226-libre

Proceedings of the 5th National Conference; INDIACom-2011 Computing For Nation Development, March 10 11, 2011

Bharati Vidyapeeths Institute of Computer Applications and Management, New Delhi

Copy Right INDIACom-2011 ISSN 0973-7529 ISBN 978-93-80544-00-7

Spintronics-A New Light in Nanoscale Devices Namit Gupta1, Kamal Chaudhary2 and Sumant Katiyal3

[email protected], [email protected] and [email protected]

ABSTRACT Spintronics is the field in which the spin degree of freedom of the electron plays an important role in addition to or in place of the charge degree of freedom in mainstream electronics. Spin of an electron, although not given sincere wattage till now can play a very vital role in understanding the transport phenomenon of charge in nano scale devices. Adding spin degree of freedom to electrons will provide significantly more capability and performance to the nano scale electronic devices [1]. Spin transport effects and spin polarization have been very well defined using ferromagnetic materials. The electrons at or very near to the Fermi surface, in the ferromagnetic materials are partially spin polarized. The larger the degree of spin polarization of these electrons, the bigger the spin transport effects. Spin based devices such as Spin-FETs , Spin-LEDs , Spin- resonant tunneling devices ,spin coherent devices and spin quantization devices are now real and at the verge of implementation. Our Tutorial paper tries to exaplore the various possibilities of research in the field of Nano structured devices, using Spintronics as the basis of nano devices.

KEYWORDS Spintronics, Spin degree of freedom, GMR, MTJ, Spin Polarization and QCA

1. INTRODUCTION Spin is a purely quantum phenomenon roughly akin to the spinning of a childs top or the directional behavior of the compass needle. The top could spin in the clockwise or counterclockwise directions, electronics have spin of a sort in which their compass needle can point either up or down in relation to a magnetic field. Thus a new kind of binary logic of ones and zeros can be led by spin of an electron. [2] Spin can be easily manipulated by an externally applied magnetic field. Another property which is shown by the spin of an electron is its long coherence or relaxation time. Once spin is created it tends to stay that way for a long time. This enables the electron charge to be transported for a longer duration of time, unlike the conventional charge states, which are easily destroyed by scattering or collision with defects or impurities or other changes. These characteristics open the possibility of developing devices that could be much smaller, consume less electricity and be more powerful for certain types of computations than is possible with electron charge based systems.

An important challenge for the nanotechnologists in the molecular computing community is to develop a design methodology and an associated two-terminal device which isolates its inputs from its outputs without requiring a clock [3]. Additionally, whatever strategy is employed should not require global analysis in order to understand the behavior of a local structure. All spin-transport devices are based on the effect that the magnetic spin of electrons can be manipulated in opto-electronic devices by magnetic fields or applied voltage. The domain of spin-transport devices is commonly referred to as magneto electronics or Spintronics. The spin transport effects are most widely used today in magnetic metallic devices such as read heads for Hard Disk Drives and in Magnetic Random Access Memories (MRAM). These device structures comprise magnetic thin films, the magnetic orientation of which can be changed with applied magnetic fields. In case of non-volatile MRAM the orientation will be preserved until the bit is switched again. Hence, a locally generated field performs the writing of a magnetic bit, by currents in the vicinity of the magnetic structure. The reading principle of magnetic bits is based on the discrimination between two distinct magnetic states in the device by measuring the (tunnel) resistance in the device. In future generations semiconductor-based spin-injection components are expected to play a significant role. Spin valve devices, Tunnel junction devices and spin injection devices are few major initiatives in this field.

2. CHALLENGES AND ISSUES The main difficulties related to spin devices are summarized as follows: Tunnel Junction Devices: In order to have optimum resistances (of about 50 Ohm) future high density tunnel junction MRAM will require thin tunnel junctions of the order of a nanometer. The spread of the junctions must be at Angstrom-level for achieving the uniformity needed for high density. This requires a high precision in manufacturing. Tunnel Junction Devices: The current bottleneck for reducing the MRAM cell size is limited by the semiconducting diode / transistor that avoid the shortcuts in the array (blocking diode). For large-scale integration, either the size of the blocking diode must be reduced or alternatively the spin device itself must perform the blocking function. Spin injection devices: Although spin injection into a semiconductor has been demonstrated, it is still to be proven that the materials and the concept are suitable for room temperature operation and large scale

