymmetry. however.1). The structure inversion asymmetry term arises from the Bychkov-Rashba spin splitting (Rashba. If bulk and structure inversion asymmetry are of similar importance. which acts on a moving electron in a transverse direction. No. 15. 1/4 0 s . 1984a. the structure inversion asymmetry leads to a splitting of the Fermi surface. having a constant magnitude. we describe the spin relaxation rate by Eq. 1/ 0 s (91) holds for all quantum-well orientations n. Note that for the  orientation the bulk and structure inversion asymmetry terms have the same form. (1997) and Schapers et al. is always normal to the plane (see Fig. the interference terms from the cross product BIA SIA can lead to spin dephasing anisotropies within the plane. 2003a).. (87) where it is used that k•n 0. Indeed.ˇ ´ Zutic.0 . (77) or to other spin relaxation mechanisms. The Bychkov-Rashba Hamiltonian is often interpreted as arising from the electric ﬁeld of the conﬁning potential. The difﬁculty in understanding the inﬂuence of the external gates is caused by the lack of the understanding of the inﬂuence of the gate ﬁeld on the asymmetry of the well. potentially providing an effective spin precession control without the need for magnetic ﬁelds (Levitov and Rashba.E. The asymmetry that gives rise to structure inversion asymmetry is the asymmetry in the band structure (including spin-orbit coupling) parameters of a heterostructure.
For degenerate holes (0) 2 is of order 1. especially at determining the zero magnetic ﬁeld spin splitting (usually seen in Shubnikov–de Haas oscillations). If electrons are thermalized.356 ˇ ´ Zutic. or interpret the splitting differently [see. (1975). 0 1 10 1983). Mod. In addition.. The spin-ﬂip scattering probability depends on the state of the holes (degenerate or nondegenerate. (92) where A is proportional to the exchange integral between the conduction and valence states. (2002a). Fabian. Vol. If holes are degenerate and the electrons’ velocity v k is greater than the Fermi velocity of the holes’. Rev. 2. The Bir-Aronov-Pikus mechanism coexists with the Elliott-Yafet and D’yakonov-Perel’ mechanisms in p-doped materials lacking inversion symmetry.7 10 5 eV (Aronov et al. E B /2m c a B and ex the exchange splitting of the excitonic ground state). We present below the most frequently used formulas when assessing the relevance of the Bir-Aronov-Pikus mechanism. Maialle and Degani. v k needs to be replaced by the thermal velocity v e (3k B T/m c ) 1/2. (95). Gate dependence of BR was also measured in modulation-doped InP/InGaAs/InP quantum ¨ wells (Engels et al. leading to inhomogeneous dephasing. and ex 4. Interpretation of experimental data on structure inversion asymmetry is difﬁcult. Furthermore. so that a large Bychkov-Rashba term can be present in otherwise symmetric quantum wells with no common atom. Bir-Aronov-Pikus mechanism 1 s 2 0 3 N aa B vk p vB Na 0 4 5 Na p . (1999) and Rowe et al. Weak antilocalization studies of InAlAs/ InGaAs/InAlAs quantum wells have recently been used to study electron density dependence of BR by Koga et al. in asymmetric AlSb/InAs/AlSb quantum wells. 76. For electron-spin relaxation due to exchange with nondegenerate holes. accompanied by spin exchange. which enhances the free hole contribution. (1998) show a constant BR 0.9 meV. The formulas are valid for the usual cases of heavy holes m v m c .. S is the electron-spin operator. 8 s. Schapers et al. At large temperatures. 1/ s is only weakly dependent on N a . No. where also the gate voltage dependence is obtained: BR ranged from 10 11 eV m at the depleting gate voltage of 1 V. and Das Sarma: Spintronics: Fundamentals and applications concluded that the induced spin splitting arises from microscopic electric ﬁelds in asymmetric atomic arrangements at the interfaces. by holes. The BychkovRashba parameter was measured in GaSb/InAs/GaSb quantum wells ( BR 0. Another potentially relevant mechanism for spin relaxation of donor-bound electrons in p-doped semiconductors is the exchange interaction with holes bound to acceptors (D’yakonov and Perel’. which is masked by Zeeman splitting at ﬁnite ﬁelds. 1997. due to its increased importance at large electron energies. Both electron hopping and hole spin ﬂip motionally narrow the precession. even for large acceptor densities. fast or slow). (93) and 1/ s N 1/3 for degenerate holes from Eq. 1997). For GaAs a B 114 Å. (94) E k /E B . and v B /m c a B . then 1 s 3 0 3 pa B v k k BT . p is where a B is the exciton Bohr radius a B the density of free holes. 3. independent of gate voltage. For an unscreened Coulomb potential 0 2 2 1 exp 2 1 . as was ﬁrst shown by Bir et al. (1990).7 107 cm s 1 .6 10 11 eV m. bound on acceptors or free. 3 Na (93) 2 /e 2 m c . Phys. The dependence on the acceptor density is essentially 1/ s N a for nondegenerate/bound holes from Eq. v B E Fh (95) Spin relaxation of conduction electrons in p-doped semiconductors can also proceed through scattering. there are experimental reports that either fail to observe the expected spin splitting due to the BychkovRashba ﬁeld. measurements of Heida et al. J is the angular momentum operator for holes. The observed values are several 10 12 eV m. The temperature dependence of s is dominated by the temperature dependence of (0) 2 as well as by p. April 2004 where E Fh is the hole Fermi energy. (2001)]. the D’yakonov-Perel’ mechanism can become more important. demonstrating that control of BR may be difﬁcult. 1973b). The exchange interaction between electrons and holes is governed by the Hamiltonian H AS"J r . in InAlAs/InGaAs/InAlAs quantum wells (20 nm).D. (0) 2 is Sommerfeld’s factor. more experimental efforts need to be devoted to this interesting topic.1. The exchange interaction provides an effective magnetic ﬁeld for electron spins to precess.. v B 1. Brosig et al. For a completely screened potential where (0) 2 1. A general trend is that the Bir-Aronov-Pikus dominates in heavily doped samples at small temperatures. for example. and r is the relative position of electrons and holes. III. typically less than 1 meV. In order to unify the different views on what exactly the Bychkov-Rashba spin splitting means and how the spin splitting can be tuned with gate voltage. to 5 10 12 eV m at 1. . In a between. 1998). Speciﬁc examples of the domain of importance for the three mechanisms are discussed in Sec. E B 4.. Model calculations of Bir-Aronov-Pikus processes for electrons in p-doped bulk and quantum wells were performed by Maialle (1996. the splittings are small. On the other hand.5 V.9 10 11 eV m for a 75-Å-thick well) by Luo et al. The three mechanisms can be distinguished by their unique density and temperature dependences. 0 is an exchange splitting pa2 rameter: / 0 (3 /64) ex/E B (with E B denoting the 2 2 Bohr exciton energy.
where the feedback from this precession can be directly observed through the modulated precession of electron spins. 1997). 2002.. lead to irreversible spin dephasing and decoherence of electron spins. 28Si. B is the Bohr magneton.1). 2002). for example. et al. which can be much faster than what is inferred from T 2 . The typical time scale for the ﬂuctuations in GaAs is given by the nuclear Larmor precession period in the ﬁeld of neighboring nuclei and is of order 100 s (Merkulov et al. provides an important mechanism (D’yakonov and Perel’. In GaAs all the lattice nuclei carry the magnetic moment of 3/2 spin. April 2004 mal or nonequilibrium distribution of the nuclear spins. (97) where g is the effective g factor of the electron. Nuclear moments also precess (and orient) in the magnetic ﬁelds of polarized electrons. 1984). There are three important regimes in which the hyperﬁne interaction leads to spin dephasing of localized electrons: (i) In the limit of small orbital and spin correlation between separated electron states and nuclear spin states. such as those conﬁned in quantum dots or bound on donors. 1984). Spin relaxation . i is the label for nuclei at positions Ri . 2002). an electron bound on a shallow donor in Si experiences only around 100 magnetic nuclei. masking the effects of temporal ﬂuctuations of B n (see Sec. spatial variations in Bn lead to inhomogeneous dephasing of the spin ensemble. 2002). The electron Zeeman splitting due to the average B n corresponds to a ﬁeld of 1 T or thermal energy of 1 K.i S•Ii B i r Ri . Eq. Fabian. The time scale for the Larmor precession of nuclear spins in hyperﬁne ﬁelds is 1 s in GaAs (Merkulov et al. since the donor states spread to greater distances.D.A. The effective Hamiltonian for the hyperﬁne interaction is the Fermi contact potential energy (Slichter. Vol. for electron spins in quantum dots (Merkulov et al. the hyperﬁne interaction is relevant for spintronics as a means to couple. respectively. and thus even in the low-doping limits ( 1014 cm 3 donors) the exchange interaction can be rather large. Merkulov. III.ˇ ´ Zutic. g 0 2. Hyperﬁne-interaction mechanism The hyperﬁne interaction. We stress that it is the free electron g factor g 0 and not the effective g that appears in the hyperﬁne interaction. given by the corresponding therRev. 2002). with the rate proportional to the rms of B n . Such processes are sometimes referred to as spectral diffusion.. experiencing the combined magnetic moments of many nuclei. who found suppression of 1/ s if the external ﬁeld is greater than B n for regime (i). in regime (iii). 2003b).. both of nuclear spin 1/2. in a controlled way. 2003c). Ensemble spin dephasing due to the hyperﬁneinteraction mechanism in an external magnetic ﬁeld has been studied by D’yakonov and Perel’ (1973b). III. (ii) Temporal ﬂuctuations in B n . which can occur due to nuclear dipole-dipole interactions. The interaction is too weak to cause effective spin relaxation of free electrons in metals or in bulk semiconductors (Overhauser.. Many useful parameters for evaluating effective magnetic ﬁelds and precession frequencies due the hyperﬁne-interaction mechanism in GaAs are given by Paget et al. 1989). Such inhomogeneous dephasing is seen by electron-spinresonance (ESR) experiments on donor states both in Si (Feher and Gere. electron and nuclear spins (D’yakonov and Perel’. Semenov and Kim. Korenev. since the electron Zeeman levels split by B n undergo random shifts (de Sousa and Das Sarma. In addition to spin dephasing. Phys. As a result. carries no spin and the hyperﬁne interaction is due to 29Si (natural abundance 4. 2002. 1973b) for ensemble spin dephasing and single-spin decoherence of localized electrons. as it is strongly dynamically narrowed by the itinerant nature of electrons (see Sec. as shown by Yafet (1961) [see also Paget et al. so this effect does not lead to motional narrowing of B n . for a complete nuclear polarization (Paget et al. 2. and n. Localized electrons are typically spread over many lattice sites (104 – 106 ). 1977).. and even for the case of conduction electrons in semiconductors (revisited by Pershin and Privman. and Das Sarma: Spintronics: Fundamentals and applications 357 4. which is the magnetic interaction between the magnetic moments of electrons and nuclei.i Ii Ri 2 . Mod. 76. spin precession due to B n is motionally narrowed. 1958). while in Si the most abundant isotope. by Gordon and Browers. due to the external ﬁeld. 1959) and in GaAs (Seck et al. Merkulov et al. electron spins precess many times before the nuclear spin ﬂips.. This effect can be removed by spin-echo experiments (in Si donor states performed. electron and nuclear spin operators expressed in the units of . It follows that the spin of an electron in an orbital state (r) experiences magnetic ﬁeld Bn 2 3 0 g0 g i n. it leads to individual spin decoherence.0023 is the free-electron g factor. The spread in the Larmor precession period due to the variance in B n in GaAs is estimated to be around 1 ns (Dzhioev..67%) or the frequent donor 31P. H 8 0 g0 3 4 n. (1977)]. No. (96).. 1953a).. This effect is much more pronounced in GaAs than in Si. (1977). S and Ii are. and the effects of the hyperﬁne interaction are considerably smaller than in GaAs. (iii) In the presence of strong orbital correlations (electron hopping or recombination with acceptor hole states) or spin correlations (direct exchange interaction) between neighboring electron states. (96) where 0 is the vacuum permeability. While the direct spin exchange interaction does not cause ensemble spin relaxation (the total spin is preserved in spin ﬂip-ﬂops). 2002). Calculations of s using the hyperﬁne-interaction mechanism were performed for shallow donor states in Si at low temperatures and magnetic ﬁelds (Saikin et al. or a smaller Larmor precession period than the correlation time for random changes in B n .i is the nuclear gyromagnetic ratio. an effect important in optical orientation (Meier and Zakharchenya.3).
1953a). who studied spin coherence time sc of a single electron spin in the regime in which the electron Larmor period due to B n is much shorter than the correlation time of the nuclear magnetic ﬁeld ﬂuctuations. Na (Feher and Kip. The temperature dependence of s is consistent with the Elliott-Yafet mechanism: 1/ s is constant at low temperatures. For instance. 2002a. Yafet. Rb3 C60 (Janossy et al. Be.. by spin echo measurements (Tyryshkin et al. For the majority of metals tested (alkali and noble). (2002.. 1955. 2003). consistent with phonon-induced spin relaxation. Orchard-Webb et al. 2002) to infer that the in-plane spin relaxation time is unusually long. (59) shows excellent agreement with experiment.. 1999a). Walsh et al. Fabian and Das Sarma. 1954. 1993). 76. 1976). 1976b). 2003). This work showed that the magnitudes of the spin-mixing probabilities b 2 . To our knowledge the longest s reported for a metal—a microsecond— was found in high-purity sodium below 10 K (Kolbe.. The spin injection technique (see Sec. Surfacescattering spin relaxation times in normal metals and superconductors are collected by Meservey and Tedrow (1978). 1981). Al (Lubzens and Schultz. 1963). (64) (Beuneu and Monod.. The initially suggested deviation from the Yafet relation for several polyvalent metals (Al. 1966. Au (Elezzabi et al. who offer reasons why mechanism (ii) should dominate spin decoherence in GaAs quantum dots of radius smaller than 100 nm. including Al (Johnson and Silsbee.1 to 1 ns. 1976). from picoseconds to microseconds.. Heersche.. C. Lubzens and Schultz. 1999c). Cs (Schultz and Shanabarger. Jedema. 2001). 1998. Cu (Jedema. The spin relaxation time of conduction electrons in metals has been measured by both CESR and spin injection techniques. 1979. In addition to CESR and spin injection. A realistic. et al. 1978).. Various band-structure anomalies (spin hot spots). graphite (Wagoner. in a 50-nm-wide quantum dot.. K (Walsh et al. Nijboer. (62). 1966. Typical values of s were found to be 0. 1970)... there are still too few experimental data to make conclusions about the merits of speciﬁc models of the hyperﬁne-interaction mechanism. Damay and Sienko. 1994). indicating impurity spin-ﬂip scattering. 1985. has been tested for many important metals and found to be valid over many orders of magnitude of g (Beuneu and Monod. and Das Sarma: Spintronics: Fundamentals and applications processes due to phonon-assisted hyperﬁne interactions were investigated in GaAs quantum dots.358 ˇ ´ Zutic. 2003b). Spin decoherence times for single-electron spins were recently computed for case (ii) by Khaetskii et al. while at high temperatures 1/ s grows linearly with increasing T. 1970. 1979).. this reference contains a useful collection of data for g). 1966a). and van Wees. can increase the atomicphysics-derived b 2 by several orders of magnitude. strongly enhancing spin relaxation in polyvalent metals as compared to simple estimates (Fabian and Das Sarma. Phys. Unfortunately.81 Various data on CESR s are collected by Beuneu and Monod (1978. Eq. Monod and Schultz.. 2001. 81 Rev. 1964. taken from atomic physics to test Eq. with the possible exception of Li (Damay and Sienko. MgB2 (Simon et al. (65). ´ Au (Monod and Janossy. Realistic estimates of hyperﬁne-interaction spin dephasing in GaAs quantum dots were given by de Sousa and Das Sarma (2003a.. Monod and Beuneu.. The majority of simple metals are believed to follow the Elliott-Yafet mechanism of spin relaxation. Mg (Bowring et al. Rb (Schultz and Shanabarger. 2002. Using YBa2 Cu3 O7 . 1996). This is supported by several facts: (i) The Elliott-Yafet processes give the right order of magnitude for s (Elliott. Jedema. Kolbe. 1999c). No. and Mg) was later resolved by spin-hot-spot theory (Silsbee and Beuneu. for example. Cu (Schultz and Latham. 2001). 1991). Vol. 1998. II) was also used to measure s for various metals. but the range of observed values is large.C) using Eq. Fabian. 2003). April 2004 . ´ 1960. but were found to be ineffective (Erlingsson et al.. Vescial et al.. 1966b). et al. while other known possible spin relaxation mechanisms lead to much greater s than what is observed (Overhauser. Pd. to be described below. a best ﬁt gives the quantitative formula. Fabain and Das Sarma. 1966b). (65). Zn (Stesmans and Witters. Spin relaxation in metals (iv) (v) The Yafet relation. IV. should not be used in the solid-state environment. Jedema. 1955. Eq. Nijboer.F). 2. 1977). 1982) and tunneling spectroscopy of metallic nanoparticles (Petta and Ralph. Matsubara et al. 1965. 1982). 2003). ﬁrst-principles calculation for Al (Fabian and Das Sarma. is satisﬁed for most metals with the known temperature dependence of s (Monod and Beuneu. Orchard-Webb et al. III. or symmetry points on the Fermi surface.. the estimated sc is 50 s. (iii) A list of selected metals includes Li (Feher and Kip. Interesting results were obtained by injecting spin into superconductors. Costache. Eq. 1983. 1971). large enough for quantum computing applications (see Sec. Filip. 1978. 1971). information about spin-orbit scattering times so (see below) in various (but mostly noble) metals at low temperatures has been also obtained from weak localization magnetoresistance measurements on thin ﬁlms (Bergmann. data were interpreted (Fi et al. Mod. Jedema. such as crossing of the Brillouin-zone boundaries. accidental degeneracy points. and Nb (Johnson. 1971). et al. et al. The Elliott relation. see Sec.. 1965. This claim is supported by the recent measurement of the spin dephasing time of about 60 ms of an isolated spin in a phosphorus donor in isotopically pure 28 Si. (ii) CESR measurements of s in various metals are numerous. 2001). 1999b. Walsh et al. Be (Cousins and Dupree. 1976a.
ˇ ´ Zutic, Fabian, and Das Sarma: Spintronics: Fundamentals and applications
about 100 s at low temperatures to 1 s close to the superconducting transition temperature. For quasiparticles moving along the c axis, s is more likely to be the usual spin-orbit-induced spin relaxation time, having values of 10–100 ps. The microscopic origin of quasiparticle spin relaxation in cuprate superconductors is not yet known. There is one more important time scale, the spin-orbit scattering time so , that is often invoked in mesoscopic transport as a characteristic of spin relaxation processes. We discuss it brieﬂy in connection to s . The spin-orbit scattering time is the scattering time of Bloch electrons by the spin-orbit potential induced by impurities. The spin-orbit part of the Fourier transform of the impurity potential can be written as ic(k k )(k k )• , where c(q) is proportional to the Fourier transform of the impurity potential. The spin-orbit scattering time then is (Werthamer, 1969) 1/ 2
Ni c k k
N EF ,
where N i is the impurity concentration, N(E F ) is the density of states per spin at the Fermi level, and the angle brackets denote Fermi-surface averaging. As a parameter so also includes the spin-orbit coupling of the host lattice, in the sense of the Elliott-Yafet mechanism. Note, however, that s so , even at low temperatures where the impurity scattering dominates spin relaxation, since the spin-orbit scattering includes both spin-ﬂip and spin-conserving processes, which, for isotropic scattering rates are in the ratio 2:1. In addition, the spin relaxation rate is twice the spin-ﬂip scattering rate, since each spin ﬂip equilibrates both spins equally. For isotropic systems, 1/ s 4/(3 so ). For a discussion of the effects of the D’yakonov-Perel’ processes on weak localization see Knap et al. (1996). We illustrate spin relaxation in metals by the case of Al, whose s has been measured by CESR and spin injection and numerically calculated from ﬁrst principles. The case is instructive since it demonstrates both the general principles of the Elliott-Yafet mechanism and the predicting power of realistic band-structure calculations of s . Spin relaxation in Al was originally observed in CESR experiments, in which s was measured at low temperatures, from 1 to 100 K (Lubzens and Schultz, 1976b). The spin relaxation rate 1/ s was found to be independent of temperature below 10–20 K; at higher temperatures 1/ s increases linearly with increasing T. The same behavior was later observed in the original spin injection experiment (Johnson and Silsbee, 1985, 1988d). Recently, s was measured by spin injection at room temperature (Jedema, Costache, et al., 2002; Jedema, Heersche, et al., 2002a; Jedema, Nijboer, et al., 2003). Unlike the CESR and the original spin injection experiments, which were performed on bulk samples, the roomtemperature measurement used thin Al ﬁlms, observing strong spin relaxation due to surface scattering. Spin relaxation in Al, as well as in other polyvalent metals, at ﬁrst appeared anomalous (Monod and BeuRev. Mod. Phys., Vol. 76, No. 2, April 2004
neu, 1979), in that a simple application of the Yafet relation, Eq. (65), yielded estimates for 1/ s two orders of magnitude smaller than the observed data. Consider Na as a reference. The atomic so / E [cf. Eq. (58)] for Na and Al are within about 10% of each other (Beuneu and Monod, 1978; Monod and Beuneu, 1979), yet the corresponding s for Al is about two orders of magnitude smaller than that for Na (Feher and Kip, 1955; Vescial et al., 1964). This anomaly extends to the g factors as 8 10 4 and for Al it is six times well. For Na, g Na 5 10 3 , while one would expect them greater, g Al to differ also by about 10%. Note, however, that the Elliott relation, Eq. (62), is unaffected by this discrepancy, as it predicts that s (Na)/ s (Al) 40. It was later suggested (Silsbee and Beuneu, 1983) that this is due to accidental degeneracies in the two-band Fermi surface of Al. A full theoretical description, supported by ﬁrstprinciples calculations, of spin relaxation in Al and other polyvalent metals led to the spin-hot-spots theory (Fabian and Das Sarma, 1998, 1999a, 1999b, 1999c). Spin hot spots are states on the Fermi surface that have anomalously large spin mixing probabilities b 2 ( so / E) 2 , arising from small energy gaps E. Quite generally, such states occur near Brillouin-zone boundaries and accidental degeneracy points, or at highsymmetry points. The condition for a spin hot spot is both a small band gap E and nonvanishing so . 82 If an electron hops in or out of a spin hot spot, the chance of spin ﬂip dramatically increases. Although the total area of spin hot spots on the Fermi surface is small, their contribution to 1/ s is dominant, due to the large value of their b 2 in the Fermi-surface average b 2 , as was shown by analytical arguments (Fabian and Das Sarma, 1998, 1999a). A realistic numerical calculation (Fabian and Das Sarma, 1999b) for Al, also showed that both the accidental degeneracies considered by Silsbee and Beuneu (1983) and states close to the Brillouin-zone boundaries dominate spin relaxation. A realistic calculation of s in Al, based on pseudopotentials and a realistic phonon description, has been performed by Fabian and Das Sarma (1999b) and compared to the experimental data available for T 100 K (Johnson and Silsbee, 1985, 1988d). Figure 16 shows both the experiment and the theory. In the experimental data only the phonon contribution to 1/ s is retained (Johnson and Silsbee, 1985); the constant background impurity scattering is removed. The ﬁgure shows a rapid decrease of s with increasing T at low T, where the agreement between experiment and theory is very good. Above 200 K (the Debye temperature T D 400 K) the calculation predicts a linear dependence s ns 24 T 1 K 1 . In the phonon-dominated linear regime the Elliott-Yafet mechanism predicts that the ratio a ph
82 At some symmetry points b may be very small. This occurs in the noble metals which have Fermi states at the Brillouin-zone boundaries, where E is large, but the corresponding so is very small due to symmetry.
ˇ ´ Zutic, Fabian, and Das Sarma: Spintronics: Fundamentals and applications
FIG. 16. Measured and calculated s in Al. The low-T measurements are the conduction-electron spin resonance (Lubzens and Schultz, 1976b) and spin injection (Johnson and Silsbee, 1985). Only the phonon contribution is shown, as adapted from Johnson and Silsbee (1985). The solid line is the ﬁrst-principles calculation, not a ﬁt to the data (Fabian and Das Sarma, 1998). The data at T 293 K are results from room-temperature spin injection experiments of Jedema et al. (2002a, 2003). Adapted from Fabian and Das Sarma, 1999b.
face. Such a situation is common in polyvalent metals, whose spin hot spots have anomalously large spin-orbit coupling. In a magnetic ﬁeld the electron spins precess correspondingly, with rates varying by L ( g/g) L , where L is the Larmor frequency. Mo2 tional narrowing leads to 1/ s ( L ) c , where c is the correlation time for the random changes in g. At small magnetic ﬁelds c p and 1/ s B 2 p . Such a quadratic dependence of 1/ s on B is a typical motional narrowing case and has been observed at low temperatures in Cu (Lubzens and Schultz, 1976b). As the ﬁeld increases c becomes the time of ﬂight through spin hot spots, in which case c 1/B. As a result 1/ s acquires a component linear in B, in accord with experiment. In an effort to directly detect phonon-induced spin ﬂips in Al, an interesting experiment was devised (Grimaldi and Fulde, 1996; Lang et al., 1996) using the Zeeman splitting of the energy gap in Al superconducting tunnel junctions. Although the experiment failed, due to overwhelming spin-ﬂip boundary scattering, it showed the direction for future research in studying spin-ﬂip electron-phonon interactions.
D. Spin relaxation in semiconductors
does not depend on T (see Sec. III.B.1). The ph calculated value is a th 1.2 10 4 (Fabian and Das Sarma, 1999b), showing that 104 phonon scatterings are needed to randomize electron spin. An important step towards extending spin injection capabilities was undertaken recently by achieving spin injection into Cu and Al at room temperature (Jedema et al., 2001; Jedema, Costache, et al., 2002; Jedema, Heersche, et al., 2002a; Jedema, Nijboer, et al., 2003); the measured data are unique in providing reliable values for spin-diffusion lengths and spin-relaxation times in these two important metals at room temperature. The measured values for Al are somewhat sensitive to the experimental procedure and data analysis: s 85 ps (Jedema, Heersche, et al., 2002a) and s 124 ps (Jedema, Nijboer, et al., 2003), as compared to s 90 ps predicted by the theory at T 293 K. The room-temperature experimental data are included in Fig. 16 for comparison. They nicely conﬁrm the theoretical prediction. Less sensitive to data analysis is the ratio a ph , for which the experiments give 1.1 10 4 (Jedema, Heersche, et al., 2002a) and 1.3 10 4 (Jedema, Nijboer, et al., 2003), ph comparing favorably with the theoretical a th 1.2 10 4 . Spin relaxation in Al depends rather strongly on magnetic ﬁelds at low T. CESR measurements (Lubzens and Schultz, 1976a, 1976b) show that at temperatures below 100 K, 1/ s increases linearly with increasing B. A speciﬁc sample (Lubzens and Schultz, 1976b) showed a decrease of s from about 20 ns to 1 ns, upon increase in B from 0.05 to 1.4 T. It was proposed that the observed behavior was due to cyclotron motion through spin hot spots (Silsbee and Beuneu, 1983). The reasoning is as follows. Assume that there is considerable spread (anisotropy) g g of the g factors over the Fermi surRev. Mod. Phys., Vol. 76, No. 2, April 2004
Although sorting out different spin-relaxation mechanisms of conduction electrons in semiconductors is a difﬁcult task, it has generally been observed that the Elliott-Yafet mechanism is relevant in small-gap and large-spin-orbit coupling semiconductors, while the D’yakonov-Perel’ processes are responsible for spin dephasing in middle-gap materials and at high temperatures. In heavily p-doped samples the Bir-Aronov-Pikus mechanism dominates at lower temperatures, the D’yakonov-Perel’ at higher. In low-doped systems the D’yakonov-Perel’ dominates over the whole temperature range where electron states are extended. Spin relaxation of bound electrons proceeds through the hyperﬁne interaction. Finally, spin relaxation of holes is due to the Elliott-Yafet processes. In bulk III-V or II-VI materials, for holes s p , since the valence spin and orbital states are completely mixed. However, in twodimensional systems, where the heavy and light hole states are split, hole spin relaxation is much less effective.
1. Bulk semiconductors
There is a wealth of useful data on semiconductors.83
83 References for selected semiconductors include the following: p-GaAs (Fishman and Lampel, 1977; Seymour et al., 1981; Aronov et al., 1983; Marushchak et al., 1984; Zerrouati et al., 1988; Sanada et al., 2002); n-GaAs (see III.D.3); p-Alx Ga1 x As (Garbuzov et al., 1971; Clark et al., 1975); p-GaSb (Safarov and Titkov, 1980; Sakharov et al., 1981; Aronov et al., 1983); n-GaSb (Kauschke et al., 1987); n-InSb (Chazalviel, 1975); InAs (Boggess et al., 2000); p-InP (Gorelenok et al., 1986); n-InP (Kauschke et al., 1987); n-GaN (Fanciulli et al., 1993; Beschoten et al., 2001).
ˇ ´ Zutic, Fabian, and Das Sarma: Spintronics: Fundamentals and applications
A comprehensive theoretical investigation of spin dephasing in bulk semiconductors (both p and n types), applied to GaAs, GaSb, InAs, and InSb, has been carried out by Song and Kim (2002) by using the ElliottYafet, D’yakonov-Perel’, and Bir-Aronov-Pikus mechanisms. The calculation uses analytical formulas like Eq. (66), while explicitly evaluating p for different momentum scattering processes at different control parameters (temperature and density), but only for nondegenerate electron systems (Boltzmann statistics) and zero magnetic ﬁeld. The main results are as follows: in n-type III-V semiconductors D’yakonov-Perel’ dominates at T 5 K. At lower T the Elliott-Yafet mechanism becomes relevant. This crossover temperature appears to be quite insensitive to the electron density, being between 1 and 5 K in most investigated III-V semiconductors for donor densities greater than 1014 cm 3 (Song and Kim, 2002). For p-type materials the dominant mechanisms are D’yakonov-Perel’ and Bir-Aronov-Pikus, with the crossover temperature sensitive to the acceptor density. For example, in p-GaAs at room temperature, the D’yakonov-Perel’ mechanism dominates below 1018 cm 3 , while at large densities the Bir-Aronov-Pikus mechanism dominates. In small-gap InSb the D’yakonov-Perel’ mechanism appears to be dominant for all acceptor densities at temperatures above 50 K. The strong disagreement with experiment found at low T (at 5 K, to be speciﬁc) points to our still limited theoretical understanding of spin relaxation in semiconductors. The discrepancy likely arises from neglect of hyperﬁne-interaction effects. Spin-relaxation times as long as 300 ns were recently obtained in bulk, 100-nm-wide GaAs at 4.2 K, placed in the proximity of quantum wells (Dzhioev, Zakharchenya, et al., 2001; Dzhioev, Korenev, Merkulov, et al., 2002). The samples were low doped ( 1014 cm 3 uncompensated donor density), so that optical orientation detected s of electrons bound on donors. At such low donor concentrations the hyperﬁne-interaction mechanism is responsible for spin relaxation. The unusually large s is attributed to the presence of additional conduction electrons in the structure, coming from the barriers separating the sample and the nearest quantum well. The hyperﬁne interaction is then motionally narrowed by the exchange interaction between the donorbound and conduction electrons. Upon depletion of the conduction electrons from the sample by resonant excitations in the quantum well, s decreased to 5 ns (Dzhioev, Korenev, Merkulov, et al., 2002), implying that the effects of the static hyperﬁne ﬁelds on boundelectron spin precessions are not reduced by motional narrowing. Spin relaxation of holes in bulk III-V materials is very fast due to complete mixing of orbital and spin degrees of freedom in the valence band. The Elliott-Yafet mechanism predicts that hole s is similar to hole p ; this is a common assumption when considering hole contribution to spin-polarized transport. Hole spin lifetime in undoped GaAs has been measured by optical orientation and time-resolved spectroscopy (Hilton and Tang,
Rev. Mod. Phys., Vol. 76, No. 2, April 2004
2002). The observed value at room temperature is s 110 fs, consistent with the theoretical assumption. Spin relaxation of conduction electrons in strained III-V crystals was studied experimentally and theoretically by D’yakonov et al. (1986). Spin relaxation under strain is enhanced and becomes anisotropic due to the strain-induced spin splitting of the conduction band, which is linear in k, similarly to the bulk inversion asymmetry in two-dimensional systems (see Sec. III.B.2). It 2 was found that 1/ s , where is the applied stress, and that s is only weakly temperature dependent. Spin relaxation of photoholes in strain crystals has been studied by D’yakonov and Perel’ (1973a), with the conclusion that the hole spin along the strain axis can relax (by the Elliott-Yafet processes) on time scales much longer than in unstrained samples, due to the lifting of heavy and light hole degeneracy. Compared to III-V or II-VI, much less effort has been devoted to investigation of s in bulk Si. The reason is that CESR is thus far the only technique capable of effective detection of spin relaxation in Si. Optical orientation is rather weak (Lampel, 1968) due to the indirect band-gap structure, while robust spin injection in Si is yet to be demonstrated. Spin relaxation in Si is slow due to the presence of inversion symmetry (the D’yakonovPerel’ mechanism is not applicable) and lack of a nuclear moment for the main Si isotope. Earlier experimental studies (Feher and Gere, 1959) were concerned with the hyperﬁne-interaction-dominated spin dephasing in donor states. A comprehensive experimental study of low-doped Si (P donors were present at the levels 7.5 1014 N d 8 1016 cm 3 ), at temperatures 20 T 300 K, was performed by Lepine (1970). Three distinct temperature regimes were observed: (20 T 75 K) Here s decreases with increasing T. The hyperﬁne-interaction mechanism dominates: electrons are bound to the ground donor states, while thermal excitations to higher states and the exchange interaction with conduction electrons motionally narrows the hyperﬁne interaction. (b) (75 T 150 K) In this temperature range s continues to decrease with increasing T, the effect caused by the spin-orbit interaction in the ﬁrst excited donor state being motionally narrowed by thermal motion. (c) (T 150 K) Here 1/ s increases with T, in accord with the Elliott-Yafet mechanism. The observed room-temperature CESR linewidth is about 8 G, corresponding to the electron spin lifetime of 7 ns. (a)
2. Low-dimensional semiconductor structures
The importance of low-dimensional semiconductor systems (quantum wells, wires, and dots) lies in their great ﬂexibility in manipulating charge and, now, also spin properties of the electronic states. Studies of spin relaxation in those systems are driven not only by the need for fundamental understanding of spin relaxation
ˇ ´ Zutic, Fabian, and Das Sarma: Spintronics: Fundamentals and applications
and decoherence, but also by the goal of ﬁnding ways to reduce or otherwise control spin relaxation and coherence in general. For a survey of spin relaxation properties of semiconductor quantum wells, see Sham (1993). Spin relaxation in semiconductor heterostructures is caused by random magnetic ﬁelds originating either from the base material or from the heterostructure itself. All four mechanisms of spin relaxation can be important, depending on the material, doping, and geometry. The difference from the bulk is the localization of the wave function into two, one, or zero dimensions and the appearance of structure-induced random magnetic ﬁelds. Of all the mechanisms, the D’yakonov-Perel’ and hyperﬁne interaction are believed to be most relevant. The most studied systems are GaAs/AlGaAs quantum wells. The observed s varies from nanoseconds to picoseconds, depending on the range of control parameters such as temperature, quantum-well width or conﬁnement energy E 1 , carrier concentration, mobility, magnetic ﬁeld, or bias.84 Spin relaxation has also been investigated in In/GaAs (Paillard et al., 2001; Cortez et al., 2002), in an InAs/ GaSb superlattice (Olesberg et al., 2001), in InGaAs (Guettler et al., 1998), in GaAsSb multiple quantum wells (Hall et al., 1999). II-VI quantum wells, speciﬁcally ZnCdSe, were studied by Kikkawa et al. (1997), who found s 1 ns, weakly dependent on both mobility and temperature, in the range 5 T 270 K. Electron and hole spin dephasing have also been investigated in dilute magnetic semiconductor quantum wells doped with Mn ions (Crooker et al., 1997; Camilleri et al., 2001). Reduction of spin relaxation by inhibiting the BirAronov-Pikus electron-hole exchange interaction through spatially separating the two carriers has been demonstrated in -doped p-GaAs:Be/AlGaAs (Wagner et al., 1993). The observed s was 20 ns at T 10 K, which is indeed unusually large. The exchange interaction was also studied at room temperature, observing an increase of s with bias voltage which increases spatial separation between electrons and holes, reducing the Bir-Aronov-Pikus effects (Gotoh et al., 2000). In the fractional quantum Hall effect regime it was demonstrated (Kuzma et al., 1998) that nonequilibrium spin polarization in GaAs quantum wells can survive for tens of s. Spin lifetime was also found to be enhanced in GaAs quantum wells strained by surface acoustic waves (Sogawa et al., 2001). A theoretical study (Kiselev and
Here is a list of selected references with useful data on s in GaAs/AlGaAs quantum wells: conﬁnement energy dependence has been studied by Tackeuchi et al. (1996); Britton et al. (1998); Ohno, Terauchi, et al. (1999, 2000); Endo et al. (2000); Malinowski et al. (2000); temperature dependence is treated by Wagner et al. (1993); Ohno, Terauchi, et al. (1999, 2000); Malinowski et al. (2000); Adachi et al. (2001); carrier concentration dependence is studied by Sandhu et al. (2001); dependence on mobility is examined by Ohno, Terauchi, et al. (1999); and dependence on magnetic ﬁeld is studied by Zhitomirskii et al. (1993).
Rev. Mod. Phys., Vol. 76, No. 2, April 2004
Kim, 2000) proposed that spin dephasing in 2DEG can be signiﬁcantly suppressed by constraining the system to ﬁnite stripes, several mean free paths wide. Theoretical studies focusing on spin dephasing in III-V and II-VI systems include those of Wu and Metiu (2000); Lau et al. (2001); Wu (2001); Bronold et al. ´ (2002); Lau and Flatte (2002); Wu and KuwataGonokami (2002); Krishnamurthy et al. (2003); Puller et al. (2003). Spin relaxation due to D’yakonov-Perel’ mechanism with bulk inversion asymmetry term in the important case of GaAs/AlGaAs rectangular quantum wells was investigated by Monte Carlo simulations (Bournel et al., 2000) at room temperature, including interface roughness scattering. Nice agreement with experiment was found for s (E 1 ), where E 1 is the conﬁnement energy. Interface roughness becomes important at large values of E 1 , where scattering increases s (see also Sherman, 2003a). Spin relaxation and spin coherence of spin-polarized photoexcited electrons and holes in symmetric p- and n-doped and undoped GaAs/AlGaAs quantum wells was investigated using rate equations (Uenoyama and Sham, 1990a, 1990b). It was shown that in these heterostructures hole spin relaxation proceeds slower than electron-hole recombination. Hole relaxation is found to occur mostly due to acoustic phonon emission. The ratio of the spin-conserving to spin-ﬂip hole relaxation times was found to be 0.46, consistent with the fact that luminescence is polarized even in n-doped quantum wells at times greater than the momentum relaxation time. Similar observations hold for strained bulk GaAs, where hole spin relaxation is also reduced. Spin relaxation of holes in quantum wells was calculated (Ferreira and Bastard, 1991; Bastard and Ferreira, 1992) using the interaction with ionized impurities and s-d exchange in semimagnetic semiconductors. It was shown that size quantization signiﬁcantly reduces spin relaxation of holes, due to the lifting of heavy and light hole degeneracy. The observed spin lifetimes for holes at low temperatures reached up to 1 ns, while at T 50 K in the same samples s got smaller than 5 ps (Baylac et al., 1995). Spin dynamics and spin relaxation of excitons in GaAs (Munoz et al., 1995; Vina et al., 2001) and ZnSe (Kalt et al., 2000) were investigated experimentally and theoretically (Maialle et al., 1993; Sham et al., 1998). Coherent spin dynamics in magnetic semiconductors was considered by Linder and Sham (1998). Spin relaxation in Si heterostructures has been investigated by electron spin resonance in modulation doped Si/SiGe quantum wells. Very-high-mobility (about 105 cm2 V 1 s 1 ) samples with n 3 1011 cm 2 free electrons forming a 2D electron gas show, at T 4.2 K, T 1 up to 30 s (Graeff et al., 1999; Sanderfeld et al., 2000) and T 2 of the order of 100 ns (Graeff et al., 1999), depending on the orientation of B with respect to the quantum-well growth direction. Spin relaxation was attributed to Bychkov-Rashba spin splitting in these asymmetric wells, estimating the corresponding BR in Eq. (88) to be around 1 10 14 eV m (Wilamowski and
1999). At the doping density N d 1 1016 cm 3 of Si donors and T 5 K.3As with N d 1 1016 cm 3 at 4. Rev. Awschalom. Figure 17 summarizes these ﬁndings. Korenev. Spin-orbit coupling in symmetric Si/SiGe quantum wells has been studied theoretically by Sherman (2003b). Example: Spin relaxation in GaAs FIG. single-spin decoherence times sc c due to the exchange interaction between electron spins on neighboring donors. Kavokin. 2002. parameter-free theoretical estimates with labels indicating the dominant spin relaxation mechanisms. Fabian. a. solid lines. albeit complex. 2004) have conﬁrmed the dominant role of the D’yakonovPerel’ spin relaxation mechanism. Vol. the timeresolved Faraday rotation data of Awschalom (2001) and Kikkawa and Awschalom (1998). 2003b). reaching its ﬁrst maximum (180 ns) at around 3 1015 cm 3 . At greater and smaller doping densities. 1997). Both optical orientation and time-resolved Faraday rotation spectroscopy have been used to measure s in bulk n-GaAs.B.2 K. as the mechanisms (ElliottYafet and D’yakonov-Perel’) effective for conduction electrons are ineffective for states localized in quantum dots (Khaetskii and Nazarov.2 ns (Eki- p-GaAs is extensively discussed by Meier and Zakharchenya (1984). April 2004 85 mov and Safarov. dotted line. Recently a lower bound. that similar to electrons bound on donors.5 T was oriented parallel to the plane of the heterostructure. 2002). acting with longitudinal phonon and impurity scattering. (2002). and Das Sarma: Spintronics: Fundamentals and applications 363 Jantsch... while s 100 ps at 4 T) (Gupta et al. et al. 3. III. 2001). A recent comprehensive study of s based on optical orientation revealed a nice. the dominant mechanism is a hyperﬁne-interaction process (Khaetskii et al. Recent studies (Wilamowski and Jantsch. Faraday rotation studies (Kikkawa and Awschalom. the optical orientation data of Dzhioev. Unfortunately. using a simple model of the exchange coupling between donor states. solid circles. 2003a. Khaetskii. The magnetic ﬁeld of 7. Si/Ge heterostructures may have enhanced rates of spin relaxation due to the leakage of the electron wave function to Ge.F). on T 1 of 50 s has been measured at 20 mK in a one-electron quantum dot deﬁned in a 2DEG GaAs/AlGaAs heterostructure by Hanson. Witkamp. From Dzhioev.. experiments on CdSe quantum dots (of diameter 22–80 Å) show strong inhomogeneous dephasing ( s 3 ns at B 0. leading to microsecond spin relaxation times. It is believed. et al. Wilamowski et al.. 2001) found even longer spin lifetimes. Merkulov et al. 2002). 2002). 1971).2. the observed s was 130 ns at zero magnetic ﬁeld. We review recent experimental results on spin relaxation in bulk n-GaAs85 and GaAs-based lowdimensional systems.7Al0. No. however. spin-relaxation time is signiﬁcantly reduced: for both a nominally undoped sample and for N d 1 1018 cm 3 . dashed vertical line. 2001.. 17. et al.. where a sudden increase brings s to an- . 2. limited by the signal-to-noise ratio. 2000.a for a brief discussion of the inﬂuence of magnetic ﬁeld on s ). Mod. IV. 2001. The spinrelaxation time rises with increasing N d at small doping levels. Phys..ˇ ´ Zutic. the metal-insulator transition at N dc 2 1016 cm 3 . 1998. who solved numerically kinetic equations in the presence of magnetic ﬁeld. 76. it was found that s 1. Only the D’yakonov-Perel’ mechanism was considered. the observed bound sufﬁces for performing elementary quantum gates (see Sec. While the actual value of T 1 may be orders of magnitude larger. Zakharchenya. s 0. A much larger spin lifetime was found by optical orientation on n-GaAs (Dzhioev et al. A comprehensive theoretical investigation of s in bulk n-GaAs is reported by Wu and Ning (2000) and Wu (2001). picture of spin relaxation in bulk n-GaAs over a large range of doping levels (Dzhioev. Bulk n-GaAs The importance of GaAs for spintronics and quantum computing applications has been recently underlined by the discovery of rather long spin-relaxation times (of the order of 100 ns) in n-doped samples. a ﬁt to the experimental data on the exchange correlation time (triangles) c . In the earliest observations of optical spin orientation of electrons. 2002. Semenov and Kim.. s then decreases until N d N dc 2 1016 m 3 . In quantum dots the relevant spin relaxation mechanism is still being debated. 2002. et al. as well as by the development of experimental techniques to manipulate spin precession in this semiconductor in a coherent manner (Awschalom. in n-Ga0. 2002.2 ns. Oestreich et al. de Sousa and Das Sarma.. The spin dephasing is argued to be strongly suppressed by cyclotron motion in highmobility samples (see Sec. which is heavier than Si and has greater spin-orbit interaction. 2002. open triangles. (2003). where for N d 1 1015 cm 3 the observed s was 42 ns. Kavokin. masking the intrinsic spin dephasing processes. (Color in online edition) Spin relaxation time in n-GaAs as a function of donor density N d (labeled as n D here) at low temperatures: empty symbols.
where the exchange correlation time c depends on B through magnetic orbital effects on the bound electron wave functions (magnetic conﬁnement reduces the extent of the bound orbital. The decrease of s with increasing T above 50 K has been found to be consistent with the D’yakonovPerel’ mechanism (Kikkawa and Awschalom. III. 3 1015 cm 3 N d N dc . Measured magnetic-ﬁeld dependence of the spin * dephasing time (here denoted as T 2 to indicate the likely presence of inhomogeneous broadening. and Das Sarma: Spintronics: Fundamentals and applications other maximum. The spin-relaxation time increases with B in the metallic regime. with increasing overlap between donor states). The behavior of s in the intermediate regime. 17) the donor states start to overlap and form an impurity conduction band—electronic states delocalize and the sample becomes metallic. The above picture is valid at T 5 K. The sample is insulating at low T and nondegenerate at high T (T 50 K. However. appears as a result of spin-orbit coupling in semiconductors lacking inversion symmetry. Zakharchenya. dropping from 130 ns at 5 K to less than 1 ns at 150 K. Beyond the critical density N dc 2 1016 cm 3 (the dashed vertical line in Fig. at B 0 and B 4 T. 1998). S1 S2 (Dzyaloshinskii. thus reducing the exchange integrals between neighboring donor states). taking s T 3 in Eq. It is thus likely that s FIG. the sample held at B 4 T shows at ﬁrst a rapid increase with increasing T. where donors are ionized. Indeed. B 2 c(B). and then a decrease at T 50 K. Conductivity is due to hopping between donor states. Eq. 19. 2002) agree well with the data. These anisotropies are ampliﬁed by increasing B and motionally narrowed by the exchange interaction. which increases with increasing N d (that is. with the correlation time c provided by the usual S1 •S2 . Gorkov and Krotov. No. At still higher doping levels s decreases strongly with increasing doping. which should be generally present for bound electrons in systems lacking inversion symmetry (such as III-V and II-VI). 1998. Figure 19 plots s (T) for an insulating sample with N d 1 1016 cm 3 at B 0 and B 4 T. would give s N d 4/3 . (74). 1998. both the temperature and the magnetic-ﬁeld dependences of spin relaxation in bulk n-GaAs have been studied by Kikkawa and Awschalom (1998). Doping levels. This new mechanism of spin relaxation. while extracting the tempera- 86 The anisotropic exchange interaction of the DzyaloshinskiiMoriya form.A. which is observed in Fig. see Sec. was proposed by Kavokin (2001) to be due to motional narrowing of the antisymmetric exchange interaction86 between bound electrons. where s decreases with increasing N d . although still being investigated (Kavokin. (66).. April 2004 . 1958.364 ˇ ´ Zutic. no satisfactory quantitative explanation for s (B) in insulating samples exists. Moriya. For the zeroﬁeld data the initial decrease of s with B is very rapid. Measured temperature dependence of the spin dephasing time for bulk n-GaAs doped with N d 1 1016 cm 3 Si donors. where isolated shallow donors are not normally ionized. In contrast. 2003). are indicated. 17. considering that 2/3 E F N d and assuming the Brooks-Herring formula for 3/2 the impurity scattering (1/ p N d /E F ). Korenev. s in the insulating samples decreases with increasing B. Adapted from Kikkawa and Awschalom. and the sample is a Mott insulator at small dopings. 18. 76. Figure 18 shows s (B) for samples with varying doping levels at T 5 K. appears to give a satisfactory explanation for the experimental data.. The Elliot-Yafet mechanism. 2. Rev. 2002b. Equation (73) for degenerate electrons explains the observed data rather well. Bound electrons are more susceptible to g-factor anisotropies (due to distribution of electron energies over donor states) and local magnetic-ﬁeld variations (due to hyperﬁne interactions). et al.1) for bulk n-GaAs at 5 K. one obtains s 2 1/N d . Figure 17 shows that it is the rather narrow window around the metal-toinsulator transition where the largest s are found. direct exchange. In addition to the doping dependence of s . reaching 150 ns. Adapted from Kikkawa and Awschalom. The data on the insulating side are consistent with the hyperﬁneinteraction mechanism: the precession due to local random magnetic ﬁelds from the nuclear moments is motionally narrowed by the exchange interaction. a behavior qualitatively consistent with the predictions of the D’yakonov-Perel’ mechanism. FIG. 1960). Vol. Phys. However. The theoretical estimates (Dzhioev. Mod. At N d N dc the D’yakonov-Perel’ mechanism dominates. varying from insulating (N d N dc 2 1016 cm 3 ) to metallic (N d N dc ). assuming 4 MeV for the donor binding energy). Fabian.
with x 0. setting the length scale for the spin dephasing. ¨ Hagele et al. Recent experiments on time-resolved Kerr rotation (Kimel et al. leading to the correlation time c . the measured T * depends on the 2 measurement ‘‘history. GaAs-based quantum wells We discuss selected experimental results on spin relaxation in GaAs/Alx Ga1 x As quantum wells. In these structures the excitonic effects . electrons are localized. since nuclear polarization typically takes minutes to develop.8. 19 is less obvious. Kikkawa and Awschalom obtained the spin diffusion (responsible for spreading) and electronic diffusion (drift by electric ﬁeld) coefﬁcients. the observation of the drift of precessing electron spins—transverse spin drift—in GaAs with N d 1 1016 cm 3 . What can be expected for s at room temperature? The answer will certainly depend on N d . The D’yakonov-Perel’ mechanism for conduction electrons was also observed in p-GaAs in the regime of nondegenerate hole densities N a 1017 cm 3 at temperatures above 100 K (Aronov et al. April 2004 87 interested in the spin lifetime of single (or a few) electrons. Mod. The temperature dependence of s for samples with N d N dc has been reported (Kikkawa and Awschalom. While Fig. The wells. For spintronic applications to make use of the large s observed in bulk n-GaAs. indicating.1 ns. where the data are still sparse. By directly analyzing the spreading and drifting of the electron spin packets in time. b. The inhomogeneous magnetic ﬁeld due to polarized nuclei causes inhomogeneous broadening of the electronic s . 18 and 19. The * measured spin dephasing time is indeed T 2 . The technique should give consistent results at small ﬁelds and large temperatures. 2001). c is smaller than 0. 2001) suggest that 5 ps s 10 ps for undoped GaAs and 15 ps s 35 ps for a heavily doped n-GaAs with N d 2 1018 cm 3 . The reason for this difference (Kikkawa. over 100 m in moderate electric ﬁelds (tens of V/cm) at T 1. as well as in heavily doped samples where the nuclear ﬁelds are motionally narrowed by the itinerant nature of electrons. Unfortunately. 2000). The result is shown in Fig. presenting the temperature and conﬁnement energy dependence of s . For a speciﬁc model of spin relaxation in bound electron states c was extracted experimentally by Dzhioev. where charge is transported via hopping.. 1983). Rev. For an informal recent review of s in n-GaAs.. after the contribution from the Bir-Aronov-Pikus mechanism was subtracted using a theoretical prediction.. for these degenerateelectron densities. 17. These results are difﬁcult to interpret. that p (T) is only weakly dependent on T.. The increase with increasing T of s at 4 T invokes a picture of motional narrowing in which the correlation time decreases with increasing T much faster than the dispersion of local Larmor frequencies. This was followed by a remarkable result of Kikkawa and Awschalom (1999). since the sample is just below the metal-to-insulator transition. where one is There is a discrepancy in the data presented in Figs. show different T * 2 under otherwise equivalent conditions. bound to impurities or conﬁned to quantum dots. s is only about 1 ns in Fig.F). and Das Sarma: Spintronics: Fundamentals and applications 365 ture dependence of p from the measurement of mobility.87 Similar behavior of s (T) in insulating samples was found in GaN (Beschoten et al. We do not know of a satisfactory quantitative explanation for these experimental results.. so that according to the D’yakonov-Perel’ mechanism s T 2. Furthermore. were grown on a single wafer to minimize sample-to-sample variations when comparing different wells. 2000). At low temperatures and large magnetic ﬁelds.6 K. et al. Fabian. the larger the s . (1998) observed the transport of a spin population—longitudinal spin drift—in i-GaAs over a length scale greater than 4 m in electric ﬁelds up to 6 kV/cm and at low temperatures. Figure 20 plots the temperature dependence of 1/ s in the interval 90 T 300 K for quantum wells of widths L ranging from 6 to 20 nm (Malinowski et al. 18 reports s 3 ns at 5 K and 4 T. but the obvious trend is the smaller the c . The reported interface roughness was less than the exciton Bohr radius of 13 nm. Closely related to spin relaxation is spin diffusion. Zakharchenya. Phys. 18 and 19. The two times c and s differ by orders of magnitude. Although this may restrict the design of room-temperature spintronic devices.2. the useful time for spin quantum computing would be extracted in the limit of very small dopings. The nuclear polarization effect is also part of the reason why the T 2 (T) at 4 T sharply deviates from that at zero ﬁeld at small T. 1998) to be very weak. (2002) by detecting the changes in the spin polarization due to longitudinal magnetic ﬁelds. The randomizing processes are spin ﬂips due to direct exchange. No.ˇ ´ Zutic. nuclear polarization develops via the Overhauser effect inhomogeneously throughout the electron spin excitation region. 76. Ensemble spin dephasing seen for insulating GaAs samples appears to be due to motional narrowing of the hyperﬁne interaction.35 and orientation along . rather than the intrinsic T 2 . From the observed mobility it was found that p (T) T 0. so in order to explain the experimental data the theory should include ionization of donors. Investigations of this type in even smaller doping limits may prove important for understanding singlespin coherence. IV. Korenev. which can be taken as a measure for the lifetime sc of the individual spins. 2. How close is s to the individual spin lifetime sc ? There is no clear answer yet. but they suggest that spin diffusion is strongly enhanced through the exchange interaction. such a regime seems acceptable for spinbased quantum computing (see Sec. reported in Figs. At low T. Take the N d 1 1016 cm 3 sample.’’ This is the reason why two different measurements. which is indeed consistent with the experimental data. For the doping levels where s is greater than 100 ns. 19. Extracting this lifetime from the experiment is not easy. Vol. The origin of s (T) below 50 K in Fig. see Kavokin (2002a). 2003) turns out to be electronically induced nuclear polarization (Kikkawa and Awschalom. It was found that the former is about 20 times as large as the latter. one is limited to both very small temperatures and small doping levels.
Sandhu et al. From Malinowski et al. The D’yakonov-Perel’ mechanism predicts. From the observed hightemperature bulk s (T) (at low temperatures s is affected by Bir-Aronov-Pikus processes) one can estimate p 1/T. The solid line is a quadratic ﬁt. No. 2000. Kimel et al. approximating the bulk data (cf. Phys.. are likely to inﬂuence future spintronic research and possible applications. the s T 2 dependence. 2000). and 1/ s TE 2 p from Eq. dominate at T 50 K (with the reported s 50 ps). At low T. Vol. in which spin precession about the intrinsic magnetic ﬁelds (here induced by bulk inversion asymmetry) increases as E 2 with increasing conﬁnement. Spin-polarized transport 1.. Similarly to bulk GaAs. 2001. As Fig. where 15 ps s 35 ps was found for a heavily doped n-GaAs). 76. p should be similar in bulk and lowdimensional structures. The increase is consistent with the 1/ s T 2 behavior. while not yet commercially viable. For the well with L 15 nm. spin relaxation in GaAs quantum wells was found to be reduced at carrier concentrations close to the metal-to-insulator transition (n 5 1010 cm 2 . Measured temperature dependence of the conduction-electron spin relaxation rate 1/ s in GaAs/AlGaAs quantum wells of varying widths: dashed curve. the condition being that thermal energy is greater than the subband separation). FIG. SPINTRONIC DEVICES AND APPLICATIONS In this section we focus primarily on the physical principles and materials issues for various device schemes. and Das Sarma: Spintronics: Fundamentals and applications FIG. The thickest well increases with the same power law. The ob1 served data in Fig. when one realizes that conﬁnement energy E 1 2 is k n .. (74)] in the bulk and wide quantum dots. over the whole temperature range. in addition to the Bir-AronovPikus mechanism. so the data presented are for free electrons. In order to make a reliable comparison with theoretical predictions (the expected mechanism is that of D’yakonov-Perel’ in two-dimensional systems). for the bulk in1 version asymmetry after thermal averaging (E k →k B T). A.. for the nondegenerate-electron densities employed in the experiment. The Bir-Aronov-Pikus and Elliot-Yafet mechanisms were found not to be relevant Rev. 2000). s depends rather weakly on T for the narrow wells with L 10 nm. solid line. while the exciton ionization is complete roughly at T 90 K. s will deviate from that in the bulk due to impurity scattering. 1/ s T 2 . 2001). April 2004 to the observed data (Malinowski et al. which. and with the quadratic dependence for the wide wells. 2000. From Malinowski et al. IV. Spin relaxation was studied using pump-probe optical orientation spectroscopy with a 2-ps time resolution and the typical excitation intensity/pulse of 1010 cm 2 . 2000).. F/I/S tunneling Experiments reviewed by Tedrow and Meservey (1994) in ferromagnet/insulator/superconductor (F/I/S) . Fabian. after being approximately constant (or somewhat decreasing) as T increases to about 200 K. 20 shows. Conﬁnement strongly enhances spin relaxation. one needs to know the behavior of p (T). Mod. which is consistent with the constant s for the narrow wells. 20.. The data are at room temperature.. The downturn for the highest-E 1 well (of width 3 nm) is most likely due to the increased importance of interface roughness at such small widths (Malinowski et al. Figure 21 shows the dependence of 1/ s on the experimentally determined conﬁnement energy E 1 for a variety of quantum wells on the same wafer (Malinowski et al.366 ˇ ´ Zutic. 1984). that 1/ s T 3 p [see Eq. showing behavior consistent with the D’yakonov-Peral’ mechanism. This is consistent with the D’yakonov-Perel’ mechanism for two-dimensional systems. Measured room-temperature dependence of 1/ s on the conﬁnement energy E 1 for GaAs/AlGaAs quantum wells. 21 are consistent with the theoretical prediction. to about 10 ps in most conﬁned structures. (79). s increases with increasing T at greater temperatures. 2. The spin-relaxation time varies from somewhat less than 100 ps for wide quantum wells. When one assumes that momentum relaxation in these elevated temperatures is due to scattering by phonons.. 21. data for a lowdoped (N a 4 1016 cm 3 ) bulk p-GaAs (Meier and Zakharchenya.
1994) and a common choice for I/S regions. 22. should serve as a caution for studies of novel. and q is the proton charge. 2002) P→P G G N↑ G N↓ / G N↑ G N↓ . Wiesendanger. (101) Here we focus on a conventional s-wave superconductor with no angular dependence in . Worledge and Geballe. T is Here the temperature. (b) normalized spin-resolved conductance (dashed lines) and the total conductance (solid line) at ﬁnite temperature. With the further simpliﬁcations of spin independence and a constant tunneling matrix element (Tedrow and Meservey.9 was measured (Parker et al. Parkin et al. 1990. Fulde. 2000a. discussed here in the context of F/I/S tunneling. Monsma and Parkin. measured polarization is not an intrinsic property of the F region and could depend on interfacial properties and the choice of insulating barrier. The factors (1 P)/2 represent the difference in tunneling probability between ↑ and ↓ electrons. 2002).88 The BCS density of states is split in a magnetic ﬁeld H. Only two of the four peaks sketched in Fig. Eq. Mod. where V is the applied bias.5 T. These experiments also stimulated more recent imaging techniques based on the spin-polarized scanning tunneling microscope (see Johnson and Clarke. F/I/S tunneling conductance is shown in Fig. nearly complete spin polarization P G 0.ˇ ´ Zutic. Fabian. Rev. Theoretical analyses (Maki.. has demonstrated that the current will remain spin polarized after tunneling through an insulator. 22..C) can be neglected. 2000b. Usually P can be identiﬁed as (Worledge and Geballe. 1960) act as pair breakers and reduce the value of . for ↑ (↓) spin parallel (antiparallel) to the ﬁeld. This conductance can be expressed by generalizing analysis of Giaever and Megerle (1961) as 1/k B T. k B is the Boltzmann constant. junctions have established a sensitive technique for measuring the spin polarization P of magnetic thin ﬁlms and at the same time. more exotic. G(V) (dI/dV) S /(dI/dV) N G ↑ (V) G ↓ (V). would require a full calculation of spin-dependent tunneling. and Das Sarma: Spintronics: Fundamentals and applications 367 G V ˜ 1 P NS E 2 4 cosh2 ˜ 1 P NS E 2 4 cosh2 H dE E qV /2 H dE . spintronic materials. 1973) using many-body techniques show that the spin-orbit scattering would smear the Zeeman-split density of states. applied parallel to the interface. where B is the Bohr magneton. 1964. Vol. In CrO2 /I/S tunnel junctions. The assumption of spin-conserving tunneling can be generalized (Tedrow and Meservey. including the appropriate boundary conditions and a detailed understanding of the interface properties. E→E B H. 22. Correspondingly. The tunneling conductance is normalized with respect to its normal-state value for an F/I/N junction. 1990. III. G(V) G( V)]. Challenges in quantifying P. (Color in online edition) Ferromagnet/insulator/ superconductor tunneling in an applied magnetic ﬁled: (a) Zeeman splitting of the BCS density of states as a function of applied bias. The degree of spin polarization is important for many applications such as determining the magnitude of tunneling magnetoresistance (TMR) in magnetic tunnel junctions (MTJ) [recall Eq. due to a shift in quasiparticle energy. it is customary to make some simpliﬁcations. while the magnetic impurities (Abrikosov and Gorkov. Maekawa et al. a good approximation for Al2 O3 /Al (Tedrow and Meservey. For each spin the normalized BCS density of states is 2 ˜ NS (E) Re( E /2 E 2 ). in terms of materials parameters. 1994. Spin polarization P of the F electrode can be deduced (Tedrow and Meservey. (2)]. (2004) have shown that by replacing aluminum oxide (a typical choice for an insulating region) with magnesium oxide. E qV /2 (99) FIG. In an actual MTJ. 2000a) to extract P in the presence of spin-orbit and spin-ﬂip scattering.. 22(b) [for P 0. (100) the spin polarization of the normal-state conductance (proportional to the weighted average of the density of states in F and S and the square of the tunneling matrix element). have been observed. extraordinarily large values of TMR ( 200% at room temperature) can be achieved even with conventional ferromagnetic (CoFe) electrodes. Different probes for spin polarization generally can measure signiﬁcantly different values even in experiments performed on the same homogeneous sample. Phys. 1973. Bruno and Schwartz. 1971b. 1994). 1971a. 2000a. eventually merging the four peaks into two. 76.. No. Wiesendanger et al. Neglecting the spin-orbit scatter- . where for simplicity we assume that the spin-orbit and spin-ﬂip scattering (see Sec. 1998) with the ultimate goal of imaging spin conﬁgurations down to the atomic level. April 2004 88 the spin polarization of the tunneling density of states in the F region at the Fermi level. (100) can be expressed as P→P N NF↑ NF↓ / NF↑ NF↓ . one can signiﬁcantly increase the spin polarization in F/I/S junctions. even when F is a simple ferromagnetic metal. While a rigorous determination of P. 1994) from the asymmetry of the conductance amplitudes at the four peaks in Fig. indicating no features due to the minority spin up to H 2. where E is the quasiparticle excitation energy and the superconducting gap. where ↑ is the electron spin with the magnetic moment parallel to the applied ﬁeld (majority electrons in F). and a review. 2.
1995. This is explained by the two-particle process of Andreev reﬂection (discussed further in Sec. 1995). Spin-dependent tunneling was also studied using a high-temperature superconducting electrode as a detector of spin polarization (Vas’ko et al. have all been shown to affect the measured P directly. Moodera et al. Gadzuk (1969)]. 2. 1994. Tanaka and Kashiwaya. Zhu et al.. a simpler N/I/S case is reviewed by Hu. Fabian. 1994).. no spin-polarized tunneling was observed using an a-Ge barrier (Gibson and Meservey. would be ineffective.A. magnetic read heads. in addition to the usual quasiparticle tunneling. Early theoretical work addressed this apparent difference. 2000). as well as identiﬁcation of the relevant spin polarization remain to be fully understood. By consid- . contributes to the I-V characteristics of a F/I/S junction (Hu and Yan. The suppression of a zero-bias conductance peak. Perhaps this is also an apt description for the present state of experiment on spinpolarized tunneling between two ferromagnetic regions. Even for MTJ’s with standard ferromagnetic metals.. 2000a).368 ˇ ´ Zutic. 1998 and Kashiwaya and Tanaka. 1975). 1999. 1998. F/I/S tunneling was also studied using amorphous Si (a-Si) and Ge (a-Ge) as a barrier. for example. for spin-unpolarized tunneling. While with a-Si some spin polarization was detected (Meservey et al. 2 and 3). Mod. Consequently P G could even have an opposite sign from P N—which. 2000b).3).... 2001). 2004). while nearly localized electrons. 1998). Similar arguments.. From the assumed parabolic dispersion of the spin subbands with ﬁxed spin splitting. While this can signiﬁcantly extend the temperature range in the tunneling experiments. Duke (1969) concludes that (with only a few exceptions) the study of tunneling is an art and not a science. was recently used to detect spin injection into a high-temperature superconductor (Ngai et al. Vol. F/I/S conductance measurements (Tedrow and Meservey. 1998. ˇ ´ Kashiwaya et al. A resurgence in interest in the study of MTJ’s. However. Co. 1999. 76. There are also several important differences between studies using high-temperature and conventional low-temperature superconductors. 2001). 2. and details of the interfacial properties. in Fe. Mazin (1999) showed the importance of the tunneling matrix elements. for inequivalent density-of-states contributions to G(V). The superconducting pairing symmetry no longer yields an isotropic energy gap. 1976). a lack of understanding of high-temperature superconductors makes such structures more a test ground for fundamental physics than a quantitative tool for quantitatively determining P. was spurred by the observation of large room-temperature TMR (Miyazaki and Tezuka. No. A sign change of the pair potential can result in G(V 0) 0 for T→0 even for a strong tunneling barrier and give rise to a zero-bias conductance peak (Hu. follow` ing a hiatus after the early work by Julliere (1975) and ¨ Maekawa and Gafvert (1982). the bias and the temperature dependence of the TMR. Stearns (1977) suggested that only itinerant. 1982). freelike electrons will contribute to tunneling. with a large effective mass. would be measured by spin-resolved photoemission. due to the indirect gap.. respectively. which have different Fermi velocities for different bands (see also Yusof et al. were generalized to more complex electronic structure. The Tedrow-Meservey technique is also considered as a probe to detect spin injection in Si. Chen et al.. The difference between bulk and the surface densities of states of the ferromagnet (probed in tunneling measurements. where optical methods. 1994) have revealed positive P—the dominant contribution of majority spin electrons for different ferromagnetic ﬁlms (for example.. Ni.. With a few exceptions (Worledge and Geballe. 1995. contribute to the total density of states but not to G(V) [see also Hertz and Aoi (1973) and. that P is not directly related to the magnetic moment (Meservey et al. giving positive P→P k (k F↑ k F↓ )/(k F↑ k F↓ ). which. and even for the BCS-like picture the density of states should be accordingly modiﬁed. 1999. F/I/F tunneling For a list of references see Tedrow and Meservey (1973. Rev. IV. electronic-structure calculations typically give NF↑ NF↓ and P N 0 for Ni and Co NF↑ /NF↓ 1/10 (Butler et al. 1999). 1994.. however. in the context of tunneling in a high-temperature superconductor). This discovery opened the possibility of using MTJ’s for fundamental studies of surface magnetism and room-temperature spin polarization in various ferromagnetic electrodes as well as suggesting applications such as highly sensitive magnetic-ﬁeld sensors. and Gd).. Stearns related the measured polarization to the magnetic moment. Good agreement between tunneling data and electronic structure calculations was illustrated by the example of Nix Fe1 x (Nadgorny et al. in contrast to the ﬁrst reports of tunneling magnetoresistance ` (Julliere. 1985). the spin polarization of the projections of Fermi wave vectors perpendicular to the interface. which only differ in the middle layer’s being an insulator and a metal. 1999. It is instructive to notice the similarity between the schematic geometry and the direction of current ﬂow in an MTJ and that in CPP giant magnetoresistance (recall Figs.. 2000). and Das Sarma: Spintronics: Fundamentals and applications ing was shown to lead to the extraction of higher P values (Tedrow and Meservey. showing. In a brief review of current ﬁndings we intend to identify questions that could arise as new materials for MTJ’s are being considered. 2000.89 and efforts to understand precisely what is being experimentally measured have continued. Zutic and Valls. 2000b). the choice of tunneling barrier (De Teresa et al. April 2004 89 In the preface to a now classic reference on spinunpolarized tunneling in solids. and nonvolatile magnetic memory applications. measured by a scanning tunneling microscope. Monsma and Parkin. 2000). Wei et al. Phys.. which can change over time (Monsma and Parkin. Oleinik et al.
A study of a similar geometry using a Boltzmann-like approach shows (Chui. 2002). Bowen et al. Rev.ˇ ´ Zutic.. with the spinconserving tunneling across the barrier. 1984) Related applications are usually in a diffusive regime..3. 2000.3MnO3 (Co/STO/LSMO. T→0. For novel materials. Hu and Suzuki. These ﬁndings. 2002). 1998a. for both ↑↑ and ↑↓ orientations. as deﬁned by Stearns (1977). Appelbaum. relevant to epitaxially grown MTJ’s (Mathon and Umerski.. near the F/I interfaces. and suppress the TMR (Zhang et al. 1997). Fabian. 1966. (2) but with the redeﬁned polarization P→P k 2 k F↑ k F↓ / 2 k F↑ k F↓ . Tsymbal and Pettifor. as expected from P 1. 1997. This effectively leads to the two-current model proposed by Mott (1936a) and also applied to CPP giant magnetoresonance geometries (Valet and Fert. 2002). in which electronic structure calculations and an understanding of the interfacial proper` ties are not available. Magnons were observed (Tsui et al. was used to explain the large negative TMR ( 50% at 5 K). were applied to ﬁt the decay of the TMR with temperature (Inoue and Makeawa. Hot electrons localized at F/I interfaces were predicted to create magnons. Mathon and Umerski. a common choice for the I region with metallic ferromagnets. April 2004 90 A formally analogous problem was considered by Grifﬁn and Demers (1971) in an N/I/S system where the twocomponent wave functions represented electronlike and holelike quasiparticles rather then the two-spin projections. Miyazaki.A. Variation of the density of states [inferred from the spin-resolved photoemission data (Park et al. Vol.. 91 For its limitations and extensions see comprehensive reviews by Moodera and Mathon (1999) and Moodera et al. No. 1997) that the spin splitting of electrochemical potentials persists in the F region all the way to the F/I interface. Slonczewski (1989) considered F/I/F as a single quantummechanical system in a free-electron picture. IV. 1970) to the spin-polarized case (Xiang et al. with P 0 and TMR 0. 1997. together with Julliere’s model. 2002). LeClair et al. 1995) as the cause of decreasing TMR with T by spin-ﬂip scattering. The bias dependence of the TMR was also attributed to the density of states by extending the model of a trapezoidal tunneling barrier (Brinkman et al. 1993.92 The resulting TMR can be expressed as in Eq. is also P N (in a free-electron picture) and i is the usual imaginary wave vector through a square barrier. Gijs and Bauer. 1998b)] within the range of applied bias in MTJ’s of Co/SrTiO3 /La0. which considered both spin-dependent and spin-ﬂip scattering. 2. 76. including determining precisely which spin polarization is relevant and the related issue of reconciling the (typically positive) sign of the observed TMR with the electronic structure (Bratkovsky. 1997.95. 2002).. which would even change sign for positive bias (raising the Co Fermi level above the corresponding one of LSMO). a nonmagnetic semiconductor used as a tunneling barrier (Kreuzer et al.2 0). III-V ferromagnetic semiconductors (Chun et al. 1971) in an antiferromagnetic NiO barrier in singlecrystal Ni/NiO/Pb tunnel junctions and were suggested (Moodera et al. It was related to the decrease of the surface magnetization (Pierce et al.. The two F regions are treated as uncoupled. 1997) that. and resonant tunneling in F/I/N/F junctions (Yuasa et al. 1997) in a tight-binding representation. ` In an approach complementary to Julliere’s.. Julliere’s formula still provides useful insights. MacLaren et al. ` Julliere (1975) modiﬁed Eq.. 1982. with the assumption that the component of the wave vector parallel to the interface k is not conserved (incoherent tunneling). The values for P extracted from F/I/S measurements are in good agreement with the observed TMR values (typically positive.. 1997). G(T) G(0) is proportional to T ln T. Jansen and Moodera. 2002).. Oleinik et al. 2003). 1999).7Sr0... (1999). or collective spin excitations. 2000. Such a loss of coherence is good approximation for simply capturing the effects of disorder for amorphous Al2 O3 . 2002). De Ter` esa et al. coherent tunneling was assumed. Pierce and Celotta. 1999). the Julliere model for the tunneling magnetoresistance has continued to be used for interpreting the spin polarization in various MTJ’s. 92 . implying ↑ ↓ and an additional voltage dependence of the TMR. with conserved k . Recent examples include F regions made of manganite perovskites displaying colossal magnetoresistance (CMR. and applied it to study F/I/F tunneling.. 1969) showed that the presence of magnetic impurities in the tunneling barrier produces temperature-dependent conductance—referred to as zero-bias anomalies. When matching the two-component wave functions at interfaces. see Sec.. Mod. magnetite (Fe3 O4 . (1998).. Using an s-d exchange (between itinerant s and nearly localized d electrons) Hamiltonian. The decay of TMR with temperature can be attributed to several causes. (100) in the limit V→0.. Phys. A different temperature dependence of TMR was suggested by Moodera et al. A quantitative understanding of MTJ’s faces challenges similar to those discussed for F/I/S tunneling. ` Julliere’s result can be obtained as a limiting case from a more general Kubo/Landauer approach (Mathon and Umerski. (102) where P k . 1999). De` spite its simplicity. 1966. it was shown (Zhang et al.. 1997. for a detailed discussion and a review of related experimental results see Duke. at V→0. Co/carbon nanotube/Co MTJ (Tsukagoshi et al. and Das Sarma: Spintronics: Fundamentals and applications 369 ering the limit of ballistic transport in CPP giant magnetoresistance90 it is possible to give a uniﬁed picture of both TMR and CPP giant magnetoresonance by varying the strength of the hopping integrals (Mathon and Umerski. 2001) and the I region was modeled by a square barrier. Through the dependence of on V the resulting polarization in Slonczewski’s model can change sign. Early theoretical work on N/I/N tunneling (Anderson.. Julliere’s formula91 does not provide an explicit TMR dependence on bias and temperature. ` However. suggesting P N 0. 1999.
Shang et al. Mod. 1996)... the observed TMR in Fe/ZnSe/Fe0. II. 1999) large TMR (up to 1000%). 1967. Loraine et al. Moodera et al.. 2001). 2001). For an epitaxially grown Fe/ ZnSe/Fe MTJ.. electrically the spin injection in GaAs quantum wells (Mattana et al.0%. increasing with ZnSe thickness. 2003). double MTJ’s were also shown to give similar TMR values and were used to determine FIG. a minor loop. Tunneling magnetoresistance is nonmonotonic with thickness in AlAs (with the peak at 1. 50 nm) tunnel junction of 200 m in diameter. with the decrease of TMR with T expected from the spin-wave excitations (MacDonald et al.. 1982.. (b) TMR curves of a Ga1 x Mnx As (x 4. The sample size is 3 3 mm2 . The important implications for MTJ’s are the possibility of large overestimates in the TMR amplitude (Moodera and Mathon. attributed to magnons. Bold solid curve. Kreuzer et al.93 To improve the performance of MTJ’s it is desirable to reduce the junction resistance..Mn)As/AlAs/(Ga. 1996. seen also in N/I/N junctions. as compared to the conventional all-metal MTJ’s. 76. Vol.3%. 1999) and a desirable hysteresis effect—at 95 Nonuniform tunneling current has been studied in nonmagnetic junctions (Pederson and Vernon. 2004). and other metallic structures involving Si. However. Ge. GaAs. All-semiconductor magnetic tunnel junction: (a) magnetization of Ga1 x Mnx As (x 4.. 50 nm)/AlAs (3 nm)/ Ga1 x Mnx As (x 3. 23. A lower barrier in F/Sm/F MTJ’s can have important implications in determining the actual values of TMR. 50 nm) trilayer measured by a SQUID at 8 K.. Magnetization shown is normalized with respect to the saturation value M s .. using a semiconducting barrier could prove an alternative strategy for difﬁcult fabrication of ultrathin ( 1 nm) oxide barriers (Rippard et al.0%.D. 1967). From Tanaka and Higo. electronic structure calculations have predicted (MacLaren et al..3) that at a ZnMnSe/AlGaAs interface they limit the spin injection efﬁciency (Stroud et al. and Das Sarma: Spintronics: Fundamentals and applications M(T)/M(0) T 3/2. 1994). and conventional MTJ’s (Moodera et al.370 ˇ ´ Zutic. 1997). A smaller RC constant would allow faster switching times in magnetic random-access memories (for a detailed discussion see De Boeck et al. sweep of the magnetic ﬁeld from positive to negative. However.5 nm). another II-VI semiconductor. 50 nm)/AlAs (1. 2002). The results are consistent with k being conserved in the tunneling process (Mathon and Umerski. CPP multilayers (Lenczowski et al. There is also a possibility of using all-semiconductor F/Sm/F single-crystalline MTJ’s where F is a ferromagnetic semiconductor..15 was limited below 50 K. The tunneling current in that regime has been shown to be highly nonuniform95 and the measured apparent resistance R m V/I (different from the actual junction resistance R J ) can even attain negative values (Pederson and Vernon. 1985. An additional decrease of TMR with T was expected due to the spin-independent part of G(T) (Shang et al. April 2004 .... 2. discussed above.. Such a temperature dependence (known as Bloch’s law and reviewed by Krey.3%. 1975) were not reproduced. Large TMR ( 70% at 8 K). has been measured in an epitaxially grown (Ga. a room-temperature effect remains to be demonstrated as the available wellcharacterized ferromagnetic semiconductors do not have as high a ferromagnetic transition temperature. 93 ` The early F/Ge/F results (Julliere.Mn)As junction (Tanaka and Higo. The standard four-probe technique for measuring I and V has been known to give spurious values when the resistance of the F electrodes is non-negligible to the junction resistance. thin solid curve. 2002). 2001) or only a small (Meservey et al. dashed curve. Boeve et al. was also obtained for TMR (MacDonald et al. 23. Correspondingly. Systematic studies of MTJ’s containing a semiconductor (Sm) region (used as a tunneling barrier and/or an F electrode) have begun only recently.. 2001). (32) infer a reduced spin-valve effect. sweep from negative to positive. Rev. Some F/Sm/F magnetic tunnel junctions have been grown epitaxially. 2002) and from Eq. No. Jia et al.6 nm)/Ga1 x Mnx As (x 3. Fabian. For a given AlAs thickness. 2000). 1998. These would simplify integration with the existing conventional semiconductor-based electronics and allow ﬂexibility of various doping proﬁles and fabrication of quantum structures. Rzchowski and Wu. We recall (see Sec. Phys. Results on ZnS. 2000. 94 Interface defects could diminish measured TMR. 2002) spin-dependent signal.. 2001. reaching 15% at 10 K for junctions of higher resistance and lower defect density94 (Gustavsson et al. 1998).85Co0. 1998). 1996. and GaN as a barrier have shown either no (Gibson and Meservey. shown in Fig. 1998). and the amplitude of TMR can be studied as a function of the crystallographic orientation of a F/Sm interface.. demonstrated a TMR of 5% at room temperature (Guth et al..
2000). In contrast.. Fabian. which could lead to spin precession even in the absence of an ¨ applied magnetic ﬁeld (Konig and Martinek. van Son et al. with no (spin-averaged) Fermi velocity mismatch. which is related to the square of the normal-state transmission. v¯ (103) single-particle Hamiltonian for spin ↑.. ferromagnetic single-electron transistors.A. yielding (de Jong and Beenakker. For an illustration of how such a technique can be used to study spin-polarized transport ´ in a wide range of heterojunctions see Melin and Feinberg (2002). 2000) was obtained with the more rigorous nonequilibrium Keldysh technique (Keldysh. 1999). 76. Bergeret et al. 2. It is instructive to note a similarity between the twocomponent transport in N/S junctions (for electronlike and holelike quasiparticles) and F/N junctions (for spin ↑. A convenient description is provided by the Bogoliubov–de Gennes equations (de Gennes. accompanied by additional boundary resistance (Blonder et al. For a spinsinglet superconductor an incident electron (hole) of spin is reﬂected as a hole (electron) belonging to the opposite spin subband ¯ . Matsuyama et al. When S is in the normal state. v ¯ are the electronlike quasiparticle and holelike quasiparticle amplitudes. 1997. is the pair potential (de Gennes. (2002). could be ignored for low-transparency junctions with conventional superconductors. In the N/S junction Andreev reﬂection is responsible for the conversion between the normal and the supercurrent. Phys. H u * v¯ H¯ E u . The transparency of this approach98 makes it suitable for the study of ballistic spin-polarized transport and spin injection even in the absence of a superconducting region (Heersche et al. (47). ↓).. 1989).C. This is a phase-coherent scattering process in which the reﬂected particle carries information about both the phase of the incident particle and the macroscopic phase of the superconductor. and ¯ denotes a spin opposite to (de Jong and Beenakker. 98 A good agreement (Yan et al. No. 2002. 1995. A spherical Fermi surface in the F and S regions. respectively. A detailed understanding of MTJ’s will also require knowing the inﬂuence of the interface and surface roughness (Itoh et al. only a fraction of the incident electrons from a majority subband will have a minority subband partner in order to be Andreev reﬂected. and u . and Z is the strength of the -function barrier. The probability for Andreev reﬂection at low bias voltage (qV ). back to the nonsuperconducting region. (100)] at low bias and T 0 and Andreev reﬂection is the dominant process. 1986). with different populations in two spin subbands.. They obtained a result that interpolates between the clean and the tunneling limits. ˇ ´ Zutic and Valls. Andreev reﬂection Andreev reﬂection (Andreev. Rev. single-particle tunneling vanishes [recall Eq.2). in which the two limits correspond to Z→0 and Z→ . 2002.96 Andreev reﬂection thus is responsible for a proximity effect where the phase correlations are introduced to a nonsuperconducting material (Demler et al. 2003). 1982. 2001.. Here H is the For instructive reviews see Lambert and Raimondi (1998). 97 .97 Grifﬁn and Demers (1971) have solved the Bogoliubov–de Gennes equations with square or -function barriers of varying strength at an N/S interface. responsible for a conversion between a dissipative quasiparticle current and a dissipationless supercurrent [see also early work by de Gennes and Saint James (1963)]. for high-transparency junctions (see the discussion of Sharvin conductance in Sec. shows the importance of Coulomb interactions. in terms of the total number of scattering channels (for each k ) 2 N k F A/4 at the Fermi level.. 1989). is assumed. 3. 1996).ˇ ´ Zutic. while in the F/N case a conversion between spin-polarized and unpolarized current is characterized by the spin diffusion length. Izyumov et al. et al. II. Hu and Matsuyama. 1995) * and by matching the wave functions at the boundaries (interfaces) between different regions. 2001. in which the I region is a quantum dot.1.. (2003). A theoretical study of F/I/F junctions. E the excitation energy. respectively. 1999). April 2004 96 Equation (103) can be simply modiﬁed to include the spin ˇ ´ ﬂip and spin-dependent interfacial scattering (Zutic and Das Sarma. which both lead to current conversion. 2001. h (104) 1 equivalent to R Sharvin . 1987). Vol. at an interface with a superconductor.. and Das Sarma: Spintronics: Fundamentals and applications 371 H 0 the two values of resistance can be used for various nonvolatile applications (Moodera et al. 1964) is a scattering process. (1982) used a similar approach. Blonder et al. IV.. A comprehensive review of tunneling phenomena and magnetoresistance in granular materials. from Eq. and double tunnel junctions is given by Maekawa et al.. Nitta.. 2001. characterized by the superconducting coherence length. Mod.↓. Zeng et al. and k F is the spin-resolved Fermi wave vector. 2003). Pannetier and Courtois (2000). Here A is the pointcontact area. the zerotemperature Sharvin conductance is G FN e2 N↑ N↓ . Rammer and Smith. 1964. For spin-polarized carriers. 2002). as discussed in Sec.. This can be simply quantiﬁed at zero bias and Z 0. Even in the spinunpolarized case it is known that the full quantumˇ ´ mechanical approach (Tesanovic et al. Fominov. 1986) can lead to qualitatively different results from the usual quasiclassical picture and from averaging out the spatial information on the length scale of the inverse Fermi wave vector. while a Cooper pair is transferred to the superconductor.. Hu. In the superconducting state all of the N ↓ and only (N ↓ /N ↑ )N ↑ scattering channels contribute to Andreev reﬂection across the F/S interface and transfer charge 2e. known as the Blonder-Tinkham-Klapwijk method. Halterman and Valls.
h ↓ (105) The suppression of the normalized zero-bias conductance at V 0 and Z 0 (de Jong and Beenakker. was used as a sensitive transport technique to detect spin polarization in a point contact (Soulen et al. Mod.1). 2002. Detection of P in a high-temperature superconductor is even possible with a large barrier or a vacuum between the F and S regions. Panguluri... 2003. including the ﬁrst direct measurements (Braden et al. Jedema et al. one of the Heusler alloys originally proposed as half-metallic ferromagnets (de Groot. shows only partial spin polarization.Mn)As and (In. Fabian.Mn)Sb.. 1995). in order to extract the spin polarization of the F region more precisely.. Parker et al. we mention some subtleties that arise even for the simple model of a spherical Fermi surface used to describe both F and S regions. not only at V 0.. Ji et al. A more complete analysis should also quantify the effects of spin-orbit coupling. The advantage of such techniques is the detection of polarization in a much wider range of materials than those which can be grown for detection in F/I/S or F/I/F tunnel junctions. using a thin-ﬁlm nanocontact geometry (Upadhyay et al. From Soulen et al. emphasized the importance of ﬁtting the conductance data over a wide range of applied bias. even when all the Fermi velocities differ. 2003) and even in a 1D tight-binding model with no spin polarization (Afﬂeck et al. ´ Melin and Feinberg. 2003).. 1999. and Maekawa. 2003). Panguluri. A large number of experimental results using spin-polarized Andreev reﬂection has since been reported (Bourgeois et al. (106). Vol.A. Nadgorny et al.5 meV. et al. P G . The roughness or the size of the F/S interface may lead to a diffusive component of the transport (Fal’ko et al.102 are on the same side of the S region. In other words.101 Large magnetoresistive effects are predicted for crossed Andreev reﬂection (Deutscher and Feinberg. Feng and Xiong. 2001. bipolar diodes and transistors.. Rev. 1983). 2001. 2002. Takahashi. for ﬁxed Fermi velocity mismatch. 1998. and Das Sarma: Spintronics: Fundamentals and applications FIG. (Color in online edition) The differential conductance for several spin-polarized materials. or semicon- 99 ˇ ´ Similar measurements were also suggested by Zutic and Das Sarma (1999) to yield information about the FSm/S interface. as proposed by Wang and Hu (2002) using resonant Andreev reﬂection and a d-wave superconductor. complications can then arise in trying to disentangle the inﬂuence of parameters Z. for a quantitative interpretation of the measured polarization. 2001... even at large interfacial barrier (referred to as a zero-bias conductance peak in Sec.. the spin polarization (nonvanishing exchange energy) can increase the subband conductance. Unlike charge transport in N/S junctions (Blonder and Tinkham. April 2004 . Fermi velocity mismatch between the F and the S regions does not simply increase the value of effective Z. 24.. Phys. et al. 2002. 2000). 2003). Mueller. No. 100 Similar results were also obtained when F and S region were separated with a quantum dot (Zhu et al.. Mazin et al. 1983) in a Grifﬁn-DemersBlonder-Tinkham-Klapwijk approach. Such structures have also been theoretically studied to understand the implica´ tions of nonlocal correlations (Apinyan and Melin. 2002) and the role of inelastic scattering (Auth et al. 1999. Data are given in Fig. A similar study. Spin-polarized drift and diffusion Traditional semiconductor devices such as ﬁeld-effect transistors. Panguluri et al. 2004) of the spin polarization in (Ga. et al. Nadgorny... Speciﬁcally. Tsoi.372 ˇ ´ Zutic. unlike 1999. 2. 101 Interference effects between quasielectron and quasihole scattering trajectories that feel pair potentials of different sign lead to a large conductance near zero bias. 76.. 4. 102 Recent theoretical ﬁndings suggest that the separation should not exceed the Fermi wavelength (Yamashita. 2003. Conversely. 2000) when the two F regions. at Z V 0 and normal incidence it is possible to have perfect transparency. an increase in Fermi velocity mismatch could increase the subgap conductance.A. 24. important additional factors (similar to the limitations discussed for the applica` tion of Julliere’s formula in Sec. where v S is the Fermi ˇ ´ velocity in a superconductor (Zutic and Das Sarma.. (2002)].. the Fermi surface may not be spherical [see the discussion of Mazin (1999). Zeng et al. IV. Note that NiMnSb.. showing the suppression of Andreev reﬂection with increasing P G .100 In a typical interpretation of a measured conductance. 1999. at a ﬁxed exchange energy. 1998).. satisfying ( v F↑ v F↓ ) 1/2 v S .. G FS /G FN 2 1 P G . separated within the distance of the superconducting coherence length. As a caution concerning the possible difﬁculties in analyzing experimental data. (106). and Fermi velocity mismatch from the nature of the point contacts (Kikuchi et al. ˇ utic and Valls. (106) with the increase in the spin polarization P G →(N ↑ N ↓ )/(N ↑ N ↓ ). 1998). 2003.. specifying what type of spin polarization is experimentally measured and also that of Xia et al. 2001). For example. The vertical lines denote the bulk superconducting gap for Nb: (T 0) 1. Z ´ in Eq. IV. G FS e2 2N ↓ h 2N ↓ N N↑ ↑ 4 e2 N .99 However.2) need to be incorporated in the picture provided by Eq. 2000). 2002).
j s j ↑ j ↓ . the second term represents diffusion. and (111) need to be solved self-consistently. Many proposed spintronic devices as well as experimental settings for spin injection (Sec. if spin precession is important for device operation.. In many important cases Eqs. Let the equilibrium spin splitting of the carrier band be 2q c . Martin (2003). . Materials considerations c qD c c q c c c. The spin resolved carrier ˇ ´ charge-current density is (Zutic et al. Pershin and Privman (2003a). (2002a). which usually reˇ ´ quires numerical techniques (Zutic et al. Heide. and spin. it follows from Eq. and Das Sarma: Spintronics: Fundamentals and applications 373 ductor solar cells rely in great part on carriers (electrons and holes) whose motion can be described as drift and diffusion. 76.A.B. while ramiﬁcations of magnetic drift for unipolar transport were studied by Fabian et al. 2001) to bipolar transport and to systems with spatially inhomogenous charge density. and double per- Rev. Finally. Equation (108) describes the spincharge coupling in bipolar transport in inhomogeneous magnetic semiconductors. and the lower sign is for holes. the ¯ bar denotes a complementary carrier (n p. 2003). Beck et al. Examples include half-metallic oxides such as CrO2 . Zutic et al. (2002a). Recently an interesting study (Saikin et al. Mod. respectively. In some cases it is possible to extract the relevant physics in limiting cases analytically. moving in the electrostatic potential which comprises both the external bias V and the internal built-in ﬁelds due to charge inhomogeneities. (108) (109) where the carrier density c c ↑ c ↓ and spin s c c ↑ c ↓ . Wegrowe. which appears after realizing that spin relaxation equilibrates spin while preserving carrier density. 2. Fabian. A spin-polarized unipolar transport can be obtained as a limiting case by setting the electron-hole recombination rate to zero and considering only one type of carrier (either electrons or holes).ˇ ´ Zutic. and we introduced the carrier charge and spin conductivities c q( c c sc s c ) and sc q( sc c c s c ). III. limited by carrier recombination and spin relaxation ˇ ´ (Fabian et al. In unipolar spin-polarized transport one does not need to consider carrier recombination. and ˜ c P c0 c is the nons equilibrium spin density. Nominally highly spin-polarized materials. Vol. (2002). (107). the upper sign is for electrons. sc is the spin relaxation time of the carrier c (not to be confused with the single spin decoherence time discussed in Sec.2). for example). j j ↑ j ↓ . while the last term stands for magnetic drift—carrier drift in inhomogeneously split bands. It also often sufﬁces to study pure spin diffusion. and is the dielectric constant. III. Transient dynamics of spin drift and diffusion was considered by Fabian and Das Sarma (2002). qD sc c qD c s c .. limited by carrier recombination. if the built-in electric ﬁelds are small. where bipolar refers to the presence of electrons and holes. II) can be described by both carrier and spin drift and diffusion. 2002). (110). 2003) was reported on a Monte Carlo simulation of quantummechanical spin dynamics limited by spin relaxation. Fabian et al.. could provide both effective spin injection into nonmagnetic materials and large magnetoresistance effects. 1987. important for nonvolatile applications. where N d and N a are the donor and acceptor densities. (2002). Consider electrons and holes whose density is commonly denoted here as c (for carriers). April 2004 . as discussed in the previous sections. Zutic et al. usually neglecting electric ﬁeld or magnetic drift. (107) where and D stand for mobility and diffusion coefﬁcients. not spin up and down. We illustrate spinpolarized drift and diffusion on the transport model of spin-polarized bipolar transport.1).103 More transparent are the equations for the total charge. Drift of the spin-polarized carriers can be due not only to the electric ﬁeld. to be solved in a self-consistent manner. (111) connecting the electric ﬁeld and charge q(p n N d N a ). Fe3 O4 . The ﬁrst term on the right-hand side describes drift caused by the total electric ﬁeld. in which quasiclassical orbital transport was carried out for the in-plane transport in III-V heterostructures where spin precession is due to bulk and structure inversion asymmetry (see Sec. Similarly. where c ( c↑ c↓ )/2 and cs ( c↑ c↓ )/2 are charge and spin mobilities. Steady-state carrier recombination and spin relaxation processes are described by the continuity equations for the spin-resolved carrier densities: • jc q c c w c c ¯ c 0¯ 0 c c 2 ˜c s sc . In addition. (2000). the set of equations is completed with Poisson’s equation. (110) Here w is the spin-dependent recombination rate. Yu and Flatte (2002a. Unipolar spin-polarized transport in inhomogeneous systems in the presence of electric ﬁelds was analyzed by Fabian et al. current densities: jc jsc c sc sc c c c qD c c qD sc s c . No. 2002) jc q c sity can cause charge currents. Spin-polarized drift and diffusion in model GaAs quantum wires was studied by Sogawa et al. 2002a... and similarly for the diffusion coefﬁcients. 2002). (2002a) and Martin (2003). Phys.. the recombination-limited drift diffusion is supplied by Maxwell’s equations. 2000. Bipolar transport in the presence of electrical ´ drift and/or diffusion has been studied by Flatte and Byˇ ´ ers (2000). (109) that spatial variations in carrier densities can lead to spin currents. but also to magnetic ﬁelds. Spatial variations in spin den- 103 Equation (107) can be viewed as the generalization of the Silsbee-Johnson spin-charge coupling (Johnson and Silsbee. CMR materials. spin dynamics need to be explicitly incorporated into the transport equations (Qi and Zhang. 2002b). In inhomogeneous devices where charge buildup is the rule. B.