Proceedings of the 5th National Conference; INDIACom-2011


There is a strong possibility of developing new architectures for the nanoscale devices, as the existing architectural models are all time tested for the CMOS circuits [2]. These new architectures will definitely be helpful for the devices to succeed in nanoscale. Quantum computing is one such architecture, which is useful for the few of the nanoelectronics applications [4]. In this proposed work it is planned to develop the theoretical models to deal with the spin property of electrons and the effect of spin polarization and spin fluctuation on the electronics properties used in the nano electronics. In proposed model the spin properties such as the spin precession, spin diffusion, spin drift, and spin relaxation will be modeled to explain the electronic properties viz. current, resistance, relaxation time of electron and magnetic properties those have significant impact on the applications of nano structures in the electronic technology. Spin relaxation (how spins are created and disappears) and spin transport (how spin moves in metals and semiconductors) [5] are fundamentally important not only as basic physics value but also because of their demonstration value as phenomena in electronics technology. For E.g. GMR (Giant magneto resistive) Short term studies of spin transport properties of semiconductor, have made us understood that the spin properties of an electron can be very useful for the application of electron and nuclear spin to quantum computing and quantum information processing. Quantum algorithm given by Peter Shor of Bell laboratories, would factor large numbers into primes, which is very typical task in conventional computers, proves an immense possibility for spin devices well suited for quantum computing/ quantum information processing. Datta Das were [6, 7, 8] the first to give a scheme for the Spintronics device based on MOS technology. In their scheme they proposed that in FET like structures made of InAlAs and InGaAs, the gate electrode produces a field that forces the electron spin for precess. And the electron current is modulated by the degree of precession in electron spin introduced by the gate field. Mark Johnson at Naval research laboratory, in his all metal spin transistor model, inspired by the GMR (a tri layer structure), tried to control the current by an applied magnetic field. Thus device can act as Switch or Spin Valve. Due to its all metallic structure its miniaturization was very much possible at nanoscale level, but a disadvantage of not acting as amplifier is also associated with it. Thus it is very obvious that the Daas-Datta model is still suitable for nanoscale devices. But the major drawback comes that what about the spin transport at interfaces, particularly at ferromagnet/semiconductor interface. I will be trying to explore this part of the device characteristics. Recently, a nano FET design given by Igor Zutic proposes the use of magnetic field instead of electric field, producing same physical effect and also resulting in spin polarized current due to spin up and spin down effect. In all the above mentioned considerations , the study regarding the spin up and spin down of an electron on application of a magnetic field has been observed, but most of the authors have suggested mainly two states, i.e. spin up and spin down states only. The basic

equation for the electron spin based on the characteristic parameters of an electron such as electron mass, charge, drift velocity, Fermi velocity, electron relaxation time is presented by Gengyun Li The above discussion indicates that future electronic devices to be used in communication systems, computing systems, Instrumentation systems etc will inevitably be built using nanoelectronics, i.e., from devices and wires with feature sizes below twenty to thirty nanometers. The SIA roadmap [10] predicts that traditional silicon-based systems will have feature sizes of below 20 nm within the decade. There are also advances being made in building computing systems using new technologies, such as molecular electronics [11]. Successfully harnessing nanoelectronics requires a rethinking of the abstractions and models that are the basis of designing all such systems. While each technology has its own unique requirements, we show that new abstractions are necessary strictly because the feature sizes are nanoscale. The transistor has proven an extremely useful building block for digital logic. In addition to its switching properties, it can be used to construct logical gates which isolate their inputs from their outputs (I/O isolation) and it has gain which promotes noise immunity [12]. Some nanoelectronics technologies have no equivalent device. In addition, as dimensions scale down, the wires used to connect the devices become the dominant source of signal delay and power consumption [10].