. and magnetic properties.. 2002) provide a method for distinguishing carrier-induced ferromagnetism. Tromp et al. relevant to device applications. Van Esch et al. for example. 2001. Ferromagnetic semiconductors (Nagaev.Mn)As layer. was later revisited suggesting extrinsic effect (Bennett et al. were reviewed by Samarth et al.Mn)Te (Escorne et al. Photoinduced ferromagnetism in a (In. structures of reduced dimensionality. 26(b) the corresponding Hall resistivity Hall R 0 B R S M. spin transistors. For example. Interfacial properties. In contrast to (In. Photoinduced ferromagnetism was demonstrated by Koshihara et al. 1992). Park et al.Mn)As (Munekata et al.Mn)As (Ohno et al... have been demonstrated to be highly spin polarized.. Consequently.Sn. The excess holes generated in the GaSb layer are effectively transferred to the p-doped (In. and Das Sarma: Spintronics: Fundamentals and applications ovskites (Kobayashi et al. Rev.Mn)As and (Ga.. 1989. Hayashi et al. 2003). as compared to the uniformly doped bulk ferromagnetic semiconductors (Nazmul et al.Mn)V compounds. 1999. The increase in magnetization. not limited to semiconductors.Mn)As layer and is absorbed in a GaSb layer. 2003) is not universally accepted. 2003). resulting in a paramagneticferromagnetic transition. from ferromagnetism that originates from magnetic nanoclusters.Cr)Te (Saito et al. 2001). electrons and holes. 2. 1974). by a built-in ﬁeld.. 1991. IV.. The presence of such nanoclusters often complicates accurate determination of T C as well as of whether the compound is actually in a single phase. Mod. (1997) in p-(In. 1996) showed the low solubility of Mn and the formation of magnetic nanoclusters characteristic of many subsequent compounds and different magnetic impurities. Chiba. which typically leads to lower mobilities and shorter spin relaxation times than in n-doped materials... Ott et al.Mn)As with high carrier density ( 1020 cm 3 ).. and (Pb. 1983).. Vol.374 ˇ ´ Zutic. optical. 2000. Fermi level. more recent interest in ferromagnetic semiconductors was spurred by the fabrication of (III. Experiments in which ferromagnetism is induced optically (Koshihara et al. (112) is shown. 1960).. valence band. conduction band.. and spin-based quantum computers (Sec. the reported room-temperature ferromagnetism in an increasing number of compounds (reviewed by Pearton et al. Typical for (III. Ohno et al. while in MTJ’s it is the surface rather than the bulk electronic structure which inﬂuences the relevant spin polarization. measured by a SQUID. 1996. 2002... (2003). A large band bending across the heterostructures separates.. 1993). a much lower carrier density in (Zn. It remains to be understood what the limitations are for using extrinsic ferromagnets and. From Koshihara et al. shown in Figs.Mn)Te (Cochrane et al.F). early work reporting ferromagnetism even at nearly 900 K in La-doped CaBa6 (Young et al.Mn)As/GaSb heterostructure. (b) band-edge proﬁle of (In. based on the exchange interaction between the carrier and the magnetic impurity spins. E F . Materials properties of hybrid F/Sm heterostructures. where they enhance the ferromagnetic spin exchange among Mn ions.. electrical. 1974). However. Such experiments also suggest possible nonvolatile multifunctional devices with tunable. respectively. and in Fig. is shown in Fig. Phys. It is possible to use selective doping to substantially increase T c .. Conclusive evidence for intrinsic ferromagnetism in semiconductors is highly nontrivial.. A high T C and almost complete spin polarization in bulk samples are alone not sufﬁcient for successful applications. E c . Unpolarized light penetrates through a thin (In. Hall M. Fabian. for reviews of halfmetallic materials see Pickett and Moodera. are given by Zvezdin and Kotov (1997) and Sugamo and Kojima (2000). Spintronic devices typically rely on inhomogeneous doping.. Most of the currently studied ferromagnetic semiconductors are p-doped with holes as spin-polarized carriers. . 1997... Ferromagnetic order with Mn doping was obtained previously. a II-VI ferromagnetic semiconductor with Curie temperature T C near room temperature (Saito et al. E v . Early work on (Ga. 2000.. whether they can be effective spin injectors.Mn)As (De Boeck et al. Fang et al. Comprehensive surveys of magneto-optical materials and applications.Mn)V compounds. and/or structures containing different materials. (Ge. known since CrBr3 (Tsubokawa. 1986). for example. can signiﬁcantly inﬂuence the magnitude of magnetoresistive effects105 and the efﬁciency of spin injection. Doping FIG. et al.Mn)As/GaSb heterostructure. 1998..104 After the initial discovery of (In. where the R 0 is the ordinary and R S the anomalous Hall coefﬁcient. 26(a). 2003) and electrically (Ohno. Hall is dominated by the anomalous contribution. 105 In magnetic multilayers GMR is typically dominated by interfacial scattering (Parkin. 76. 2002). 2003). most of the research has focused on (Ga. No. 1997). as discussed in the previous sections. Wang et al. 2002). 1997.. 1997.Mn)As/GaSb heterostructure: (a) light-irradiated sample displaying photoinduced ferromagnetism—direction of light irradiation is shown by an arrow. in (Sn. 25 and 26. for example. 25. April 2004 104 properties and the possibility of fabricating a wide range of structures allow spintronic applications beyond magnetoresistance effects. suggests that transport properties can be effectively controlled by carrier doping. Oiwa et al.Mn)Te (Story et al. spin lasers.
Chiba. Manipulation of the g factor in a GaAs/ AlGaAs quantum well in Fig. Electrically induced ferromagnetism was realized by applying gate voltage V G to change the hole concentration in d 5 nm thick (In. A different type of photoinduced magnetization was measured in ferromagnetic (Ga.. Oiwa et al.D. where x varied gradually Rev.Mn)Te (Awschalom et al. 2002). 2003.1 and V (Moriya et al. doping with magnetic impurities can give even g * 500. the corresponding data for R Hall Hall /d M [recall Eq. Inset. (2002) showed that in MnGe ferromagnetism can be manipulated at higher temperature and at signiﬁcantly lower gate voltage (at 50 K and 1 V). II. In semiconductors g factors. by changing the polarization of a circularly polarized light. (b) Hall resistivity at 5 K before (dashed lines) and after (solid lines) light irradiation.. Related experiments on modulation-doped GaAs/Al0.Mn)As. 1976. 26. In Fig. as an electric analog of the manipulation of M from Fig.Mn)VI materials have shown that nonequilibrium spin-polarized carriers can change the orientation of magnetic spins in (Hg. 2.Mn)Te quantum wells (Boukari et al. R Hall curve after application of 125 V..ˇ ´ Zutic. 28. while. April 2004 . 2001). while illumination in p-i-p structures would destroy ferromagnetism.106 In a Faraday geometry (Sec. 125. (112)] show that the ferromagnetism can be switched on and off. 1987). From Koshihara et al.. FIG. channel was grown on top of a InAs/(Al. 2003). It was demonstrated that illumination by light in p-i-n diodes would enhance the spontaneous magnetization. Weisbuch and Herman.. can be very different from the free-electron value. 27.. 2000. 2003) suggest that the main contribution of carrier spin to such rotation is realized by generating an effective magnetic ﬁeld through the p-d exchange interaction. clear hysteresis at low ﬁelds ( 0.3Ga0. one can modulate the Hall resistance and thus the induced magnetization by up to 15% of the saturation value (Oiwa et al. No. In GaAs-Fe composite ﬁlms an observation of room-temperature photoenhanced magnetization was used to demonstrate that a magnetic force can be changed by light illumination (Shinshi et al. which determine the spin splitting of carrier bands (and consequently inﬂuence the spin dynamics and spin resonance).Mn)As/GaSb at 5 K observed before (open circles) and after (solid circles) light irradiation. Application of V G 0. Phys. Subsequently. I. II. 1997. Fabian. 76.40 for x 0. R Hall vs ﬁeld curves under three different gate biases.3. dotted line. The combination of light and electric-ﬁeld control of ferromagnetism was used in modulationdoped p-type (Cd.7As have shown that by applying V G one can shift the electron wave function in the quantum well and produce 1% change in the g factor (Jiang and Yablonovitch. 2003). and 125 V results in a qualitatively different ﬁeld dependence of R Hall measured at 22. the variation of Al concentration changes the g factor (Chadi et al.7 mT) as holes are accumulated in the channel (V G 125 V). relies on the results for a bulk Alx Ga1 x As. Solid line show a theoretical curve.3.Mn)As 106 Previous studies in paramagnetic (II...D. Magnetization curves for (In. and Das Sarma: Spintronics: Fundamentals and applications 375 FIG. Additional experiments on photoinduced magnetization rotation (Munekata et al. (Color in online edition) Electric-ﬁeld control of ferromagnetism. solid line. R Hall curve measured at V G 0 V before application of 125 V.B.5 K (sample B): almost horizontal dash-dotted line.Ga)Sb/AlSb and GaAs substrate... 2002).3).. 1989. Subsequent work by Park et al.44 for x 0 and g 0. rather than by spin-transfer torque. From Ohno. 1977) to g 0. With strong spinorbit coupling in narrow-band III-V’s they are 50 for InSb and 15 for InAs. paramagnetic response when holes are partially depleted from the channel (V G 125 V).Mn)Te (Krenn et al. et al. 27. Mod.. as discussed in Sec. 26. 1985) and in (Cd. as discussed in Secs. in an optimized Alx Ga1 x As quantum well. Below a metal gate and an insulator the (In. dashed lines. Vol.Mn)As used as a magnetic channel in a metal-insulator semiconductor FET structure. the same curves shown at higher magnetic ﬁelds.
2003. The anisotropy of g factor (g tensor) allows voltage control of both the magnitude and the direction of the spin precession vector . No. (1998).2 T with EuSe (Moodera et al. 28. no precession is observed.. 1967. corresponding to g 0. was explained by the inﬂuence of the magnetic ﬁeld on the height of the barrier formed at the N/F interface (for 107 At zero magnetic ﬁeld EuSe is an antiferromagnet.108 could be modiﬁed by an applied magnetic ﬁeld. C. II. (2002). 2001). Egues. Giazotto et al. (1998)...E. These spin-ﬁltering properties of the Eu chalcogenides. Joshi et al. Oberli et al.. The large spin splitting of the Eu chalcogenides was subsequently employed in the absence of applied ﬁeld with EuS (Moodera et al. Petukhov et al. Perel’ et al. In an applied magnetic ﬁeld two quantum point contacts.. 2003). (2003). (2002b). (2002b). 79) also have their spin-dependent counterparts. Mendez et al. 1969.. used together with one-electron quantum dots. 113 QPC was also used to locally create and probe nonequilibrium nuclear spin in GaAs/AlGaAs heterostructures in the quantum Hall regime (Wald et al. 2. 2003).. exchange splitting of a conduction band in bulk EuS is 0. Borda et al. At V G 2 V. and nearly 100% spin polarization was reached at B 1. 1999).110 Several other realizations of spin ﬁltering have been investigated. 1968. Ionicioiu and D’Amico. and the angular precession frequency B gB/ can be used to determine the g factor. and nanocrystals (Efros et al. Avodin et al. It is also possible to manipulate g factors dynamically using time-dependent V G (Kato et al. 111 These include the works of Voskoboynikov et al. Vurgaftman and Meyer (2003). Filipe et al.376 ˇ ´ Zutic. Koto. April 2004 . the barrier height was lowered by 25% at 2 T). et al.. Ohno (1998). much larger changes were measured—when V G is changed. where N is a normal metal and F is a ferromagnet. Nagaev. Koga et al. 2001. and at moderate ﬁelds it becomes a ferromagnet with T C 5 K. 1983). (1995).3. (1999).. 2002). Kiselev and Kim (2001). 2002. Mesoscopic spin ﬁlters have also been suggested (Frustaglia et al. 2003). Cacho et al. Figure 28(a) gives the time-resolved Kerr rotation data (the technique is discussed in Sec. 1990).C. Spin ﬁlters EuSe.1) was ﬁrst realized in N/F/N tunneling.. in which spin ﬁltering can be tuned by an applied bias. (1995). Deshmukh and Ralph. A choice of particular atomically ordered F/Sm interfaces was suggested to give a strong spin-ﬁltering effect (Kirczenow. where t is the delay time between the circularly polarized pump and linearly polar* ized probe pulses. (1989). (2003). (2003). Rev. 2001. de Andrada e Silva and La Rocca (1999). Mod. (b) displacement of the electron wave function towards the back gate into regions with more Al concentration as a positive voltage V G is applied between back and front gate. Brehmer et al. 109 This method could be realized using single-electron transistors or quantum point contacts 110 See. see Sec. 1999). (2002). which can be described as * exp( t/T2 )cos( t). Guo et al. 2000. Ting ` and Cartoxia (2002). van Dijken et al. an emitter. Aleiner and Lyanda-Geller (1991). and a collector fabricated on top of a high-mobility 2DEG in GaAs/ AlGaAs.112 discussed in more detail in Sec. Phys. 1994). Kato. IV. across the structure. (1998. The change in I-V curves in the N/F/N structure. 1998.. such as EuSe107 and EuS. leading to an increase of g.. The structures studied are typically double-barrier resonant tunneling diodes (for an early spin-unpolarized study see Tsu and Esaki. either with Zeeman splitting or using ferromagnetic materials.36 eV (Hao et al. 112 See Monsma et al. Predicted quantum size effects and resonance tunneling (Duke.. 76. 114 Suggested by Sharvin (1965) as a technique to study Fermi surfaces. Vol.. can act as spin polarizer and analyzer (Potok et al. were proposed as the basis for a method to convert single spin into single charge measurements109 and provide an important ingredient in realizing a quantum computer (DiVincenzo. (2002). Upadhyay et al. From Salis. 2001). 2001) provide effective spin ﬁltering and spin-polarized currents. Rippard and Buhrman (2000). 108 At zero magnetic ﬁeld. 1993).. Kasuya and Yanase. Governade et al. for example.113 In a ballistic regime and at T 70 mK (mean free path 45 m sample size 1. et al. 1988). p. Fabian. III). see also van Houten et al.5 m) magnetic focusing114 with B results in base-collector volt- Solid-state spin ﬁltering (recall the similarity with spin injection from Sec. 1973). in quantum dots (Recher et al. and Das Sarma: Spintronics: Fundamentals and applications FIG. (1998).. IV. relying on spin-orbit coupling111 or hotelectron transport across ferromagnetic regions. Voltage-controlled spin precession: (a) time-resolved Kerr rotation measurements of electron spin precession in a quantum well at different gate voltages V G with Al concentration of 7% at 5 K and B 6 T. and we discuss here a particular realization. the electron wave function efﬁciently samples different regions with different g factors (Salis. limited by the spin-orbit coupling and interfacial spin-ﬂip scattering. It was shown by Esaki et al. T 2 is the transverse electron spin lifetime with inhomogeneous broadening. 2001). Slobodskyy et al. Zeeman splitting in semiconductor heterostructures and superlattices (enhanced by large g factors. Petukhov (1998). (1976) that magnetic tunneling through (ferro)magnetic semiconductor ´ Eu chalcogenides (von Molnar and Methfess. 2001.F.
due to the large in-plane ﬁeld B . 12 and 13) were employed for injecting and detecting spins in semiconductors. can be tuned by the gate voltage.. and the collector transmission coefﬁcient T C is related to the collector conductance by g C (2e 2 /h)T C . where is the Landau-level ﬁlling. In Eq. D. Rev.115 Several spin diodes have recently been proposed or demonstrated with the goal of either maximizing the sensitivity of the I-V characteristics to spin and magnetic ﬁeld. p-n heterostructures have combined Cr. age peaks. where m * is the effective mass. which has a diode property due to the existence of a built-in ﬁeld in the contact. which. Osipov et al. In a magnetic ﬁeld both the mobile carriers and the recombination centers have an equilibrium spin polarization due to the Zeeman splitting. (113) we note a recurring form for a spin-valve effect.. one with 1 and the other with 1. Spin diodes FIG.. No. Fabian. when the separation of the two quantum point contacts is an even multiple of the cyclotron radius m * v F /eB .ˇ ´ Zutic. These results are illustrated in Figs. (c) focusing peak height at B 6 T with spin-selective collector conductance (g C 0. The resulting effect of spin ﬁltering modiﬁes the collector voltage V C (Potok et al. (2003).. (2) or to spin-charge coupling due to nonequilibrium spin [recall Eqs.3. and v F the Fermi velocity. Interestingly. 2003) where. Spin light-emitting diodes (recall Figs. 29. II. Johnston-Halperin et al. Solomon (1976) found a variation of 0. (43) and (114)]. comparing E as a quantum point contact at 2e 2 /h (dashed curve) and E as a quantum dot with both leads at 2e 2 /h (solid curve). Mod. while resonant tunneling diodes have been where 0 1 parametrizes the imperfections in focusing. (1992) is based on a junction between two coplanar AlGaAs/GaAs 2DEG’s. The trick to modifying the current is to decrease (even saturate) the spin polarization of either the electrons or the centers by electron spin resonance. but also in other semiconductors where electrical detection would be desirable. that is. The emitter E can be formed into either a quantum dot or a quantum point contact. using a small perpendicular magnetic ﬁeld. As discussed in Sec. Solomon (1976).D. Magnetic tunneling diodes have been used for spin injection from a ferromagnetic to a nonmagnetic semiconductor.or Eu-based ferromagnetic semiconductors and InSb (Viglin et al. 2003). Indeed. is accompanied by a spin ﬂip. due to the current-induced dynamic nuclear polarization. Adapted from Folk et al. Gates marked with ‘‘ ’’ are left undepleted when E is operated as a quantum point contact. 2. 1972). Van Dorpe. Vol. 29(a) and (b) on a slightly modiﬁed structure (Folk et al.. 76. similar to TMR in Eq. The measured signal involves the product of two different spin polarizations.01% of the saturation current at small biases where recombination in the space-charge region dominates. or facilitating spin injection and its detection through semiconductor interfaces comprising a magnetic semiconductor as the injector. 1997. for example. Another mesoscopic spin ﬁlter with fewelectron quantum dot (GaAs/AlGaAs-based) was used to demonstrate a nearly complete spin polarization which could be reversed by adjusting gate voltages (Hanson. Liu. which changes the conductance of the quantum point contact.. in p-GaMnAs/n-GaAs p-n junctions (Kohda et al. et al. the current is also time dependent. Electrons are focused from E to C through the base region B. in turn. VC h/2e 2 I E 1 P I E P T C . 2002). The current in a p-n junction depends on the recombination rate. the two regions have opposite spins at the Fermi level. April 2004 115 Spin diodes can also probe fundamental properties of electronic systems. and Das Sarma: Spintronics: Fundamentals and applications 377 for example. 1998). (3)] of the emitter current I E . P I E and P T C are the spin polarization [recall Eq. by applying a gate voltage. (Color in online edition) Mesoscopic spin ﬁlter: (a) micrograph and circuit showing the polarizer-analyzer conﬁguration used in the experiment of Folk et al. where optical methods are ineffective. proposed and demonstrated a silicon p-n junction whose current was modiﬁed by changing the spin polarization of the recombination centers. The collector C is a single point contact. (113) Spin diodes are inhomogeneous two-terminal devices whose electronic or optical properties depend on the spin polarization of the carriers. 2001... Similar experiments could be used to detect nonequilibrium spin due to (potential) spin injection in Si. 2002.. (b) base-collector voltage (V C ) showing two focusing peaks. ..5e 2 /h). Such devices were envisaged long before the emergence of spintronics. Phys. et al. 2003. The current crossing such a junction. 2003). The diode demonstrated by Kane et al. Vandersypen. one can cause the emitter to form either a quantum dot or a quantum point contact. depends on the relative orientation of the spin of the carriers and the centers (Lepine. Effective spin ﬁltering.
II. A magnetic unipolar diode ´ has been proposed by Flatte and Vignale (2001) to simulate the working of ordinary diodes. 76. No. unpolarized holes.3) and spin ﬁlters (Sec. but with homogeneous monopolar doping (either donors or acceptors. not through the spin effects described here. (2002) have proposed the magnetic bipolar diode described below.E.. The large-bias or high-injection limit is the regime where the injected carrier density becomes comparable to the equilibrium density. determines the FIG.. 2001b) and the spin ˇ utic et al. large forward current ﬂows. Rev. 30 (also see Fig. IV. The MBD is the prototypical device of bipolar spintronics.4). where the Shockley theory (Shockley. These devices offer opporsolar cell (Z ´ tunities for effective spin injection. empty circles. IV.. 1980. Zutic et al.C. the intrinsic carrier density n i .. with the spin-up and spin-down carriers playing roles similar to those of the electrons and holes in bipolar diodes. Si-based p-i-n diode sandwiched between two ferromagnetic metals was suggested to allow controlling the device performance by an externally applied magnetic ﬁeld (Dugaev et al. typically 1 V. 2002) were performed by self-consistently solving for the driftdiffusion. which are enhanced over those of metallic systems by the exponential dependence of the current on bias voltage. The magnetic bipolar diode116 (MBD) is a p-n junction diode with one or both regions magnetic (Fabian ˇ ´ et al.. 2002b. or spin capacity—the effect of changˇ ´ ing. The n region (right) is nonmagnetic. I-V characteristics of the diodes. Zutic et al.2 the terms majority and minority refer to the relative carrier (electron or hole) population and not to spin. by which we mean that it has a spin-split conduction band with the spin splitting (Zeeman or exchange) 2q k B T. continuity. The role of inhomogeneous doping in the p-n junction is played by the inhomogeneous spin splitting of the carrier band.. dependent on the splitting. 2003) lie in the combination of equilibrium magnetism and nonequilibrium spin and effective methods to manipulate a minority carrier population. 2002a) jn n0 e qV/k B T 1 P n P n0 1 . not both)..4. but electrons can be spin polarized by a spin source (circularly polarized light or magnetic electrode).C).C. 2002a. (2002a) for the magnetic case. While the numerical calculations are indispensable in the high-injection limit. the current drops signiﬁcantly. which are ordinary diodes in a magnetic ﬁeld.. The I-V characteristics of such diodes depend on the magnetic ﬁeld through small orbital effects on diffusion coefﬁcient. The advantages of magnetic bipolar spintronic ˇ ´ devices (Fabian et al. The p region is magnetic. Fabian. II. and Das Sarma: Spintronics: Fundamentals and applications demonstrated as effective spin injectors (Sec. 2002a). 2001a). Here and in Sec.117 valuable insight and analytical formulas can be obtained in the low-injection limit. Zutic et al. as well as carrier recombination and spin-relaxation equations. a subﬁeld of spintronics in which both electrons and holes take part in carrier transport.A. 30.A. 30. Zeeman splitting can be signiﬁcantly enhanced by the large g * factors of magnetically doped (Sec. IV. (114) where V is the applied bias (positive for forward bias) and n 0 ( ) (n 2 /N a )cosh(q /kBT) is the equilibrium i number of electrons in the p region. A scheme of an MBD is shown in Fig. 1950) for ordinary p-n junctions was generalized by Fabian et al. IV. 8). This occurs at forward biases comparable to the built-in potential. The dependence of the electric current j on q and P n was obtained by both numerical and analytical ˇ ´ methods.. Mod. The n region is nonmagnetic. nonequilibrium spin density (Zutic et al. If the relative orientation is antiparallel ˇ ´ (bottom).D. Vol. 2002). Examples of nonmagnetic bipolar spintronic devices are the spinˇ ´ polarized p-n junction (Zutic et al. spin ampliﬁcation (see Sec. indicated by the spin splitting 2q of the conduction band. IV. 2. Adapted from Zutic et al. but spin polarized by a spin source: Filled circles. 2002. It is straightforward to generalize these considerations to include spin-polarized holes (Fabian et al. The p region (left) is magnetic. at the edge of the depletion layer.3) or narrow-band-gap semiconductors (Sec. Numerical calculations (Zutic et al... If the nonequilibrium spin in the n region is oriented parallel (top ﬁgure) to the equilibrium spin in the p region. spin-polarized electrons. . by voltage. Equation (114) generalizes the SilsbeeJohnson spin-charge coupling (Silsbee.B). Using an MBD with a ferromagnetic semiconductor slightly above its T C is also expected to give large g * factors. Johnson Not to be confused with the usual magnetic diodes. 2002a. 2003). The interplay between the equilibrium spin of polarization P n0 tanh(q /kBT) in the p region. while either electrons or holes (or both) are spin polarized (see Sec. 2001b). April 2004 116 117 The small-bias or low-injection limit is the regime of applied bias in which the density of the carriers injected through the depletion layer (the minority carriers) is much smaller than the equilibrium density of the majority carriers.3). To illustrate the I-V characteristics of MBD’s. 2002. consider the low-injection limit in the conﬁguration of Fig. Phys. (Color in online edition) Scheme of a magnetic bipolar diode. II. and the nonequilibrium spin source of polarization P n in the n region. discussed in Sec. ˇ ´ Finally.. The most useful effects of the spin-charge coupling in MBD’s are the spin-voltaic and the giant-magnetoresistive effects. The electron contribution to the total electric current is (Fabian et al..378 ˇ ´ Zutic. and the acceptor doping N a .
D. For parallel nonequilibrium (in the n region) and equilibrium spins (in the p region). (2003) have shown that the current in p-CoMnGe/n-Ge magnetic heterojunction diodes can indeed be controlled by magnetic ﬁeld. and Das Sarma: Spintronics: Fundamentals and applications 379 and Silsbee. (114) (Zutic et al.. In the latter case. Indeed. 2002a). The GMR effect should be useful for measuring the spin reˇ ´ laxation rate of bulk semiconductors (Zutic et al. 31. In unipolar systems. Van Dorpe. Spin splitting 2q on the p side is normalized to k B T. 2002a). originally proposed for ferromagnet/ paramagnet metal interfaces. The advantage of the spin-charge coupling in p-n junctions.. Arata. (1968)119 were perhaps the ﬁrst to show that a ferromagnetic p-n junction. where optical detection is ineffective. making the ﬂow of one spin species greater than that of the other.. P n 1. Equation (114) can be understood qualitatively from Fig.8 V for the diode from Fig.118 FIG. This is an example of the spin-voltaic effect (a spin analog of the photovoltaic effect). 2002a). (Color in online edition) Giant magnetoresistance (GMR) effect in magnetic diodes. Vol... in which nonˇ ´ ˇ ´ equilibrium spin causes an emf (Zutic et al. If.. Kohda et al.. The current is symmetric in . et al. Rev. Adapted ˇ ´ from Zutic et al. electrons experience a larger barrier and the current is inhibited.. and to observe the above predicted phenomena. MBD’s can display an interesting GMR-like effect. 31 by the strong asymmetry in j. and fully polarized. ˇ ´ which follows from Eq. 76. 2003). Fabian. (114)] a nondegenerate carrier den- (ii) This could be a way to detect spin injection into Si.3). allowing for the exponential enhancement of the effect with increasing V. is the nonlinear voltage dependence of the nonequilibrium carrier and spin densities (Fabian et al. Recently. Materials parameters of GaAs were applied.. magnetic p-n junctions have already been demonstrated. the current is a strongly asymmetric function of . Heavily doped p-GaMnAs/n-GaAs junctions were fabricated (Ohno. Arata et al. 1 – 10 %. The solid curve corresponds to a switchedoff spin source. 2002). n region. This is demonstrated in Fig. Mod. 2002. 2001. as well as for detecting nonequilibrium spin in the nonmagnetic region of the p-n junction. Zutic and Fabian. as a function of 2q /k B T for both the unpolarized. increasing the current. P n0 P n 0. unless the effective g factor is 100 at B 1 T (Sec. no current ﬂows through the depletion layer. Incorporation of (Ga. To have functioning MBD’s at room temperature. With spin source on (the extreme case of 100% spin polarization injected into the n region is shown). 2.B). No. For a strong spin-charge coupling [recall the discussion of Eq. 30). II. In the opposite case of antiparallel relative orientation. Although practical MBD’s are still to be fabricated and the predicted effects tested. shown by the dashed curve.Mn)As layer in the intrinsic region of p-i-n GaAs diode was shown to lead to an efﬁcient photodiode. Phys. In addition. The use of ferromagnetic semiconductors is limited by their T C (Sec. Figure 31 plots j. 2000.. the difference between j for parallel and antiparallel orientations.. could act as a diode.5. 2001. If P n 0.ˇ ´ Zutic. 2003). The corresponding GMR ratio. 30 (Fabian et al. In equilibrium. 30. 2003).. the relative change is 66%. to the case of magnetic p-n junctions. for example. 2002. increasing exponentially with increasing due to the increase in the equilibrium minority carrier density n 0 ( ). where transport is due to the majority carriers. Current/spin-splitting characteristics (I ) are calculated self-consistently at V 0. several important challenges have to be met: (i) Zeeman or exchange splitting needs to be sufﬁciently large to provide equilibrium spin polarization. the current will depend on the sign of P n0 • P n . which also includes the contribution from holes. Tsui et al. This may be difﬁcult at room temperature. Wen et al. as the electron currents from both sides of the junction balance out. Johnston-Halperin et al. The balance is disturbed either by applying bias or by selectively populating different spin states. In the ﬁrst case j is a symmetric function of . as the majority carrier density is ﬁxed by the density of dopants. Field for bringing this reference to our attention.. 2002. 1985). Current can ﬂow even at V 0 when P n 0. can also be calculated analytically from Eq. et al. IV. P n 0 and V 0. Liu. such a modulation of the current is not likely. It would be interesting to see such devices combined with a spin injector in the bulk regions. (114) as 2 P n P n0 /(1 P n P n0 ) (Fabian et al. in which the Mn ions function as recombination centers (Teran et al. 2003) to demonstrate tunneling interband spin injection. the direction of the zerobias current is controlled by the relative signs of P n0 and Pn . most electrons cross the depletion layer through the lower barrier (see Fig. displaying large GMR. 30). the effective barriers for crossing of electrons from the n to the p side is different for spin-up and spin-down electrons (see Fig.. 119 . based on the ferromagnetic semiconductor CdCr2 Se4 doped with Ag acceptors and In donors. The current depends strongly on the relative orientation of the nonequilibrium spin and the equilibrium magnetization. P n 0. April 2004 118 We thank M. as opposed to metals or degenerate systems.
requiring high carrier mobility and long spin relaxation (easier to realize for spinpolarized electrons). The device is based on spin injection and spin detection by a ferromagnetic source and drain. Vol. where the Coulomb blockade becomes important (Takahashi and Maekawa. will need to be understood. 1990. There is also a large category of spin singleelectron transistors. Fabian. (i) The effective spin injection of spin-polarized carriers from the ferromagnetic source into a 2DEG is nontrivial (see Sec. 1989). The parameter BR should be effectively controllable by the gate. Spin single-electron transistors can be viewed as an extension of magnetic tunneling (see Sec. E. which. Ballistic spin-polarized transport should be realized through the channel with uniform BR by eliminating undesirable electric ﬁelds due to interface inhomogeneities.. (1996). ﬁrst realized by Ono et al. If is the angle between k and the source-drain axis.E. The samples need to be smaller than the spin diffusion lengths. The average spin component along . strong spin relaxation in the depletion layers... The effects of actual device structures. 2002a) for 11 eV m. where the current modulation factor is cos2( /2). The average spin cessing at the angle at the drain in the direction of magnetization is s (t )cos s0 m•(k n). and Martinek et al. Another option is using Si heterostructures. ´ and later investigated by Barnas and Fert (1998). III. The structure inversion asymmetry should dominate over the bulk inversion asymmetry. (2002). the modulation effect decreases. such as two. and a spin parallel to the in-plane k. IV. IV. and Das Sarma: Spintronics: Fundamentals and applications (iii) (iv) (v) sity is desirable. (2002). Various spin transistors that contain metallic (and insulating) regions have been proposed (Johnson. a three terminal device which offers spindependent ampliﬁcation. and the hot-electron spin transistor. etc.. 1988. An effective integration is needed of magnetic and nonmagnetic structures into single devices (Samarth et al. 76. Spin ﬁeld-effect transistors Datta and Das (1990) proposed what became the prototypical spintronic device scheme.and/or three-dimensional spin ﬂow.B.2). is not easily realized in many other ferromagnetic semiconductors that are typically heavily doped (Sec.. 1990) that 0 using a one-dimensional electron motion along channel as a waveguide. The largest effect is seen for 0. in which bulk inversion asymmetry is absent. Consider a 2D electron gas conﬁned along the plane Rev. 2 BR k and the injected spin at s s 0 cos( t). Equation (88) determines the evolution of the expectation value for a spin perpendicular to the plane. quasi-onedimensional channel of an ordinary ﬁeld-effect transistor. You and Bader. The precession axis of always lies in the channel plane (see Sec. s s•k/k: ds n /dt 2 BR ks . and the spin precession rate must be large enough (ii) (iii) (iv) . Bauer et al. the probability of ﬁnding the spin in the direction of magnetization m. The Datta-Das SFET is yet to be realized. of the unit vector n. 2003). Zvezdin et al.1m e . There are at least four important difﬁculties in observing the proposed effects. Das et al. ds /dt 2 BR ks n . 2003..D. in the asymmetric. Three important cases are discussed in detail: the Datta-Das spin ﬁeld-effect transistor. and m c 0..A.Cr)Te. setting the scale for ballistic transport. The structure of the Datta-Das SFET is shown in Fig. 2003). Taking typical values (Nitta et al. 1988) like InAs. Eq. current moduBR 1 10 lation should be observable at source-drain separations of L 1 m.2). s s•(k n)/k. Note that does not depend on the momentum (or energy) of the carriers.. in which the structure inversion asymmetry dominates the spin precession (Luo et al.. with the spin s pre2 BR m c L/ . Koga et al. Mod. (2002). Phys. s n s•n. (88). but by gate bias. As a result. The attractive feature of the Datta-Das SFET is that spin-dependent device operation is controlled not by external magnetic ﬁelds. the Datta-Das spin ﬁeld-effect transistor (SFET). the magnetic bipolar transistor. the small magnitude of the spin-orbit interaction makes BR in Si probably rather weak. 2000. For a review of spin single-electron transistors see Maekawa et al. is constant. However.B). 1. Korotkov and Safarov (1999). Ciorga et al. 1993a. (115) By combining two magnetic p-n junctions in series one can obtain a magnetic bipolar transistor (Sec. April 2004 where BR is the structure inversion asymmetry coefﬁcient appearing in Eq.4). IV. Spin modulation of the current becomes ineffective if transport is diffusive. No. It was therefore proposed (Datta and be limited by further conﬁning the Das. so the current is modulated by 1 cos2 sin2( /2). The device will work best with narrow-gap materials (Lommer et al. the electron will reach the drain at time t Lm c /( k cos ). so the results (unlike those for bulk inversion asymmetry) are insensitive to the relative orientation of n and the principal crystal axes. As the spread in the momenta increases. interface contacts. 2. II.2) to double tunnel junctions. and on spin precession about the built-in structure inversion asymmetry (BychkovRashba) ﬁeld . 1. where the source is labeled with zero.380 ˇ ´ Zutic. Spin transistors We review several proposals for spin transistors that have at least one semiconductor region and that aim at integrating spin and charge transport within traditional device schemes of the ﬁeld-effect and junction transistors. while likely in (Zn. spin-dependent band offsets and band bendings.. 1997. (88). 1998). which controls the spin precession rate.
Model calculations using the tight-binding formulation of H SIA (recall Sec. These four factors present a great challenge to fabricating a Datta-Das SFET at room temperature. 1997. Holes form mostly the base current. It is not surprising that the conductance through the transistor. Materials considerations discussed in Sec. nonmagnetic emitter E. . limiting the design to special materials and very clean interfaces. and in an extensive review by Bournel (2000). (2003). 30.. has been considered by Schliemann et al. ﬂowing to the emitter. while the B-E junction is reverse biased (V BC 0). Wang. (2002b) and Fabian ˇ ´ and Zutic (2004). was investigated by Winkler (2004) using k•p model calculations.2) were reported by Pareek and Bruno (2002). Ciuti et al.53Ga0.B. 2002b.. Applying the generalized Shockley theory to include spin effects (Fabian et al. Such a three-terminal device can be thought of as consisting of two magnetic p-n junctions connected in series. in the present orientation-dependent bulk inversion asymmetry. IV. In an MBT also depends on the spin splitting 2q (see Fig. Electrons ﬂow from E to B.. I B . 2004.D also apply to an MBT in order to provide a sufﬁcient equilibrium polarization in a magnetic base P B0 . and structure geometry. and Das Sarma: Spintronics: Fundamentals and applications 381 ( BR 10 11 eV m) to allow at least a half precession during the ballistic transport. Mod. even in the absence of ferromagnetic regions which are replaced by a rotating external magnetic ﬁeld of uniform strength (Wang. This additional dependence of in an MBT is ˇ ´ called magnetoampliﬁcation (Fabian and Zutic. is attached to the base/collector junction Rev. the emitter has a nonequilibrium polarization P E from a spin source. including the bulk inversion asymmetry. I C /I B (see Fig. similar to the magnetic diode case in Fig. the modulation of BR by biasing voltage (iii) has been already convincingly demonstrated in In0. Adapted from Fabian et al. where they either recombine with holes (dashed lines) or continue to be swept by the electric ﬁeld in the B-E depletion layer towards C. Only the spin polarization of electrons is assumed. see also Miller et al. and collector C. The directions of the current ﬂows are indicated. While nonmagnetic. For more discussion of the Dresselhaus bulk inversion asymmetry and the Bychkov-Rashba structure asymmetry see Sec. ˇ ´ Fabian and Zutic. 2002a). where the transistor can amplify currents.48As quantum wells (Nitta et al. and Guo. to the collector (Fabian et al. Phys. The current ampliﬁcation (gain) 32) is typically 100 in practical transistors. 76. through the base. 32. 2004).. a theory of MBT was developed by Fabian et al. In the so-called forward active regime. depends rather strongly on the crystallographic orientation of the two-dimensional channel (Łusakowski et al. in general. 2003).ˇ ´ Zutic. Grundler. The relative orientation of the gates’ magnetization leads to magnetoresistance effects. (2003) and Lebedeva and Kuivalainen (2003). simpliﬁed schemes of MBT [not including the effect of nonequilibrium spin ( P E 0)] ´ were also considered by Flatte et al. II. III. 2004. (1997) in a push for silicon-based spintronics. (2002b) proposed a ferromagneticoxide-semiconductor transistor. leading to equilibrium spin polarization P B0 tanh(q /kBT). No. and another ferromagnetic metal. Vol. 2004) in order to make possible an effective control of by P E .52Al0. 2. The Datta-Das SFET has generated great interest in mesoscopic spin-polarized transport in the presence of structure inversion asymmetry. (1999).B. the E-B junction is forward biased (here with voltage V BE 0 lowering the built-in potential V bi ). Carriers and depletion regions are represented as in Fig. 2003). Later. FIG. (2002).b. a spin detector. carrier lifetimes. with a nonmagnetic source and drain. April 2004 to modulate the current ﬂow. but with two ferromagnetic gates in series above the base channel. Faˇ ´ bian and Zutic. Further theoretical investigations on the theme of the Datta-Das ´ spin transistor can be found in Nikolic and Freericks (2001) and Matsuyama et al. 32) and the nonequilibrium polarization P E . The conduction band has a spin splitting 2q . In both conﬁgurations the aim is to control current ampliﬁcation by spin and magnetic ﬁeld. Fabian. Hu et al. 2000) (for GaAs/ AlGaAs 2DEG. A scheme of a particular MBT is shown in Fig. while the spin source is provided by a ferromagnetic spin injector attached to the emitter.. (Color in online edition) Scheme of an n-p-n magnetic bipolar transistor with magnetic base B.2. 2004). Conduction and valence bands are separated by the energy gap E g . 32. 2. in the presence of both bulk and structure inversion asymmetry. In this proposal (also called SPICE for spin-polarized injection current emitter) the semiconductors have no equilibrium spin. 30. 2003). 2002b. Distinct SFET’s have also been suggested. Magnetic bipolar transistor The magnetic bipolar transistor (MBT) is a bipolar transistor with spin-split carrier bands and.. such as the doping densities. The current ampliﬁcation I C /I B can be controlled by P B0 as well as by the nonequilibrium spin in E.. This ratio depends on many factors. An important prediction is that nonequilibrium spin can be injected at low bias all the way from the emitter. 1999. Initial experimental investigations of magnetoresistance in the Datta-Das SFET systems were performed by Gardelis et al. An SFET that can operate in the diffusive regime.47As/In0. However. an injected spin source (Fabian et al. A related device structure was already proposed by Gregg et al. Recently spin precession in the Datta-Das SFET... diffusion coefﬁcients.
Zutic and Fabian. In the limit of slow carrier recombination (Fabian et al. However. or collector. 2003). inﬂuence how many of the injected electrons can overcome the B/C Schottky barrier and contribute to the collector current I C . both elastic and inelastic scattering processes determine the effective spin-dependent mean free path. resulting in a spin-ﬁltering effect which can be described in terms of a spin-dependent mean free path (Rendell and Penn.and energydependent interfacial properties. which are injected into a base region as hot carriers.D). IV. 3(a)] surrounded by a silicon emitter (E) and collector (C) with two Schottky contacts.120 On the other hand. Si (solid) and GaAs (dashed) materials parameters were applied. 2002). Fabian. 2000). The MBT is. Hong and Mills. (116) for the case of P E 0]. in effect. a semiconductor bipolar transistor. Below we discuss two representative examples. 2000). and electronic structure. The sensitivity of the current to spin can be used to measure the injected spin polarization. since its functionality depends on tunability of the band structure of the emitter. unless at least two regions are magnetic. The advantage of the MBT. which also has an emitter/base/collector structure. hot-electron transistors are also of interest for their ability to sense magnetic ﬁelds.D. 2001).. 2000. a magnetic heterostructure transistor. 2. In Si it is the spin injection at the emitter-collector depletion layer which controls the current. Hot-electron spin transistors different purposes. Smith et al. Spin transistors that rely on transport of hot (nonthermalized) carriers have the potential of serving several Rev. Mod. scattering in the base region strongly depends on the relative orientation of the magnetizations in the ferromagnetic layers. but can be tuned during the device operation by magnetic ﬁeld or spin injection signals. cosh q /k B T 1 P E P B0 . The calculated dependence of the gain on the spin splitting for P E 0. A subsequent variation—a ballistic electron-magnetic microscopy. The scattering processes in the base. On the one hand.382 ˇ ´ Zutic. 1999. their possible memory applications. The only remaining effects on the I-V characteristics come from the sensitivity of the carrier densities in the equilibrium magnetic regions to [see Eq. which also uses an STM tip to inject hot carriers. 121 Similar to other hot-electron spin devices. Phys. 76. Adapted from Fabian et al. typically has a sizable current gain—a small change in the base current leads to a large change in the collector current (see Sec. 1997. scattering processes. 33. there is no spin-charge coupling in the spacecharge regions. 1988. Generally. an implication of the spin-voltaic effect ˇ ´ ˇ ´ (Zutic et al.. 1995. 2002b. only a small current gain 2 (due to large current in a metal base) was predicted in magnetic tunnel-junction-based devices (Hehn et al. The gain is very sensitive to the equilibrium magnetization in Si. a spinvalve transistor and a magnetic tunneling transistor. is capable of resolving magnetic features at a 10-nm length scale (Rippard and Buhrman. relevant to spintronic devices. 2002). . formed at E/B and B/C interfaces. 3. (116) Both magnetic ﬁeld (through ) and nonequilibrium spin affect the gain. 2002. The challenges to demonstrate the predicted phenomena in MBT are similar to those of magnetic bipolar diodes. IV. given in units of thermal energy k B T. and Das Sarma: Spintronics: Fundamentals and applications FIG. however.9 is shown in Fig. If no spin source is present ( P E 0). The spin-valve or Monsma transistor provided an early demonstration of a hot-electron spin transistor and realization of a hybrid spintronic device that integrates metallic ferromagnets and semiconductors (Monsma et al. while the rapid carrier recombination in GaAs prevents more effective control of the transport across the base...2). Pappas et al. As the spin-charge coupling is most effective across the depletion layer (see Sec. 1998). Calculated gain dependence of an MBT as a function of base spin splitting 2q . Electrons with spin which has magnetic moment opposite (antiparallel) to the magnetization of a ferromagnetic layer typically are scattered more than electrons with parallel magnetic moment. they could be used as a diagnostic tool to characterize spin. Levy and Mertig. No.. together with the reverse bias V BC . April 2004 120 These efforts are motivated in part by the success of (spininsensitive) ballistic electron-emission microscopy in providing high spatial and energy resolution of the properties of metal/ semiconductor interfaces (Kaiser and Bell. for GaAs and Si materials parameters. 2002b). base.. Bonnell. this coupling is essential for the current in Si.9.. 122 Another realization of a spin-valve transistor combines a GaAs emitter with a Si collector (Dessein et al. In particular. A three-terminal structure121 consisted of a metallic base (B) made of a ferromagnetic multilayers in a CPP geometry [as depicted in Fig. see Sec. 1991.E. Vol. The nonequilibrium spin polarization in the emitter is P E 0. Similar to the physics of GMR structures (Gijs and Bauer.122 Forward bias V EB controls the emitter current I E of spin-unpolarized electrons. and their potential as a source of ballistic hot-electron spin injection.. 2000). 33. 1980. IV.. is that the band structure is not built in. the term transistor characterizes their three-terminal structure rather than the usual functionality of a conventional semiconductor transistor.
2003b). 2001.2 layer. Together they form a magnetic tunnel junction.. after passing an F1 layer of thickness t acquire an effective transmitted polarization P N1 N↑ N↓ N↑ N↓ e e t/l ↑ t/l ↑ e e t/l ↓ t/l ↓ . Different coercive ﬁelds in the regions 1 and 3 ensure independent switching of the corresponding magnetizations in the applied magnetic ﬁeld. 2003).. Schematic energy diagram of a magnetic tunneling transistor. deﬁned in Eq. not limited to energy selection (via Schottky barriers) of the carriers injected into the base and collector regions. This was ﬁrst achieved (Jiang et al. van Dijken et al... In GaAs. Mod. 1998. which can be determined by measuring the base layer thickness dependence of the collector current (see Rippard and Buhrman. need to be addressed to better realize the potential of hotelectron transistors. raised by the operation of the spin-valve transistor. i 1. van Dijken et al. region 2 the Al2 O3 tunnel barrier of height . there are indirect minima at L points at higher energy (Blakemore. 2003b). Phys. Sato and Mizushima. 12. Several important challenges. (118) For electrons with sufﬁciently high excess energy. important for the device performance. 76. 2000) demonstrate a capability for magnetic-ﬁeld sensors. After an initial decrease of MC with electron energy. 2003c.. injected from the emitter. Sato and Mizushima..4 K and B 2.3 V larger than the base/collector Schottky barrier there is an onset of hot-electron transport into L valleys FIG. (117). 1982)... 2002a. more similar to the original spin-valve transistor. Yamauchi and Mizushima. where ↑↑ (parallel) and ↑↓ (antiparallel) denote the relative orientation of the magnetizations. 2000. is not sufﬁcient to demonstrate spin injection in a semiconductor collector. 2003c).ˇ ´ Zutic. April 2004 123 where N ↑ and N ↓ represent the number of transmitted electrons with majority or minority spin and l ↑ and l ↓ are the corresponding attenuation lengths (the polarization P N2 has an analogous form). and Das Sarma: Spintronics: Fundamentals and applications 383 sometimes also referred to as the attenuation length. the ratio I C /I E . van Dijken et al. at V EB 0. shows a nonmonotonic behavior with V EB (van Dijken et al. 2001. Unpolarized electrons. the emitter was nonmagnetic (Cu) while the base was a magnetic multilayer (F1/N/F2) (van Dijken et al. after a background subtraction.. A particular realization is depicted in Fig. 34. For conclusive evidence spin detection in a collector region is needed.. combined with a variable V EB .123 The magnetoresistive response is usually expressed using magnetocurrent (MC). In another realization of a magnetic tunnel transistor. a scattering process (inﬂuencing the mean free path) does not necessarily remove the electron from the collector current. 1998. These challenges include increasing the small collector current and determining whether the spin injection of hot carriers into semiconductors is feasible.. it would be desirable to fabricate structures in which semiconductor regions played an active role. often referred to as magnetic tunneling transistors. Vlutters et al. From van Dijken et al. At large V EB bias. Fabian. 2003a. 34. The resulting strong spin-ﬁltering effect can be inferred from the transmitted hot carriers with a spindependent exponential decay within the Fi . 2001. Furthermore.. Measurements at T 1. in addition to the direct conduction-band minimum at the point [recall Fig. The resulting magnetocurrent can be expressed as (van Dijken et al. measured in various hot-electron spin transistors (Monsma et al. 125 .. The magnetocurrent MC. The addition of a tunnel junction. A large magnetocurrent alone. in which a GaAs collector served as an n-type spin aligner and InGaAs/GaAs was used for a quantum well. 1995. 2. Region 1 is the emitter. The large values124 of MC ( 200%) and the sensitivity of 130% per G measured at room temperature (Anil Kumar et al. 1999.5 T. The attenuation length. (117) analogous to the expression for GMR or TMR structures [recall Eq. (119) Analogous to the spin LED from Fig.. deﬁned as the change in collector current. can be substantially increased over that of the spin-valve transistor (Sato and Mizushima. showed majority spin injection with P circ 10%. Region 4 is a semiconductor collector that has a Schottky barrier at the interface with the base. inﬂuenced by the conduction-band structure of a collector.. 2003c). An alternative class of hot-electron transistors. Jiang et al. 2003b). 2003b) MC 2P N1 P N2 / 1 P N1 P N2 . No. accompanied by an increase in MC (van Dijken et al. 2003c. Vol.. 34. 2001. Rev. allows exploration of hot-electron transport over an energy range of several eV. 6(a)]. and region 3 the base. van Dijken et al. 1997. 124 These values substantially exceed the CPP giant magnetoresistance value for the same magnetic multilayer used in the base. has a tunneling junction instead of a Schottky barrier emitter (Mizushima et al. 2002b) can therefore differ from the effective mean free path. 2002b. normalized to the minimum value MC I C↑↑ I C↑↓ /I C↑↓ . (1)]. 2003) using optical detection with a spin LED structure125 added to the collector in Fig.
A spin to charge conversion allowed for an electrical readout of the spin. quantum dots). 2002. 2001). microseconds or longer in operational experimental conditions (de Sousa and Das Sarma. Elzerman et al. 2003) or spin-selective single-electron-transistor spectroscopy (Ono et al. 2003b).8 V and at T 77 K. while with Eq. The control of spin dynamics and entanglement [many-spin quantum correlations (Nielsen and Chuang. the quantum analog of the binary bit in a classical digital computer. 35) is to manipulate the spin states of a single electron using external magnetic ﬁelds (or microwaves) for single-qubit operations and to utilize the quantum exchange coupling between two neighboring electrons to carry out two-qubit operations. i. 1998). 35.. the measured MC exceeds 3400%. 2000)] at the single-spin level in a semiconductor quantum dot structure is a formidable task. No. and Das Sarma: Spintronics: Fundamentals and applications ` which is analogous to Eq. (2002). Levy (2002). Fabian. (119) the polarization of the transmitted electrons can be estimated to exceed 90%. Kane (1998. Rev. et al. State-of-the-art techniques to measure a single spin in a solid. a quantum system can be in a superposition of all the basis states. Spin qubits in semiconductor nanostructures A potentially revolutionary idea in spintronics is the possibility of using the two-level nature of electron spin to create a solid-state quantum computer (DiVincenzo. 2003a).126 A common theme in these proposals is the idea of manipulating the dynamics of a single (or a few) electron spin(s) in semiconductor nanostructures (e. April 2004 126 spin-based quantum computing employ GaAs quantum dots (Loss and DiVincenzo. II. Sec. even with a ferromagnet only 3 nm thick (van Dijken et al.j (t)Si •Sj where the ﬁrst summation. effectively performing (via a unitary evolution) many classical computations in parallel. are still not sensitive enough for quantum computing operations. At V EB 0. Privman et al. At the end of the trapping time the spin was measured by quickly shifting the Zeeman resolved spin states towards the Fermi energy. 2001). While the dimensionality (2 n ) of the Hilbert space of n electron spins is the same as the number of conﬁgurations of a corresponding classical system. Loss and DiVincenzo (1998). Future studies of hot-electron spin transistors are expected to result in increased spin injection even at room temperatures and to utilize other semiconductor collectors. which has not been achieved even at mK temperatures. The scheme works according to the timedependent Hamiltonian H(t) B g Bi (t)Si . Single-qubit operations are realized by single-spin precessions (circles). 2003). Mod. The ﬁgure shows two electrons localized in the regions deﬁned by the gates (shaded). such as magnetic resonance force microscopy (Sidles et al. describes the local exchange interaction (J is the exchange coupling). Vol. 1995.j J i. to be contrasted with typical picosecond coherence times for charge or orbital states in solids. Burkard et al. 2004). 2. 1995. but with the redeﬁned deﬁnition of spin polarization. Das Sarma. Meier et al. Several ‘‘spin-based’’ quantum computer schemes have been proposed and extensively studied.384 ˇ ´ Zutic. 2002). 2000. Interest in spin-based quantum computing will . 2003. A qubit is essentially a controllable quantum two-level system (Nielsen and Chuang.. (2) for TMR using Julliere’s model.. (2003). Piermarocchi et al. although impressive experimental advances have recently been made (Fujisawa et al. Barbic. The real motivation for using the spin as a qubit is its long coherence time. DiVincenzo (2000). The basic idea (see Fig. with coupling between electron spins—via the exchange interaction—allowed by tunneling between the dots.. The basic unit in a quantum computer is the quantum bit (or qubit). 1998) or Si (or Si-Ge) systems (Kane. Phys. 2000). Mamin et al. 76. et al.. A theoretical analysis of spin injection and spin ﬁltering in magnetic tunneling transistors was given by Rashba (2003a) who extended the approach for ballistic spin injection (Kravchenko and Rashba. Friesen et al. for example. Electrons are localized in electrostatically deﬁned quantum dots. The current architectures for FIG. Skinner et al. (1999).g.. It would be particularly desirable to demonstrate hot-electron spin injection in Si and facilitate an integration with the complementary metal-oxide semiconductor technology. (2003). (2000). 2001). (2003). A single electron with unknown spin was trapped in a quantum dot for a few milliseconds. recently a single shot readout of the spin of an individual electron has been demonstrated using an electrical pump-probe measurement (Kouwenhoven. with the reasonable hope that the predicted behavior will extend to many-spin systems. over all neighboring spin pairs. Troiani et al. 2003. performed by applying local magnetic ﬁelds (here perpendicular to the page) to each dot. with different variations. Hu and Das Sarma (2000. Vrijen et al. Das Sarma. Two-qubit operations are done through the exchange interaction indicated by the dashed curves. F. (Color in online edition) The Loss and DiVincenzo (1998) proposal for spin-based solid-state quantum computing. (2003). Nielsen and Chuang. (2000). Koiller et al. 2003).. The tunneling is controlled by gate voltage.2) to include the effects of hotelectron transport and inelastic scattering. 2000.. requisite for practical quantum computation. while the second describes the single-spin operations by local magnetic ﬁelds.. (2002..C. Variations to this scheme are described by Burkard et al. However. See. (1998). Hanson. 2002. Witkamp.
1968). 128 There still remain many challenges in accurately predicting Curie temperature. Smet et al. The early work of Lampel (1968). Future progress in spin-polarized transport will be largely driven by the materials advances. and relaxation. Electron-nuclear spintronics is likely to become relevant for quantum computation and for few-spin manipulations. A Small signals attributed to spin injection have already been reported (Jia et al. which can beneﬁt from long nuclear spin coherence times (even lasting minutes). with the basic principles well understood. In the context of semiconductors. for example. The broad subject of spin-based quantum computing. single-spin relaxation and decoherence. the control of spin in silicon would raise hopes for seamless integration of spintronics with the current information technology. V. calculation of which from ﬁrst principles is a nontrivial task that has not been accomplished even for the elemental metals. see Sec. the interaction between electron and nuclear spins (Fleisher and Merkulov. Phys.D. There are many challenges and open questions to be tackled by future research. First principles results suggest that dominant models of ferromagnetism in semiconductors cannot be used to explain a variation of Curie temperature across differˇ ´ ent materials (Erwin and Zutic. for example. which is 23% below the threshold current of the spin-unpolarized . Nagaev (1983). Paget and Berkovits. is beyond the scope of this review. and their relation to the ensemble spin dephasing. (2003). An important and still debated issue is spin relaxation and decoherence of localized or conﬁned electrons.. the small magnitude of the spin-orbit interaction and the absence of inversion symmetry lead to relatively long room-temperature spin lifetimes (of about 10 ns. is by itself exciting and worth pursuing. here two InGaAs quantum wells. emitting circularly polarized light (Sec. 1984b. even though many proposed spintronic device schemes may turn out to be impractical in the end. whether they can achieve higher mobility. Kikkawa and Awschalom. The role of the resonant radio waves is played by periodically optically excited electron spins (Kalevich et al. which is related to the areas of quantum measurement and quantum decoherence (Zurek.. need to be pursued further in the context of quantum computing. April 2004 127 comprehensive transport calculation in the actual devices with realistic electronic structure of the studied materials would provide valuable insights into both the spin polarization being measured and how it is reduced from the moment it was generated. the fundamental spin physics. Sanvito et al. is a systematic understanding of the effects of magnetic interfaces and materials inhomogeneities on spin-polarized transport. emphasizing both the fundamental aspects of spin dynamics. 2.5 A cm 2 .128 and a simple fabrication of high quality interfaces with nonmagnetic materials. 1981. higher Curie temperature. 76.3As (Ekimov and Safarov. While dynamic nuclear polarization induced by electron spin can often be a nuisance for detecting intrinsic spin dynamics (Sec. OUTLOOK We have reviewed selected topics on spintronics. III. et al.7Al0. 2002. considering all-semiconductor structures rather than the hybrid structures with metallic ferromagnets will depend on the improvements in ferromagnetic semiconductors. What is missing. Important materials advances have been realized in improving the compatibility of Si/III-V structures (Sieg et al. (2003). (2002). Rudolph et al. Timm (2003). even in the currently available materials.. in particular a robust spin injection into silicon. Fuchs. 1998). Witkamp. Mod. which are effectively unpolarized. as shown ﬁrst in p-doped Ga0. suggesting that the existing control of spin in GaAs or in III-V ferromagnetic semiconductors might be extended to Si. In addition. Furthermore.D. The laser is a vertical-cavity surfaceemitting laser (VCSEL). The same can be said of the g factor. II. an NMR signal can be detected optically by measuring changes in the circular polarization of photoluminescence due to resonant variations of the nuclear ﬁeld (D’yakonov and Perel’. Das Sarma et al. in both metals and semiconductors. transport.ˇ ´ Zutic.. for example. relaxing some constraints on the operational length and time scales. 2004). optically pumped in the gain medium. extracted from the pump power for the lasing operation. Vol. 2000. Salis.3). 1972). and the potential applications. On the other hand. The electrons recombine with heavy holes. have demonstrated the operation of a spin laser. their importance lies in stimulating interesting experimental and theoretical research. Spin relaxation and spin dephasing of conduction electrons is a mature ﬁeld. What is needed are accurate bandstructure-derived calculations of spin relaxation times. and Ekimov and Safarov (1972) established the basic principles for a series of experiments that demonstrated various realizations of an all-optical NMR. 2003) is of fundamental importance for spintronics. 1980. 1986. 2001). where the hyperﬁne-interaction mechanism dominates. Konig et al. A ﬁrst step towards understanding single-spin relaxation is the recent experiment of Hanson. While the current push for spintronics is driven by the prospect of technological applications. 1984). et al. Bhat et al. 2003). and Das Sarma: Spintronics: Fundamentals and applications 385 only increase as we understand more about spin coherence and relaxation from other spintronic studies. (2003) in a oneelectron quantum dot. Fabian. 1984. Furthermore. Kalevich. 1984a. Rev. which has a longstanding tradition in the solid-state community. The range of potential spintronic applications goes beyond the use of large magnetoresistive effects. No. Vagner. For reviews of ferromagnetic semiconductor theories outside the scope of this article see. Dietl ¨ (2002). (2003). Fleisher and Merkulov.. with 50% spin-polarized electrons.127 While GaAs is of great technological importance.. An NMR of the nuclear spin polarized by spinpolarized photoexcited electrons has already been used to detect the nonequilibrium electron spin in Si (Lampel. 1996). was found to be 0. III.B). (2002).1). The threshold electrical current.
Caux. 1781–1796 [Sov. Johnson. B.. D. 2002. 89. I. F. This work was supported by the Defense Advanced Research Projects Agency.-S. H. Zagoskin. Phys.. Chiba et al. and M. F. the driving of a spin-polarized current can lead to coherent microwave oscillations in nanomagnets (Kiselev et al. 2003).. We thank D. respectively. Phys. Harley. and that the spin injection spot and the gain medium be within the spin transport length. 2002.. 2000. M. for a ﬁxed pump power. Aleiner. Smith. Adachi. rather than applied magnetic ﬁeld. 2003. K. B. Kikkawa. 2. R. and A. D. The direction of magnetization is ﬁxed (pinned) in the thick Co ﬁlm and free in the thin Co ﬁlm. Ji et al. A.. 113303. 2001. Munekata. T.. 1960. ‘‘Conductance and electron spin polarization in resonant tunneling. L. T. Rev. Furthermore.. B 68. ‘‘Magnetoresistive switch effect in metal/ semiconductor hybrid granular ﬁlms: extremely huge magnetoresistance effect at room temperature. McCombe. 36). M. T. ‘‘Spin-polarized electron transport at ferromagnet/semiconductor Schottky contacts. Eksp. which have much smaller magnetization than their metallic counterparts. Lett. N.. I. 1243–1253 (1961)].’’ Phys. W. de Sousa. 53. ACKNOWLEDGMENTS FIG. Nadgorny. Buhrman. 1999. D. Moodera. D.’’ Semicond. V. 035340. C. Adapted from Albert et al. E. Fiz. Erwin.. Ohno. J. No. R. T. Ralph.’’ Phys. Urazhdin et al. Awschalom. E. Rev. Goldman. Marcus. Johnson. Smith. Rev.1) has already been realized in several experimental geometries. We thank R. Safarov. P. Emley. Smith.’’ Phys. van ´ Driel. V. M. C. Myers. 226802.’’ Pis’ma Zh.. Gorkov.. Oiwa. it has been demonstrated that reversal of magnetization in (In. ‘‘Quantitative study of magnetization reversal by spin-polarized current in magnetic multilayer nanopillars.. Parkin.. 2003). M. Matsukura. D. Kaminski and D. D. S. Vol. 91–95 (1991)]. nanoconstrictions (Myers et al. J. I. Jonker. E. S.. A. and A.. 76. L. Iordanskii. These include nanowires (Wegrowe et al. Rev. and D. using a gate voltage in a ﬁeld-effect transistor structure (Chiba et al. L. H. Technol. and L. are expected also to require much smaller switching currents. M. a room-temperature operation and an electrically pumped laser should be viable as well. Parkin. Jonker. Jr. Rashba. While the experiment was conducted at 6 K. and Y. 89–92 [JETP Lett. M. The common feature of all these geometries is a need for very large current densities ( 107 A cm 2 ). Kavokin. Venkatesan. 2004). Mn)As can be manipulated by modifying the carrier density. 322–326. Tanaka for providing us with illustrative ﬁgures from their published works. 2004) to demonstrate spin-transfer torque (together with other cooperative phenomena) in ferromagnetic semiconductors. 2004. Wei for useful discussions. J. Soulen. S. G.. electrons ﬂow from the thin to the thick Co ﬁlm. R. I. H. C. Eksp. S. C. H.. Teor. Fisher. and H. Mazin. 2003). Mod. 2002. Akinaga. Albert et al. M. 1998. Yamanouchi et al. T. B. the emission power of the laser changed by 400% upon changing the degree of circular polarization of the pump laser. (b) differential resistance dV/dI of a nanopilar device as a function of applied ﬁeld. Albert. C. London). Ongoing experiments (Munekata. Fabian. Ralph. J. von Molnar. and J. H. Lyanda-Geller. S. D. A. Y. I. Halterman. E. Sherman. The requirement is that the spin relaxation time be longer than the carrier recombination time. Munekata. can also be used to manipulate magnetization of ferromagnetic materials brings the exciting prospect of a novel class of spintronic devices. 17. H. D’yakonov. Abrikosov. 2002. and the National Science Foundation—Electrical and Communication Systems. Emley. 1961. The reason for the decrease in threshold is the selective coupling of spin-polarized electrons to photons with one helicity. Albrecht. Rippard et al. J. Albrecht. J. Priour for help with preparation of the manuscript. Teor. 36. B 62. K. B. E.’’ Phys.386 ˇ ´ Zutic. (Color in online edition) Nanopilar device: (a) schematic diagram of a nanopilar device operating at room temperature. 2000. 27). In addition to reversal of magnetization. Fiz. K. For positive current.. point contacts (Tsoi et al. B 66. S. I. Silsbee. 2000. Y. M. and nanopilars (Katine et al. Rippard. M. (c) dV/dI of the device as a function of applied current. A. S. and R. Schmidt. 36–39. C. Maekawa. M. O. I. J. see Fig.. van Dijken.. April 2004 129 . P. S. Kelly et al. A. Ohno. 2002. Ohno.. and D. Hu. J.129 The demonstration that the ﬂow of spin-polarized carriers. R. all involving metallic ferromagnets. Marcus. Based on the ﬁndings in electric-ﬁeld-controlled ferromagnetism (see Fig. Spin-transfer torque (Sec. P. S. T. Kavokin. Afﬂeck. 39. The Principles of Nuclear Magnetism (Oxford University Press. Buhrman. N. T. Ohno. and Das Sarma: Spintronics: Fundamentals and applications VCSEL. Stiles.B. 1433–1445. Sci. D. C. ‘‘Spin relaxation in n-modulation doped GaAs/AlGaAs(110) quantum wells. Kikkawa.. JETP 12. S. Rev. H.. J. 1991. Y.’’ Physica E (Amsterdam) 10. 2003. ‘‘Contribution to the theory of superconducting alloys with paramagnetic impurities. 2002. K. The arrows in panels (b) and (c) represent the direction of magnetic ﬁeld and current sweeps.. 2003.’’ Zh. REFERENCES Abragam. X. the US Ofﬁce of Naval Research.. 1999. Bussmann. Valls. H. I. which is a key element in realizing various magnetoresistive applications. 2003). 53. M. B. D. B. ‘‘Andreev scattering and Josephson current in a one-dimensional electron liquid. ‘‘Electron spin injection at a Schottky contact. A.
Rev. Lett. E. 246804. M.. A. ¨ Beck. 85. Y. ‘‘Spin injection into semiconductors. B 60. JETP 35. S. F. Mermin. ‘‘Observation of spinpolarized-electron tunneling from a ferromagnet into GaAs. de Haan.’’ Zh. J. Fiz. 1228–1231 (1964)]. F. Fert.. JETP 57. van Lierop.. L. S. 1976. Titkov.’’ Phys. Lett. 80.. 2002. van Driel.. Phys. M. Friederich. Phys. Magn. Phys. F. J. 267. D. 4096–4099. 1966. F. Young. 1958. and N. V. 3464–3467. 71. van ’t Erve. Ivchenko. Jansen.. Lodder. 93. Rev. 1966. C. April 2004 Avodin. 035310. Lett. Petroff. 1992. ‘‘Spin-dependent transport in a Luttinger liquid. ‘‘Spin transport by giant ambipolar diffusion. M. Bennett. Lett. Lett. E. ‘‘Manipulating and storing spin coherence in semiconductors. Mod. Thierry-Mieg. 54. 1992. 30. L. N. and R. V. F. M.’’ Phys. 335–341. C. Melin. G.’’ Solid State Commun. 2001. J. Eksp. 75. and G. and R. U. Johnston-Halperin. B 68. Phys.. ‘‘Temperature dependence of electroluminescence and I-V characteristics of ferromagnetic/non-magnetic semiconductor pn junctions. ‘‘Spin transport in interacting quantum wires and carbon nanotubes. ‘‘Low-ﬁeld magnetocurrent above 200% in a spin-valve transistor at room temperature. E. Anil Kumar. Hache. J. Lett.’’ J. von Molnar. W. and F. Teor. F. H. R. G. Appl. D. Andreev. Jakob. N. 373–389. Kikkawa. 2001. Rev. No. P. T. Fert.... 1999.. J.. 2001. E. 214. Ando. and A. L.’’ Phys. Teor.. Rev.’’ Phys. Today 52 (6). ‘‘Longrange proximity effects in superconductor-ferromagnet structures. S. Brousseau. 2472–2475. Jr. P. G. A. G. Kiesel. Fisk. Janko. S. Fiz. G. M. 33–38. B 57. T. A. 2003. F. van Wees. 1058–1061. M. Pikus. Pikus. L. 1983. G.. W. B. 846–850 (1959)]. Lacerda. M. . 370–376 [Sov. ‘‘Spin-dependent transport through the ﬁnite array of quantum dots: Spin gun. 2003. Eksp. H. S.. Bergeret. Arata. 9353– 9358. 1982. and G. 95–97. (S. ´ Apinyan. 812–815. A. 1–6. G. Mater. M.-Ph. Chazelas. and H. 1976. 513–516.’’ Phys. Brataas.’’ Physica E (Amsterdam) 12. 17. P. R. and J.’’ Surf. V. Lett. 680–687 (1983)]. 86. B 25.. ‘‘Electronic properties of two-dimensional systems. 35. A. 1823–1825 [Sov. Lett.’’ Appl. Balakirev. Bastard.. P. 1999.. G. P. 82. 76. V. 1982. Rev. Alvarado. Broto. and R. T. J. 76. 437– 672. and A. M. Eitenne. Mansﬁeld. R3213–R3216. ‘‘Tunneling potential barrier dependence of electron spin polarization. Pratt. 274–289. K. A. Danishevskii. Berkeley. 200. C. 1976a. Barbic. ‘‘Nanoscale Zeeman localization of charge carriers in diluted magnetic semiconductorpermalloy hybrids. Fiz. JETP 19. Stern. D. Mishra. Averkiev. M. O. E. W. S. 84. M. ‘‘Studies of spin-orbit scattering in noblemetal nanoparticles using energy-level tunneling spectroscopy. ‘‘Giant spin relaxation anisotropy in zinc-blende heterostructures.. Aronson.. Fowler. Bauer. B. ‘‘Spin injection and polarization of excitations and nuclei in superconductors. A. Jones. J. Efetov. ‘‘Spin polarization of magnetoresistive materials by point contact spectroscopy. Bakun. Sartan.’’ cond-mat/0306709. 2002. Ohno. 2001.. Young. A. X.. Appelbaum. 90.’’ Physica E (Amsterdam) 10. Grimaldi. and S. D. Magn. S.and giant-magnetoresistive materials.. A. and A. 2002.. Amand. ‘‘Perpendicular transport of spinpolarized electrons through magnetic nanostructures. and R. F. A. 2000. 698–700 (1976)]. Aronov. Lett. Magn. Yu. and A. ‘‘Giant magnetoresistance of (001)Fe/(001)Cr magnetic superlattices.. Solid State Physics (Saunders.. 9987–9994. Aronov. Hughes. ‘‘ ‘s-d’ exchange model of zero-bias tunneling anomalies. 1995. ‘‘Spin relaxation of conduction electrons in p-type III-V compounds.’’ Eur. B. 32–34 (1976)]. A. 024403. Rev. K. Philadelphia). Sci..’’ Phys. Rev. 1988. Anderson.’’ Phys. 24.’’ Zh. Aronov. Auth. Bianchi.-Peterburg) 10. G. J. Phys. 1177–1180 [Sov. 46. ‘‘The thermal conductivity of the intermediate state in superconductors. B.’’ Phys. ‘‘Circular photogalvanic effect in optically active crystals. Henderson. Baylac. Felser. Dareys. E. A. Mater. and J. Fabian. Volkov. Matter 48. 1703–1706. J. 17. ‘‘Emission of spin waves by a magnetic multilayer transversed by a current. Rev. L. and Das Sarma: Spintronics: Fundamentals and applications 387 Alvarado. B: Condens. ´ Atanasov.’’ J. Lett. Marie. Berger. D. E.-C.. ‘‘Electron spin and optical coherence in semiconductors. B 64. 2000.’’ Z. Zhang. ‘‘Spin injection in metals and polarization of nuclei.’’ J. J. A. ´ Awschalom. Sipe. M. and J. Eksp. 24.. Malzer. Mod. F. A. J. G.ˇ ´ Zutic. E. C. A.. Baibich.. P. Awschalom.. 1387–1390.’’ Phys. N. Y. Dmitrieva. 57–60. 15 582–15 584. ‘‘Hole spin relaxation in n-modulation doped quantum wells. 2003. Awschalom.’’ Solid State Commun. Egger. 5–16. Ansermet.’’ Fiz. 37–39 [JETP Lett. Rev. Ohno. and V. Bazaliy. and V. and Y. ´ Berciu. ‘‘Microscopic theory of non local pair correlations in metallic F/S/F trilayers. I. Beschoten.’’ Phys..’’ J. Aronov. ‘‘Spin-ﬂip scattering times in semiconductor quantum wells. B 54.. Rev. P. N..’’ Phys. 2. Lett. 1996. F. S.. N. Warnock. Dohler. T. J. A. Phys. D. Vlutters. Ferreira. Asnin. Rev. I. and S.: Condens. Eksp. 91–95. Vol.. 68. A. B. and L. Rev. and W.. V. A. Matsukura.. and P. Block.. 1209–1215 [Sov. Z. Teor. ´ Barnas. D.’’ Phys.’’ cond-mat/0310632. ‘‘Coherent control of photocurrent generation in bulk semiconductors. M. and V. 1987. A. Fiz. Rev.’’ Phys. 1998. B. 2003. and J. 58. Rev. Phys. JETP 44. 193–196 (1976)]. ‘‘Current-perpendicular (CPP) magnetoresistance in magnetic metallic multilayers. Creuzet. E. 1979.’’ Physica E (Amsterdam) 10. Phys. 1998.. Teor. 1995. G.’’ Pis’ma Zh.’’ Phys. ‘‘On the theory of paramagnetic resonance and paramagnetic relaxation in metals. Egger. 61. 288–291. P. Bass. Rev. Poluprovodn. Poggio. B.’’ Zh.. DenBaars. Fiz. D. Phys. and C.’’ Rev. B. ‘‘Low-temperature magnetic spectroscopy of a dilute magnetic semiconductor. Rev. Balents. S. 3928–3930.’’ Phys. Rogachev. Balents. Nguyen Van Dau. Rev. G. E. and K. Andreev. and B. L. M. 1996. 1964. L. Semicond 10. Bacquet. D. Kuperin. ‘‘Localized magnetic states and Fermisurface anomalies in tunneling. ‘‘The origin of weak ferromagnetism in CaB6 . Lett. ‘‘Magnetic resonance diffraction using the magnetic ﬁeld from ferromagnetic sphere. M. 6027–6050. Matter 10. Keller. 407–411. 565–570. Pikus.’’ Phys. ‘‘Modiﬁcation of the Landau-Lifshitz equation in the presence of a spin-polarized current in colossal.’’ Zh.’’ Phys. A. Eksp. ‘‘Spin-torque transistor. E. J. Teor. Phys. Gerasimenko. Magn. 2003.. 1999. L1–L6. Ashcroft. 1998. Renaud. Bergmann. Tekh. ‘‘Magnetoresistance oscillations due to charging effects in double ferromagnetic tunnel junctions. Phys. 1170–1184 [Sov. 1976b. Phys. 91. Golub. P. R.
Lett. M.. R. and J. and V. ‘‘Room temperature spin relaxation in GaAs/AlGaAs multiple quantum wells. New York). A. Supercond. Brosig. B 27. R10 135–R10 138. G. V. Chun. Chauss. 1971. Prinz. and J. A.’’ Phys. A. Gandit. C. 91. Campagna. 2422–2425. H.’’ Phys. Y. Aronov. Ed. R. Scanning Probe Microscopy and Spectroscopy (Wiley. Tinkham. Rev. Bauer.’’ Phys. ‘‘Nuclear induction. ‘‘Zero-ﬁeld spin splitting in InAs-AlSb quantum wells revisited. 924–926. ‘‘Spin battery operated by ferromagnetic resonance. ‘‘Scattering approach to parametric pumping. Lett. and H.’’ Phys. Yu. K. 4515–4532. Rev. Hu. P. Rev.’’ Phys. P. 060404. P.’’ IEEE Trans. P. Vol. 112– 118. Phys. N. 026602. B 8. 2002. F.. Mod. Borghs..’’ Phys. and D. Beuneu. Lett. Loss. and P. Rev. D. P. Phys. Bussmann.’’ Phys. W. Barthelemy. B 66. M. Rev. Bussmann. R. ‘‘Dilute magnetic semiconductors in the low carrier density regime. 1915–1921. M. Tserkovnyak. D. C. Bruno. N. 77. T. Martin. Teor. W. 70. Tatarenko. P. Bir.. MacLaren. 2001.. Ferrand.. and B.. Bowring. and M. Rowell. 1978. and W. E. B 60. Bratkovsky. X. Grison. 2001. 85. R. and B. Bloch. P. Brar. Burkard. F. Cameron. 67.’’ Zh. E. 22. 1982. Siegmann. A. Grevatt. 2000. Fert.’’ Appl. J.. Pikus. Wang. Klapwijk. H. A. Malinowski. Cheng. Wang. Rev. ‘‘Electron-electron spin-ﬂip scattering and spin relaxation in III-V and II-VI semiconductors. Halperin. Ch. G. Bibes. Schwarz. 965–986. Loss. J. D. D. Cheng. 064517. M. 71–83. E. G. Malinowski. Kennett. ‘‘Precession and motional slowing of spin evolution in a high mobility two-dimensional electron gas. Rev. M. G. Lett. Warburton.’’ Phys. Engel. 2344–2347. Burkard. B 63. M. 33. ‘‘Switching of vertical giant magnetoresistance devices by current through the device.Mn)As. M. J. S. H. Bruno. D. Saurenbach. D.. Phys. 1975. I. and G. G. R13 989–R13 992.. M. 2003. Lett. Mater. and J. B 59. Phys. ‘‘Nearly total spin polarization in La2/3Sr1/3MnO3 from tunneling experiments. K. Appl. Dollfus. X. 1973.’’ Phys. G. Beech.’’ Phys. 2140–2142. Wan. ‘‘Light and electric ﬁeld control of ferromagnetism in magnetic quantum structures. ‘‘Conduction electron spin resonance in magnesium. ‘‘Geometrically constrained magnetic wall. 2476–2478. 3161–3178.. ‘‘Room-temperature electron spin relaxation in bulk InAs. P. B 63. ‘‘Resonant tunneling through ErAs semimetal quantum wells. Harley.’’ Appl.’’ Appl. Fiz. Marsden. Rev.’’ Fortschr. 625–630.. Magn. Karimov.. Bratkovsky. ‘‘Spin coherence and dephasing in GaN. G. Dynes. A. Y. Rev. 2001. Moshchalkov. R. B 58.’’ Phys. and N. and A. J. E. Contour. Bertolini.’’ J. 933– 935. L.’’ Phys. Tinkham. 1946. and X. 092402. 4828–4830. J. 233–235. ‘‘Metallic to tunneling transition in Cu-Nb point contacts. 88. Magn. Rev. Butler. 1984. R. 69. Bronold. 1998. 2003.. Zinn. 2002. 2000. Lett. Bhat. H. Liu. and X. and M. Lemaitre. ˇ ´ Boeve. Rev. Zhang. Hesto.’’ Phys. 705–712 (1976)]. and D. G.’’ Phys. Phys. I. ‘‘Reduction of electron tunneling current due to lateral variation of the wave function. B 39. Phys. R. 236601. 1983. and T. ‘‘Monte Carlo study of spin relaxation in AlGaAs/GaAs quantum wells. 15. J. C.’’ Phys. L. D. 233206. Blonder. M. 1969. Bournel. 90. Miller. 5432–5435. Cousins. 226-230. Phys. 73.. De Boeck. Phys. P.. Phys. Osipov. Cotti. Awschalom.’’ Phys. Prinz. Rev. S. Sulpice. 2000.’’ Phys. Rev. Brinkman.’’ Phys. Blonder. F. F. 460–474. C. Smithard.. and J.. Brouwer. Olesberg. K. Flatte. Lett. G. Boguslawski. R123– R181. Rev. Ensslin. Chau. T. Schwartz.. P. J. Bhat. V. Z. G. T. A. M. ‘‘Tunneling of electrons in conventional and half-metallic systems: Towards very large magnetoresistance. Nemeth.’’ Phys. Rev. T. P. Lett.. Schulthess. ‘‘Magnetic ﬁeld splitting of the density of states of thin superconductors. Eksp.-F. Rev. Brandt. T. F. 597–599. 2003.’’ J. 89. S.’’ Phys. 1995. 1999. 33. 1982. J. 1–167. Perozzo. G. Blakemore. Status Solidi B 43. H. A. Rev. Zarand. ´ Borda.. Lett.’’ Phys.388 ˇ ´ Zutic. D. ‘‘Transport properties of epitaxial Fe-GaN-Fe tunnel junctions on GaAs.. J. ‘‘Coupled quantum dots as quantum gates.. G. T. Phys. Harley. J. No. 1970. 1980. ‘‘The Elliott relation in pure metals. Ibbetson. G. 207204. Brataas. Smith. ¨ Binasch. J. and D. Anane.. B 18. Allen. E. C. . Xiong. S. Zhu. Grunberg.. K. S..’’ Appl. ‘‘Semimagnetic semiconductors. ´ Boggess. Shields. 1998. E. and D. Kroemer. 1268–1270. Rev. Lett. Halperin. J. A. A. A. 1989. ‘‘Transition from metallic to tunneling regimes in superconducting microconstrictions: Excess current. Lau. Wasiela. 1998. 1999. R. J. 1333–1335. B 66. charge imbalance. A. J. 82. Busch. Gutmann. ‘‘Optically injected spin currents in semiconductors.. B. Rev. ‘‘SU(4) Fermi liquid state and spin ﬁltering in a double quantum dot system. A. Parker.. R.. 2000. P. K. 75. B.’’ Appl. ‘‘Magneto-electronique dans des dispositifs ´ a semiconducteurs.. O. W. 41. ‘‘Tunneling conductance of asymmetrical barriers. Simmons. Zhang. and H. ‘‘Transport in superconductor/ferromagnet/ superconductor junctions dominated by interface resistance. 83. and G. P. JETP 42. von Delft. Saxena. W. Monod. M. 2002. 53. Lett. S. B 56. ‘‘Observation of electron polarization in photoemission. S.-P. (Paris) 25. W. ‘‘Semiconducting and other major properties of gallium arsenide. E. 2346–2348.. ´ ´ Bournel. J. and A. 2003. S.’’ J. A. A. and Das Sarma: Spintronics: Fundamentals and applications Hu.’’ Adv.. O. and J.. Z. Cassan. and J. Rev. A. Appl. 121202. P. Boukari. Hofstetter. 193–256. E.’’ Ann. Y. P. M. L. 1999. Cibert. A. Magn. and W. Berciu.’’ Phys. Kossacki. A. Rev. A. Lett. 76. S. 2. I. Zhang. A.’’ cond-mat/0307656. Bonnell D. Sipe. 77. Phys. J. 389–391. Gaj. ‘‘Spin relaxation of electrons due to scattering by holes. ‘‘Spin extraction from a nonmagnetic semiconductor. and V. 2000. C. Brand. H. B 25. Sanvitto.. C. 2002. R.. B. E. ‘‘Magnetic-ﬁeld dependence of electron spin relaxation in n-type semiconductors. ‘‘Direct measurement of the spin polarization of the magnetic semiconductor (Ga. Rev. Ritchie.-G. Fabian. Bourgeois.. Phys. M.. ‘‘Spintronics and quantum dots for quantum computing and quantum communication.. R. 34. Lett. 48. Thomas. B 63. M.’’ Appl. 2070– 2078. Bowen.’’ Phys.. and supercurrent conversion. S. 2002. Rev. B. 2001. R. 056602.’’ J. Y. 1997. A. 1382–1397 [Sov. ‘‘CPP giant magnetoresistance of NiFeCo/Cu/CoFe/Cu multilayers. 2425–2428. 2001. S. DiVincenzo. Dietl.’’ Solid State Commun. C. and P. A. Palmstrøm. Nguyen. Samarth. M. L... Brehmer. and B. F. A. 1999. and T. ‘‘Enhanced magnetoresistance in layered magnetic structures with antiferromagnetic interlayer exchange.. M. A. J. and D. Phys. Britton. Wilkens. April 2004 Braden.