3. LITERATURE REVIEW Spintronics (sometimes also referred to as magneto-electronics although we prefer the spintronics terminology because a magnetic field or the presence of a magnetic material is not necessarily essential for manipulating spins) is the emerging field [l3,14] of active control of carrier spin dynamics and transport in electronic materials (particularly, but not necessarily limited to, semiconductors). In some sense, existing technologies such as GMR-based memory devices and spin valves are elementary spintronic applications where the role of spin, however, is passive in dictating the size of the resistance (or tunneling current) depending on the spin direction controlled by local magnetic fields. Spintronics is projected to go beyond passive spin devices, and introduce applications (and possibly whole new technologies) based on the active control of spin dynamics. Such active control of spin dynamics is envisioned to lead to novel quantum-mechanical enabling technologies such as spin transistors, spin filters and modulators, new memory devices, and perhaps eventually quantum information processing and quantum computation. The possibility of monolithic integration on a single device of magnetic, optical, and electronic applications, where magnetic field and polarized light control spin dynamics, is an exciting new spintronic prospect for creating novel magneto-electro-optical technology. The two important physical principles underlying the current interest in spintronics are: The inherent quantum mechanical nature of spin as a dynamical variable (leading to the possibility of novel

Spintronics-A New Light in Nanoscale Devices


spintronic quantum devices not feasible within the present-day charge-based electronics) The inherently long relaxation or coherence time associated with spin states (compared with the ordinary momentum states). The fact that carrier spin in semiconductors can be easily manipulated noninvasively by using local magnetic fields, by applying external electric fields through controlled gates, and even by shining polarized light is an important impetus for developing spintronics applications. In spite of the great current interest in the basic principles and concepts of spintronics a large number of obstacles need to be overcome before one can manufacture spintronics applications. For example, a basic spintronics transport requirement is to produce and sustain large spin-polarized currents in electronic materials (semiconductors) for long times. This has not yet been accomplished. In fact, it has turned out to be problematic to introduce spin polarized carriers in any significant amount into semiconductor materials. Similarly, for quantum computation one requires significant and precisely controllable spin entanglement as well as single spin (i. e. a single Bohr magneton) manipulation using local magnetic fields. Currently there is no good idea about how to accomplish this. It is clear that a great deal of basic fundamental physics research will be needed before spintronics applications become a reality. A feasibility study of logic circuits utilizing spin waves for information transmission and processing is given by Wang et al. As an alternative approach to the transistor-based architecture, logic circuits with spin wave bus do not use charge as an information carrier. They have described the general concept of logic circuits with spin wave bus and illustrate its performance by numerical simulations based on available experimental data. Theoretical estimates and results of numerical simulations on signal attenuation, signal phase velocity, and the minimum spin wave energy required per bit in the spin bus are obtained. The transport parameters are compared with ones for conventional electronic transmission lines. The potential value of spin wave bus is, however, an interface between electronic circuits and integrated spintronics circuits. The logic circuits with spin wave bus allow us to provide wireless read-in and read-out. There is a growing interest in novel nanometer scale devices and architectures to address the shortcomings and drawbacks inherent to the traditional CMOS-based architecture [15]. Spintronics is one of the most prominent approaches in offering an alternative route to the traditional semiconductor electronics. Recent breakthroughs in the experimental study and control of the spin dynamics in semiconductor nanostructures [16-18] opens new possibilities for spin utilization in information processing. There are some intriguing ideas on possible spin-based logic devices [5-8] taking advantage on the additional degree of freedom provided by spin. It will be extremely beneficial in terms of power consumption to avoid the use of electric current in spin-based circuits. As a possible solution, it was proposed the single spin logic architecture, where the interconnection

between the spin-based cells is accomplished via exchange coupling [19].

4. SPIN RELAXATION AND DECOHERENCE The great promise of spintronic technology is based upon the fundamental ability of electron spins in electronic materials to preserve coherence for relatively long times. A typical electron remembers its initial spin orientation for a nanosecond. This time scale is indeed long when compared with the typical times femto seconds- for electron momentum relaxation. Perhaps a more revealing quantity than spin lifetime (which is usually called spin relaxation time T1 or spin decoherence time T2, depending on the context of the experiment) is the spin diffusion length Ls which measures how far electrons diffuse in a solid without losing spin coherence. The important fact that Ls is typically a micrometer makes spintronics a viable option for future micro- and nanoelectronics; any information encoded in electron spins will spread undisturbed throughout the device. Clearly, the longer the spin lifetime, the better and more reliable will be the spintronic devices. The study of spin relaxation is thus of great importance for spin-based technology (Review of the current understanding of spin relaxation processes in electronic systems is given in Ref. [2]). Initial measurements of spin lifetimes were conducted in metals like Na or Li by conduction electron spin resonance (CESR) technique [20]. The most important outcome of these experiments concerned the magnitude of T1 (nanoseconds) and its temperature behavior, T1 is constant at low temperatures (below, say, 50 K) and is increasing linearly with increasing temperature at elevated temperatures (above, say, 200 K). These two observations have shaped the theoretical understanding of the processes behind spin relaxation in metals. It is now generally accepted that electron spins in (nonmagnetic) metals decay by scattering off impurities (at low temperatures where T1 is constant) and phonons (at higher temperatures where T1 grows linearly with increasing temperature). The spin-flip probability of such processes is finite because of the finite spin-orbit interaction induced by either host ions or impurities (this is the so called Elliott-Yafet mechanism of spin relaxation [21]). Das-Datta has performed the first realistic calculation of T1 in a metal (aluminum) [22]. Their calculation not only provides the first direct proof of the validity of the Elliott-Yafet mechanism, but also shows that by engineering the band structure of metals (or semiconductors) it is possible to tailor spin relaxation (e.g., T1 can be changed by orders of magnitude by doping, straining, alloying, or changing dimensionality). An important development came with the discovery of spin injection by Johnson and Silsbee [23]. In the original experiment spin-polarized electrons were injected from a ferromagnetic electrode (Permalloy) into a nonmagnetic metal (aluminum), and the spin diffusion length was monitored. This method of measuring Ls (and thus spin lifetime) has a great potential since, unlike CESR, spin injection does not need an applied magnetic field which, in some cases, radically affects



spin relaxation processes. In addition to providing a useful method for measuring spin relaxation, the Johnson-Silsbee spin injection experiment brought about a whole new field of electronics: spintronics. Indeed, spin injection is the most natural way to integrate spin dynamics with electronic transport in electronic devices. There is no need for magnetic field or radiation to excite spin-polarized electrons. One only needs ferromagnetic electrodes. The last truly fundamental obstacle in the progress towards integrating the new spintronic with the traditional semiconductor technology has been recently overcome with the discovery of spin injection into a semiconductor [24, 25]. In addition to be able to create the population of spin-polarized carriers, it is required to find out a way to monitor and control the dynamics of spin processes in electronic materials. This quest has been pioneered by Kikkawa and Awschalom [26]. In a typical experiment spin-polarized electrons in a semiconductor like GaAs are excited by a circularly polarized light and then the electrons spin evolution is monitored at small (picoseconds) time intervals. Several new exciting results came from such experiments on semiconductors [27], a dramatic (two orders of magnitude) increase of electron spin lifetime with increased doping, unusually large (hundreds of micrometers) spin diffusion length, and the ability to optically control nuclear spin polarization (with electron spins acting as intermediaries between light and nuclear spins). The physics of injection of spin-polarized electric currents from magnetic materials into semiconductors through solid state interfaces is currently attracting much attention because of its fundamental interest and its potential relevance to the realization of spintronic semiconductor devices that utilize the electron's spin degree of freedom as well as its charge to store, process and transmit information.[27,29] High spin-injection efficiencies have been reported from magnetic to non-magnetic semiconductors at low temperatures.[30,31] Efficient injection of spin-polarized electrons from ferromagnetic metals (F) to semiconductors (S), although desirable for room temperature semiconductor spintronic devices, has not yet been achieved experimentally because the large mismatch between the resistivitys of metals and semiconductors is an obstacle to spin injection[32,35], and the interpretation of relevant spin-transport measurements has been controversial.[36,42] It was suggested in the seminal work of Datta and Das that spin injection may have particularly interesting ramifications in F/S/F double hetero junctions and that a unique transistor that relies on manipulation of the electron's spin instead of its charge may be feasible. Calculations of the spin-conductance of two-terminal F/S/F systems have been reported in the semi-classical diffusive and ballistic regimes. [32, 51, 52] However no theory of transport in F/S/F systems has treated the effects of spin injection and quantum coherence together within the same theoretical framework. It is proposed to explore the interplay between spin injection and quantum coherence in ballistic F/S/F systems theoretically within the Landauer formalism of transport. Spin-dependent