R. 1999. 39.’’ Phys. A. and M. X. ˇ ´ Das Sarma.48As heterostructures as B→0. I. Fayﬁeld. Lett. and A. ‘‘Temperature-dependent magnetization in diluted magnetic semiconductors. A.. J.’’ Pis’ma Zh.’’ Phys. Sato. Samarth. West.’’ 58th DRC (Device Research Conference) Conference Digest (IEEE. and H. Pioro-Ladriere.. B 63. Hu. M. 4781–4783. E. S. Das Sarma. and R. 89. J. R. 15.’’ Phys.’’ Philos. McGuire. B. O. Zawadzki. 6039–6045. and A. Rev. T. Idzerda 2003. Das. ‘‘Issues. McGuire. D. Nawrocki.Mn)As tunnel junction. 155201. N. Rev. 207–215. Chiba. Sachrajda. N.. B 11. Ogale. 2821–2826. and R. Zutic. M. J. 100408. Chadi. Ohno. Amand. V. Jbeli. P. Phys.’’ Phys. A. Matsukura. Bhattacharya. J. ‘‘Electron interaction on the giant magnetoresistance in the perpendicular geometry. Piscataway) 95–98. Chadi. S. C. 665–667. Pomeroy. and L. Petroff. Davis.. 1966. 2001. Lampel.Mn)As/GaAs/(Ga. 27. ‘‘Evidence for two current conduction iron. Sienko..’’ Phys.3MnO3 /YBa2 Cu3 O7 x interface: Role of Andreev bound states. Fabian. and transport. I. B 66. Lett. 1975. ‘‘Tunneling into ferromagnetic quantum Hall states: Observation of a spin bottleneck. T. M. J. Lett. 134–138. H. Scalbert. Burnham. J. 2002a. W. I. Pfeiffer. Jaffe. Ashoori. Cohen-Tannoudji. 2002. A.’’ Phys. X. Eksp. F. Egelhoff. Chung. 464– 465. Strom-Olsen. Johnston-Halperin. ‘‘Ferromagnetic imprinting of spin polarization in a semiconductor. L. S. A. X. and P. ‘‘Oscillatory effects and the magnetic-susceptibility of carriers in inversionlayers. Plischke. 2003. T. 2000b.’’ Science 301. Baumberg. Hwang.’’ Phys. D. 1997. and N. 1984a. 2001. Lett. 2003. R3832–R3835. concepts. 066601. J. Yu. S. Fabian. ‘‘Electrical manipulation of magnetization reversal in a ferromagnetic semiconductor. and J. 76.’’ cond-mat/0403500. and E. Zutic. S. Hu. J.’’ Phys. Marie. White.. and J. ‘‘Applications of spin dependent transport materials. I. R169–R177. Lett.. 1411–1414. H. and I. Chui. ‘‘Electron and hole spin relaxation in modulation-doped CdMnTe quantum wells. B 67.. H. Lett. S. Dordrecht/Plenum. 603–606. Rev. M. Lett. 2. S. X. Magnetic Interactions and Spin Transport (Kluwer Academic. 78–81 (1984)]. ‘‘Properties of a 2D electron-gas with lifted spectral degeneracy. A. and J. A. R..’’ Appl. C..ˇ ´ Zutic. 7574–7588. 2004.. April 2004 Ciuti. Sci. G. P. and Y. Matsukura. Dargys. J. E. Drouhin. Chun. p. 2000a. P. Burnham. and I. Zutic. Cacho. R. J. ‘‘Nuclear polarization in InSb by a dc current. Damay. A. 56. Reifenberger. W. 36. Chye. 943–945. 1976. B.’’ Phys. D. F. B.. 156601. Rev. Rev. Mag. and K. G. Singh. A. W.. ‘‘Current-driven magnetization reversal in a ferromagnetic semiconductor (Ga. 66–69 [JETP Lett. Gerardot.. 19. 10. Munoz.. Clark.. and N. 516–523. Fabian. and C. Appl. 2118–2121. 2002. M. and R. Rev. Eds. Rev.Mn)As into InAs quantum dots.’’ Phys. Rev. and I. Sham.’’ Phys. Potashnik. and A. Clark. W. Lett..’’ Phys. Zutic. ‘‘Optical spin resonance and transverse spin relaxation in magnetic semiconductor quantum wells. 2000c. B 9. ‘‘Theoretical perspectives on spintronics and spin-polarized transport. 289–295. and B. A. W. J. and J. Pohm. B 64.’’ Phys. R. Cibert. P. 5758–5765.. Cortez. ¨ Cochrane. S. 88. Cousins. 977–983. ‘‘Evidence for spin splitting in Inx Ga1 x As/In0. Bychkov. Hong. 1999. M. 1974. Phys.’’ Phys. G. S. 1976. Fiz. ‘‘Bias dependence in spin-polarized tunneling. ‘‘Tunable negative differential resistance controlled by spin blockade in single-electron transistors. ‘‘Electronic analog of the electrooptic modulator. ‘‘Spin transmission in metallic trilayers. D. J.. B 55. Rev. Awschalom. D. Biswas. D. Y..’’ Superlattices Microstruct. and M. Phys. Campbell. ‘‘Spin relaxation of conduction electrons in n-type indium antimonide at low temperature. B 52. ‘‘Electron spin resonance of conduction electrons in beryllium.’’ Phys. 1967. R. H. S. B 13. Datta. J. 201301. and challenges in spintronics.. 2001. 4466–4469. Chazalviel. B. and H. cond-mat/0006369. C 17. Lett. H. ‘‘Universal scaling of ballistic magnetoresistance in magnetic nanocontacts. M. 56.... Rev. 33. ‘‘Spintronics: electron spin coherence. Miller. Ferreira. B. Das. P.’’ Appl. and G. Magn. Wolf (North-Holland. Flack. Camilleri. Rashba.’’ Am. 1984b.’’ Phys. F. Rev. Sham. and E. B 66.’’ Phys. Chui. J. D. 1965. Peretti. and L. R. Amsterdam). Datta. and T. Vol. Rev. Wolf.. D. D.’’ Phys. 39. and Das Sarma: Spintronics: Fundamentals and applications 389 Bychkov. Kostler. 2002.. C. J. ˇ ´ Das Sarma.. ‘‘Spin relaxation of conduction electrons in Alx Ga1 x As. B 12. Kita. D’yakonov. 085331.’’ Appl. . in Progress in Optics.’’ Phys.52Al0. A. D.. J. ‘‘Spin-polarized transport across a La0. 5600–5603. Kaminski. Y. S. 1989. Awschalom. Chui. ‘‘Organic luminescent devices and magnetoelectronics. 2002b. M.’’ Phys. S. T. F. Teor.’’ Solid State Commun. Zutic. 2003. Cullen. Tatarenko. R. C. X. D. 3258– 3261. Rev. Rev. P. Lassailly. 80. 7358– 7360. 2002.. 1975. O. 2177–2179. B 39. 89. 165216. Wojtowicz. S. C. L. I.’’ J. J. Vol. ‘‘Spindependent properties of a two-dimensional electron gas with ferromagnetic gates. A. Rev. M. Rev. Rev. Chan. B 66. Wu. Schiffer.7Sr0. M.-N. Feher. and I. B 13. H. Yamanouchi. K. C. Z. 212508. T. S.. 1997. Hawrylak. entanglement. Phys. Y. ‘‘Spin-polarized tunneling in hybrid metalsemiconductor magnetic tunnel junctions. Krebs. R. M. 2002. 287203. Garcia. 1963.. No.. Rev. 1995. Chtchelkanova. Rashba. Daughton. 89. Clark. and T. ˇ ´ Chen. Phys. Yu. Ciorga. A. Venkatesan. Ciuti. P. ‘‘Conduction-electron spin resonance in metallic lithium. S. K. Crooker. F. S. Gerard. H.. D. T. Voisin.. H. ˇ ´ Das Sarma. Rev. Mod. P. Y. Gomez. 3013–3021. Semenov. ‘‘Spintronics. Ohno. Y. Hu. 2001. J. Dupree. Phys. Hu.. Fabian. E. D. Das Sarma. ‘‘Spin electronics and spin computation. Chiba. 1995. Phys. Fert.. Lett. 1555–1562.’’ Phys. ‘‘Spin injection from (Ga.. Rev. ‘‘Spin polarization dynamics in n-doped InAs/GaAs quantum dots. S. C. D. 119.’’ Physica E (Amsterdam) 13.’’ J. 2002. S. Greene. Ku. Fabian. E. ‘‘Spin ﬁltering of free electrons by magnetic multilayers: Towards an efﬁcient self-calibrated spin polarimeter...’’ IEEE Trans. Rev. T. 2002.’’ J. J. 508–511. White. 81. Teppe. D 32. Bussmann. ‘‘Magnetization studies of (GeTe) 1 x (MnTe) x pseudobinary alloys. Samarth. P. and A. edited by E. Rev. ‘‘Luttinger-Kohn Hamiltonian and coherent excitation of the valence-band holes.. 93. Smith. S. ‘‘ 1 conduction electron g factor and matrix elements in GaAs and Alx Ga1 x As alloys. 25. 83. ˇ ´ Das Sarma. M. and K. H. 74. C. S. and R. 1990..’’ Phys. New York).
Vaures.. Lett. A.. and D. Ehrenreich (Academic.. DiVincenzo. 1991. 1955. 1. W. Teor. . S. Dediu. ‘‘Technology and materials issues in semiconductor-based magnetoelectronics. D’yakonov.’’ Phys. A. and B. and S.’’ Phys.. 5453–5456. M. ‘‘Spin orientation of electrons associated with the interband absorption of light in semiconductors. R. de Sousa. Bender. 82. 2. ‘‘Spin-polarized quasiparticle injection devices using Au/YBa2 Cu3 O7 /LaAlO3 / Nd0. P. J. 1297–1300. R. ‘‘Elementary excitations in the vicinity of a normal metal-superconducting metal contact. ‘‘Possibility of orienting electron spins with current.’’ Zh. No.’’ Solid State Commun.. F. 110–112 (1986)]. 1986. B 68. V. I. New York). Chen. R. 206–208 [JETP Lett. R. van Engen. Dupree. Matacotta. New York). 17. R.. 76. I.3MnO3 : New ideas on spin-polarized tunneling. 74. 13. M. ‘‘Superconducting proximity effects in magnetic metals. Phys. Gurney. Phys. Perel’. 459–460. ‘‘Current-induced spin orientation of electrons in semiconductors. 7778– 7781. Arnold. A. and G.. Oesterholt.’’ Pis’ma Zh. T. Perel’. Semicond. De Boeck. S.. A. JETP 33. Superconductivty of Metals and Alloys (Addison-Wesley. K. and V. G. and Das Sarma: Spintronics: Fundamentals and applications de Andrada e Silva. Johnson.’’ Appl. Vol. Y. B 68. and D. Rev.7Sr0. A. edited by T.’’ Phys. Greene.’’ Phys. 1997. Contour. 2000. P. P. B 43.’’ Pis’ma Zh. S. 558–562. 60. J. Taliani. Weimann. R. de Sousa. A. JETP 63. 467–469 (1971)]. 657–660 [JETP Lett. G. S. in Handbook of Semiconductors. P. Y. A. Shreekala. B. ‘‘Gate control of spin dynamics in III-V semiconductor quantum dots. P. B 67. ‘‘Electron-spin polarization by resonant tunneling. C.216. Rev. de Jong. 2000. Sharma. DiVincenzo. I. 15 174–15 182. M. De Teresa. A. G. Eksp. L.’’ Phys. J. A. and G. Borghs. D. Sci. 122. ‘‘Theory of nuclearinduced spectral diffusion: Spin decoherence of phosphorous donors in Si and GaAs quantum dots. Boeve. J. 266803. 1967.. A. M. M. 1997. 85. J. 487–489. Wilhoit. Phys. Tela 13. de Sousa. Rev.’’ Semicond. Mahajan (North-Holland. Lett. 1996. von Klitzing. 68. and V. 8. Rev. ‘‘The range of g factors and the breakdown of motional narrowing in conduction electron spin resonance.’’ J. Technol. Demler. 2003. Litvinov. Perel’.’’ Phys. Barbanera. ` F. ‘‘New class of materials: Half-metallic ferromagnets. 20. A. 1963.’’ Appl. D. M. J. M. 2002. P. ‘‘Quantum computation. B. 4. Lett... V. H. Rev. Phys. I. 3. Duke.. ‘‘Electron spin coherence in semiconductors: Considerations for a spin-based solid-state quantum computer architecture. Pai. 1969. 1989.’’ Phys. and V. 1995. Lett. Rev. 771–783. Perel’. K. Vol. ‘‘Using single quantum states as spin ﬁlters to study spin polarization in ferromagnets. and R. B 38. and H. Das. P. Venkatesan. 1971a. A. Boeve. G. Saint James.’’ Fiz. 2002.’’ Semicond. Fiz. 1999. Dugaev. ‘‘The physical implementation of quantum computation.’’ Fiz. V. April 2004 Dieny. C. Dietl. and C. E.. ‘‘Feasibility of optical orientation of equilibrium electrons in semiconductors. H.. Kumar. 1971c.’’ Phys. Das Sarma. D. I. 4288–4291. W. ‘‘Nanometer-scale magnetic MnAs particles in GaAs grown by molecular beam epitaxy. 90. 1997. Fiz. Phys. de Gennes. B 59. V. Ralph. Lobb. 4785–4787. 2002.’’ Zh. Rev. 178–181 [Sov. M. 1994. V.390 ˇ ´ Zutic.. S. Fert. Phys. Das Sarma. ‘‘Evaluation of vacuum bonded GaAs/Si spin-valve transistors. 115322. C. D’yakonov. M. Appl. P. ‘‘Quantum computing and singlequbit measurements using the spin-ﬁlter effect. 1988. 100. B. Mueller. M. Motsnyi.. ‘‘Spin relaxation of conduction electrons in noncentrosymetric semiconductors. Dobers. L. Van Hoof. Borghs. 5155–5157. 377–392. DiVincenzo. Moss and S. Teor. and J. M. T. 13. and S.3MnO3 heterostructures. Dresselhaus. L. and M. Rev.. Solid State 13. J. I.. Rev. Lodder. and D. W. 1718–1720.. C. J. D. 1053–1059 (1971)]. Bassani. Ramesh. Marushchak. de Groot. G. and A. Phys.’’ Phys. 71. 87. C. Lyonnet. 2002. C. and V. Van Esch. 181–184. J. ‘‘Ferromagnetic semiconductors. and G. D’yakonov. ‘‘Spin relaxation of two-dimensional electrons in noncentrosymmetric semiconductors. Technol.. M. D. 655–661 (1986)]. 2003b. Status Solidi 24. Barnas. Seitz. Rev. J. 13. 342–354. 1251. Das Sarma. ‘‘Transition from Sharvin to Drude resistance in high-mobility wires.’’ J. 1999. and G. and F. C. C. Lagae. Rev. K. and G. Feinberg. Van Roy. Vygranenko.’’ Phys. Appl. E. 48. Fiz. ‘‘Modeling of magnetically controlled Si-based optoelectronic devices.’’ Appl. ‘‘Andreev reﬂection in ferromagnet-superconductor junctions. Lett. E.’’ Physica E (Amsterdam) 16. M. Barthelemy. 2744–2746. ‘‘Spin-orbit splitting of electronic states in semiconductor asymmetric quantum wells. Titkov. 3581–3585 [Sov. B 55. F. V. (S. p.. ‘‘Spin-orbit coupling effects in zinc blende structures. N. Beasley. Lett. M. Sci. V.. 17. and S. Phys. R.’’ Fortschr. Holland. Dietl.’’ Phys. 1971b. 033301. 300. H. Boeck. 1954–1965 [Sov. 155330. and F. J. 1999. M.375... Phys. 76. de Groot. R..’’ Phys. 2000. Z. 255–261. R. and S. Borghs.. Murgia. Z. D’yakonov. 1657–1660. Reading MA). Kachorovskii. 2003a. P. I. B.. J. R. Fabian. Eksp. 144– 146 (1971)].’’ Science 270. 1994. D’yakonov. Bruynseraede. and V. Yu. Tekh. Seneor. 35A.. Tverd. G. I. D. D’yakonov. S.’’ Phys. 1971d. D’yakonov. Teor. ‘‘Room temperature spin polarized injection in organic semiconductor. 2024–2027. 1971e. and D. Perel’. Parkin... 3023–3026 (1971)]. Vieira. I. Fiz.. Teor. Eksp. Phys. ‘‘The effect of strain on the spin relaxation of conduction electrons in III-V semiconductors. Beenakker.. De Boeck. 1123–1133 [Sov. 1986. de Jong. Poluprovodn.-Peterburg) 20. M. B 49. I. C. V. 580–586. Dessein. I. ‘‘Electron-spin resonance in the two-dimensional electron gas of GaAs-Alx Ga1 x As heterostructures.. Rev. Mod. Rev.’’ Phys. 89. M. W. T. I. 1983 Dutch Patent No. I.. 50. 151–152.’’ Phys. 275–279. Delaey. La Rocca. H.. La Rocca. Montaigne. Lett. M. B 55. R. 2003c. R15 583–R15 585. de Andrada e Silva. Canadian Patent No. ‘‘Giant magnetoresistive in soft ferromagnetic multilayers. R. ‘‘Inverse tunnel magnetoresistance in Co/SrTiO3 /La0. Rev. M. R. edited by F. G. Lett. Supplement 10. Maur.. P. Liu. Talyansky. Phys. Janner. 16 293– 16 299. in Solid State Physics.’’ Phys. Deutscher.’’ Phys. K. 13. C. Eksp. Deshmukh. Perel’. 602. I. Dessein. and K. 1983. F. Buschow.. 1995... Rev. ‘‘Coupling superconducting-ferromagnetic point contacts by Andreev reﬂections. de Gennes. and A. Speriosu.7Sr0. and V. Dong. P. Z. G. Lett. Turnbull. Mueller.
‘‘Organic-based magnets: Opportunities in photoinduced magnetism. Vol. Lee. and H. 2335– 2339 [Sov.. R. and K. B 66. and D. Eksp. A. P.’’ Phys. Fiz. P. Safarov. 87.. and V. 1765–1768 (1997)]. 19.. 161308. and S. Teor. M. 125301.. Rev.’’ Phys. R. Byers.-Peterburg) 39. C. Das Sarma.. B. Kouwenhoven. Chem.. N. Phys. 96. M. Rev. I. Luth. 1954. 2002.-Peterburg) 45.’’ Phys. Solid State 39. Sueoka. J. K. 28.. 205422. Katzer. Korenev. ˇ ´ Erwin. Katzer. and V. D. Rev. Lange. Y. 1973a. Eksp. S. Teor. 2001. 76. B 55.’’ Phys. P. ‘‘Manipulation of the spin memory of electrons in n-GaAs. Solid State 45. 98. P. M. 1975–1979 [Phys. B. Perel’. 1976. Perel’. Jonker. M. A. Solids 4. Ekimov. C. Dyson. Tela (S. Buhrman. in Optical Orientation. Phys. I. 293–297 [JETP Lett. 950–955 (1975)].’’ J. R. Gregg. B 67. M. Gammon. edited by F. 241–255. R. Eksp. R. Ekimov. 38. Vandersypen. I. 915. 1958. Ghazalli. J. Stuttgart). Fabian. and interface structure of Fe/ GaAs. J. and M. B 65. Leroux-Hugon. Rev. ‘‘Inﬂuence of an electric ﬁeld and deformation on the optical orientation in semiconductors.. Elzerman. I. ‘‘The Hanle effect in nonuniformly doped GaAs. I. and I. Rev. and D. Zakharchenya. H. T. Dzyaloshinskii. I. 2001.. I.’’ Phys. ‘‘First-principles study of nucleation. . R. Korenev. and P. Stepanova. Poluprovodn. Kavokin. 2003. T. V.. 177–183 (1973)]. 88. Fiz. 1974. Elezzabi. M.’’ Jpn. J. R1958–R1961. 2002. Dzihoev. B. Tarucha.. M. Zakharchenya.’’ Zh. I. Dzhioev. D. 8. B. 245204. Suter.. ´ Esaki. Endo. V.. S. Zvezdin. B. 2002. J. D’yakonov. Lett. M.’’ Phys. 852–854.’’ Fiz. Dzhioev. 15.. and Das Sarma: Spintronics: Fundamentals and applications 391 D’yakonov. Tekh. pp. 11–71. D’yakonov. V. ‘‘Theory of the Hanle effect in optical orientation of electrons in n-type semiconductors. and V. B.’’ Phys.’’ Phys. ‘‘Theory of the effect of spin-orbit coupling on magnetic resonance in some semiconductors. ‘‘Optical effects due to polarization of nuclei in semiconductors. I. 2. Gammon.’’ Fiz. p. Hanson. 1998. 1967.. D. Perel’. ‘‘Magnetointernal ﬁeld emission in junctions of magnetic insulators. Korenev. I. I. 74... A. L. V.. J.’’ Nature Mater. P. Semicond. 182–185 [JETP Lett. and K.. ‘‘Long electron spin memory times in gallium arsenide.’’ Phys. Zakharchenya (North-Holland. M. M.-Peterburg) 7. I. Teor. Bracker. ‘‘Coupling mechanisms for optically induced NMR in GaAs quantum wells. and M. A. I. and R. K. Y. 1973b. G. Egelhoff. Kavokin. K. W. Pierce. Gammon. ‘‘Few-electron quantum dot circuit with integrated charge read out. N.. L. Rosen. Theory of electron diffusion and the skin effect.ˇ ´ Zutic.. Katzer. 85. No. 15. Poluprovodn. I. Efros.. Escorne. Phys. S. Fabian. 81. J. 256801. and D. ‘‘Band-structure effects in the spin relaxation of conduction electrons.. Phys. B 64. and D. Stiles. F. Rev. Fiz.’’ Pis’ma Zh. Appl. Tischler. Rev. in Proceedings of the 12th International Conference on the Physics of Semiconductors. V. Mukasa. D’yakonov. F. Appl.’’ Pis’ma Zh. Rev. P. Tverd. I. Meier and B. Epstein. A. M. Pappas. I. Teor. Greidanus. M. Phys. Schefﬂer. J.. I. 1996. cond-mat/0401157. R. R. Dzhioev. 1998. L. L. Tela (S.’’ Phys. Jr.. 251–254 [JETP Lett. V. Franceschi. Lenzmann. M. JETP 38. Johnson. 1195. Al. J. 7. Tekh. J. Perel’. 1955. 1984.. Tverd. Katzer. and V. and M. Rev. Zakharchenya. S. Vol. ‘‘Optical orientation in a system of electrons and lattice nuclei in semiconductors Theory. Part 1 39. Mod. and V. I. Erwin. Erlingsson.’’ MRS Bull. K. 492–499. Lazarev. D. Rev. Dzhioev. T. Rev. ‘‘Paramagnetic ion-doped nanocrystal as a voltage-controlled spin ﬁlter. D. Th. ‘‘Experimental and theoretical approach to spin splitting in modulation-doped Inx Ga1 x As/InP quantum wells for B→0. I. P. C. S. Korenev. E. Amsterdam). 3220–3223. S. V..’’ Phys. 2225–2263 (2003)]. and S. 2002. A. 13..’’ Zh. Eksp. Pilkuhn (Teubner. I. Zvezdin. 349–359. 2003. P.’’ Phys. A. Gammon. B. Rev. L. and D. 2000. C. A. 208–210 (1976)]. 80. Eickhoff. Dzhioev. and L. B. ¨ ¨ Engels. 5075–5079. ‘‘Electron spinrelaxation times in p-type -doped GaAs/AlGaAs double heterostructures. Lett. 5624–5627. I. Modern Problems in Condensed Matter Science. G. L. V. A. M. Willems van Beveren. I. 67. Bland. 1551–1553 (1974)]. and V. L.’’ Pis’ma Zh.. ‘‘Direct measurement of the conduction electron spin-lattice relaxation time T 1 in gold. and M. ‘‘Spin relaxation of conduction electrons in polyvalent metals: theory and a realistic calculation. 179–181 (1972)]. Stepanova. M. Phys. ‘‘Optical detection of dynamic of nuclei in semiconductors. Phys. 2004. growth. Efros. Stiles. ‘‘Tailoring ferromagnetic chalcopyrites. Rashba. Nazarov. 2003. J. ‘‘Even-odd effects in magnetoresistance of ferromagnetic domain walls.’’ Pis’ma Zh. S. V. B 66. A. 177–179 (1971)]. Flinn. ‘‘Spin polarization of injected electrons.’’ Phys. A. V. Ekimov. H. B 65. L. Rev. M. 1999a. (S.’’ Fiz. K. 1997. fractal magnetism and beyond. Schapers. Teor. 1974. I. Semicond. ‘‘A thermodynamic theory of ‘weak’ ferromagnetism of antiferromagnetics. II. Lett. 153409. D. 198–201 (1970)]. Rev. 2153–2160 [Phys. ‘‘Spin diffusion of optically oriented electrons and photon entrainment in n-gallium arsenide. Teor. (London) (in press). 266–279. 397–401. I. Fiz. Lazarev. Lett. and S. 12. Fiz. Safarov. Zakharchenya.’’ Science 296. edited by M. spintronics. Eksp. P. 2002. Rev. and V. B 67. S.. Safarov. Safarov. V. J. Lett.’’ Phys. Zakharchenya. 200–203 (2001)]. Eksp. 4578– 4581.. S. 77. and M. B. 12. P. 1971. von Molnar. L. Lett.. Rev. A. 10. ‘‘Nucleus-mediated spin-ﬂip transitions in GaAs quantum dots.. Perel’. (S. 1997. JETP 40. 2003. D. V. D’yakonov. S.. D. R. ‘‘Optical orientation of carriers in interband transitions in semiconductors. Gor’kov. Alvarado. L. ‘‘Electron spin resonance absorption in metals.-Peterburg) 10.. 362–376 [Sov. I. B. 2001. ‘‘Spin-dependent perpendicular magnetotransport through a tunable ZnSe/Zn1 x Mnx Se heterostructure: A possible spin ﬁlter?’’ Phys. and V. 206601. Freeman. Korenev.. Rev. J. G. Zutic. April 2004 Egues. S. Elliott. 13. S. S. Fiz. 1970. ‘‘Low-temperature spin relaxation in n-type GaAs. Merkulov.. 74. Fabian. J. 1972. Falko. 257–261 [JETP Lett. I. ‘‘Optical orientation and the Hanle effect of neutral and negatively charged excitons in GaAs/Alx Ga1 x As quantum wells. F. Berkovits. L. Zakharchenya. Meltser. 350–353 [Sov. ‘‘Observation of optical orientation of equilibrium electrons in n-type semiconductors.’’ Fiz. 1912–1924 [Sov.’’ Phys. Dzero.’’ Phys. P. and V. F. L. I. Das Sarma. 100402. Johnson. Y. I. 195306. 2002.
and W.. B 65. ˇ ´ Fabian. and Das Sarma: Spintronics: Fundamentals and applications Fabian. T. 78. I. Parkin. 165301. ‘‘Spin-polarized quasiparticle transport in cuprate superconductors. ‘‘Conditions for efﬁcient spin injection from a ferromagnetic metal into a semiconductor. W. Zakharchenya (North-Holland. E.. Das Sarma. Fert..’’ Phys. Kohlhepp. Lambeth. G. A. J.. pp. and L. V. ‘‘Theory of semiconductor magnetic bipolar transistors. de Groot. Amsterdam). C. J. R. ˇ ´ Fabian.’’ Science 299. J. Marcus. Dutta. ‘‘High ﬁeld superconductivity in thin ﬁlms. Modern Problems in Condensed Matter Science. 2002a.D. Fleisher. 1967. T. B.’’ J. Phys. J. Fiz.. 82. 8. Lett. 4192–4195. Rev.. E. P. 114. Lambert. 045316. P. Filip. and A.’’ Phys. M. D. 1815–1817. 1990. F. J. 1999b. 1973.. and N. 2002.. van der Weide. ´ Flatte. 87.’’ Phys. ˇ ´ Fabian. 1984b. V. ‘‘Electron spin resonance absorption in metals. 337–348. M. and S. Z. B. P. I. Vignale. Lampel. A. P. Zutic. T. A.-F. Sci.’’ Pis’ma Zh. Rev. 1968. A. Flensberg. Filipe.. ˇ ´ Fabian. Volkov. P. Gere. ‘‘Phonon-induced spin relaxation of conduction electrons in aluminum. 76. J. Lett. 82. D. Lett. J.’’ Phys. and G.. M. Lee. Ya. Xiong. 679–682. Rev. A. 667–719. 8.’’ Phys. Fisher. G. and I. A. and S. and I. A. 135–137. Rev. E. G. ‘‘Surface.’’ J.’’ Phys.’’ Phys. Phys. Vol. J. Smits. ‘‘Nuclear polarization via ‘hot’ conduction electrons. 1999. M. .. Das Sarma. and S. 4220–4223. Phys. Lett. Peretti. ‘‘Unipolar spin diodes and transistors. Lett. Rev. 2002.’’ Phys. ‘‘Spin-polarized current ampliﬁcation and spin injection in magnetic bipolar transistors. Das Sarma. 1996. 6554–6565. Zutic.’’ Adv. and G. A.. ‘‘Electron spin resonance experiments on donors in silicon: II. and A.. T. 169–172. ´ Flatte. Rev.. G. 2. F. Salamon. M. E. R. 2002. Merkulov. and E. 1273–1275..’’ Appl. T. J. and M. R.. C. Fabian. B 67. M.. 2000.. C. P. April 2004 Filip. Hentschel. M. van Driel. R. Mod. Technol. M. boundary. J.’’ Phys. ‘‘Spin transport in inhomogeneous magnetic ﬁelds: a proposal for Stern-Gerlach-like experiments with conduction electrons. Rev. Sarma. ‘‘A gate-controlled bidirectional spin ﬁlter using quantum coherence. M.. 2001. ‘‘Spin relaxation of photoelectrons in p-type gallium arsenide. Schmidt. D. Borghs. J. M. Yu. Savage. B 64. Hopster. 820– 831. M. Pierce. W. Ossau.. Jedema. Fishman. Fert. A. Lagally. J. 1993. Moustakas. 184420. S. Fang. and thin-ﬁlm magnetism. Phys. 84. R. 1999c. Hathaway. Teor. B 53. G. I. 2003. Fraser.. Fiederling. and S. 8340–8344..’’ Phys. de Jonge. edited by F. 1998. 4740. Ferreira. ‘‘Practical design and simulation of silicon-based quantum-dot qubits. 81. ‘‘Proximity and Josephson effects in superconducting hybrid structures. C. Lett. 1299–1340. S. Fabian. 1955. Schuhl. 2002b. 245308. No. and S. Feher. J. Lett.’’ Phys. Appl. 1984a. 1190–1192. Bader. M. ‘‘Two-current conduction in nickel. A. B. 1211–1214. 91. ‘‘Detection of electrical spin injection by light-emitting diodes in top. Feng. ‘‘Spin relaxation of conduction electrons. Campbell..’’ Phys.’’ eprint cond-mat/0211639. Rev. J. Rev. Vol. A. 173–258.’’ Phys. and R. 80. Merkulov. Yeh. Victora. B 66. Byers. Prinz. 1977. 3. ‘‘Spin-dependent transmission of electrons through the ferromagnetic metal base of a hot-electron transistorlike system. ‘‘Experimental search for the electrical spin injection in a semiconductor. Rev. Rev. Rev. G. I. S.’’ Phys. Johnston-Halperin. B. ‘‘Theory of magnetic bipolar transistor.’’ Phys. Joynt. H.’’ Nature (London) 419. Fal’ko. G. Res. 2003.’’ Phys. 2003. Jaffres. G... Ossau. Fu. B 17. Fulde. Zutic.-J. and size effects at critical points. Reuscher. Zakharchenya (North-Holland. and I. 69. and L.’’ Appl. Bastard. Z. G. 2001. Y. and N.’’ Phys. 2001. B 62. M. LeClair. T. Lett. Phys. D. ‘‘Andreev reﬂections and magnetoresistance in ferromagnetsuperconductor mesoscopic structures. thesis (Universiteit Twente). Friesen. I. 532–538 (1999)]. ‘‘Quantum transport in nonuniform magnetic ﬁelds: Aharonov-Bohm ring as a spin switch. 2003. Fanciulli.. M. H. 224516.’’ Phys. W. D. Shkrebtii. 9996–9999. W. Schuller. H. K. J. Kleim. Fert. V. Safarov. 1991. E. 024436. Rev. R. V.’’ Appl. Phys. 787–790.. Das Sarma. A.. A. Lett. Lei. Umansky.’’ Appl. ‘‘Magnetic bipolar transistor. D. 2003.. Eriksson. and S. Lett. D. ‘‘Diffusion equation and spin drag in spin-polarized transport.’’ Phys. Rev. and V. 1959. 2160–2162. 1999. Folk. and D. 115314.. Rev. 497–503 [JETP Lett.’’ Nature (London) 402. 83. E. Rev. in Optical Orientation. 69. I. 1708–1715. Rev. ‘‘Tunneling resonances and Andreev reﬂection in transport of electrons through a ferromagnetic metal/quantum dot/superconductor system. Feher. Sipe. B 69. and H. J. D. electron spin relaxation effects. Fiederling. Lett. Fujisawa. Rev. J. S. Zutic. 121301. E. Frustaglia. Jensen. Experimental.. and K. ‘‘Conductionelectron spin resonance in zinc-blende GaN thin ﬁlms. Phys. B 66. 84. B. 19. Meier and B.’’ J. 85–87. Richter.’’ Phys. 2425–2428. Hoving. and S. M. G. Das Sarma. Rev. Rev.and side-emission conﬁgurations. G.’’ Phys. A.. and S. Eksp. Gronsky.. Lassailly. M. B 67. Fabian. Mortensen. Tokura.. 15 144–15 147. C. W. Mater.. de Wijs.. 21. 1959. G. J. C. I. and S. and J.. ‘‘Allowed and forbidden transitions in artiﬁcial hydrogen and helium atoms. Huang. Nagle. 2001. K. T. B 48. Lampel. ‘‘Interfacial. Koopmans. and G. P. ‘‘Spin-ﬂip scattering of holes in semiconductor quantum wells.. D.’’ Ph. K. Modern Problems in Condensed Matter Science. 2002. J. Amsterdam). M. ´ Flatte. 2003. I. and T. Vol. Awschalom. van Wees. Schmidt.392 ˇ ´ Zutic. P. edited by F. B 64. pp. L. Austing. A. Hirayama. Lett. 198–208. Rev. 2004. ‘‘Theory of the bipolar spin switch. Potok. A. M. 2000. Falicov. ‘‘Spinpolarization in half-metals. Y. Waag. Fleisher. V. N. 9687–9691. Drouhin. B 16. and I. interface. Rugheimer. Kip.-F. Grabs. Molenkamp. in Optical Orientation. 256602. 5. Feher.. 98. ‘‘Spin diffusion in semiconductors. 278– 281. 1999. A.’’ Appl. C. J.. ‘‘Theory of spinpolarized bipolar transport in magnetic p-n junctions. 22. and A.. and R. Fominov. D. ‘‘Injection and detection of a spin-polarized current in a light-emitting diode. 2002. H. 2004. Meier and B. Vac. J. and S. Lett. Molenkamp. Rev. Tarucha. and H. G.’’ Phys. M. G.’’ Phys. F. B 43. G. Rev. Swagten. Y. 83. Rev. 1245–1256. ‘‘Spininjection device based on EuS magnetic tunnel barriers. H. Phys. ‘‘Quantum interference in electron-hole generation in noncentrosymmetric semiconductors.
Merrick. 119–128. A. W. and G. Pidgeon. 1952. ‘‘Ferromagnetic resonant tunneling diodes as spin polarimeters. Gruber. L. and M. Lett. Lett. A. 6074–6077. Grifﬁn.’’ J. Reuscher. ‘‘Study of superconductors by electron tunneling. and R.’’ Phys. Megerle. T. ‘‘Tunneling in the normalmetal-insulator-superconductor geometry using the Bogoliubov equations of motion. A.’’ Acta Phys. Titkov. H. 2002. and W.. p.. Lett. S. L. and A.. Eksp. 4063–4077. V.’’ Phys.’’ J. Ganichev. 2002. Phillips.’’ Phys. R935–R983. D. Bhattacharya. 951–952. Bauer. Lett. D. D. 347–350 [Sov. and W. C. L. Ivchenko. ‘‘Cuprate/manganite heterostructures.. Griswold.’’ Phys. T. Gu. 84. Prettl. 14. J. Barner. Eroms. G. Ossau. and G.ˇ ´ Zutic. H. F. W. Krotkov. D. Clarke. Schille.’’ Phys. Kip.’’ J. ‘‘Magnetoresistance in excess of 200% in ballistic Ni nanocontacts at room temperature and 100 Oe. Triques.’’ J. Tekh. J. ‘‘Spin relaxation and antisymmetric exchange in n-doped III-V semiconductors. . ‘‘Electrically detected magnetic resonance of two-dimensional electron gases in Si/ SeGe heterostructures.-P. B 67. J. M. Caballero. K. S. L. M. 2003. Phys. M. Rev. Zulicke. O.. A. Phys. 75. Supercond. V. D. 58. and R. Kleinsasser. Rev. ¨ G. Phys. Rev. Phys. ‘‘Effect of electron-hole interaction on electron spin relaxation in GaAs/AlGaAs quantum wells at room temperature.. 2003. Gim. Grundler. Kraus. Ivchenko. B 4. M. D.’’ Phys. ‘‘Scattering theory of the Johnson spin transistor. W. 175.. F.. and C. 2923–2926. Wegscheider. Va’sko. Fiederling. Phys. Lett. S. Vervoort. 2001. Ganichev. D. Furdyna and J. Schroll.. Fiz. W. and K. A. D. Appl.. J. Rev. Linﬁeld. 081302. A. 2000. Pol. and C. 182. 2. B 55. S. B. V.. Prettl. 86. V. I. Kein. Appl. W. Weiss.. Weiss. Mater. S. Gardelis. 416–426. Rev. Fazio. Ritchie. 13 242–13 250. L. W. G. E. Phys. Da Costa. 2003. Magn. G. P.. Ganichev. 1985. Marushchak.. J. F.. Goldman. D. B 68. N. and Das Sarma: Spintronics: Fundamentals and applications 393 Furdyna. Rev. 36–40 [JETP Lett. Teor. ‘‘Direct measurement of quasiparticle evanescent waves in a dirty superconductor. T. edited by J. 1999. P. Wegschelder. T. M. Rev. Fiz. Fulde.. V. P. S. Kamada. S. B 58. R. Browers. N. New York). Ounadjela. C.. 20. and P. M.’’ Phys. Matter 15. Kossut (Academic. 216–218 (1986)]. 1997. J. 78. U. ‘‘Properties of amorphous germanium tunnel barriers. Wegscheider. Magn. 1996.’’ Phys. P.’’ J. L. Schuh. Brandt.’’ Appl. Roussignol. G.’’ Phys. P. ‘‘Microwave spin resonance absorption by conduction electrons in metallic sodium. B 68. Fabian.’’ Adv. Smith. Glazov. Gijs. W. F. Bel’kov. P. Danilov. Teor. M. Skvarla. Stutzmann.’’ Appl. 403–405 (2002)]. Fulde. B 60.’’ Phys. Y. A. S. Semicond. Rev. 1999. N. ‘‘Spin-sensitive bleaching and monopolar spin orientation in quantum wells.. Kraus. M. Weiss. Gruzdov. Belkov. (S. R29–R64. 1971. 1. L. ‘‘Precession spin relaxation mechanism caused by frequent electron-electron collisions. W. 122. W. ‘‘Pumping spin with electrical ﬁelds. 2003. B 65. 3394–3398. and D. ‘‘Spin injection and transport in magnetic-superconducting oxide heterostructures. Graeff. and F. V.: Condens. 2550–2553. Thompson. Lett. Poluprovodn. M. E. Giaever. 1–9. S. and H. N. 82.’’ Phys. V. H. and J. Bichler. K. K. and W. Taddei. ‘‘Filtering spin with tunnel-coupled electron wave guides. Prettl. Garbuzov. P. ‘‘Electron spin manipulation using semimagnetic resonant tunneling diodes. L. A. ‘‘Measurement of the lifetime and of the spin-relaxation time of electrons in semiconductors by the optical-orientation method. and W. M. C. A. Molenkamp. D. Z. ¨ Governale. Allen. 87. Ivchenko.. Ando. Rev. J. T. 1988. B 59. ‘‘Spin-orbit scattering effects on the phonon emission and absorption in superconducting tunneling junctions. Guettler. and P. 1584– 1596. V. 7764–7767. Murdin. Sollinger. V. Grimaldi. P. T. 2000. Vol. D. C. A. 77. R.’’ Fiz. 140403. Abstreiter. D. N. and P.’’ Phys. Gregg. 2202–2208. D. ‘‘Optical polarization in Inx Ga1 x As-based quantum wells: Evidence of the interface symmetry-reduction effect. 2449–2451. Rev. ‘‘Spin photocurrents in quantum wells. April 2004 Goldman. Y. 2002. Eksp.’’ Pis’ma Zh. L. Ganichev. 476–478 [JETP Lett.. Sparks. Schmidt. Grimaldi. S. 24–26 (1971)]. Bauer. W. J. E. Barnes. R. M. Slater. H. 69–82. Appl. 1998.. R.’’ Phys. A 97. Phys. 64. A. and V. Tarasenko. Nikolaev.. Belkov. V. 1988. Phys. Rev. Lett. 88.’’ Phys.. R10 179–R10 182. S. M. V.’’ cond-mat/0301519. Wegscheider. 2001. M. Gadzuk. 140507. Kirschman. M. 76. and E. and E. 2000. L. A. and V. Demers.-Peterburg) 20. R..’’ Pis’ma Zh. 90. Phys. 2002. D. 46. A.. Beltram. ‘‘Spin-galvanic effect. and F. ‘‘Spin splitting of the conduction band of InP. Rev. W. K. Magn. A. R. F.. 1997. E. J. H.. D. P. Lett.. K. W. Sogawa.’’ Phys. S. 1971. and K. Iwamura. 1961. Harmand. Voisin. C.-W. M. and L. 2003. Ganichev. Viart. ‘‘The art of spin electronics. Boese. 1101–1103. ‘‘Microwave spin echoes from donor electrons in silicon. Holzmann. ‘‘Band-structure effects in the ﬁeldinduced tunneling of electrons from metals. 4358–4561. Va’sko. A. Murzyn. E. Garcia. Glazov. Fazio. V. Rev. Zhao. A. I. 2002. Prettl. and W. M. C. M. ‘‘Current injection into high-temperature superconductors: Does spin matter?’’ J. B. Rev. ‘‘Conversion of spin into directed electric current in quantum wells.. Vol. G. Rev. 155324. and J. Taddei. L. A. Beregulin. Rev. 15 523–15 530. 1969. Giazotto. 153–156. Phys. 1101–1111.. Rev. Geux. Mater. Governale. Rondi.. 1999. 88. and Y. N. and W.. Tarasenko. 1999. S. 2003.. Gester. Gaj. P. Brataas. 1986. and W. A.’’ J. P. Meservey. 283–290. ‘‘Perpendicular giant magnetoresistance of magnetic multilayers. B 66. in Semiconductors and Semimetals. Gorelenok.’’ Phys. 25. Gordon. Schneider... M. Ekimov. Larkin. Weiss. 13. Loloee.’’ Phys. Munoz. Gorkov. 1958. Prettl.’’ Phys. 82.. Kittel. S. Gotoh. 75.’’ J. ‘‘Theory of screening of the phonon-modulated spin-orbit interaction in metals. Rev. Schafﬂer. D. 2001.. Pratt. A. Danilov. ‘‘Spin-valve effects in a semiconductor ﬁeld-effect transistor: A spintronic device. ‘‘Diluted magnetic semiconductors. Rev. M. 033203. ‘‘D’yakonov-Perel’ spin relaxation under electron-electron collisions in QWs. 285–445.. Appl. C. 368– 370. Nikolaev. D. E. W. Ivchenko. 2001. Cooley. 200. ‘‘Spingalvanic effect due to optical spin orientation in n-type GaAs quantum well structures.’’ Nature (London) 417. Safarov. G. I. No. Ivchenko. F. Mod. Rev. C. E. J. 1997. 13. 286. Kagawa. Gibson. R. E. Magn. ‘‘Large Rashba splitting in InAs quantum wells due to electron wave function penetration into the barrier layers. and J. Ivchenko.’’ Phys. 057401.
7207–7210. Rev.. ‘‘Optical pumping in semiconductors. H. 1580– 1582... G. Rev. Montaigne. 1997. B. Hughes. Vandersypen. Dinia.. and O. B.. 2. X. W. Petrou. J. Yanovsky. 1998. 2. and H. and R. ‘‘Potentiometric measurements of the spin-split subbands in a two-dimensional electron gas. and Y. Schuhl. W. Kioseoglou. J. A. ‘‘Spinresonant splitting in magnetically modulated semimagnetic semiconductor superlattices. A.. Johnson.. Rev. Ploog. M. A. 1972. and S. 2002. H. L. 79. Awschalom. Aoi. Gupta. Bland. Jonker. 2001. Hanson. ‘‘Uber magnetische Beeinﬂussung der Polarisation der Resonanzﬂuoreszenz. Rev. Magno. T. Rodina. M. N. Lett. D. P.’ ’’ Appl. P. Lett. Johnson. 203–206. 91. and B. 2002. Otuka.-Q. Vink. R. Kuipers. M. and A. C. A. Rev. 1994. Phys. Lett. ‘‘Semiconductor few-electron quantum dot operated as a bipolar spin ﬁlter.. ‘‘Spin-dependent current transmission across a ferromagnet-insulator-twodimensional electron gas junction. Rev.’’ IEEE J. ‘‘Spin currents in magnetic ﬁlms. B.’’ Phys. T. M. Rev. Rev. 3487–3489.-M. ‘‘Subpicosecond spin relaxation in GaAsSb multiple quantum wells. ‘‘Hot-electron three-terminal devices based on magnetic tunnel junction stacks. Sipe..... 1985. R. W. Hammar. V. Happer.’’ Phys. 1332–1334 [Low Temp. Mod. 2003. Kuhn. Hayashi. Phys. and K. Ploog. and M. 1063–1068. B 64. K.. 4156–4158. Bogachek. ‘‘Proximity effects at ferromagnet-superconductor interfaces. L. ‘‘Transport properties of YBa2 Cu3 O7 /ferromagnetic interfaces. 2001. C. Hall.. Rev. Johnson. J.. Willems van Beveren. B 68. 1117– 1138. 2001. A. Rev. 2002. Nestle. Kost.’’ Appl. Shimada. Phys. 2131. D. T.. and U. Meservey. 196802. B. Halterman. M. and K. Tang and M. Roukes. A. Lett. R. van Wees. Sel. A.’’ Phys. A. and L. Eddrief. and M. (Paris) 10. B 59.. T. L.’’ Phys. Nishinaga.. D. and G. 2591–2593. Ruhle. 197201.’’ Appl. ‘‘Optical pumping. 3296–3305. Quantum Electron. van den Berg. 2003. Eberl.’’ cond-mat/0311414. 2003. Greene. van Beveren. K. Gupta. Awschalom. B. A. H. T.. U. J.. Schmerber. ¨ ¨ Hagele. 2000. Phys. Alivisatos. H. Y. X. P. ‘‘Analysis of the transport process providing spin injection through an Fe/AlGaAs Schottky barrier. L.’’ Phys. J. 2001. H.. S. Heide. ‘‘Coherent control of exciton density and spin. 2001. Hermann. W. A. F. 769–775. Gu. Lett. 1994... Hammar. 2001. E. T.. I. Monzon and H. ´ Hache. Magnetic Multilayers and Giant Magnetoresistance (Springer... P. 2002. B 42. 93–105. Rev. Johnson. B. 144411. bulk GaAs.’’ Phys. 8235–8243. B 66. N. and A. and H. 10 421–10 424. Rev. S. P. Kouwenhoven. B. ‘‘Nonmagnetic spinguides and spin transport in semiconductors. N. 75.’’ Rev. B 57. R. Vandersypen. Selvig. Phys. van t Erve. A. 1973.394 ˇ ´ Zutic. X.. Guertler. 1999. Guth. Vol. J.. Yanovskii.’’ Appl. N.’’ Phys. G. L. Schapers.’’ Phys. ¨ Heberle. ‘‘Efﬁcient electrical spin injection from a magnetic metal/tunnel barrier contact into a semiconductor. 78. ‘‘Spin transport in GaAs. C. 87. 73. Phys. and M. Lett. D.. Y. Kawazoe.’’ J. Itskos. E. Cryst. Covington. 146601. J. ‘‘Optical orientation and femtosecond relaxation of spin-polarized holes in GaAs. Growth 175Õ176. 81. T. G. Klapwijk. Hass. Lett. Rev. Lu. 2002. J. Kalinenko. Tsuchiya. Jonker. W. ¨ Hanle. Hanson. Rev. M. ‘‘A reply to the comment by F. Gurzhi. ‘‘Quantum beats of electron Larmor precession in GaAs wells. P. B. Phys. Binder. E. Bennett. Hehn. M. Tang. ‘‘Spin-polarized electron transport in . R.. Nizk. ‘‘Tunnel magnetoresistance in magnetic tunnel junctions with a ZnS barrier.’’ Phys. Yang. Atanasov. and K. J. Lett. 89. Elzerman. B 65. Hertz.’’ Phys. W. I. Hilton. Phys. H. and A. Bennett. ‘‘(GaMn)As: GaAs-based III-V diluted magnetic semiconductors grown by molecular beam epitaxy. C. H. G. Jonker. ‘‘Spin-ﬁlter effect of ferromagnetic europium sulﬁde tunnel barriers. S. 1997. B 56.’’ Z. L. Gurzhi. 3252– 3256. Berlin). A. J. Moodera. Borghs. R.’’ Physica C 235-240. Rev. M. Leonard. Zhao. April 2004 Hanbicki. L. J. 30. and Das Sarma: Spintronics: Fundamentals and applications Guo. Yang. J. N. J. Top.’’ Phys. B 61. T. 1999. Johnson. E. ‘‘Enhancement of spin injection from ferromagnetic metal into a two-dimensional electron gas using a tunnel barrier. J. van Driel. ‘‘Spin-orbit interaction in a two-dimensional electron gas in a InAs/AlSb quantum well with gatecontrolled electron density. Kopeliovich.’’ Appl. D. Xu. O. K. R. No. B 66.’’ Phys.’’ Phys. I. ‘‘Response to comment on ‘Efﬁcient electrical spin injection from a magnetic metal/tunnel barrier contact into a semiconductor. Lett. Lett. 2000. 80. 2000. 1999. 27. and A.. T. G. Temp. Li. J. 125307. J. Feldmann. N. Rev.’’ Phys.. Lett. 44. and A. and V. ‘‘Spin dynamics in semiconductor nanocrystals. H. Valls. Rev. Witkamp. 1996. V. B 8. 82. P. N.’’ Phys. R. ‘‘Spin coherence in semiconductor quantum dots. R. D. 88. L. 2002. Hao. 1998. L. Heida.’’ Appl. Rev. M. M. 1240–1242. W. Phys. Ruhle. M. Hanbicki.’’ Phys. W. R. A. A. 11 911–11 914. 985–986 (2001)]. Safarov.. Ed. Y. C. Kalinenko. Gustavsson. Rev. L. Itskos. J.. ‘‘Spinguide source for the generation of the high spin-polarized current. 1990. M. W. A. Hartman. Phys. 1905–1906. 184422. Peng. 27. Hershﬁeld. T. Rev. ‘‘Charge and spin transport through a metallic ferromagnetic-paramagneticferromagnetic junction. P. 155312. M. I. 1924. ‘‘Structural and transport properties of epitaxial Fe/ ZnSe/FeCo magnetic tunnel junctions. B 64. J. ‘‘Zeeman energy and spin relaxation in a one-electron quantum dot. and Y. ¨ Heberle. Kopeliovich. Phys. and L.’’ Phys. A. van Driel. A. George. P. A. 2002. 5024–5025. B. Lett. 72. 2001. M. Ostreich. Johnson. J. and A. Petrou. and K. and M. and M. ¨ Heersche.. K. Rev. Mod. 4092–4094. G.. V. Yasar. M. N. Mallory. M. Holmes. T.. Kohler. 83. P. K.’’ Ann. Efros. Takayanagi. ‘‘Observation of coherently controlled photocurrent in unbiased. Landman. Lampel. V. Fabian. Nitta.’’ Appl. T. 014509. Kouwenhoven. 169–249. H. P. A. R. H. and M. M. Tanaka. R.-L. 306–309. R. B 64.. J. and H. R. C. A. M. Lett. L.’’ Fiz. Hammar. Hammar. J. 84. F. Hanbicki. and D. and M. Kostoulas. 1997.’’ Phys. Chow. ‘‘Detection of spinpolarized electrons injected into a two-dimensional electron gas. and C. Lett. and H. J. P. Lett. 066806. ‘‘Spin-dependent tunneling from transition-metal ferromagnets. Kioseoglou. R. ‘‘Observation of spin injection at a ferromagnetsemiconductor interface. 125113. 2001. 161307.’’ Phys.. T. Hirohata. Blaumberg. 78. 3887–3890. Phys.. Rev.’’ Phys. Elzerman. 2003. 76. Hammar.’’ Phys. G. Etgens.