electron transmission at the interfaces, Rashba spin-orbit coupling [55, 58] and quantum interference are treated in a unified way. Quantum coherence can have unexpected implications for spin injection and that some intuitive concepts that have played a key role in the development of the field of spintronics but are founded on semi-classical physics no longer apply. For example, it has been shown that in the ballistic quantum coherent regime a pronounced spin valve effect (a change in resistance when the magnetization of a ferromagnetic electrode is reversed) can occur without any spin-polarization of the current owing through the semiconductor. This surprising phenomenon is an inherently quantum spin valve effect since it has no analog in the semi classical ballistic and diffusive transport regimes that have been considered previously. It will be essential to take account of the phenomena that is introduced in interpreting spin injection experiments in the quantum regime and in schemes for quantum computation that involve spin injection. While focusing on F/S/F hetero junctions, the quantum spin valve effect that has been introduced is general and should also occur in all-metal and all-semiconductor coherent quantum systems. Calculations of ballistic electron spin transport in ferromagnetic metal /semiconductor/ ferromagnetic metal structures in the coherent quantum regime. Results demonstrate that in the coherent quantum regime the relationship between spin transport and conductance measurements (a key experimental probe of spintronic phenomena) is qualitatively different than in the semi classical regime that has been studied experimentally to date: In the quantum regime a comparison of the conductance of a heterostructure with parallel and anti parallel magnetizations of magnetic contacts can no longer be regarded as an unequivocal indicator as to whether or not spin injection is taking place; it should be supplemented by other probes in studies of coherent spin injection. Moreover, predictions that coherent quantum system should exhibit an unexpected quantum spin-valve effect that occurs even in the absence of a net spin current owing through the device. These surprising conclusions do not rely on the semiconductor specific Rashba spin-orbit coupling that has been included in Hamiltonian model, but are general consequences of quantum interference.[57,58] They should apply to all-metal and all semiconductor systems as well as to the ferromagnetic metal /semiconductor/ferromagnetic metal hetero junctions that have been discussed. Differences between the coherent quantum regime and the semi-classical regime such as those that have been described should also occur if potential barriers are present at the interfaces between the ferromagnetic and non-magnetic parts of the structure. They should be taken into consideration in interpreting spin injection experiments in the quantum regime and in schemes for quantum computation that involve spin injection.

5. SPIN POLARIZED TRANSPORT The goals of employing both spin and charge transport (spin-polarized transport) in potential novel device applications



imposes intrinsic limitations on their design; they should consist of either heterostructures or inhomogeneous materials. While similar design constraints have been extensively investigated and well understood in the case of pure charge transport in conventional electronics, it is not clear how the spin degrees of freedom will behave in transport across interfaces in a heterostructure or through an inhomogeneous material. For example, by placing a semiconductor in contact with a nonmagnetic metal a Schottky barrier is formed whose properties will govern charge transport across the semiconductor/metal junction. Thus spin degree of freedom or spin wave analysis becomes relevant here. Currently there is no physical understanding for the corresponding spin-dependent Schottky barrier relevant for spin-polarized transport across interfaces. This is an important issue as some of the proposed spintronic devices [59] rely on the direct electrical spin injection from a ferromagnet into a semiconductor [23, 24]. The situation is further complicated by the possibility of spin-flip scattering at magnetically active interfaces. These considerations have to be included in assessing the feasibility of various spintronic devices because they imply that the degree of carrier spin-polarization can be strongly modified during transport across semiconductor/ferromagnet interfaces. Fabricating hybrid structures which would combine a semiconductor and a superconductor would allow investigating some of the aforementioned features and determining the degree of an extrinsically induced carrier spin polarization in the semiconductor. This could be realized by using Andreev reflection which governs transport properties at low applied bias, in this two-particle process an electron incident to the interface at the semiconductor side is accompanied by a second electron of the opposite spin. Both electrons are then transferred into the superconductor where they form a Cooper pair. The probability (measured by, e.g., conductance) of such processes strongly depends on the amount of spin polarization and the spin transparency of the interface [60, 61]. Material inhomogeneities can also act favorably and be tailored to give desired effect for spin-polarized transport. Such a p-n junction [62] could be used as a building block for novel spin transistor applications which would utilize both spin and charge degrees of freedom. The use of single electron or single-spin based devices inherently possesses fundamental drawbacks associated with scheme reliability and fabrication tolerance. To overcome these limitations, it was recently proposed the utilization of spin waves as a collective physical phenomenon for information transmission [63]. Spin wave is a collective oscillation of spins in an ordered spin lattice around the direction of magnetization. The phenomenon is similar to the lattice vibration, where atoms oscillate around its equilibrium position. Potentially, it is possible to use ferromagnetic films as spin conduit of wave propagation or referred to as Spin Wave Bus (SWB), where the information can be coded into a phase of the spin wave. The key advantage of the SWB is that information transmission

is accomplished without an electron transport. Besides, there are other significant advantages: Ability to use superposition of spin waves in the bus to achieve useful logic functionality; A number of spin waves with different frequencies can be simultaneously transmitted among a number of spin-based devices; The interaction between spin waves and the outer devices can be done in a wireless manner, via a magnetic field.

The excitation of spin waves can be done by the local magnetic field produced by micro- or nano-scale antenna, while the detection of the spin wave is via the inductive voltage produced by propagating spin waves. The concept of SWB is promising for implementation in different spin-based nano architectures [63]. A first working spin-wave based logic circuit has been recently experimentally demonstrated [64].

6. SPIN BASED QUANTUM COMPUTATION Among all possible spintronic devices, by far the most revolutionary is the proposed spin-based quantum computer (QC) which has the promise to vastly outperform classical computers in certain tasks such as factoring large numbers and searching large databases [65]. In QCs electron or nuclear spins are used as the basic building blocks. The spin-up and -down states of an electron or a nucleus provide the quantum bit (qubit), in analogy with 0 and 1 in a classical computer. However, as a quantum mechanical object, a spin can have not only up and down states, but also arbitrary super positions of these two states. These arbitrary super positions can very well be explained by spin wave theory. This inherent parallelism and other quantum mechanical properties such as entanglement and unitary evolution are the fundamental differences between QCs and classical computers. In the search for appropriate hardware for a QC, many proposals have been put forward [65]. The spin-based solid-state models have been given by [66]. One of the first proposals [67] suggests using quantum dot- trapped electron spins as qubits. Here a single electron is trapped in a gated horizontal GaAs quantum dot, with pulsed local magnetic field and inter-dot gate voltage governing the single-qubit and two-qubit operation. Another proposal replaces the quantum dot electrons by donor electrons [68]. Here varying the gyro magnetic ratio in a compositionally modulated SiGe alloy allows electron spin resonance for single qubit operations and exchange interaction for two-qubit operations. One important advantage of electron spins is their maneuverability, electrons are mobile and can be manipulated by both electric and magnetic fields. Aside from electron-spin-based QC models, there are also nuclear-spin-based proposals; such as the one using nuclear spins of phosphorus donor atoms in Si as qubits [69]. Here external gates are used to tune the nuclear magnetic resonance frequency, and donor electrons are the intermediaries between neighboring nuclear spins, introducing two-qubit operations through electron exchange interaction and hyperfine



interaction. The main advantage of nuclear spin qubits is their exceedingly long coherence time, which allows many coherent operations. Indeed, bulk solution Nh4R is one of the most advanced QC architectures [65], even though its ensemble-average character prompts some researchers to question [70] whether it really possesses all the quantum mechanical powers needed for tasks such as factoring. The major difficulties facing various QC models are achieving precise control over unitary evolutions and maintaining quantum mechanical coherence. While traditional electronic devices deal with large numbers of electrons at a time, in spin-based QCs one has to be able to precisely control spins of individual electrons. Furthermore, the electron spins need to be essentially isolated from their environment so that their dynamics is governed by quantum mechanics. If this isolation is imperfect, the spins' quantum information will leak into their environment, and the dynamics of the spins will become irreversible and classical, so that the QC operation will be disrupted. Spin decoherence has many different channels such as spin interaction with boundaries, impurities, and host nuclei, even with external controls. For example: one common approach to tune the exchange interaction between electrons or electrons and nuclei is to use electrical gates which are connected through a transmission line to the outside. External noise such as Johnson-Nyquist noise can thus cause fluctuations in the gate voltage, which in turn cause errors in the exchange. The rate of this error can be as large as a few MHz [66]; which corresponds to the limits of the currently available error correction schemes. Another error during exchange is caused by inhomogeneous magnetic fields [71]. In essence, the different Zeeman couplings of two neighbor electrons causes mixing of the two-electron singlet and triplet states, therefore preventing the electron spin states from complete disentanglement for swap. This error is proportional to the square of the inhomogeneity [71], and can usually be corrected. Indeed, there is an existing scheme which can circumvent this error [72]. Furthermore, it has been proposed [73] that one can utilize certain decoherence-free subspace of four quantum dots as qubits: relying completely on exchange for all operations and eliminating the use of any external magnetic field. Such a scheme is more difficult to realize experimentally, but it does provide the advantage of smaller decoherence because of fewer noise channels. Although, the concept of designing of a quantum level electronic device is simple, its realization is not. Two issues motivate quantum nano scaling devices: - Quantum mechanical concepts must be applied to solve intractable (NP-complete) digital logic problems. - From a device (Digital device) miniaturization point of view, the size limit of a bit of information is important. Recently, this issue has attracted increased attention, due to the current development of nanotechnology and the design problems of semiconductor and metal devices that are approaching the quantum size limit. Consequently, the idea of quantum size device computing, in which the elements that carry the