X. A.’’ Phys. G. 109–148 (2002)]. Strand. 32. 87.47As/In0. 161304. 2001. Nitta. and B.’’ Phys.’’ Usp. A 64. 3793–3810.. Kobayashi.. Filip. F. 713–716. and X. B.. T. D. Hirsch. Rev.’’ Phys.’ ’’ Appl. H.-Z. Yu. ‘‘Determination of the spin polarization of half-metallic CrO2 by pointcontact Andreev reﬂection. Heersche.-M. C 13. Symmetry and Optical Phenomena. Maekawa. E. van Wees. A. J.. Lett. Giant Magneto-Resistance Devices (Springer. Jia. J. Das Sarma. (Springer. Rev.’’ Phys. Yan. 2003. 2003. F. G. Jedema. Osaka. 1980. 2000. Hong. Rev. 83. 7736–7739.. thesis (Rijksuniversiteit Groningen). Forro. 2000. T. J. and T. Jedema. Hirota. Itoh.’’ J. 5585–5588. F.. ‘‘Effects of spin-ﬂip and magnon-inelastic scattering on tunnel magnetoresistance. A. R. 1999. R. 106601. 2002. ‘‘Electrical detection of spin accumulation and spin precession at room temperature in metallic spin valves. Rev. Y. 113–154 [Phys. Jansen. 2130.’’ Nature (London) 416. C.’’ Phys.. Rev. S. April 2004 Ivchenko. Mater. ‘‘Resonant and nonresonant conductionelectron-spin transmission in normal metals.. Rev. 89. and I.’’ Phys.. B. R12573–R12576. 345– 348. Jansen. F. B. 1997. Makeawa. and T. M. ‘‘Zero-ﬁeld spin splitting in an inverted In0. 1980. Byers. 2002. D. Anguelouch. C. J. 4707–4709.’’ Phys. J.. Magn. and B. Chauvet. Hu.. 2002b. Nauk 172. J. J. Zawadzki. Khusainov. Izyumov. ‘‘Spin-accumulation-induced resistance in mesoscopic ferromagnet-superconductor junctions. 216601. 2002. Berlin). and L. F. and B. and Das Sarma: Spintronics: Fundamentals and applications 395 ferromagnet/semiconductor hybrid structures induced by photon excitation. L. ‘‘Midgap surface states as a novel signature for d x 2 x 2 -wave superconductivity. F. Hansen. J. B 62. 1994. M. Pekker.’’ Phys. Filip. Magn. Chien. van Wees. 1999.’’ cond-mat/0108339. Takayanagi. Phys. G. Jedema. 2001. Jedema. ‘‘Origin of the zero-bias conductance peaks observed ubiquitously in high-T c superconductors. and K. J. ‘‘Current-induced spin-wave excitations in a single ferromagnetic layer. Magn.’’ Phys.. 2003. ´ Isakovic. B... 9047–9050. B 63. and B. 2003. Shibata. L. Palmstrøm. Inoue. J. 041307. J. J. D.. B 64. ‘‘Electrical spin injection and accumulation at room temperature in an all-metal mesoscopic spin valve. J. Yu. 1993. T. Rev. J. A. and H.-M. J. Rev.. B 21. B. P. Hu. 276601. B. ‘‘Spin-polarized transport in a two-dimensional electron gas with interdigital-ferromagnetic contacts. and B. Schultz. Klude.’’ Phys. Shi. 1526– a b 1529. Filip. R. 1999. J. Das Sarma. B 61. Rev.’’ J. 085319. Costache. 2001. 041307. Pikus. J. Stiles.’’ Phys. van Wees. M. Filip. F. J.’’ Phys.’’ Phys. Hu. F.ˇ ´ Zutic. and E. 1996. 2001.. ‘‘Competition between superconductivity and magnetism in ferromagnet/superconductor heterostructures. ¨ ¨ Hubner. New York). Ruhle. ‘‘Hilbert-space structure of a solid-state quantum computer: two-electron states of a double-quantum-dot artiﬁcial molecule. ‘‘Conduction electron spin resonance in Rb3 C60 ..n0m surfaces of d-wave superconductors. A 61. J. Suzuki. Lett.’’ Phys. 2001. 2001. E. 1632–1639.. Supercond. B 57. R. J.’’ Phys. Rev. Wyder. J. J. C. van Wees. Baselmans. R. ‘‘Electrical detection of spin precession in a metallic mesoscopic spin valve. 90. ‘‘Mesoscopic SternGerlach device to polarize spin currents. Jansen. 1091–1094. B 60. 2000. and S. van Wees. ´ ´ Janossy. Superlattices and Other Heterostructures. 81. O. L.’’ Phys. and A. M. and M. Phys. Hu. A. B 67. S. 2002. Cooper. Hu. Takahashi. H. 1999. Das Sarma. J. J. 1999. Jedema. Lett. 125333. Hommel. and S. 2001. Filip. Mod. M. ‘‘Conductance quantization and magnetoresistance in magnetic point contacts.-R. M.. C.’’ Nature (London) 410. B 60. ‘‘Electrical spin injection in metallic mesoscopic spin valves. 5589–5600. 198. Phys. 2001. E. Rev.’’ Phys. B 63. Phys. Inomata. S. 16 549–16 552. and H. Ionicioiu. ‘‘Spin Hall effect.. M. Jedema. Vol. Carr. A. Hoving. ‘‘Spin injection and spin accumulation in all-metal mesoscopic spin valves.’’ IEEE Trans. Fabian. F. C.’’ Phys. Y. A.. Q. Maekawa. 042312. and P. and B.’’ Nature (London) 416. Proshin. Inoue. Ji.’’ Phys. Soc. E. Bhat. A. ‘‘Effects of randomness on tunnel conductance and magnetoresistance in ferromagnetic tunnel junctions. H. 68. 1266–1276. . R. S.. B 60.’’ Phys. Jedema. Y. Rev. Lett. Crowell. ‘‘Spin-valve effects in nickel/silicon/nickel junctions. ‘‘Spintronics (communication arising): Spin accumulation in mesoscopic systems. Nijboer. Lett. C. G. D. and J. W. H. ‘‘Spin injection and spin accumulation in permalloycooper mesoscopic spin valves. and M.53Ga0. Heersche. 72. 84. Klapwijk. J. Kumazaki. Chien. Rev. Takayanagi. A. ‘‘Decoherence and dephasing in spin-based solid state quantum computers. 2. 86. Xiao. Lett. 71. Phys.’’ J.. 1999.’’ Phys. Hu. 2nd ed. T. 1834–1837. 2002.52Al0. Matsuyama. van Wees. Pfeffer. and Y. Gupta. van Wees. Hu. 2002. R. J. ‘‘Negative spin polarization of Fe3 O4 in magnetite/manganite-based junctions. Y. Filip. A.. Baselmans. Rev. Rev. A. Imamura. Rev. D’Amico. Akazaki. Lett. Usp. 81. Sakakima. J.’’ Phys. J. Nijboer. 76. P. 104425. 062301. 1003–1006. and W.48As heterostructure: Band non-parabolicity inﬂuence and the subband dependence. Jpn. 90. Rev. 6073–6118. 15. ‘‘Gate-controlled electron spin resonance in GaAs/Alx Ga1 x As heterostructures. 167–169. No. ‘‘Spin-based quantum computation in multielectron quantum dots. ‘‘Direct observation of optically injected spin-polarized currents in semiconductors. A. J. and P.’’ Appl. B. and S.D. X. ‘‘Magnetoresistance in doped magnetic tunnel junctions: Effect of spin scattering and impurity-assisted transport. Lett. 27–35. J. and S. van Gelder. A. C. Lett. F. 5162–5164. 45. Rev. T. G. J. ‘‘Spin injection across a heterojunction: A ballistic picture. 1998. 2002. A.. van Driel... A. M. Ji. A.’’ Ph. ‘‘Pointcontact spectroscopy in metals. A. H. Hu. Moodera. Rev.. Chou. W. Nitta. Yang. H. A. ‘‘Predicted giant magnetic moment on non. H. M. Yablonovitch. N. E. J.’’ Phys.. Rev. Baselmans. 2000. T.-R. Jensen. T.-M. ´ Janossy.. Heersche. Lett.. B 64. V. 066803.. C.. X.. de Sousa. J. 810. and S. J. Lett. Rev.. 2002a. L.’’ J. and G. W. Rev. N. Hu. Hu. T. J. and S. H. Y. Sipe. Mills. Jedema. Jiang. D. and D. C. Fiz.. R.. Rev. S. ‘‘Spin dependence of the inelastic electron mean free path in Fe and Ni: Explicit calculations and implications. C. Strijkers. and S. B 67. Rev. ‘‘Optical pumping in ferromagnet-semiconductor heterostructures: Magneto-optics and spin transport. R. Rev. A.’’ Phys. ‘‘Comment on ‘Efﬁcient electrical spin injection from a magnetic metal/tunnel barrier contact into a semiconductor. J. H.
3934–3939..’’ J.Mn)As. Petrou. Johnson.’’ Phys. S. Wei. X. E.. 2001. A. 256603. ‘‘Evidence for an electric-ﬁeld-induced phase transition in a spinpolarized two-dimensional electron gas. Johnson. 1988a. A. T. 1993b. Sci. 60. Rev. 1980. M. ‘‘A silicon based nuclear spin quantum computer. E. New York). 2000. ‘‘Spin injection into metals and semiconductors. ‘‘Aharonov-Bohm oscillations and spin-polarized transport in a mesoscopic ring with a magnetic impurity. in Magnetic Interactions and Spin Transport.. 1988b. Solid State 22. and S. 5326–5335. T.’’ Phys. D. Coldren. G. Lett. V. D. and R. Solid State 28. 1988d. 2002a. Rev. 298–309. A. Glembocki. Jonker..’’ Phys. Kalevich. 55. Silsbee. Johnson.’’ Appl.. Tverd. Parkin. Kawakami.’’ Phys.’’ Appl. H. Kane.’’ Nature (London) 393.. T. Jonker. and R. Rev. S. 67. M. B. 115. Phys. Fabian. Phys. Supercond. J. B. 892–893 (1981)]. 7264–7267.. Wolf. and X. 63. Johnston-Halperin. Blewett. E. 2000. R. 3594–3597. ‘‘Electron spin injection and detection at a ferromagnetic-paramagnetic interface. Kaiser. B.. Pierece. Status Solidi B 221.’’ Fiz. Patent No. Tanaka. 8180–8183. Wagner. and Das Sarma: Spintronics: Fundamentals and applications Jiang. Rev. 1991. Sahoo. U. Kashiwaya. Fleisher. Kul’kov. I. J. Dordrecht/Plenum.. N. 1998. Rev.’’ Phys.’’ Fiz. 2001. Silsbee. Johnson. and M. Awschalom. Johnson. J. Idzerda (Kluwer Academic. and Y. K. G. M. Phys. Prog. 2003. R. M. 2. 63. Silsbee. M. Bell. and R. I. Wang. Johnson. Lofgreen. Park. 740–748. Lett. and L. Rev. 6141–6152. Gossard.. 1987. B 35.. Rev. Lett. A. 125112. B 62. 90.. Wang. ‘‘Spin coupled resistance observed in ferromagnet-superconductor-ferromagnet trilayers. Johnson. Rev.. R. P. and R. 2142–2145. 1460–1462. Phys. ‘‘Highly spin polarized tunnel injector for semiconductor spintronics using MgO(100). Mod. Johnson. Rev. 48. D. H. W. M. and D. Johnson.. W. Rev. H. ‘‘Charge and spin diffusion in mesoscopic metal wires and at ferromagnet/nonmagnet interfaces.. 377... Shelby. 67.’’ Phys. 1947–1949 (1986)]. L. 1524–1526 [Sov. and R. Lett. Rev. G. Lett. Holm. Jonker. 77. 1992. Phys. Lett.. T. ¨ Kalt. Fleisher. 2002b. M. O. Appl.’’ Phys. Hanbicki. Johnson. R. S. D. U.. van Dijken. 2003. J.. A. Silsbee. 2003a. ‘‘Robust electrical spin injection into a semiconductor heterostructure. Shelby. V. H. Y. D. Lett. Parkin. Kioseoglou. ‘‘Silicon-based quantum computation. H. ‘‘Manifestation of the sign of the g factor of conduction electrons in resonant cooling of the nuclear spin system of a semiconductor. K. K. edited by A. Tela 22. R. 79.’’ cond-mat/030445.749 (23 February 1999). . 1999. 1988. 54A. B. Phys. M. 1981. ‘‘Analysis of anomalous multilayer magnetoresistance within the thermomagnetic system.. 041306. M. G. 2000. Lett.. 477–480. ‘‘Coupling of electronic charge and spin at a ferromagnetic-paramagnetic metal interface. Rev. Phys. 075322. Rev. and J.’’ Phys. 1990. 76. ‘‘Tunneling effects on surface bound states in unconventional superconductors. Brown. 2000. Johnson. 1975. D. and R. S. ‘‘Enhanced carrier lifetimes and suppression of midgap states in GaAs at a magnetic metal interface. T. K. Harris. Johnson. Park. ‘‘Optical detection of hot-electron spin injection into GaAs from a magnetic tunnel transistor source.’’ Phys. S. 515– 564. Phys. T.. Jayannavar. Furis. M. 1406–1409.’’ Nature (London) 416.S.. T. Jonker. Cheong. M. Solid State 23. 075320. S. Itskos.’’ Phys. Tela 23.’’ Phys. 2001. Lett. ‘‘Quantifying electrical spin injection: Component-resolved electroluminescence from spin-polarized light-emitting diodes. ‘‘Thermodynamic analysis of interfacial transport and of the thermomagnetoelectric system. Silsbee. and Y. Vol. ‘‘The role of spin orientation and relaxation in exciton-exciton scattering.’’ Rep. H.. Johnson. P. Kalevich. B 46. 4959–4972. 1985. ¨ Kainz. Bennett. ‘‘Direct investigation of subsurface interface electronic structure by ballistic-electronemission microscopy. Tela 28. J. Tverd. Phys.. ‘‘Spintronics (communication arising): Spin accumulation in mesoscopic systems. 1993a. 5312–5325. B 68. T. Kioseoglou.. Rev. D. 2004. and A. V. 17.’’ preprint. and V. 5. B. 1959. Young.’’ Phys.. 1790–1793.874. R. Petrou. Harris. X. M. R. and A. Erwin. V. D. Galbraith. B. M. 809–810. April 2004 Jonker. and S. S. Solomon. Johnson. and I. 2003b. 225–226. Petukhov.’’ Fiz.’’ Semicond. D. and R. 1988c. B. Rev.’’ Phys. M. S.396 ˇ ´ Zutic. 2003. R. S. D. ‘‘Anisotropic spin splitting and spin relaxation in asymmetric zinc-blende semiconductor quantum structures. ‘‘Interfacial charge-spin coupling: Injection and detection of spin magnetization in metals.. ‘‘Manifestation of the sign of the g factor of conduction electrons in resonant cooling of the nuclear spin system of a semiconductor. West. and A. 1986. Clarke.. Kaplan. 79.. ‘‘Bipolar spin switch. Bank. 575–577. 4886–4889.. Kalevich. C. ‘‘Spin-polarized scanning tunneling microscope: Concept. pp. T. Silsbee. M. V.’’ Phys. J. G. Jiang.’’ J. ‘‘Dynamic nuclear polarization by spin injection. M. Phys. L. J.’’ Appl. M. Johnson. Rev. ` Julliere. K.’’ Fortschr. R. 70. ‘‘Spin nomenclature for semiconductors and magnetic metals. ‘‘Electrical spin injection and transport in semiconductor spintronic devices. Rev. ‘‘Spin accumulation in gold ﬁlms. design. S.. 60. B. B 64. H. 3462–3465 [Sov. 1023–1041. 65. Lett. 703–705 (1980)]. Pfeiffer. Winkler. 133–137. Stiles. 1994. D. T.’’ J.’’ Phys.. Technol. D. B 37. ‘‘Spin-polarized Zener tunneling in (Ga. D. B 37. M. M. and J.’’ Phys. and K. Rev. ‘‘Tunneling between ferromagnetic ﬁlms. 14. Trondle. B 67. H. C. M. Lett. and V.. Joshi. Wachter. No. and R. G. Macfarlane. Jonker. B. J. Macfarlane.’’ MRS Bull.’’ Phys. Kul’kov. G. 1208–1211 [Sov. 273–281.’’ Phys. Rossler.’’ Phys. 1997. 28. 3098–3100. B 65. Y. ‘‘Spin injection: A survey and review. E. ‘‘Optically induced nuclear magnetic resonance in semiconductors. Byers. 320– 323.. 2002. 2000. and preliminary results from a prototype operated in air. 1641–1724. Kane. B.. Kane. Johnson. Tverd.. Petrou. ‘‘Ferromagnetnonferromagnet interface resistance. G. M. 2003. Chtchelkanova.’’ Science 260. ‘‘Spin injection experiment.. ‘‘Application of the diffusion-modiﬁed Bloch equation to electron spin resonance in ordinary and ferromagnetic metals. Hanbicki. Phys. Johnson. K. Appl. 1680–1682..
027903. 3572–3580. 139–141. Rev. ‘‘A spin Esaki diode. Lett. X. D. Quantum Theory of Solids (Wiley. Kavokin. Kimel. A. New York).. 684–696. and Y. 380–383. Kivelson. 2001. T. ‘‘All-optical magnetic resonance in semiconductors.’’ Phys. A. Yanase. K. Pavlov.. V. Phys. Glazman. private communication. Rev. N. Rokhsar.. Takayanagi. k. D. Part 2 40. H. Khaetskii. G. Kikkawa. V. R10 610–R10 613. B 35.’’ Phys. ‘‘Ideal spin ﬁlters: A theoretical study of electron transmission through ordered and disordered interfaces between ferromagnetic metals and semiconductors.’’ Jpn. V. N. V. and H. Tanaka. 2001. ‘‘Anisotropic exchange interaction of localized conduction-band electrons in semiconductors. H. Rev. B 64. 2003. Kikkawa. 1996. F. Kimel. 1963. Datta. Rev. Awschalom. 1987. V. and S. Rev. and T. 677–680.’’ Rev. J. Koga. J. A. ‘‘Current-driven magnetization reversal and spin-wave excitations in Co/Cu/Co pillars. 046801. C. and S. Samarth.’’ Phys. J. Rev. 47. Pisarev. 195329. Kimura.. G. S. J. Knap. ‘‘Spinﬁlter device based on the Rashba effect using a nonmagnetic resonant tunneling diode. B 61. J. J.. Eksp. C. N. Rev.. ‘‘Electron spin decoherence in quantum dots due to interaction with nuclei. J. C. D. 2003. X. 134425. Awschalom. ‘‘Spin coherence in semiconductors: storage. L. Kikuchi. Rev. 1976. 1996. Loss. Malajovich. Introduction to Solid State Physics.. and D. Rev. 2002a. V. Kobbi. Gossard.. Gupta. Fiz. V. Kauschke. 054422. Rev. No. and J. K..’’ Phys. and D. J. Sankey. Polarized Electrons (Springer. Tokura. 2001.’’ Phys.. Bertho. ‘‘Anomalous transport phenomena in Eu-chalcogenide alloys. Immamura. Albert. Rev. ‘‘Conductance quantization and Andreev reﬂection in narrow ferromagnet/superconductor point contacts. ‘‘Resonant spin ampliﬁcation in n-type GaAs. Rev. ‘‘Spin-polarized current-induced magnetization reversal in single nanowires. and D. ‘‘Symmetry of anisotropic exchange interactions in semiconductor nanostructures. ‘‘Microwave oscillations of a nanomagnet driven by a spin-polarized current. Ohno. ‘‘Progressive suppression of spin relaxation in two-dimensional channels of ﬁnite width. et al. 88.’’ Phys. A. J.’’ Phys. Phys. Hoffer.. ‘‘Electron spin evolution by interaction with nuclei in a quantum dot. Takayanagi.’’ Phys. Rasing. Keldysh.’’ Phys. Smorchkova.’’ Phys. C. Kikkawa. B 53. L1274–L1276. Katine. Pikus. 2002a. and D. and D. M. Schoelkopf. 2002b. Mod. 1999. F. Rev. D. L. 126601. Mod. 473–476. C.’’ Phys. and reduced dimensionality. Kavokin. J. 221–227. 3149–3152. Kittel. 78. Ohno. 88. Maekawa. B 67..’’ Zh. 1284–1287. Terakura. 27–31. ‘‘Bogoliubov quasiparticles. R. Kittel. and S. ‘‘Spin current in ferromagnet-insulator-superconductor junctions. Gridnev. and T. A. Kelly. Rev. 1964. 12 639– 12 642. Beasley.. L.. Kikkawa. I.’’ Phys. and L. Lett. Nitta. . V. ‘‘Exchange in silicon-based quantum computer architecture. Hu. C. 125316. Nazarov. 3912–3924. Buhrman. N. W. Status Solidi A 190. Rasing. (Paris) 11. V. 2003. K. Rev. Cardona. and Y. 2002. 2001. and K... V. B 62. and Das Sarma: Spintronics: Fundamentals and applications 397 Kashiwaya. B 68.. (Wiley. Takahashi. G. 1018–1026 (1965)]. 2002. B 63.’’ Nature (London) 395. A. K. F. Driscoll.. Buhrman. T. M. Teor. E. R. ‘‘Rashba spin-orbit coupling probed by the weak antilocalization analysis in InAlAs/InGaAs/InAlAs quantum wells as a function of quantum well asymmetry. Kim. Akazaki.ˇ ´ Zutic. A. Kavokin. T. S. and D. R. 2000. 194–201. Kiselev. V. ‘‘Spin relaxation in semiconductor mesoscopic systems. Kastler.’’ Physica E (Amsterdam) 10.’’ Phys. Ansermet. D. M. S. V. 40. 2003. B 41. ‘‘Lateral drag of spin coherence in gallium arsenide. Ralph. ‘‘Spin relaxation in semiconductor quantum dots. D. 2001. Koiller. Lett. ‘‘Spin-ﬂip transitions between Zeeman sublevels in semiconductor quantum dots. V. F. F. Koga. Nazarov. P. and M. Fabian. H. A. V. Pavlov. 1968. and spin-charge decoupling in superconductors.’’ cond-mat/0212347. Truong. A. Kirczenow.’’ Phys. Iordanskii. ‘‘Ultrafast dynamics of the photo-induced magneto-optical Kerr effect in CdTe at room temperature. B 60. Rev. B 63. Khaetskii.. ‘‘Optical manifestations of electron spin transport and relaxation in semiconductors. D. Khaetskii. 1999. M. and D.’’ Phys. K..’’ Phys. 775... 2. and L. N. T. Kiselev. Matsukura. 2000. I.. and A. 2000. Takamura. Nitta. ‘‘Room-temperature ultrafast carrier and spin dynamics in GaAs probed by the photoinduced magneto-optical Kerr effect. Emley. A.. and D. M. C. spinons.-P. J. B.. Lett. Awschalom... C. Saweda. D. N. Phys. D. LitwinStaszewska. R... D. J. Khaetskii.’’ J. Phys.’’ Phys. V. Rev. 2000.’’ Nature (London) 397.’’ Appl.’’ Phys. Rev. Levy.’’ Science 299. V. 075305. 1998. S. J.’’ Science 277. Glazman. Y. 020508. Kobayashi. 84. W. 76. ‘‘Spin relaxation of holes in the split-hole band of InP and GaSb. 1990.. 235201. Myers. and Y. A.. B. A. Rev. 1998. 89. 186802. J. A. Kiselev. Gridnev. Vol.. Myers.’’ Phys. 1997. R. 1950. Bentivegna. W. Kikkawa. V. 4313– 4316. A. Awschalom. Pikus. Das Sarma. Ralph. A.. N.. Mestres. J. 2002.’’ Nature (London) 425.. Y.. 2002b. E. Kato. Wegrowe. S. Lett... J. T. 3843–3853. 88. B 65. Awschalom.’’ Phys. I. and K. Kim. D. I. ‘‘Quelques suggestions concernant la produc´ ´ tion optique et la detection optique d’une inegalite de population des niveaux de quantiﬁcation spatiale des atomes— ` ` application a l’experience de Stern et Gerlach et a la resonance magnetique. Rev. Kessler. B 64.. J. 2001. A.’’ Phys. V. 2000. New York). J. transport. ‘‘T-shaped ballistic spin ﬁlter. M. Phys. Khaetskii. T. R. A. V. and H. Mosser. B 61. April 2004 Kikkawa. ‘‘Diagram technique for nonequilibrium processes. F. ‘‘Room-temperature magnetoresistance in an oxide material with an ordered double-perovskite structure. E.’’ Physica E (Amsterdam) 9.. C. 80. Kasuya. and D. Bentivegna. A. V. W. R. Loss. Awschalom. ‘‘Gigahertz electron spin manipulation using voltage-controlled g-tensor modulation. Y. V. and M. 13 115–13 120. Zduniak. K. 2001. Appl. Phys. Krivorotov. V. M. Skierbiszewski. 1201–1204. Kohda. V. 255–265.’’ Science 287. 11 693–11 696. S. A. ‘‘Room-temperature spin memory in twodimensional electron gases. JETP 20. 2003.. Lett. Lett. New York). 1515–1527 [Sov. Rev. 7th ed. 2001. K. Yoshida. Robert.-E. ‘‘Weak antilocalization and spin precession in quantum wells. Pisarev..
’’ Zh. M. and H. C. H. Levitov. Y. April 2004 Lamari. B 3. 1989. Phys. Eisenmenger. N. Rashba. C.. Rev. Sham. Rev. Mater. B 67. Wang. and J.’’ J.. Levy. 93. 161301. JETP 61. ‘‘Spin life-times of electrons injected into GaAs and GaN. 78. ‘‘Photoinduced magnetization in dilute magnetic (semimagnetic) semiconductors. Y. I. Gregg. MacDonald.’’ Phys.. A. K. W. Koshihara. 47–111.’’ Appl. 2002. Raimondi. 107201. and Das Sarma: Spintronics: Fundamentals and applications Koiller. 90. Xu. Lepine. pp. Guo. ¨ Konig. Mortazavi. Lett. Urano. MacLaren. Lepine. 1195. 2003. G. 163–211. Z..’’ Phys. T. 1397–1399. Rev. M. Bønsager. 2001. Lett. Kirschman. ‘‘Ferromagnetic order induced by photogenerated carriers in magnetic III-V semiconductor heterostructures of (In. B 59. Lett. ´ Levitov. 412–416. J. C. A. Bullock.. and G. Olesberg.’’ Phys. A. C.. Rev. . Khandelwal. Zawadzki. F. ¨ Lommer. Hirasawa. Long. Venkatesan. Kravchenko. 2002. H. ‘‘Spatially resolved spin-injection probability for gallium arsenide. H. M. Lambert. 90. and P. 121. D. ´ Lau. 2002. 9845–9864. 121310. Jenniches. 436–441. M. S. and U. Mortazavi. Phys. Lett. S. J. P. W. New York). S. Phys.’’ Phys. 88.. M. Martinek. pp.. 166602. C. S. Krenn. 1994. Lang.’’ Phys.. ‘‘The electric conductivity of a laminated metal system (alternating magnetic and nonmagnetic layers). 6682– 6685. and L. Appl. Jungwirth.’’ Phys. S. ‘‘Electronspin decoherence in bulk and quantum-well zinc-blende semiconductors. Phys. Choi. W. LeClair. P. Rev. 1971.. N. Lebedeva. and P. S.’’ Phys. Lett. M. 1518–1521... ‘‘Spin splitting in semiconductor heterostructures for B→0. 2000. Iye. B 2. Rev. 76. and R. J. C. J. 2003. B 64. P. I. Kreuzer.. 55. S.398 ˇ ´ Zutic. J. Rashba. P. and E.. Lett.’’ Physica E (Amsterdam) 2. Levy. D. E. K. Papaconstantopoulos (Springer. and S. edited by D. M. Rev. Rashba. and R. 2001. Krishnamurthy. Gardelis. 2003.’’ Phys. V. S. G. H. Teor. 268. and H. Phys. J. 1999.’’ Phys.. 91. W. in Spin Dependent Transport in Magnetic Nanostructures. S. ¨ Konig. Schuh. 1985.-F. W. ‘‘Theory of the coherent spin dynamics in magnetic semiconductor quantum wells. ‘‘Spin polarization of injected electrons (reply). 147902. W. 703–727 [Sov. 108. and A. W..’’ Science 292. Emery. ‘‘Magnetization reversal and magnetoresistance in a lateral spin-injection device. B 39. H. Bauer. Salamo. W. 1510–1513. Barnes. and D. ‘‘Spin injection into a ballistic semiconductor microstructure. E. ‘‘Dissipationless spin transport in thin ﬁlm ferromagnets. J. D. W. 75. Lett. 686–690.’’ Phys. ‘‘Universal quantum computation with spin-1/2 pairs and Heisenberg exchange. Krey.. J. 2003. W. C. Krinichnyi. Appl. 2546–2549. ‘‘Interaction-driven spin precession in quantum-dot spin valves. J. and J. Bland. Rossler. Lett. Maekawa and T. Rev. Shinjo (Taylor and Francis.’’ Appl.Mn)As/GaSb.’’ J. I. ‘‘2-mm waveband electron paramagnetic resonance spectroscopy of conducting polymers. Lett. J.’’ Phys..’’ Phys. 096803. Korotkov. ‘‘On the signiﬁcance of quantum effects and interactions for the apparent universality of Bloch laws for M s (T). D. A. Magn. B. 728–731. Loraine. 187202. J. Oliver. Rev. P. Phys. 320–323. R. 2003. 133–137 (1973)]. ‘‘Ultraslow electron spin dynamics in GaAs quantum wells probed by optically pumped NMR. ‘‘Disentangling the exchange coupling of entangled donors in the Si quantum computer architecture. Mod. J. M. 2003. J. LaBella. Gijs. A. Jansen. Kolbe. H. 89–92. Z.’’ Zh. and G. 20. H. 2004.: Condens. Das Sarma. Lenczowski. M. 2003. and M. 8682– 8684. Rev. West. Wegscheider. D. W. and G. A. D.’’ J. Bauer. Munekata. V.. LaBella. Lett. van de Veerdonk. D. Appl.’’ Appl. and K. ‘‘Phononinduced spin ﬂip in extremely thin superconducting Al tunneling junctions in high magnetic ﬁelds. Q. Kaltenegger. Teor. Nazarov. Linﬁeld. 1998. U. ‘‘Modeling of ferromagnetic semiconductor devices for spintronics. G. Y. J.. 2003.. N. 2002. L. 87.. J. Eksp. 5154–5159. Flatte. W. Weiss. Fiz. 1997. ‘‘Nonequilibrium spin distribution in a single-electron transistor. M. H. W. Allen. N. F.. Schliemann. Barrett..’’ J. 277–291. Phys. and V. N. 89. V. 173–221. J. R.’’ Science 281. and I. Pugh. 2000.. T. Rev.’’ Phys. L.. ‘‘Optically induced magnetization in a dilute magnetic semiconductor: Hg1 x Mnx Te. Fabian.’’ Phys. Takagi. Newman.. Korenblum. Mertig. Thibado. Emery. 2002. S.. M. 1996. Swagten. W. H. B 67. H. P. Ding. ‘‘Band structure and density of states effects in Co-based magnetic tunnel junctions. 165329. I.. Rev.’’ Phys. H. ‘‘Currentdistribution effects in microstructures for perpendicular magnetoresistance experiments. Dietl. Fulde. C. M. and M. Ya. J. Krenn.. van Schilfgaarde. Rev. M. Rev. J. Thompson. W. Phys. 2004. 2002. 1968.. 1998. Flatte. Janssen. Oiwa. 1998. Giesbers. V.. 2003. G. H. X. 1761–1763. L.’’ Phys. ‘‘Spin-dependent recombination on silicon surface. Malcher. Y. P. R. and J. J. Matter 10. I. J. A. and M. private communication. Safarov.. and A. ‘‘Nuclear dynamic polarization by optical electronic saturation and optical pumping in semiconductors. 491–493.. S. 1988.’’ Phys. T. S. 60. L. V. 603–626 (2002)]. Lampel. 1970. Kravchenko. 2001. P. Singh and D.’’ J. Moodera. Spalek. Hu. Katsumoto. Oliver. ´ Lau. Appl. Vol. Met. 88. 2429–2439. Kouwenhoven. Bichler. B. 1984.. 85. W.’’ Phys. Lett. de Jonge. Fiz.’’ Science 292. edited by S. Berlin).. H. and E. M. C. Lee. Ritchie. ‘‘Effect of the doping concentration on the zero-ﬁeld splitting and Rashba parameter in a p-InAs MOSFET. Ding. E. J. N. A. and M. 2002. in Electronic Structure and Magnetism of Complex Materials. V. J. and E. 10 918–10 934. Wieldraaijer. Rev.’’ Synth. B. ‘‘Spin relaxation time of conduction electrons in bulk sodium metal. Thibado. MacDonald. A.. J. H. Kuivalainen. Lett. 1972. Rev. 067401. 77. Salamo. 2002. Kohlhepp. F. ‘‘Spin resonance of localized and delocalized electrons in phosphorus-doped silicon between 20 and 30 K..’’ Phys.. Pfeiffer. Smith. Rev. 1999. Linder. Rev. T. J.’’ J. 229–236 [Sov.’’ Phys. Sun. B. Phys. Bullock. 4582–4584. 2. and J. Lett. B 6. Davis. 83. Ya. E. 901–941. Lett. Magn. ‘‘Dynamical spin-electric coupling in a quantum dot. I. S. Rev. B 67. 80. F. Sirisathikul. J.-C. No. van de Vin. H. 89. Eliashberg. and N. Eksp.. Venkatesan. H. ‘‘Phase-coherent transport in hybrid superconducting nanostructures. M. 4617–4620. ‘‘Gatecontrollable spin battery. Kuzma. ¨ Konig. M. ‘‘Spin polarized tunneling through singlecrystal GaAs(001) barriers. G. Phys. D. 115324. JETP 94.. B. Rev. ‘‘Magnetoelectric effects in conductors with mirror isomer symmetry. Moser. ‘‘Classical properties of low-dimensional conductors: Giant capacitance and nonOhmic potential drop. Rev. Phys. ‘‘Tunability of electron spin coherence in III-V quantum wells. 83.
F. B 68. H. B 84. 2003. Rev. Imamura. D. Luo. H.’’ Phys. and D. Rugar. J. W. and B. ¨ Schon. pp. MacDonald. Nguyen Van Dau. Margulis. ‘‘Ballistic spin transport and spin interference in ferromagnetic/InAs(2DES)/ferromagnetic devices. and B. Schoeller. 1976a. Marushchak. M. Phys. Samarth. A. 1427– 1430. Barnas. and T. 1988. and P. B 62. F. C. Stiles. Rev. ‘‘Va` lidity of the Julliere model of spin-dependent tunneling. 89. Zakharchenya 1984. B. D. edited by S. Martin. 2035–2038 (1983)]. H. McCombe. F. Takahashi. 76. Fert. Vogl. A. Optical Orientation (North-Holland.. II. Mod. B 68. ‘‘Spin relaxation in GaAs/Alx Ga1 x As quantum wells. 1984. Eds. ‘‘Electron-spin relaxation in p-type quantum wells via the electron-hole exchange interaction: the effects of the valence-band spin mixing and of an applied longitudinal electric ﬁeld. N. 7685– 7693. 36. ‘‘Layer KKR approach to Bloch-wave transmission and reﬂection: Application to spin-dependent tunneling. Rev. Simmons. 5161– 5163..’’ Phys.’’ Phys. 179–187. J.’’ IEEE Trans. 2003. Z. Zhang. ‘‘Electron spin resonance in graphite. 1982. K. and S. X. and P.’’ Phys. I. 014421. Hu. J. 2003. Appl. 1999. 707–708. and A. Phys.’’ Phys.’’ Phys. Butler. 87. I. Munekata.ˇ ´ Zutic.. Dietl. B 55. Sham. M. 1967–1974. Fang. Sugihara. and X. 770–772. 1991. Bristol). N. A 57. Golubov. Shinjo (Taylor and Francis. 1964. S. and D. and K. ‘‘Effect of spin injection on the CESR at a ferromagnet-semiconductor contact. p. Lett. Mamin.’’ Phys. Long and J. M. Lepine. ‘‘Effect of bulk inversion asymmetry on the Datta-Das transistor. C. I. Majewski. W. Rev.. Rev. 10 142–10 145. Awschalom. Vol..-M. and K. Masterov. and V. Stiles. Theor. N.. 13.’’ Phys.’’ Phys. ‘‘Quantum computation with quantum dots. 44–48. H. H.’’ Phys. Rev. D. Phys. P. Guivarc’h.. Magn. Mattana.’’ Fiz. J. 120–122. T. Dederichs. 220403. 1999. 90. Y. 7576–7578. X. S. and D. New York). A. and P. S.. Rev. B. F.. 1999. and H.’’ J. B 56.. Butler.’’ Phys.’’ Prog. E. 15 776–15 788. Tekh. H.’’ Phys. Mathon. 014402. Loss. ‘‘Observation of an anomalous frequency dependence of the conduction-electron spin resonance in Al. 1971.. B 38. Munekata. ‘‘Observation of the zero-ﬁeld spin splitting of the ground electron subband in GaSb-InAs-GaSb quantum wells. Tsuzuku. ‘‘Spinpolarized tunneling and spin injection in superconductorferromagnet junctions. F. Levy. Maekawa. 2003. M.’’ Phys.’’ Phys. 90. S. Jezequel. Tela 25. Rev..’’ Phys. ‘‘Pauli paramagnetism and superconducting state. 1998. Davies (IOP. 11 827–11 832. J. 047901.’’ Phys. ‘‘Temperature dependence of itinerant electron junction magnetoresistance. Harley.’’ Phys. J. ‘‘Electrical detection of spin accumulation in a p-type GaAs quantum well..’’ Physica B 193. George. Maialle. T. J. Solid State 25. Fabian. 13 771–13 777. 1104–1106. Zhang. Umerski. New York). I. Mal’shukov. Matsuyama. Lubzens. J. A. Umerski. P. A. MacDonald. H. Rev. and A. B 60. B 63. Rev. A.. ‘‘Spin relaxation of electrons in p-doped quantum wells via the electron-hole exchange interaction. 2000. T. Wrobel. Maialle.. Takahashi.’’ Phys. Rev. Luo.’’ J. Jaffres. 1997.. Maki. J. ‘‘Spin bottlenecks in the quantum Hall regime..’’ Phys. 1117–1121. A. Phys. A.. 2003. April 2004 Malinowski.. C. G.-G. ´ Łusakowski. B 47... ‘‘Tight-binding theory of tunneling giant magnetoresistance. B 41. ‘‘Theory of tunneling magnetoresistance in a junction with a nonmagnetic metallic interlayer. Titkov. B 44. B 4. 207604. A. Rev. and M. 2003. Lubzens. J. Rev. ‘‘Theory of tunneling magnetoresistance of an epitaxial Fe/MgO/Fe(001) junction. Rev. Stepanova. D. Phys. Rev. H. Mazin. ‘‘Spin-dependent recombination on injection of electrons from a ferromagnet into a semiconductor.. Chao. 11 845–11 851. 945– 950. S.’’ Phys. Mazin. Tang. Rev. and S. Rev.. S. R. I. 1999. ‘‘Spin accumulation in ferromagnetic singleelectron transistors in the cotunneling regime. B 14. Schultz. Rev. and P. B 66. .. and L. ‘‘Ballistic spin injection and detection in Fe/semiconductor/Fe junctions. ‘‘Persistent sourcing of coherent spins for multifunctional semiconductor spintronics. and Das Sarma: Spintronics: Fundamentals and applications 399 ‘‘Effect of silicon crystal structure on spin transmission through spin electronic devices. Z. F. J. Appl. S.Peterburg) 13. 2001.’’ Fiz. 2002. 166601. and H. 143–236.’’ Phys. V. ‘‘Electron tunneling between ferromagnetic ﬁlms. D. 18. 1976b. and M. Britton. 155322. B 54. Rev.. 2002. M. 11 810– 11 819. H. 1996. Sci. Wang. Nadgorny. Malajovich. B 59. B 65. 705–708. 81. P. 2. Berry.. J. V. Ritschie. Lett. 081201. 1993. ‘‘Quantum computing with spin cluster qubits..’’ Phys. B 56. H. (S. A. Wunnicke. 2003. ‘‘Spin relaxation of conduction electrons in moderately doped gallium arsenide crystals. Makovskii. J. G. and R. R. de Andrada e Silva. Rev. Rev. 2001. F. and G. Rev. and U. Rev. J. K. Lett. in Physics of Semiconductors 2002. Fang. Mathon. R. 3537–3542 [Sov. J. S. ‘‘Electric-dipoleexcited electron spin resonance in InSb. Phys. A. 1998. Maekawa and T. 91. R. and A... MacLaren. 233307. ‘‘Spin-current generation and detection in the presence of an ac gate. Imamura. 3262–3265. and G. Maialle.-G.’’ Mater. A.. Maekawa. 32. B 67. D. Rev. D. T.’’ Phys. Martinek. J.. Grundler. ‘‘Conduction-electron spin resonance in aluminum at 79 GHz. Meier. and W. B 66. ‘‘Detection and manipulation of statistical polarization in small spin ensembles. ‘‘Exciton spin dynamics in quantum wells. Matsubara. J. Eng. and L.’’ Phys. 1192– 1193 (1979)]. J. Budakian. 1997. J. ¨ Maekawa. Meier.. Rev. A. Schultz. Tverd. 2002. J.’’ Phys. Semicond. Merkt. ‘‘Spin-drift transport and its applications. 1285–1288. Chui. and U. T. 1994. Lett. Rev. D. Lett. 83. H. Chu.’’ Nature (London) 411. ‘‘How to deﬁne and calculate the degree of spin polarization in ferromagnets. 13 034– 13 039.’’ Phys. A. Z.. Rev. A. H. 1990.. edited by A. and A. Lett. B. 1997. F. No... MacLaren. A. 2002. DiVincenzo. 2001. M. Maekawa.’’ Phys. 2042–2044 [Sov. 83. 305. S. Jungwirth. Meier. 5470–5478. O. Lett. Gafvert. Rev. 1979. Poluprovodn. I. 2001.. 024416. Margulis. ‘‘Effects of inversion asymmetry on electron energy band structures in GaSb/InAs/GaSb quantum wells..’’ Phys. Grevatt. Kasner. 29–36. in Spin Dependent Transport in Magnetic Nanostructures... R. Rev. and M. Umerski. Mavropoulos.. Loss.-M. Mathon. and D. ‘‘Probing spin polarization with Andreev reﬂection: A theoretical basis. Wagner. Degani. L..
1976.. P. Magn.. ‘‘Interface magnetism and spin wave scattering in ferromagnet-insulator-ferromagnet tunnel junctions. Moodera. Goovaerts. Rev. Y..Mn)As/AlAs/(Ga.’’ Phys. Lett. Moodera. Gibson. and X. 35. Magn. and J. Meservey. 2003. 139. Hamaya. ‘‘Transport theory of multiterminal hybrid structures. Moodera. and J. ‘‘Geometrically enhanced magnetoresistance in ferromagnet-insulator-ferromagnet tunnel junctions. Esaki.’’ Phys. R. 200. and M. Roukes. and M. Moodera.. Tedrow.. D. Borghs. Ser. 2002. 1999. and C. Tedrow. London. L. Miyazaki. B 3. M. The Theory of Atomic Collisions.’’ Appl. 1–6. F. 199. Feinberg.’’ Eur. K. Low Temp. Kinno. ‘‘Giant magnetic tunneling effect in Fe/Al2 O3 /Fe junction. 101–114. C. ‘‘The resistance and thermoelectric properties of the transition metals. Hao. 205309. 407–409. ‘‘Electron spin relaxation by nuclei in semiconductor quantum dots. M. Mater. Meservey. S. (Paris) 43. P. Van Dorpe.’’ Appl. E. ‘‘Spintunneling in ferromagnetic junctions. ‘‘Tunneling characteristics of amorphous Si barriers. S. 1960. Borghs. and G. Lett. Monsma. Tedrow... 1995. F. Gossard. 2941–2944. D. Das. ‘‘Surface relaxation times of conduction-electron spins in superconductors and normal metals. ‘‘Conduction electron spin resonance in gold. Lett. London. Phys. Wong. Meservey. Rev. ‘‘Large magnetoresistance at room temperature in ferromagnetic thin ﬁlm tunnel junctions.. L. 91–98. Mathon. Mizushima. New York). 8235–8243. 5260–5263. ‘‘Perpendicular hot electron spin-valve effect in a new magnetic ﬁeld sensor: The spin-valve transistor. 1995. 2000. Vol. W. ‘‘Gate-controlled spin-orbit quantum interference effects in lateral transport. I. E. J. E. 80. P. Tang. D.’’ Phys. P.. X. Monzon. D.’’ Appl. L231–L234. 77. T. 84. Merkulov. 29.’’ preprint. F. 1997.’’ IEEE Trans. 3273–3276. 1999. 2003. A 156. R. Mater.. Soc. Shinjo (Taylor and Francis. D. A. Meservey. 024426. Janossy. Lett. ‘‘Strong Hall voltage modulation in hybrid ferromagnet/ semiconductor microstructures. R.Mn)As trilayer structures. ‘‘The electrical conductivity of transition metals... Mathon. 53.. Rowe. 1995. Brooks. 3500–3504. D. R2893–R2896. Phys.’’ J. . Munekata. S. F. M. R. 41. 146802. P.’’ Phys. P. and R. R. Roukes. M. J. 853–856. Wang. 74. and P. B 19. 1998. Mater. S. J. Rev. Lodder. 1997. Chamon. R. 143–236. 245319. Lett.. Magn.. D. A 153. X.’’ J. Sci. 381–432. T. 3887–3895. 2000b. Rev. E. Phys. Dieny. J. and R.. ‘‘Conduction-electron spin ﬂip by phonons in metals: Analysis of experimental data. H. ‘‘Resonant magnetotunneling in GaAlAs-GaAs-GaAlAs heterostructures.. ‘‘Optical investigation of electrical spin injection into semiconductors. ‘‘An adiabatic quantum pump of spin polarized current. Lett. Monsma. and M. Motsnyi. Rev. Moodera. Rev. P. Goovaerts. L. Moodera. and B. and F. 90. Rev. 1993. Rev. L. S.. Nassar. B.. 2002. Mott. Monsma. Meservey. F. 1996.’’ J.. A. L.’’ Proc. 858–860. April 2004 Monzon. and Das Sarma: Spintronics: Fundamentals and applications ´ Melin. 1982. 805–808. 1998. Phys. ‘‘Temporal evolution of spin-polarization in ferromagnetic tunnel junctions. ¨ Miller. Lodder. D. Monod. K. J. Rev. Massey. Van Roy.’’ Phys. J.’’ Phys. 1971. Kinder. V. Mireles. Parkin.. 883–885. J.’’ Appl. 248–273.’’ Proc. S. Lett. J. Meservey. Marcus. M.. R. Zumbuhl. Moriya. and G. P. Rev. edited by S. and A. 1982. Rosen. Popma. Tanaka. S. R. and R. Tedrow. Oxford). Hao. R. ‘‘Spin injection and the local Hall effect in InAs quantum wells.. Vlutters. Safarov. 2002.’’ Phys. G. 1110–1114. 1983. 1936a. 1999. 708–710. R.’’ Phys. Rev. 89. and D. De Boeck.. ‘‘Conduction electron spin-ﬂip scattering by impurities in copper. J. 5022. F. 311–316. LeClair. 911–915. and J. 1998. Moodera. Yamauchi. 076807.’’ Science 281.. S. LyandaGeller. 33. N. Roukes. Kirczenow. and J. and S. 3rd ed. B 68. Rev. V. ‘‘Effect of electrical spin injection on magnetization reversal in (Ga. J. ‘‘Electrical spin injection in a ferromagnet/tunnel barrier/semiconductor heterostructure. Lett.. B 64. Appl. 1563–1570. Fabian.. and W. Mott. 71. ‘‘Electron spin polarized tunneling study of ferromagnetic thin ﬁlms. Monod. E.’’ Appl. J. Rev. Beuneu. Magn. L. Lett. B. 26. and H. B 42. G. J. van de Veerdonk. J. No. A. G. 81. De Boeck. Mucciolo. J. Paraskevopoulos. C. Monod. I. Moussa.. 120. Phys. C. ‘‘Anisotropic superexchange interaction and weak ferromagnetism.. ‘‘Spin polarized tunneling in ferromagnetic junctions. Parkin. 77. R.. Appl. ‘‘Variation of the electron-spin polarization in EuSe tunnel junctions from zero to near 100% in a magnetic ﬁeld. ‘‘Ballistic spin-polarized transport and Rashba spin precession in semiconductor nanowires. H. J. 2. Phys. N. T. J. M. B 33. 69. Phys. Rev. J. L. ‘‘Spin polarization of tunneling current from ferromagnet/Al2 O3 interfaces using copper-doped aluminum superconducting ﬁlms. ‘‘Surface effects in magnetic crystals near the ordering temperature. Ram-Mohan. L. R. 1988. Phys. Magn. A.’’ Phys.’’ Phys. W. 699–717.. A. Nowak. R. M. C. 1979. in Spin Dependent Transport in Magnetic Nanostructures.’’ J. Rev.’’ Phys. A. and S. 265–267.’’ Phys. N.’’ Phys. J. 1978. Ser. Lett. 2002. 70. V. F.. Van Roy. 2000a. V. and K. and P. Monzon. Meservey. ‘‘Room temperature-operating spin-valve transistors formed by vacuum bonding. B 65. and M. and S. and G. S.. Magn. Schultz.’’ J. B 26.. F. Solin. T. Rev. G. 37.’’ J. Kinder. Rev. and N. P.’’ Annu. R. Phys. G. Motsnyi. Meservey. 720–722. 2003.’’ Phys. Efros. S. Soc. C. 1965. S. Mills. Monsma. Marcus. T. T. Phys. Maekawa and T. and H. Campman. J. Lett. ‘‘Response of an extraordinary magnetoresistance read head to a magnetic bit. 1936b. M. M. Rev. Mendez.. Johnson. A. C. 2003.. Magn. Lett. 2002. Safarov. 1977. 74. Tezuka. Moriya. Mater.’’ Phys. S. ‘‘Magnetoelectronic phenomena at a ferromagnet-semiconductor interface.400 ˇ ´ Zutic. Oiwa. Mater. Rev. W.. (Clarendon. Mott. ‘‘Electron-spin polarization in tunnel junctions in zero applied ﬁeld with ferromagnetic EuS barriers. 94. Magn.’’ Phys. ‘‘Correlation between spin polarization of tunnel currents from 3D ferromagnets and their magnetic moments. Miyazaki.. T. and A.’’ Phys. and G.’’ J. 76. J. Lett. 368–382. 3087– 3089.. 632–635. J. and S. I. Nowak. Goldhaber-Gordon. L. 2001. I. ‘‘Energy-dependent hot electron transport across a spinvalve. Phys. Moodera. M. Magn. F.. Mod. 393– 401. J.’’ J. K. Lett. pp. and R.
R. Tsymbal. ‘‘Rotation of ferromagnetically coupled Mn spins in (Ga. Ohno. Osofsky. L. ‘‘Total scattering cross section and spin motion of low energy electrons passing through a ferromagnet. B 61. 2000. R. F. Arsenov. 81. 137202. Soulen. P. Muto. and N. Sci. B 62. N.’’ Phys. NY).. von Molnar. R. Mazin. Myers. Burgermeister.. 1999. and W. A. Supercond. K. 308–312. Iye.’’ Phys. 2003. S.. Lett. T. Jpn. 74. Rev. Lett. Hasenberg. L. Ohno. Nazmul. Fabian. 2002. 2003. I. A. B. M. Perez. ‘‘Magnetoresistance of Ni/NiO/Co small tunnel junctions in Coulomb blockade regime. H. Takayanagi. Ohtani. K. H. Ploog. Morales. Bratkovsky.52Al0. D. 1907–1909. Y. T. Shaw. and I. Gambino. Rev. A. Outuka. Shimada. ´ Nikolic. ¨ W.’’ Phys. 2664–2667. ¨ ¨ Oestreich. Mn)As diluted magnetic III-V semiconductors. M.. Mitsumori. Akazaki. S. Rev. A. B 67. H. Young. ‘‘Scanning tunneling spectroscopy under pulsed spin injection. ‘‘MBE growth and electroluminescence of ferromagnetic/non-magnetic semiconductor pn junctions based on (Ga. Oiwa. 2002. B. 78. ‘‘Manifestation of local magnetic domain reversal by spin-polarized carrier injection in (Ga. Segmuller.’’ Phys.’’ Nature (London) 408. Mitsumori. ‘‘Interface contributions to spin relaxation in a short-period InAs/ GaSb superlattice.’’ Phys. and D. and T. 1996. Phys. Lett. Matsukura. 1970. T. H. . F. and K.. P. Phys.’’ Science 281. ´ Ohno. E. J.. E. Mitsumori. A. Rev. 2004. 839–841. ‘‘Electron spin relaxation beyond D’yakonov-Perel’ interaction in GaAs/AlGaAs quantum wells. Louie. and D. 4196–4199. W. Munoz. Ohno. A. Rev. 951–956.’’ Appl. 1997. L. Jr. and H.’’ Phys. T. van de Veerdonk. I. Osipov. ‘‘Ferromagnetism and high Curie temperature in semiconductor heterostructures with Mn delta-doped GaAs and p-type selective doping. Singh. Y. Heimbrodt. and L. Abe. B. M. Ohno. N. G. Beschoten. Tanaka. and J. 1996.. W. H. Phys. D. Growth 111. T.’’ Semicond. Ralph. D.. Ashenford. ‘‘Transport spin polarization of Nix Fe1 x : Electronic kinematics and band structure. Lampalzer. Moriya. H. B. E. Rev. Brender.Mn)As. F. Dietl. Harris. K. ‘‘Structural and electronic properties of Co/Al2 O3 /Co magnetic tunnel junction from ﬁrst principles. T. J. Y. F. and H. R. D... Yokoyama.. B. 1998. T. 76. and H. 159-160. R. H. ´ Nikolic. 65. Stroud. Rev. Munekata. S.’’ Phys. Kobayashi. Katsumoto. Oberli. 2000. M. T.’’ J. 2001. and S. ‘‘Origin of high transport spin polarization in La0. F.’’ cond-mat/0111144. Hubner.-C. Slupinski. V. Lunn. ‘‘P-type diluted magnetic III-V semiconductors. R. C. ‘‘Making nonmagnetic semiconductors ferromagnetic. ‘‘Efﬁcient nonlinear room-temperature spin tunneling-emission in ferromagnetsemiconductor heterostructures with extended penetration depth.. Matsukura.. J. R. H. M. H. Nadgorny. Laprade. Nagosa.. W. Skelton. and S. A.53Ga0. A. 1999. 285–297. 1983. ‘‘Diluted magnetic III-V semiconductors. Phys. Cheng.’’ Appl.’’ Nature (London) 402. Orchard-Webb. Nagaev. Oiwa. Lett. Rev. J. Rev. 790–792. C. Ohtani. ‘‘Spin injection. and A. 16. Phys. F. J. and S. E. Moriya. and L. Adachi. M. E. Moscow).. Ngai. J. Surf. Ohno. Allen. 325–332. Cousins. 2003. E. Lett. H. Quantum Computation and Quantum Information (Cambridge University. 1313–1317. Lau. H. Munekata. 1251– 1253.Mn)As by hole spins. 83. S. Sci. R. Status Solidi 41. Riesen. I... 2001. Technol. W. 3963–3969. ‘‘All-optical picosecond switching of a quantum well etalon using spin-polarization relaxation. Stolz. Boggess. 2001. 1335–1338. Tarucha. 69. Nielsen.. A. Flatte. 4228–4231. Mazin. M. Supinski. E. Nadgorny. A. Wei. 867–870. J.. Esaki. J. E. ‘‘Current-induced switching of domains in magnetic multilayer devices. M. F. Arata. 201301. ‘‘Electrical spin injection in a ferromagnetic semiconductor heterostructure. 1992. V. 1995. Slupinski. C.’’ Phys. K. 1989. Cryst. and B..7Sr0. 411– 414. von Molnar. Siegmann. Vina.’’ Science 285. Rev. B 63.. L. Rev. Ohno. Matsukura. Klar. Murakami. J. J. Qadri. B 64. ‘‘Dissipationless quantum spin current at room temperature.’’ Phys. Terauchi. B.’’ Appl. ‘‘Spin relaxation in intrinsic GaAs quantum wells: inﬂuence of excitonic localization. 16. Rev. R.’’ Phys. Munekata. ‘‘The effect of concentration gradients of impurities on the temperature dependence of the spin-lattice relaxation time in lithium and beryllium.Mn)As. and K. J. H. 2000. R. Sugahara. Hubner.’’ Phys. 1998. F. Ruf. I. I. K. 944–946. J. 363– 365.’’ J. Mod. 2003.’’ Phys. 88.’’ Science 301. D. 817–820. D. D. Oiwa. D. 2. M. 3449–3451. Y. L. Soc. and J. R. Terauchi. Smits. and H.. C. Tackeuchi.. R.. B 60. Lett. E. S. Y.’’ Phys. Y. D. J.. Rev. and P.3MnO3 : Direct evidence for minority spin states. C.48As heterostructure. H. Matsukura. End. Nitta.. Austin. 2003. Phys. G. V. G. 3952– 3959.’’ Appl. Chuang.’’ Appl. S. Klar. ‘‘Mesoscopics in spintronics: Conductance ﬂuctuations of spin-polarized electrons.Mn)As thin ﬁlms. and Y. S. Ohno. Cambridge. Yu. Ohno. Weber.’’ Science 297.. and Y. Lett.. D.. H. M. H. E. Ohno. Ohno. and H. H. Y. P. Soulen. Zhang. T. and D. ‘‘Spin relaxation in GaAs(110) quantum wells. ‘‘Current rectiﬁcation by Pauli exclusion in a weakly coupled double quantum dot system.Mn)As: A new diluted magnetic semiconductor based on GaAs. 1991. E. S. Y. W. R3788–R3791. 184433. F.. K. W. L. M.. Voltz. ´ ¨ Munekata. 439–442. Chang. 66. M. T. ‘‘Gate control of spin-orbit interaction in an inverted In0. S. Broussard. 2002. J. Wang.. ‘‘(Ga. D. 1011–1015. M. L. 63. and R. Mukovskii. A. ´ Olesberg. Katine. F. S. Y. Watts. and Y. A. Oiwa. 84. Ruhle. B. ‘‘Electric-ﬁeld control of ferromagnetism. Perovic. W. Matsukura. ‘‘Magnetotransport properties of p-type (In. Penney. Chang. Lett. Ohno. I. Altunkaya. ‘‘Electron transport through a circular constriction.’’ J. H. Pettifor. Enoki.47As/In0.. Tokura. and L. Chiba. Matsukura. S. T. B. Vol. J. Yu. and H. 1348–1351.. A. 1999. and T. spin transport and spin coherence. and M. Chen. Ohno. and Das Sarma: Spintronics: Fundamentals and applications 401 Munekata. 2003.’’ Phys.’’ Physica E (Amsterdam) 6.’’ cond-mat/0307030. and T. Osofsky. Ohno. Ono. Supercond. Nakamura. ‘‘Spin injection into semiconductors. Smithard. Nishikawa.’’ J. 1849–1852. R. No. 2000. R.. J. 1999. Moriya. Hagele. 241308. Y. Ruhle. 2000. Tseng. Chang. Heimbrodt. Munekata. Lett. A. Y. Y. Ono. M. I.ˇ ´ Zutic. J... A. 1999. Physics of Magnetic Semiconductors (Mir. private communication. B 51. ‘‘Effect of optical spin injection on ferromagnetically coupled Mn spins in the III-V magnetic alloy semiconductor (Ga. A. Munekata. P. Awschalom. 68. Lett. Rev. G. Oleinik. Buhrman. L. April 2004 ¨ ¨ Oestreich. K.. Freericks. 2000. O. S. 17. 4247–4257. Shen. Ohno. H. 1995.
V. Rev. Young. 52. Amand.Mn)As interface. I. T. Furis. 2000. ‘‘Andreev reﬂection and pair breaking effects at the superconductor/ ferromagnetic (Ga.’’ Phys. M. S. J. B 15. Nadgorny. Mod. Park. 2003. P.’’ (preprint). 2003a. Kim. P. Jiang. 89. Struct. and J. Samokhvalov. T. Felder. Yassievich. No. Petrou. M. Rev.’’ Phys. 13. . 2003. 66. ‘‘Spin-dependent tunneling through a symmetric semiconductor barrier. Pershin. E. Fiz. H. M. Li. Ott. A. Rev. and A. 201304.’’ Appl. T. Privman.. B. Amsterdam).’’ Phys.-H. P. Vonlanthen. G. 661–680. Rice. R. W. Pederson. and P. A. Kwon. D. J. Miller. R.. Brown. Maekawa and T. W. C. 1967. V. and T. Rev. Lett. Rev. Kioseoglou. S.. Lett. S.. Osipov. Teor. 1971. Panchula. Hebard. L. Fabian. A. Y. and Das Sarma: Spintronics: Fundamentals and applications Osipov.. Fisk. L.’’ preprint. Lett. and B. H. B. Bhadra. 85. ´ Pejakovic. Jr. Ramesh. A. Paget. T. R. Parkin. 1–13.’’ J. Vol. S. R. 93. Englewood Cliffs. 1998. Hopster. 411–415. Ambrose. P. A 247. Samant. Lett.. Ambrosini. A. 2003b. Ganichev. L. ‘‘Microwave-absorption at a junction between the ferromagnetic semiconductor HgCr2 Se4 and the semiconductor InSb. and H. 1984.402 ˇ ´ Zutic.’’ Phys. and M. Ivanov. Lampel. P. S.’’ Phys. P.. S. Liu. D. Lu.. ‘‘Effect of ﬁlm resistance on low-impedance tunneling measurements. N. N. Abernathy.. C. N.’’ Phys. H. Rev. Venkatesan. 2002. 2003. 2004. ‘‘Magnetically engineered spintronic sensors and memory. P. X. Z. Wilson. 201307..’’ J. and A. S. 794–796. S. C. 504–507. Kamper. Eksp.’’ Phys. J. Appl. Luntz. Y. Y. Theodoropoulou. 2152–2155..’’ cond-mat/0403451. ‘‘Low ﬁeld electron-nuclear spin coupling in gallium arsenide under optical pumping conditions.. 1977. Lett. 353–359. Optical Processes in Semiconductors (Prentice-Hall. N. Kur. 237–271. A.’’ Phys.. R. Panguluri. A.. J. Pershin. I. D. J. 58. X. ‘‘Point contact spin spectroscopy of ferromagnetic MnAs epitaxial ﬁlms. Osofsky. D. G. R. Panchula. E. P. V. A.’’ Phys. P. 90. and V. 256603. B. ‘‘Propagation of spinpolarized electrons through interfaces separating differently doped semiconductor regions. Lett. A. R. and B.’’ Appl. Shinjo (Taylor and Francis. S. 71. 057202. Brundle. T.’’ J. G.. Paget.’’ Pis’ma Zh. Phys. Panguluri. G. and K. Renucci. ‘‘Exchange-biased magnetic tunnel junctions and application to nonvolatile magnetic random access memory. S.-H. et al. 1991. and T.. M. M. Furdyna. Zakharchenya (North-Holland. 76. 2000. 1998b. Y. 2004. ‘‘Giant magnetoresistance in antiferromagnetic Co/Cu multilayers. R. V.’’ Proc.’’ Phys. Furdyna. D. S. Vescovo. Phys. S..’’ Physica B 281Õ282. A. Mattson. Park. Venkatesan. edited by F. F. Pearton. Jonker.’’ Phys.-H.. J. K. S. 1152– 1154. X. 1997. K. 52. B. V. Phys. Kochev. ‘‘Wide band gap ferromagnetic semiconductors and oxides. Mamaev. P. B. Rev. V.. J. L. D. E. D. Parkin. 77. Parker. 209–219. 1969. Pannetier. 2002. Vernon. T. D.’’ cond-mat/0306041. I. ‘‘Oscillatory magnetic exchange coupling through thin copper layers. Samant. E. K. Vescovo. Overberg. and P. ‘‘Polarized photoemission spectra of the strained semiconductor layers. S. S. 2002.. D. G. 86. Parsons. Privman. R. 423–427. G. ‘‘Unusual magnetism of hexaborides. ‘‘Magnetic properties at surface boundary of a half-metallic ferromagnet La0. Kwon.. A. Gerard. 3989–3991. Lett. M. V. edited by S. Berkovits. 2002. Bennett... 23. Oskotskij. 2001. J. J. M. Subashiev. Pareek. Roche. J. G. Xiong. Y. Rice. K. ‘‘Polarization of nuclei in metals. N. and S. Bel’kov. C. Liu. 29–31. R. and W.. ‘‘Andreev reﬂection and proximity effect. I. 1999. 1991. 118. K. 66. 1953–1956. B 67. D.. A. Lett. ‘‘Spindependent electron attenuation by transmission through thin ferromagnetic ﬁlm. Kitamura. Chun.’’ Science 295. Viglin. Wojtowicz. Roche. 8. J. ‘‘A group-IV ferromagnetic semiconductor: Mnx Ge1 x . R. in Spin Dependent Transport in Magnetic Nanostructures. 2000.’’ Appl. C. 241305. Phys. Jonker. Erwin. 81. Zysler. M. 10. E.’’ Phys. Bruno. Wojtowicz.. P. 2003. W. Jbeli. B 68. X. Lett. 651–654. S. Rev. Modern Problems in Condensed Matter Science.3MnO3 . I. J. and D. C. L. and J. Spanos. C. 1993. and L. Supercond. and V. M. Rev. F. B. P. ¨ Pappas. B 65. E. Sapoval. and R. 5780– 5796. V. 92. Vol. D. and Y.. Roche. Degiorgi. Tsoi. Safarov. E. Lett. J. S. Phys. E. P.’’ Nature (London) 392. Norton. R. I.. Lett. Z. S. S. Ramesh. Courtois. ‘‘Photoinduced magnetization in the organic-based magnet Mn(TCNE) x •y(CH2 Cl2 ). B. ‘‘Origin of enhanced magnetoresistance of magnetic multilayers: Spin-dependent scattering from magnetic interface states. K. T. C. Viglin. B. P.. L. 1641–1644. Nadgorny. 386–389 (1990)]. and V. Nadgorny. Low-Dimens. F. T. R.. pp. and A. Kaiser.. Yang. Hellberg. B. Park.. ‘‘Spin-injection. 996–998 [JETP Lett. Mazin. 1Õ2. P. Samarth.’’ Phys. Sullivan. Lett. W. Meier and B. Rev. I. Lett. IEEE 91. Gavilano. Rev. S. Kim. D. 1953b. and I. P.’’ J.’’ Phys. 381–421. Scheuerlein. Mazin. O’Sullivan. G. Parkin. H.. and A. D. Rev. Bucchigano.7Sr0. T.-J. V. P. Itskos. Epstein. V. Tarasenko. G. and A. M. Overhauser. A. 1953a. P. P. Samarth. Rev. ‘‘Band-to-band optical pumping in solids and polarized photoluminescence.. Watts. J. Appl. Rev. H. ‘‘Spin polarization of CrO2 at and across an artiﬁcial barrier. A. Abraham.. B. ‘‘Measurement of spin polarization by Andreev reﬂection in In1 x Mnx Sb epilayers. S. Parkin. R. C.. Lim. V. NJ). 1634–1637.’’ Phys. 2. I. 2710–2712. J. New York).’’ Phys. ‘‘Spin relaxation quenching in semiconductor quantum dots. Park. 5828–5833. Prettl. 2000. in Optical Orientation. A. C. 2003. Kim.-J. P. Samokhvalov. Paillard. R. 77–87. Smith. Panguluri. J. ‘‘Direct evidence for a half-metallic ferromagnet. ‘‘Electrical spin injection across air-exposed epitaxially regrown semiconductor interfaces. Y. 1990. Rooks. Fowler. ‘‘Giant room temperature tunneling magnetoresistance in magnetic tunnel junctions with MgO(100) tunnel barriers. 1991. P. 689–700. 2002. 599–615. Overhauser. Phys. and V. A. S. Samant.. Thaler. Parkin. Hanbicki. P. and F. Phys. S. 88. Ren. ‘‘Spin coherence in a twodimensional electron gas with Rashba spin-orbit interaction. Parkin. 196601. 88. A. ‘‘Paramagnetic relaxation in metals.. S. Boatner. pp. N.’’ Phys.. Perel’. Pankove. Rev. M. ‘‘Investigation of heterostructure ferromagnetic semiconductorsemiconductor in the millimeter and submillimeter microwave range. P. Rev. April 2004 Park. Miller. Kaiser. Beyers. Parkin. ‘‘Focusing of spin polarization in semiconductors by inhomogeneous doping. Low Temp. C. 1998a. Semicond. Marie.
. J. Solid State 3.7 Al0. and Al. Modern Problems in Condensed Matter Science.. 1991. V. 1955. E. 1995..’’ Appl.’’ Fiz. Modern Problems in Condensed Matter Science. S.. Hey.’’ Phys. Unguris. and K. Rashba. B 26. Rev. 1998.’’ Phys. 100. 2000. Pratt. E. S. H. J.. 2003. ‘‘Complex impedance of a spin injecting junction. 266602. Y. Rashba. D. C. 081304.3 As heterostructures. Siegmann. Slichter. I. 107205.’’ Rev.’’ Phys. ‘‘Spin-dependent elastic scattering of electrons from a ferromagnetic glass. 1988. 13–17.3 As heterojunctions at B 0 and B 0 due to inverstion asymmetry. Daughton.. Y. Rashba. Modern Problems in Condensed Matter Science. and D.ˇ ´ Zutic. L. Pfeffer. 141–146. 8. Petta. D. Rammer. ‘‘Orbital mechanisms of electron-spin manipulation by an electric ﬁeld. T. 23. Landwehr and E. A. 2001. Lett. E. Amsterdam). and V. T. R. 1735–1749 [Sov.’’ Phys.. 2002. in Optical Orientation. P. Meier and B. de Groot. W. Lett. ‘‘Electrical spin injection from ferromagnetic MnAs metal layers into GaAs. 15. Potok. ‘‘Half metallic magnets. G.’’ Phys. I. 1995. edited by F. I. 2566– 2574.. Rashba.’’ J. C. 035315. Ralph. I. Steel.’’ Appl. G. 66. ‘‘Spin splitting of conduction energies in GaAs-Ga0. 1995. Y. 2 and 4 m multilayer magneto-resistive memory cells. ‘‘Properties of semiconductors with an extremum loop.. Magn. V. H. and Das Sarma: Spintronics: Fundamentals and applications 403 Pershin. S. Piermarocchi. 1995.’’ IEEE Trans. Celotta. M.. Fabian.. 2003. 89. Sheka. ‘‘Diffusion theory of spin injection through resistive contacts. M. R. V. Comstock.’’ Phys. B 55. Privman. Rev. London. 1998.’’ Fiz. O. Petukhov.. ‘‘Relaxation times in magnetic resonance. Lett. J. Rev. 513–527. I. and N. 385–410. 58–63. Zhang. E. L. ‘‘Theory of electrical spin injection: Tunnel contacts as a solution of the conductivity mismatch problem. Kawaharazuka. H. Phys. Sheka. Phys. 3117–3119. 1109 [Sov. Rev.. Phys. J. Privman. Rev. Rashba. Phys. 3060–3063. 91.’’ Phys. B 66. Tela 2. Yoo. 1976. H. Zakharchenya (North-Holland. in Landau Level Spectroscopy.’’ Phys. 241315. ‘‘Conduction-band spin splitting and negative magnetoresistance in A 3 B 5 heterostructures. 2002. 175–179. ‘‘Transmission coefﬁcients of GaAs/ ErAs resonant tunneling diodes in strong magnetic ﬁelds.. W. 2001. Lett. 80. 1987. Mod. 9447–9464.. A. Daughton. and G.7 As0. B 29. and V.. 2003a. and H. and J.’’ Phys.’’ Phys.91 Al0. and J. D. R. 109. De Boeck. Krahn. V. Ramsteiner. Hur. Preskill. and V.. L.. J. 76. Prinz. Amsterdam). Ya. Zakharchenya (North-Holland. Tela 3. 1997. B 52. 167402. A 454.’’ Phys. A. pp.’’ Phys. ‘‘Spintronics: Sources and challenge. Lett.’’ Proc. M. B 67. J. 323–359. M. Rev. Matter 7. Prinz.. G. Pickett. Today 54 (5).. Bass. Pierce. N. Vol.’’ Phys. Pohn. V. Slaughter. Phys.. Rev. G. ‘‘Combinational resonance of zonal electrons in crystals having a zinc blende lattice.. and J.’’ Phys. Rev. Sci. 1960. and V. ‘‘Spin diffusion at ﬁnite electric and magnetic ﬁelds. R5312–R5315. W. 2329–2331.’’ Phys. M. Phys. 1257–1267 (1961)]. Supercond.’’ Eur. 27. Sherman. Surf. F. Tverd. Rev. B 68. Vol. 3. 241310. 2003c.’’ Phys. A. Rev.. W. Cyclotron and combinational resonance in a magnetic ﬁeld perpendicular to the plane of the loop.. ‘‘Bychkov-Rashba spin splitting and its dependence on magnetic ﬁeld in InSb/In0. ‘‘Threshold properties of 1. J. Petukhov. and W. 2002. Moodera. P. Sham. D. G. 1984. Pfeffer. 89. Rev.. J. E.. Solid State 2. ‘‘Spin splitting of conduction subbands in III-V heterostructures due to inversion asymmetry. B 67. P.. I. 2002c. B 68. A. Today 48 (4). R. Qi. E. p. J. Pierce. Rev. A. 2575–2577. 1998.. Rev. R.’’ Phys. 2002. van Roy. I. E. ‘‘Spin relaxation of conduction electrons in semiconductors due to interaction with nuclear spins. Ploog. E.5 5 m cells. Tverd. J. ‘‘Electron spin relaxation in a semiconductor quantum well. van Leuken. Celotta. ‘‘Resonant enhancement of tunneling magnetoresistance in double-barrier magnetic heterostructures.’’ IEEE Trans. Huang. Zawadzki. M. Mourokh. ‘‘Photoemission of spinpolarized electrons from GaAs. I. ‘‘Spin currents in thermodynamic equilibrium: The challenge of discerning transport currents. 131–206. Rev. I. J. Pines. P. ‘‘Optical RKKY interaction between charged semiconductor quantum dots. R16 267–R16 270. Amsterdam). Folk. ‘‘Spin-dependent transport in metal/semiconductor tunnel junctions. T. P. Y. and H. and D. 2003. 87.’’ Phys. ‘‘Quantum computation in quantum-Hall systems. and A. Mod. Pierce. Daweritz. 1224–1238 (1960)].’’ Phys. edited by F. No. Rashba. Phys.’’ Nano Lett. A. H. Pfeffer. and S. C. 2002a.09 Sb asymmetric quantum wells. 266801. ‘‘Perpendicular giant magneRev. J. and E. Lett. M. Rev. Meier and B. E. G. E. and D. I. A 239.. T. Chantis.’’ Phys. 1982. ‘‘Inelastic scattering approach to the theory of a tunnel magnetic transistor source. 16 928–16 935. Titkov. R. and C. D. Loloee.. 7359–7362. 1991.. 8. Lee. Rev. 155309. Jr. ‘‘Magnetoelectronics. and G. 2002d. 1961. Kventsel.’’ Phys. A 129.. Zawadzki. Chen. ‘‘Spin-polarized transport. and W. . N. B 68. Zawadzki. I. I. R.’’ Phys. 385–390. Rev. J.’’ Phys. and W. D. I. C. L. Smith. Pohn. E. P. S. Hao. 2002b.’’ Science 282. 89. edited by G. Lett. ‘‘Quantum ﬁeld-theoretical methods in transport theory of metals. 126405. 24.. Vol. 5484–5500. Rashba. Rashba. B 62. Rev. P.. A. ‘‘Detecting spin-polarized currents in ballistic nanostructures. Prins. Puller... 2003. C. and R. 1984. 1999. Rashba. Demchenko. 39–44. P. Ser. 1998. Vol. ‘‘The design of a one megabit non-volatile M-R memory chip using 1. Mehra. 1. 259–294. I. Efros. Rev. E. Pikus. April 2004 toresistances of Ag/Co multilayers. T. E. Pfeffer. ‘‘Reliable quantum computers. Pikus. Lett. Rashba (North-Holland. pp. Vagner.’’ J. Rashba. B 51. I. Lett. 052407.. P. 1986.. private communication. G. 1660–1663. R14 332–R14 335. I. and F.. Zhu. H.’’ Phys. Horing. E. P. 123. Schroeder. I. in Optical Orientation. Meier. Soc. ‘‘Spin-orbital band splitting in symmetric quantum wells... 2. Ni40Fe40B20 . Marcus. Grahn. 58. M. E. B 13. J. van Kempen. G. R. Rashba. and V. Pikus. A.-F. ‘‘Studies of spin-orbit scattering in noble-metal nanoparticles using energy-level tunneling spectroscopy. 1014–1020. ¨ ¨ Kastner. Rashba. J. Rev. 2003b. B 59. Umansky. H. 695–700. and J. M. J. ‘‘Spin splitting of conduction subbands in GaAS-Ga0. 1988. Magn.: Condens.
B 69. Awschalom. N. Lett. R. Potemski. Rev. 2002.’’ Phys. 2057–2060. Rev. Shanabarger.. A.’’ Nano Lett. Titkov.. and B. A. Rzchowski. Loss. 83. Theurich. B 56. J. and P. G. Albert. Saikin. G.’’ Phys. H. ‘‘Laser threshold reduction in a spintronic device. Potashnik. 046805. K. 88. 92. Lett. and V. J. Molenkamp. ¨ Schapers. Perrella. 5884–5887. Driscoll. 10 805–10 808. and Ni. C.. and V. Chaikina. Arata. Wilamowski.’’ Appl. S. S. Rev. I. Miltat. ‘‘Electrical control of spin coherence in semiconductor nanostructures.. Schep.. Gould. M. W. 2. Molenkamp. N. Lett. April 2004 Sanada. Rev. Appl. I. Rev. M. L. Ermakova.. W. E. ‘‘Hybrid ferromagnetic/semiconductor heterostructures for spintronics. Chun. Schmidt. Schomig. Tverd. Sukhorukov. Sato. Nehls. D. Lunde. Y. ‘‘Spin relaxation of conduction electrons in p-type semiconductors. Ando. Lett. G. Titkov. Y. 2000. Cheng. ‘‘Relaxation of shallow donor electron spin due to interaction with nuclear spin bath. F. ‘‘Ballistic electron magnetic microscopy: Imaging magnetic domains with nanometer resolution. L. Seck. E.. ‘‘Spin-valve transistor with an Fe/Au/Fe(001) base. Rendell. 85. P. Phys. . and E. Lett. Rev. G. 2003. Salis. Schmidt. and F. ‘‘Phase coherent precessional magnetization reversal in microscopic spin valve elements. Vol. 90. ‘‘Ultrathin aluminum oxide tunnel barriers. C. J. ‘‘Origin of beat patterns in the quantum magnetoresistance of gated InAs/GaSb and InAs/AlSb quantum wells.’’ J. Rev. and K. B 67. ‘‘Ferromagnetism in II-VI diluted magnetic semiconductor Zn1 x Crx Te.. 2001. and H. 2001. P. N. T. 1981. Salis. Buhrman. Phys.. Privman. Cleaver. W. Silva. Kayanuma. Appl. Slobodskyy. P. J. A. M. D. S. Tela 23. Phys.. H. Schliemann. Schafﬂer. ¨ Sanderfeld. P. Phys. R. Kim. D.’’ J. Appl. I. and J. 2003. 90. Hollfelder. D. A. W. 94. A. J. Bauer..’’ Fiz. 227203. R. A. Ensslin. M. S. Y. physics and experiments. 90. G. W. and M. Yamagata... ‘‘Gateable suppression of spin relaxation in semiconductors. W. Yamagata.. M. B 62. ‘‘Effect of the heterointerface on the spin splitting in modulation doped Inx Ga1 x As/InP quantum wells for B→0. ‘‘Interface resistances of magnetic multilayers. Ando. and L. R.’’ Phys.’’ Phys. A. 2000. 84. ‘‘Spin injection in the non-linear regime: band bending effects. C. F.’’ Phys. B.. Lett. S.’’ Phys.’’ Phys. ‘‘Observation of electron spin resonance in rubidium and cesium. Fuchs. 85–104... J. S. Kaka. V. W. 1938– 1940 (1981)]. J. 201307. N. 4324–4333. A. 2002.’’ Phys. ‘‘ESR investigations of modulation-doped Si/SiGe quantum wells. 2001. and R. 2001. and L. 2004. and N. Rev. C. D. R4790–R4793. G. 2003. ‘‘Bias dependence of magnetic tunnel junctions. and X.’’ Solid State Commun.’’ Appl. 2002. Schmidt.’’ J. H. 207202.. Kikkawa. 2003.’’ Appl. Gibbs. 1001–1003. Buhrman. Oka. S. 79. B. Zayets. Schultz. M. 1615–1617. 971–974. K. and A.. A. Rippard. W. ¨ Rudolph.. Crozat. and R. K. Lett. Freitas. J. H.. 2002. Schiffer. R. Lett. Saito. C. and R. 073301. Rev. Schumacher. Y. 127. Th. 2.’’ Phys. H. 1962– 1965. 3337–3342 [Sov. W.’’ Solid State Commun. Lett. Hansen. Saito. ‘‘Quantum dot as spin ﬁlter and spin memory. Pufall. Lett.’’ Pys. 2003. F. 15. 1157–1159. and D. Rippard. J. Freitas. Sci. Phys. W. Z. 623–626. Co. 2003. ‘‘Spin dependence of the electron mean free path in Fe. A. Loss. Rev. ‘‘Spin relaxation of conduction electrons in degenerate p-type gallium antimonide crystals. Rev. Chappert. Huberman. Buhrman. Sousa. and B. Ohno. Ferrand. Baumberg. Stradling. 1980. and G.’’ Phys. ‘‘Spin injection into a single self-assembled quantum dot. Lange. and L. and D.’’ Thin Solid Films 369. 7422–7427.’’ Semicond. T. Y. W. Suzuki. Rev. Grabs. ‘‘Optical manipulation of nuclear spin by a two-dimensional electron gas. 86. van Wees. 312–315. Sousa. Rev.. Rev. Molenkamp.. Rev.’’ Phys. I. ¨ and H. A. Sandhu. Rippard. 1965. N. R. Rev. Rev. Bacher. P. Schmidt. G. Ohno. J. Lett. M. Ohno. 1973. Rev. 90. Molenkamp. A. 310–321. van Hoof. 49.. Klocke. Th.. G. Forchel. Molenkamp.’’ Phys. Phys. ‘‘Effect of external magnetic ﬁeld on electron spin dephasing induced by hyperﬁne interaction in quantum dots. 13. A. 1997. C.. 45. V. 651–655. Fabian. W..’’ Phys. 81. G. Soc. Latham.. Saikin. J. 1966. ‘‘Fundamental obstacle for electrical spin injection from a ferromagnetic metal into a diffusive semiconductor.’’ Phys. ‘‘Quasiballistic magnetization reversal. Luth.. Rev.404 ˇ ´ Zutic. S. S. 173– 179. and K. ‘‘Direct-current induced dynamics in Co90Fe10 /Ni80Fe20 point contacts. R. G. Solid State 23. 619–622.’’ Phys. C. Lett. 91. Privman. Schmidt. S. and H. 8085–8087. C. 1997. Chen. Matsukura. and K. 017201. Komova. Phys.’’ Phys. Rev. K. S. and D. Z. 15. Mozyrsky. G.’’ Phys. Phys. W. Semenov. N. H. S. Miltat. ‘‘Spin-dependent hot electron transport in Co/Cu thin ﬁlms. Mod. E. 027201. G. Lett. Kelly. Filip.. and R. M. J. Egues. and C. 148–151. 146801. ‘‘A new scheme for measuring itinerant spin polarization. H. Schultz. and M. 1769–1775. ‘‘Semiclassical Monte-Carlo model for in-plane transport of spinpolarized electrons in III-V heterostructures. 178–181. B 63. M. 4516–4518. Wyder. A. 1999. No. D. Penn. Sakharov. M.’’ Phys. 1998. Russek. 2000. 87. H.’’ J. Zaets. and T. Khitrova. A. H. J.. and D. and K. Rev. ‘‘Non-ballistic spin ﬁeld-effect transistor. Lett. A. C. 1980. G. Richter. J. Gossard. Jantsch.’’ Nature (London) 414. Engels. ‘‘Observation of electron spin resonance in copper. H. 2000. C. Rippard. 017204. S. P. Lett. 2788–2790. Oestreich. W. Awschalom. S. ‘‘High-ﬁeld spin resonance of weakly bound electrons in GaAs.’’ Appl. P. H. Ferrand. R.. 86. Rowe. 2677–2680. ‘‘Roomtemperature ferromagnetism in a II-VI diluted magnetic semiconductor Zn1 x Crx Te. Grabs. and J. B 56. J. Rev.. Technol. D. P. 035311. Ferguson. and Das Sarma: Spintronics: Fundamentals and applications Recher. Bauer. R. B 61. W. 16. ‘‘Large magnetoresistance effect due to spin injection into a nonmagnetic semiconductor. P. Wu. 2001.’’ Phys. 76. W. 75. Hill. C. 82. Schumacher. and R.’’ J.. Safarov. Kato. and P. C.. Phys. the Bir-Aronov-Pikus mechanism. Scifres.’’ cond-mat/0206347.. Mironov. D. Hagele. 2002.. D. C. Sanvito. J. Shen. 2004. J. Ohno. 2001. Supercond. Lett. R. Chappert. 2003a. Heberle. Samarth. W. Mizushima. Y. ‘‘Relaxation of photoinjected spins during drift transport in GaAs.. Hillebrands. G. Phys. ¨ Seufert. 2003b. W. I. S. V. E.. 2150–2153. Lett. P. Fassbender. ‘‘Density functional calculations for III-V diluted ferromagnetic semiconductors: A review. V. ‘‘Spin injection into semiconductors. 2000.’’ Phys. H. M. 2002. G. R. D. M. Lett. 17. Gould. Ku. J. Jpn. and L. Richter. 2002. A.
T. ‘‘Current-driven excitation of magnetic multilayers. Reson. Witters. R. Solin. Carlin. Ni and its alloys. Hines. Appl.. A. M. J. M. Heremans. Yannoni.: Condens.. and R. H. Lee. Shockley. 167–171. and B. Magn.. J. A. Muranyi. P. Hoen. Wieck.. Stein. J. von Klitzing. 1965. S. J. P. B 27.. ‘‘A possible method for studying Fermi surfaces. and V. H. 20. 83. M. 5805–5810. Rev. Rowe.. ‘‘Minimum of spin-orbit coupling in two-dimensional structures.’’ Phys. Smirl.. Lett. L. K. 246601. 1998. ‘‘Gate-voltage control of spin interactions between electrons and nuclei in a semiconductor. Stearns. Gould. Lett. A. D. Rev. A. Becker. 1998. Soulen.’’ Phys. ‘‘Model calculation of the frequency and temperature dependence of the electron-spinresonance linewidth of aluminum. Lett. B 61.. 2001. Broussard. Discl. Zhang. A. Rev.. 81. Santos. R. ‘‘Theory of the detection of currentinduced spin polarization in a two-dimensional electron gas. 68.’’ J.’’ Phys. Pashkin. April 2004 Slobodskyy. R. Rev. 281–286. W. Shinjo (Taylor and Francis. 3159–3163. A.’’ Phys. B 64. 1989. Jansen. J. E. Maekawa and T. Lett. B. H. Phys. Najmaie.’’ Phys. and H. M. ´ C. Rev. and K. 155305. Ostreich. Deutschmann. C. Vol. A. ‘‘Hydrogenic spin quantum computing in silicon: A digital approach. and K. 1530–1532. T. Bull. 249–265. Moodera. C. L. and C. E. B 66. ‘‘Temperature dependence of magnetoresistance and surface magnetization in ferromagnetic tunnel junctions. Lett. Si. J. No. ‘‘Spin-dependent recombination in a silicon p-n junction. 215–217. Rev. New York). Rev. Molenkamp. A. Rugar. Beuneu. 2003a. Tsai. B 68. ´ ´ ´ ´ Simon. J. Langdo. G. A. R. ‘‘Enhanced room-temperature geometric magnetoresistance in inhomogeneous narrow-gap semiconductors. W. Si. Magn. Cheng. Slichter. J. 76. M.. C. A. Phys. L1–L7. 6995–7002. J. P. Nowak. H. ‘‘Random-spin-orbit coupling and spin relaxation in symmetric quantum wells. Magn. and B.’’ Phys..’’ Phys. Petrovic. J. and Das Sarma: Spintronics: Fundamentals and applications 405 Seymour. 2001. Rev. 1998. P. ‘‘Spin-dependent interface transmission and reﬂection in magnetic multilayers. 2003. Song. ‘‘Spin relaxation of photogenerated electron distributions in GaAs. Electrons and Holes in Semiconductors (Van Nostrand. H. Ando. 1767–1770. S.. S. L.ˇ ´ Zutic. 2002. in Spin Dependent Transport in Magnetic Nanostructures. Shimokohbe. 984–985 [Sov. Phys. Lett. L. 159. Rev.’’ Phys.. R. ‘‘Scattering theory of ballistic-electron-emission microscopy at nonepitaxial interfaces. ‘‘Nonmagnetic semiconductors as read-head sensors for ultra-highdensity magnetic recording. Kane. R. ‘‘Spin-transport dynamics of optically spin-polarized electrons in GaAs quantum wires. B 23. and L. Rev. and G.. Ambrose. Rev. and H.’’ Zh. Shinjo. Rev. C. Teor. Mod. Ting. S. and K. S. D. Sharma.’’ J. Byers. A. ‘‘Electron spin resonance on GaAs-Alx Ga1 x As heterostructures. Phys. D. 3623–3625. M. Munekata. pp.’’ Bull. Magn. F. ‘‘Spin injection into a Luttinger liquid. B 24. Fabian. Janossy. 3111–3113. H. W. Sham. Magn.’’ J. J.. Boeckl. D.’’ Rev. D. Slonczewski. A. Sherman. H. Davenport. 2001.. C. Lapertot. J. Keyes. V. Solin. Mater. Hines. R. Co. Kato. E. (Springer. 2. T. Rev. Princeton. L. Matter 5. Budko.. Rev. 1–46. J. 85–88. and S. B 67. Lett. N.’’ Phys. 2000. A. R. H. 48. ‘‘High minority-carrier lifetimes grown on low-defect-density Ge/GeSi/Si substrates. Eksp. . B 39.’’ Phys. E. T. 5.. Alfano. Solomon. B.. N. B 63. S. 1996. S. A. Sieg. Chamon. 2682– 2692.’’ Phys. Rev. NJ). Y.. Rev. B 61. Rev. 1983.. L. 047002. 87. G. Taraschi. Mater. B. ‘‘Electrical spin injection into semiconductors. 2000. 1993. Jr. D. R. S. Tanaka. ¨ Sidles. T. M. Schonhammer.. and J. Lett. 2003. Bruland. Smith. 2002. 87. Nadgorny. 3425–3427. 1976. C. T. 80. J. edited by S. J. 1989. Stesmans. A. J.’’ Phys. ‘‘Voltage-controlled spin selection in a magnetic resonant tunneling diode. 2002. M. ‘‘Magnetic resonance force microscopy. D. 90. Lett. 2002. M.. S. Slobodskyy. Slonczewski. Wegscheider. Phys. and K.’’ Phys. 1997. P. S.. 82. Lett. M.. Sogawa. 19. 1981. E. S. ‘‘Spin relaxation in semiconductor quantum wells. and C. 2003. Junnarkar. ‘‘Spin coherence in semiconductor heterostructures.’’ Phys. K. Feher. Currie.’’ Appl.. 1980. R2917–R2920. L. ‘‘Quantum interference control of ballistic pure spin currents in semiconductors. Abstreiter. 67. and R. Y. K. ‘‘Fermi-surface structure effects revealed by the observation of conduction-electronspin resonance in Zn. 90.’’ Appl. V.. ‘‘Measuring the spin polarization of a metal with a superconducting point contact. Goel. 3rd ed.’’ Nature (London) 415. 13 914–13 922. Santos. 136603.. J. Schmidt. Moodera.’’ Phys. ¨ ¨ Sham. Zuger.’’ Appl. S. 1977.’’ Science 289. E. R. Ya. and J. F. 2003. G. H.. Sogawa. J. and J. 1981. 2000. ‘‘Conductance and exchange coupling of two ferromagnets separated by a tunneling barrier. Sipe. 1976.’’ Phys. and P. V. S. Fiz. 159902(E) (2003). G. Narayanamurti.’’ Phys. Thio. Shang. Kozhevnikov. 90. T. 1983. H. 284–285. D. T. Ando. R. 130–133.’’ Phys. S. Rev. Stiles. Shinshi. Ya. D. Silsbee. 276601. 655–656 (1965)]. 51. Nowak..’’ Phys. 2001. W. Silver. O. ‘‘Novel method for the study of spin transport in conductors. 1998.. Coey. Berlin). ‘‘Magnetic bubble tunnel detector. Slonczewski. B 58. J. Ploog. Eshlaghi. Rev. R. Mod. R. Smet. Phys. S. Phys. Canﬁeld. 2003. Ertl. J.. Sharvin. 045323. 79. Barry. C. 388–393. Sherman. Fitzgerald.. ‘‘Light-driven microcantilever actuator based on photoenhanced magnetization in a GaAs-Fe composite ﬁlm. Garbini. T. Rev. Forro. F. Garaj. Lett. M.’’ Solid State Commun. ‘‘Spin relaxation of conduction electrons in bulk III-V semiconductors.. van Driel. ‘‘Anisotropy of superconducting MgB2 as seen in electron spin resonance and magnetization data.’’ Phys. Kim. Ringel. Skinner. Principles of Magnetic Resonance. D. Silsbee.’’ Physica E (Amsterdam) 2. A. 1950. F. 2003b. Lett. E.. A. A. Weimann. R. G. Y. T.. C. and M. 4012–4014. 73. 1995. Osofsky. Silsbee.’’ Phys. J. J. R. JETP 21. F. and J.’’ Science 282. 5535–5539. 035207. M. Smith. I. 2.. Chung. 161303.. Q. 1996. M. J. C.’’ J. L. R. A. Q.. T. v. Kogan. S. Stevens.. 1998. 209–211.. Phys. Klitzing. 78. Rev. M. T. 035321. ‘‘Spin conductivity and spin-charge separation in the high-T c cuprates. ‘‘Simple explanation of tunneling spinpolarization of Fe. 2328–2330.’’ IBM Tech. ‘‘Transport and lifetime enhancement of photoexcited spins in GaAs by surface acoustic waves.. 3191–3194. J. A51–A60. ‘‘Adiabatic charge and spin transport in interacting quantum wires. C. Rev. and F. Noguchi. 087901. Bhat.
81.’’ Phys. Lett. T.’’ Phys.. ‘‘Spin injection and detection in magnetic nanostructures. Slaughter. Brocks. 90. and P. Foxon. Samarth (Springer. ‘‘Highﬁnesse optical quantum gates for electron spins in artiﬁcial molecules. Cartoixa. 27. ‘‘On the electro-dynamic qualities of metals: Effects of magnetization on the electric conductivity of nickel and of iron. L. M. Meservey. P. Munoz. M. Esaki. Y. Walker.’’ Phys. J. and S. ‘‘Quantum transport and surface scattering. ` Teran. and N. and D. 036602. W. Rev. Lett. H. Lett. Rev. Magn. Takahashi.. DeHerrera. ‘‘Investigation of radiative recombination from Mn-related states in Ga1 x Mnx As. Lett. ` Ting. 7505–7507. F.. 3200–3202. 162. B.’’ Phys. and J. Y.’’ Appl. E. F. Crowell.. ‘‘Theory of tunneling spectroscopy of d-wave superconductors. 777–779. Berlin). 91. R. R1865–R1896. Stroud. K. S.. Lett. ˇ ´ ´ Tesanovic. ` Ting. P. Wyder. Rev.. 2002. D. R. 1391–1393.. 2000. Soc.. Koyama. M. and B. Tedrow. P. X.’’ Appl. M. Garcia. and Das Sarma: Spintronics: Fundamentals and applications Stiles. M.’’ J. 80. Rev. Zangwill. ‘‘Current-driven resonances in magnetic multilayers. 4198–4200. E.’’ Proc. and P. Co. Lett. 493(E). K. 46–49.’’ Phys. Petukhov. Naji. ‘‘Effect of Coulomb blockade on magnetoresistance in ferromagnetic tunnel junctions. and P.’’ Phys. Park. T. 563–565. Isakovic.. 2001. Meservey.. ‘‘Spin polarization contrast observed in GaAs by forcedetected nuclear magnetic resonance. M. 89. Penn. Jaric. J. and S. Lett. B 61.’’ Appl. Monzon. 2003. R. Matter 15.. 969–1016.. Fulde. 31–92. 26. A. and X. 570–578. R. ‘‘Ferromagnet (MnAs)/III-V semiconductor hybrid structures. Rev.. Rev. P. Tatara. Lett. and D. M. H.’’ Nature (London) 406. Timm. 238. Stiles. B 7. G.’’ Rev. Sun. M. D. Phys. 17. Y. J. Tanaka.. Schultz. 052409. ‘‘Recent developments in magnetic tunnel junction MRAM. Phys. C. P. Lett. ‘‘Roomtemperature electron spin dynamics in GaAs/AlGaAs quantum wells. and J. R.’’ Phys. Bass. Maekawa. ‘‘Carrier-concentration-induced ferromagnetism in PbSnMnTe. V.’’ Phys. 2003. Frankel. and R. 1981. 57. Filip. M. 562–564. Jansen. Tang. 1794–1796. ‘‘Generation and detection of phase-coherent current-driven magnons in magnetic multilayers. L.. Rev. H. Higo. M. M. W.. 89. Janesky. 2. 36. 1998. M. Rev.-Y. ‘‘Bloch equations with diffusion terms. 3451–3454. 76. Chiang. Lett. Z. Tehrani. 74. 1998. 166602. 2002. G. F. G. L. Wyder. Jansen. and P. and U.. P. Smith. Fabian.’’ Phys. ‘‘Reduction of spin injection efﬁciency by interface defect spin scattering in ZnMnSe/AlGaAs-GaAs spin-polarized light-emitting diodes. Tedrow. A.’’ J. Nishikawa. Rev.-C. G.’’ Appl. 866–868. and Y. Lett. T. S. Rev.. Phys. 2000.’’ Appl. Fainchtein. Gd.’’ Phys. D. Vol. Phys. Troiani. ‘‘170 nm nuclear magnetic resonance imaging using magnetic resonance force spectroscopy. H. 56. J. C. 1999. B. Suhl. Bobbert. Phys.’’ IEEE Trans. V..’’ Phys. Tsoi. Tedrow. 87. Z. Whig. J. Rev.’’ Phys. ‘‘Spin-polarized electron tunneling. 2003. Wyder.’’ Phys. Tanaka. Maekawa. Torrey. Imamura.. 3911–3915. F. Kojima 2000.’’ Phys. and R. and B.’’ Phys. D. D. G. Tsuei..’’ Phys. S. R. ‘‘On the magnetic properties of a CrBr3 single crystal. ´ Strand.. ‘‘Magnetic ﬁeld splitting of the quasiparticle states in superconducting aluminum ﬁlms. 026602. R. Meservey. Yamashita. 2002. 2000.. and R. 016401. R. R. Tsoi. 246803. S. Magn. Eaves.-Y. 173–243. J. A. F. Rev. Phys. Tsubokawa. P.. Zhao. Engel. 2001. Meservey. E. C. J.. Lett. P.’’ cond-mat/0206215. 1986. 2002. New York). B. ‘‘Domain wall scattering explains 300% ballistic magnetoconductance of nanocontacts. Phys. Dietz. Tsu. J. T. and P. 1971b. J. 919– 921. D. A. Phys. Patane. B. Kelly. 546–550. Bland (Springer. 27. Awschalom. London 8.. 15. 1729–1732. pp. A... Rev. ‘‘Spin polarization of electrons tunneling from ﬁlms of Fe. Lett. and S. 80. C. E. and M. M. Tromp. Roukes. 25. 2752–2757. and O. Cartoxia. Sugamo. 4281–4284. in Semiconductor Spintronics and Quantum Computation. 1973. M. 2003. Rev. Rev. 2004. Rev. S. Jpn. Thurber.. Tsoi. L..-W. M. 1971. 2000. ‘‘CaB6 : A new semiconducting material for spin electronics. 2000. 318–326. Bass. Chiang. 2003.. B 67. and D. S. 1994. R. C.. and A.. 2760– 2763. N. Gallagher. Rev. X. 83. Molinari. Technol. Smith. Rev.. Z.. R. Tanaka. Lett. D.’’ Phys. ‘‘Multiple magnon excitation in NiO by electron tunneling. 1270–1272. Eds. 014407.. A. Soc. Wolff. A. in press). Tsoi. Wada. 2003. ‘‘Bidirectional resonant tunneling spin pump. Campion. 1664–1668. J. ‘‘Spin-dependent tunneling into ferromagnetic nickel. 80. ‘‘Excitation of a magnetic multilayer by an electric current..’’ Phys. 2003. M. Lett. 81.. Phys. Chen. 83. and P. 68. and S. Loss. ‘‘Spin imbalance and magnetoresistance in ferromagnet/ superconductor/ferromagnet double tunnel junctions. Meservey. M. P. R. 1960. R. W. P. 1971a. ‘‘Spin-current interaction with a monodomain magnetic body: A model study. J. Takahashi. Palmstrøm. Thomson. and S. 206802. M. R. in Utrathin Magnetic Structures III. Phys.. M. 1996.. Wyder. 104. V. D. Hohenester. 2002. Mod. C. Tedrow. 89. and P. 87. New York). ‘‘Pairing symmetry in cuprate superconductors. Rev. Tackeuchi. Lett. Lett.. ‘‘Dynamical nuclear polarization by electrical spin injection in ferromagnet-semiconductor heterostructures. April 2004 Tedrow. 192– 195.’’ J. and Y. Takahashi. M. M. Heinrich and J. Jedema. H. 1758–1761. A. Rev. and X. Calder. ‘‘Disorder effects in diluted magnetic semiconductors. 1999. S. Jansen. ‘‘Tunneling in a ﬁnite superlattice. Lett. 72. Mod. Rev. Seck. ‘‘Anatomy of spintransfer torque. van Wees. 2030–2033.. T. 1970. Ni. R. D. Maekawa. A. R. D. Stiles.. 2002. Lett. Takahashi. ‘‘Resonant interband tunneling spin ﬁlter.’’ Phys.-C.: Condens. Durlam. and S. Hanbicki. and N.’’ Phys. Phys. G. Story. Maekawa. 83. Lett. . D.’’ Phys. Galazka. Tsoi. V.’’ J. Z. ‘‘Calculation of spindependent interface resistance. 327–341. Rev. Rev. C. and L. Rep. Maekawa. B 62.’’ Phys.’’ Appl. M. 1973. 2001. G. edited by D. C. L. edited by B. Kirtley.’’ Phys. 1857. Tsoi. A. 22.406 ˇ ´ Zutic. van Gelderen. A. B 66. Appl. Tsui. Reson. 1956. Magneto-optics (Springer. T. 2002. 336–340. I. A. ‘‘Large tunneling magnetoresistance in GaMnAs/AlAs/GaMnAs ferromagnetic semiconductor tunnel junctions.. Lett. J. 1986. 797–799. Sci. Zhao. Sze. ‘‘Joule heating generated by spin current through Josephson junctions. Thurber.. and L. Jonker. Rev. ‘‘Proposal for in situ enhancement of electron spin polarization in semiconductors. and R. Physics of Semiconductor Devices (Wiley. M. No. 1995. M. A. Lett. Harrelli.’’ Phys. New York.’’ Semicond. T. ‘‘Direct observation of spin-mixing in superconductors. 2002.. 82. Kashiwaya. Lett.
B 48. S. ‘‘Comment on ‘Observation of spin injection at a ferromagnet-semiconductor interface. 775–777.. R. van Wees. M. V. Upadhyay. Rev. V. R. Sawicki.’’ Appl. and L. and I. A. Valet. 572–574.’’ Phys. A. No. B 42. 2003. Phys. Rev. Kim. M. . and L. Phys. R. Rev.Peterburg) 10. M. 2003. 829–834. and V. A. K. and S. 15 397–15 400. Grupp. Sham. Mod. 2003. 87..’’ Appl. V. P. A. X. S. Goldman. Rev. 1134–1137. P. Upadhyay. Struct. B 39. 7114–7123. ‘‘Coherent electron focusing with quantum point contacts in a two-dimensional electron gas. 146803. 81. Lett. Lodder. 127. Williamson. Beenakker.’’ J. Jr. S. and L. L. and D. Tsymbal. W. ‘‘Spin dynamics and spin-dependent interactions in semiconductor heterostructures. B. I. Reznitskikh. T. Rev.’’ Phys. Semicond. ‘‘Coulomb drag. 1523–1525 (1991)]. Buhrman. 1997. ‘‘Critical current suppression in a semiconductor by injection of spinpolarized carriers from a ferromagnet. N. 1990b. Van Esch. Raitsimring. 3881–3883. Vekua. Vol. 3364–3366. S. Bass. A1286–A1290. Tyryshkin. 2003. 1997. V. 2695–2698 [Sov. and S. and G.. 78. A.. B.. R.’’ Phys. Bari. Nikolaev. Borghs. S.. 80. L502–L504. Nikolaev. J. Yu. A. ‘‘Modeling of spinpolarized electron tunneling from 3d ferromagnets..’’ Phys.. M. P.. ‘‘Electron spin relaxation times of phosphorus donors in silicon. Lett. D’Amico. 2001. Vina. Van Roy. Rev. Kochereshko. 74. J. 1989. Larkin. and G. C. Viglin. Goldman. van Dijken. S. 197203.Mn)As Zener diode. 73. P. Vagner. Fleisher.. X. O.’’ preprint. Van Dorpe. D. Mooij. 13 103–13 112. H. S. T. van Kempen.. 354–357 [Sov. J. V.. Tekh. Tela 30. and J. Daiminger. V. Rev. Pettifor. Samokhvalov. N. Lett. De Boeck. Parkin. Rev. K. 959–964.’’ Phys. De Boeck..’’ Appl. Parkin.. 376–383. van ’t Erve. J. Rev. S. ‘‘Coherent transport of electron spin in a ferromagnetically contacted carbon nanotube. M. Samokhvalov.’’ Phys. N. M.. Birge.’’ Nature (London) 401. 8556–8575. 1993. S. W. V. P. 1997. ‘‘Spin-lattice relaxation time of conduction electrons in sodium metal. Rev. He. S. 1976. B 66. van Loosdrecht. W. Roy. 38. S. G. and S. ‘‘Magnetoresistance of half-metallic oxide nanocontacts. P. J. W. and J.’’ Superlattices Microstruct. A. Lett. P. Lett. Uenoyama.. 90. 83. Lyon. P. 10. Uraltsev. N. magnetoresistance. 2003. V. J. and J. J. 3247–3250. 1990a. 134. 2002b. S. E.. Semicond.’’ Physica B 298. and O. 2003. and spin-current injection in magnetic multilayers.ˇ ´ Zutic.. K. and I. ‘‘Hole relaxation and luminescence polarization in doped and undoped quantum wells. H. Parkin. Das. F. F. A. and J. Rev. V. Vasilev. A. and S. Jiang. 024416. 026601.. Z. Osipov. 83. Lett. ‘‘Spin ﬁltering by ultrathin ferromagnetic ﬁlms. 951–953. van Dijken. N. Phys. Lett. N. O. Viglin. Lee. ‘‘Nonmonotonic bias voltage dependence of the magnetocurrent in GaAs-based magnetic tunnel transistors. 91. Liu. Vlutters. and A. Methfess. Rev. 5023. ‘‘Increase in the transmission of microwave power through a ferromagnetic semiconductor HgCr2 Se4 in a strong electric ﬁeld. Schumacher.’’ Phys. Martin.. Vas’ko. E. P. C. Larkin. and H. Astashkin. and A. Voskoboynikov. Herlach. Jiang. Vescial. Rev. G. 3070–3073. T.. M. ‘‘Nuclear spintronics. G. ‘‘Highly efﬁcient room temperature spin injection in a metalinsulator-semiconductor light emitting diode. G.’’ cond-mat/0308244. Straka. E. ‘‘Giant magnetocurrent exceeding 3400% in magnetic tunnel transistors with spin-valve base. A. Safarov. Foxon. Tverd. Sandler. Lett... Kraus. and J.. (S. 2003c. B 58. X. 64. Parkin. Lett. A. P. M. and C. Lett. B. P. ‘‘Spindependent hot electron transport in Ni81Fe19 and Co84Fe16 ﬁlms on GaAs(001). Kraus. Louie. 1967.. Motsnyi.. Harris. Rev. B 68. Alphenaar. S. van Steenbergen. 76. H. 1964. I. K. Phys. M. Solid State 33. D. F.. S. L. Jiang. ‘‘Current-driven magnetic excitations in permalloy-based multilayer nanopillars. ‘‘Optical orientation of holes and electrons in strained layer InGaAs/GaAs quantum-wells.’’ Phys..’’ Appl. Dzhioev.’’ Phys. Buhrman. R. 1999.’’ Phys. Ago. Vignale. Palanisami. J.. E. ´ von Molnar. A. 2271–2273. Vas’ko. Rev. van Dijken.’’ Phys. 1Õ2. S. Motsnyi. 58. Nijboer. Coey. S. Zakharchenya. Wyder. L411–L417. M.’’ Phys. Phys. and R. A. ‘‘Theory of the perpendicular magnetoresistance in magnetic multilayers. ‘‘Room temperature operation of a high output current magnetic tunnel transistor. G. 2000. van Wees. Tsukagoshi. A. E. P. A. and A. 1997. Rev. Aichmayr. I. A.’’ Solid State Commun. Appl. V..’’ Appl. ‘‘Modeling of spin-dependent hot-electron transport in the spin-valve transistor. 2001. A. A. Ma. J. S. 97–100. D. Phys. 2001. ‘‘Interplay between the magnetic and transport properties in the III-V diluted magnetic semiconductors. ‘‘Spindependent electronic tunneling at zero magnetic ﬁeld. F. 1998. Lett. ‘‘Very high spin polarization in GaAs by injection from a (Ga. and J. M.. Phys. Borghs. ‘‘Probing ferromagnets with Andreev reﬂection. R. 1993. Lett. C. F. ‘‘Differential conductance of the ferromagnet/ semiconductor interface of DyBa2 Cu3 O7 /La2/3Ba1/3MnO3 heterostructures. M. D. J. 84. Rev. R. Broekaart. Pratt. V.. V.’’ Phys.’’ Fiz. J. van Dijken. J. 2003b. Urazhdin. Liu. and S. Fabian. V. S. 094417. N. G. and S. B 56. van Son. Nordman. 844–846.’’ Phys. B. J. S. D. De Boeck. ‘‘Giant negative magnetoresistance in ferromagnetic Eu1 x Gdx Se. ‘‘Carrier relaxation and luminescence polarization in quantum wells.’’ J. Part 2 42. V. Jansen. P. 82. Rev. Lett. P.: Condens. 89–94. 954–956. Parkin. Uenoyama. X. Phys. and R. G.’’ Appl. and A. C. Verbanck. 1999. L. 1987. and A. G. 2. L. Goovaerts.’’ Phys. van Dijken. 2003a. A. VanderVen. 193207. Van Bockstal. 210–212 (1946)]. Bogaerts.’’ Fiz. and R. J.. Phys. J. A. ‘‘Boundary resistance of the ferromagnetic-nonferromagnetic metal interface. L. 1998. ‘‘Hanle effect in optical orientation of electrons in n-type semiconductors. T. Sham. Jiang. P. Low-Dimens. A. 2002a. C. ‘‘Magnetization-dependent rectiﬁcation effect in a Ge-based magnetic heterojunction. Wellmann.. and P. Fert. Appl.’’ Phys. 13.’ ’’ Phys. ‘‘Comparison of magnetocurrent and transfer ratio in magnetic tunnel transistors with spin-valve bases containing Cu and Au spacer layers. P.. Lett. 2003. V. W.’’ Jpn. Van Dorpe. X. Phys.. S. B. Matter 9. Phys. K. Grietens. Forchel. 1991. April 2004 van Houten. and Das Sarma: Spintronics: Fundamentals and applications 407 Tsui. 7099–7113. ‘‘Generation in the millimeter band for the semiconductor-ferromagnetic semiconductor heterostructure. Versluijs. Louie. Jiang. J. 1998. B 65. Poluprovodn. G. T.. Phys. Borghs..’’ Phys. Osipov.
D. T.. Phys. 1998.’’ J. Z. ‘‘Spin resonance of charge carriers in graphite. Wexler. M. Tran. Rev. Brownell.. 2542– 2545. Wangsness. Mor. Rev. B 62. ‘‘Role of spin-dependent interface scattering in generating current-induced torques in magnetic multilayers. A 62. Winkler. R. and C. H. Appl. 1067–1074.’’ J. H. 2003. McEuen. R.-E. Pinch.. van der Vaart. T. Jantsch. von Philipsborn. G. H. Schmidt. and T.’’ Phys. Lee. Yeh. B. ‘‘Device application feasibility of single-crystal CdCr2 Se4 . 2003. Malissa. Rev. Rossler. G. 816–822. Munekata. and Ga1 x Alx As. A. C. J. J. A. Rev. Z. H.’’ Phys. Wang. Buhrman. 045317. and D. C.’’ Phys. P. Rev.. W. R. 4786–4789. Phys. J. G. H. in Superconductivity. and B. Werthamer. B 67. D. Phys. Waintal. B 59. ‘‘Quantum spin ﬁeld effect transistor. K. Ga1 x Inx As. Ossau.-J. ¨ Wilamowski. R. Yeh. Rev.’’ Physica E (Amsterdam) 12. Treger. Worledge.’’ J. T.-C. 447–451. and R. F. J. Lett. Slupinski. 1999. H. 89.. Walsh. and Klapwijk theory. C. 1998. 2000. 1953. Rev. ‘‘Spin-polarizing properties of the InAs/(AlSb)/GaMnSb/(AlSb)/InAs ferromagnetic resonant interband tunneling diode. 36. 1963. ‘‘Local dynamic nuclear polarization using quantum point contacts. 1966b. M. ‘‘Tunneling spectroscopy study of spin-polarized quasiparticle injection effects in cuprate/manganite heterostructures. 14. Rev. Rev. Fischer. ‘‘Directional tunneling and Andreev reﬂection on YBa2 Cu3 O7 single crystals: Predominance of d-wave pairing symmetry veriﬁed with the generalized Blonder. 181–183. 1969. Rev. W. Wolf. and D. P. Daughton. Schuetz. Daugh´ ton. 626–632. Solomon. New York). Chtchelkanova. Strasik.. D. and D. Walsh. J. ‘‘Spin manipulation using magnetic II-VI semiconductors. S.. .’’ IEEE Trans. 2001. Wang..’’ Phys. Guittienne. A. Vol. Wang.’’ Phys.. 1966a. Supercond. ‘‘Ultrafast softening in InMnAs. 247–250. J.. Ansermet. Nordman. ‘‘ESR studies of the Bychkov-Rashba ﬁeld in modulation doped Si/SiGe quantum wells. I.. and J.. Frank. H. Oiwa. Mod. H.’’ Phys. ‘‘Spin splitting in the electron subband of asymmetric GaAs/Alx Ga1 x As quantum wells: The multiband envelope function approach. B 69. and C. April 2004 Wei. 1011–1014. 12 514–12 520. C.’’ Phys. 67–70. 85. J. W. D. Vurgaftman. von Molnar. 1977. Rev. and T. 45. and H. Wen. DiVincenzo. ‘‘Resonant Andreev reﬂection in a FIDI’D’ system where F is a spin-polarized and DI’D’ is a d-wave superconductor containing a 100 110 grain boundary. Vrijen. ‘‘Prototype spin-dependent tunneling isolators integrated with integrated circuit electronics. Weisbuch.’’ Phys. L. 85. and J. P.’’ Proc. Wang. 035328. Wagner. Garrigus.’’ Europhys. ‘‘g value of potassium conduction electrons. and C. Rev. H. C. S. ‘‘The dynamical theory of nuclear induction. Rev. ‘‘Observation of vacuum tunneling of spin-polarized electrons with the scanning tunneling microscope. T. M.. Wagoner. ‘‘Evidence and evaluation of the Bychkov-Rashba effect in SiGe/Si/SiGe quantum wells. 1.’’ Phys. Wolf. Vasquez. B 47. S. B 69. 2000. R. Weger. ‘‘Thermokinetic approach of the generalized Landau-Lifshitz-Gilbert equation with spin-polarized current. Ruf. 647–653. and V. Z.. Lett... T. Waag. Awschalom. W. 2002. Y. R. and U... and Das Sarma: Spintronics: Fundamentals and applications Voskoboynikov. P.’’ Science 294. Parks (Dekker. Vinter. pp. ‘‘Spintronics: A new paradigm for electronics for the new millennium. 16.. London 89. Y.’’ Phys. 118. New York). 1960.’’ Phys.. 65. Wegrowe. A. 4. S. Vol. Molenkamp. ‘‘Experimental realization of a quantum spin pump. Bloch.’’ IEEE Trans.. Rev. 2000b. 91. P. 81. Y. C. 1999. G. 1994..’’ Phys. 2002. ‘‘Optical detection of conduction-electron spin resonance in GaAs. N. S. 5350–5352. Treger. Lett. Rev. and D. L.’’ preprint. E. Fabian. R. Wong. R. 15 375–15 377. R. H.. and P. 12 317–12 327. 258301.’’ Phys. 2001. A. Hermann. C. Fui. 2003. A. A. Y. K. 414–417. Magn. 702–704. J. M. and F. P.. K. V. E. 2000a. R. Wiesendanger. Rupp. ‘‘Current-induced magnetization reversal in magnetic nanowires. 195315.. C. Jiang.. and K. 2003.-S. Wann. Rev. D. Richards. Lett. B. J. Ploog. 2000. B 58. Potok. ‘‘Spin orientation and spin precession in inversion-asymmetric quasi two-dimensional electron systems. W. Jantsch.’’ Phys. Rev. Wilamowski.’’ Phys. A.. Winkler. Rev. Foxon. U. J. W. C. edited by R. Balandin. ‘‘Negative spinpolarization of SrRuO3 . Khodaparast. Tinkham. Rev. 728–739. Yablonovitch. 2004. Wilamowski. W. Rev. a ferromagnetic semiconducting spinel. 76. 73. D. J.’’ J.. M. L. 1999. and W. Wei. No. E. L. Umansky. 439–442. J. 2748–2751. Richards. Kouwenhoven. N. Ralph. P. T. P. Lett. Geballe. B. Magn. 291–298. Q. Worledge. and J. P. ‘‘Nanoscale CMOS.’’ Phys. Lett. D.. in Scanning Probe Microscopy. 71–95. Gambino. 5182–5185. Wang. Weisbuch.408 ˇ ´ Zutic. Appl. Roychowdhury. F. 2002. P. Rev. Guntherodt. Lange. Schmidt. Guntherodt. ‘‘Spindependent tunneling in double-barrier semiconductor heterostructures. Watson. and M.’’ Phys. R. K. D.’’ Phys. Supercond. 1993. and P. Kono. ‘‘Observation of extremely long electron-spinrelaxation times in p-type -doped GaAs/Alx Ga1 x As double heterostructures. and D. ‘‘Maki analysis of spin-polarized tunneling in an oxide ferromagnet. 1999. 1991. ‘‘Suppression of spin relaxation of conduction electrons by cyclotron motion.’’ Proc. N.. Quantum Semiconductor Structures—Fundamentals and Applications (Academic.-C. Rev..’’ Phys.. 8405–8407. ‘‘Nuclear polarization in homogeneous InSb by a direct current. 581–588. Welser. 537– 570. Lett. Guo. ¨ Wissinger. S. Wegrowe. M. Qian.. M. ¨ ¨ Wiesendanger. M. edited by R. and C. N. Myers. L. Schneider. 2000. and L. A. Geballe. J. L. R. D. and H. 1990. B 15. 142. ‘‘Spintronics: A spin-based electronics vision for the future.. W. C. Kelly. ‘‘Electron-spin-resonance transistors for quantum computing in silicon-germanium heterostructures.. Wiesendanger (Springer. Meyer. D. Rev. Roukes. 092408. G. and R.’’ Phys. M. ‘‘Spin-injection quasiparticle nonequilibrium in cuprate/manganite heterostructures.. Gruber. Rupp.’’ cond-mat/0305017. pp. Marcus. Y. Z. Reuscher. IEEE 87. and H. J. Jaccard. New York). Rossler. X. Jantsch.. 1998. ‘‘g values of rubidium and cesium conduction electrons. B 62. and W. 125209. W. 927–941..’’ Phys. G. B 62. 2002. Hershenov. Phys. Rev. L. ‘‘Size effect and non-local Boltzmann transport equation in oriﬁce and disk geometry. 2002. C.’’ Phys. Jusserant. 2. Hu. C. Liu. 012306. B 67. M. 1968.-R. M. E. 321–370. and J. 91. A. 15. Wei. B. D. Brouwer. 1488–1495. Wald. Wang.’’ Phys.. J. R. H. 1966. 2004. Schmidt. Soc. L. B 66. Tondra. 132. T.
Takahashi. B. Zakharchenya. Rev. ´ Yu. Rev. Xiong. B 64.’’ Phys.. Maekawa. 58. 137–139 (1971)]. 439– 444 (1993)]. V. Z. P. 115319.. ‘‘Spindependent transparency of ferromagnet/superconductor interfaces. Ting. Phys. Smith. Y. 821–824.’’ Phys. Lett. ‘‘Electric-ﬁeld dependent spin diffusion and spin injection into semiconductors. No. D. T. 85. and H. ‘‘Average forces in bound and resonant quantum states. N. Rev. B 66.. Hu. Turek. I... ‘‘Modeling of tunneling spectroscopy in high-T c superconductors incorporating band structure. W.’’ Pis’ma Zh.’’ Phys. C. D. W.’’ Phys. W. Zhang. P. 2003.’’ J. ‘‘Tunneling. H. Xiao. B 68. S.. Fiz. H. 2002. Mavropoulos. B 65. S. Fleisher. B 61.’’ Phys. Z. and spin-charge separation in superconductors. Zhitomirskii.’’ J. 201202. Y.’’ Phys. Guo. M. ‘‘Parametric quantum spin pump. Zeller. Phys. B. relaxation of spin-polarized quasiparticles. and T. E. and M. E. M. D.’’ Phys. D. and P. P. J. Teor. and C. Rev. Hall. C. Pfeffer. ¨ Schonherr. Hershﬁeld. and Y. C. Fiz. P.-h. Phys. T. 514– 521.. 2000. and S. 9558–9563.ˇ ´ Zutic. D. 1961. Rev. G. and A. Chem. E. ¨ ¨ Zhao. Ohno. Parkin. ‘‘Quenching of magnetoresistance by hot electrons in magnetic tunnel junction. V. Coffey. 166603. A. L. 1334–1341. Zhu.’’ Phys.. Wu. von Klitzing. G. G.’’ Phys.. Teor. ‘‘Andreev reﬂection in ferromagnet/superconductor/ ferromagnet double junction systems. Yamashita.. S. B 65. 2002.’’ Phys. Shi. B 67. 2003. Kirpichev. Grundler. 99–104. Phys. L. N. Sun. W.. Wu. in Solid State Physics. Torelli. P. Monch. Miyakawa. 235313. Sarrao. T. Zhu. B 59. 1971. Zhao. 2003. Zheng. 2000. 13. Fabre. Vasquez. Rev. W. Filin. B 52. Zerrouati. p. B 58. E.-R. ‘‘Spin Hall effect in the presence of spin diffusion. Landry. Oseroff. 412– 414. K. 393–396. B 64.’’ Pis’ma Zh. S. E. B 61. ‘‘Spin-polarized quasiparticle transport in ferromagnet-d-wavesuperconductor junctions with a 110 interface. Lampel. Appl. 1995. 76. D. 235302. Q. Yamanouchi. Vakili. W.. Maekawa. B 68. R. H. and S. 174504. and P. J. 2. Seitz and D. M.’’ J. Rev. L.. M. B. Fu. D. Y. and M. 2001. 2004. 2002. Eksp.. 3744–3747. 2000. Mizushima. and C. T. Rev. 2945–2955. and I. K. A. I. Rev. Li. K. ‘‘Electron transport in normal-metal/superconductor junctions.. Matsukara. 2002. Fabian. 14 759–14 764. Rev. 79. ´ Yu. Supercond. ‘‘Kinetic theory of spin coherence of electrons in semiconductors. ‘‘Optical detection of spin relaxation of 2D electrons during photoexcitation.’’ Nature (London) 428. Rev.’’ Phys. A. Kelly. Johnston-Halperin. Yafet. ‘‘Giant magnetoresistance in organic spin-valves.’’ Phys. and S. J. V. Xiang. and H. Yeh.. B 66. V. J. R. Wu. Pfeffer. O. Imamura. ‘‘Spin-coherent transport in ferromagnetically contacted carbon nanotubes. W. Zhao. P. . M. ‘‘Effect of optical orientation of electron spins in a GaAs crystal. E.’’ Phys. ‘‘Kinetics of spin coherence of electrons in an undoped semiconductor quantum well. Lett. f. 1997. 2000. and H.. ‘‘Can D’yakonov-Perel’ effect cause spin dephasing in GaAs(110) quantum wells?’’ Solid State Commun.. X. X. Mod.’’ Appl. G. and S. Wang. Kuwata-Gonokami. ‘‘Room-temperature spin injection from Fe into GaAs. Ott. You.. A. and D. 2001. 1999. P. and K. ‘‘D’yakonov-Perel’ effect on spin dephasing in n-type GaAs. S. Yamashita. P. Paget. and D.-f. 1999. B 67. D.. R. B. F.’’ Nature (London) 427. H. I. Flatte. 2. Bacquet. Y. 125202. April 2004 Young. G.. Rev. M.. L. Lett. Y. Rev. and J. Maekawa. Rev. ‘‘Ground-state properties of ferromagnetic metal/ conjugated polymer interfaces. H. and tunneling directionality. J. and C. Solids 21. New York). 87. ‘‘Nonequilibrium superconductivity under spin-polarized quasiparticle currents in perovskite ferromagnet-insulator-superconductor heterostructures. Vardeny. 2003. Xie. Bishop. Yan. ‘‘Effects of density of states on bias dependence in magnetic tunnel junctions. 234–237. Rev. and I.’’ Phys. T. 1999... Veshchunov. 2001. L.. Wu. M.’’ Phys. J. 3144–3146. and M. W.’’ Phys. Young. J. Frandon. 1998.’’ Nature (London) 397. B. 094515. G. Ramsteiner. 2002. C. B 37. E. Wassermeier. Metiu. Rev.. Y. Zasadzinski. Z. Rev.. Du. Wunnicke. 2002. H. Bauer. ‘‘Average forces in bound and resonant quantum states. ‘‘Spin-polarized resonant tunneling in magnetic tunnel junctions. 2002a. Lin. J. 14. V. Yafet. T. Ploog. Status Solidi B 222. Z.’’ Phys. Yusof.’’ Phys. H. Z. ‘‘Non-equilibrium Green’s-function approach to electronic transport in hybrid mesoscopic structures. ‘‘Hyperﬁne interactions due to orbital magnetic moment of electrons with large g factors. B. F. Chiba.’’ Phys. B. Rev. Flatte. B. F.-X. Rev.’’ Phys. B. Fisk.. Dzhioev. Rev.’’ Science 297.. Vol. Ohno. Q. Thompson. D. Zawadzki. Ning. Wu. Nagahama. Saxena. 10 522–10 526. 2004. Lett. ‘‘High-temperature weak ferromagnetism in a low-density free-electron gas. 172509. 80. S. B 66. ‘‘Spin transport and relaxation in superconductors. 58. V. I. 523–534. Zhu. B 65.-Z. 113306. Muhl.’’ Phys. H. 1998. H. 14. 2002. R. Samoilov. S.. and Das Sarma: Spintronics: Fundamentals and applications 409 Wu. Lett. and K. 87. K. 1963. gap symmetry.’’ Phys.. J. 5215– 5217. Sun. 1993. M. Rev. 174407. and S. Goodrich. H. Marley. Turnbull (Academic.’’ Phys. Rev. Yuasa. 89. Lett. and H. Vinzelberg. H. ‘‘Current-induced domain-wall switching in a ferromagnetic semiconductor structure. Timofeev. Imamura. Z. ‘‘Spin-lattice relaxation in p-type gallium arsenide single crystals. Levy. 509–512. Q. J. Dederichs. R. Zhang. and K. H. D. Phys.. Zysler.. S. Takahashi. 429–434 [JETP Lett. 539–541. ‘‘Crossed Andreev reﬂection in structures consisting of a superconductor with ferromagnetic leads.. edited by F. Z. ‘‘Theoretical approach to the spin-dependent hot-electron transport in a spin valve. B 58.-Y. and S. ‘‘Voltage controlled spintronic devices for logic applications. 195–197 [JETP Lett.’’ Appl. and J. 2001..’’ Phys. 2000. group velocity. 2002b. Shepel. Rev. Awschalom. Zawadzki. Rev. 3632–3638. Vol. S. 80. G. Yamashita. Zhu.-R. J. Claro. 016601. Rusanov. and N. M. Ahn.. Yamauchi. 245–259. and K. Zeng. 2002. 024516. H. 2003. B 60. Rev.. 235313.. Schneider. H. Ohno. T. and C. ‘‘Spin diffusion and injection in semiconductor structures: Electric ﬁeld effects. ‘‘Ballistic spin injection from Fe(001) into ZnSe and GaAs. Y. Friedman. Z. ‘‘Andreev reﬂection through a quantum dot coupled with two ferromagnets and a superconductor. Xia.’’ Phys.-C. 13. Wang. and R. G. Kostial.’’ Phys. Takahashi. Eksp. and F. 2002. Bader. 1934–1939.. M. 121. 1598–1600. B. B 68. Suzuki. S. ‘‘Anisotropic electrical spin injection in ferromagnetic semiconductor heterostructures. J. Rev. 1988.. 241306.-P.
‘‘Proceedings of Spintronics 2001 International Conference on Novel Aspects of Spin-Polarized Transport and Spin Dynamics.’’ J. 4. T. Mod. J. 2003. ˇ ´ Zutic. and M. I. ‘‘Spin-voltaic effect and its implications. Zudov. 2.. 431–436 (2003)].’’ Phys. W. Rev. ‘‘Spin-polarized transport in inhomogeneous magnetic semiconductors: theory of magnetic/nonmagnetic p-n junctions. and the quantum origins of the classical. 1999. Lett.. and S. Ed. A. Washington. ˇ ´ Zutic. 79. 2001b. and A. Valls. Phys. 715–775. G. 2003. ˇ ´ Zutic. Kono. Das Sarma. Vol. 1555–1566. M. Das Sarma. I. Khval’kovskii. Phys. Fabian. J.’’ Mater. ‘‘Spin polarized tunneling in ferromagnet/unconventional superconductor junctions.’’ Phys. T.. Mishchenko. I. 066603.. April 2004 . 092401. R16 322.. Zurek. M.. Eds. 53–58 [Tech.’’ Phys. and M. Munekata. New York). T. Zwierzycki. W.’’ Appl. 2003. K. Phys. 76. J..’’ J. 221–223. Kelly. Rev. and S. JIM 44. Valls. M.’’ Phys. Lett. H... 15. 2002. Rev.’’ Phys. Das Sarma. ‘‘Spin injection through the depletion layer: a theory of spin-polarized p-n junctions and solar cells. 1–104. 2002. I. J. Das Sarma. 2001a. Rev. Fabian. Xia. Phys. and S. K. and S. I. B 61. A.. ‘‘Spin injection through an Fe/InAs interface. Fabian. A. Supercond.’’ Appl.. I. Bristol/ Philadelphia). ˇ ´ Zutic. Phys. Fabian. I.410 ˇ ´ Zutic. and C.’’ Appl. I. B 66. Fabian. ˇ ´ Zutic. J. Fabian. Rev.. Modern Magnetooptics and Magnetooptical Materials (Institute of Physics. Rev. V. 6320–6323. ‘‘Spin-polarized transport and Andreev reﬂection in semiconductor/superconductor hybrid structures. D. Rev. S. 15. Zvezdin.. ‘‘DNA spintronics. ‘‘Tunneling spectroscopy for ferromagnet/superconductor junctions. 81. K. Lett. Zwolak. 2002b. A.’’ Rev. J. ˇ ´ Zutic. E. B 60. Rev. H. and I...’’ Phys. Kotov. Sanders. 82. and J. Lett. Di Ventra. M. ‘‘A proposal for a spin-polarized solar battery. Bauer. 48. 1558– 1560. H. Sun. and O. ‘‘Ultrahigh ﬁeld electron cyclotron resonance absorption in In1 x Mnx As ﬁlms. 5–12. 161307. Phys. einselection.. 2000. J. 2002. 1999. Fiz. 2003.. Stanton. 75. Supercond. Mod. 2062–2065. I. Zvezdin. and O. ˇ ´ Zutic.C.’’ Zh. Ikaida.’’ Phys.. 2003. ‘‘Current-voltage characteristic of a spin half-metallic transistor. Miura. J. Thornton 2001. I.. ‘‘Proposal for all-electrical measurement of T1 in semiconductors. 2002a. ‘‘Decoherence. A. N. ˇ ´ Zutic.. D. ‘‘Novel aspects of spin-polarized transport and spin dynamics. Matsuda. and Das Sarma: Spintronics: Fundamentals and applications Ziese. B 60. B 67. ˇ ´ Zutic. Tekh. B 64. 121201. G. Y. 925–927. and S. 1997. ˇ ´ Zutic. Trans. and V. 88. Spin Electronics (Springer. Y. No. Das Sarma. P. Turek.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.