information are atoms, has attracted the attention of many scientists. Quantum dot cellular automata (QCA) is one of the emerging nanotechnologies that exhibit a small feature size, high clock frequency, and ultra low-power consumption [74,75]. QCA provides an alternative way of computation, in which the logic states (0 and 1) are defined by the positions of the electrons. Due to the significant error rates in nanoscale manufacturing and nanotechnologies, including the QCA, there is a critical need to maintain extremely low device error rates [76]. In the manufacturing of QCA, defects can occur in the Synthesis and deposition phase. However, defects are more likely to take place during the deposition phase [77]. QCA devices are also prone to transient faults caused by thermodynamic effects, radiation, and other effects, as the energy difference between the ground and the excited state is small [82]. The concurrent testing of faults in QCA and QCA-based sequential circuits has not been addressed in the literature, till now. Thus addressing the method for designing a novel architecture for concurrent testing of latches/Flip flops for molecular QCA using reversible logic is an important issue. Reversible computation in a system can be performed only when the system comprises of reversible gates. Reversible circuits do not lose information, and can generate unique output vector from each input vector and vice versa (i.e., there is a one-to-one mapping between the input and the output vectors). Landauer has shown that for irreversible logic computations, each bit of information lost generates kT ln 2 joules of heat energy, where k is Boltzmanns constant and T the absolute temperature at which the computation is performed [78]. Bennett showed that kT ln2 energy dissipation would not occur if a computation is carried out in a reversible way [79]. The testing properties of reversible logic are utilized in a 1-D array of molecular QCA in [76], which is considered to be the earliest effort toward applying reversible logic in molecular QCA. In the manufacturing of QCA, defects can occur during the synthesis and the deposition phase, more likely during the deposition phase [77]. Researchers assume that QCA cells have no manufacturing defects, and in metal, QCA faults occur due to cell misplacement. These defects can be characterized as cell displacement, cell misalignment, and cell omission [80]. Researchers have shown that molecular QCA cells are more susceptible to missing and additional QCA cell defects [81]. The additional cell defect is because of the deposition of an additional cell on the substrate, and the missing cell defect is due to the missing of a particular cell. The testing of QCA was addressed for the first time in a seminal work reported in [77], where the defect characterization of QCA devices was investigated, and it was shown how the testing of QCA was different from conventional CMOS. The modeling of QCA defects at molecular level was done for combinational circuits in [81]. Fault characterization was done for single missing/additional cell defect on different QCA devices such as MV, INV, fan-out, crosswire, and L-



shape wire. The test generation framework for QCA was presented in [80]. It was shown that additional test vectors can be generated for detecting QCA defects that remain undetected by the stuck-at fault model. Bridging fault on QCA wires was also addressed. The first approach will be for the concurrent detection of single missing/additional cell defect model or unidirectional faults. Some model has been presented in this respect by [83], but there are strong possibilities of improvements in this model. In unidirectional faults, there are only either 1 0 or 0 1 faults, and both types of faults cannot occur simultaneously. For unidirectional faults, comparing the number of 1s in the inputs with the number of 1s in the outputs could be used for fault detection. But the proposed approach is not going to be suitable for the detection of bidirectional multiple faults, for example, if we are expecting the outputs to be 011, but due to bidirectional faults, the outputs are flipped to 101, the parity of the outputs is still the same. One of the possible solutions to concurrently detect multiple faults is to regenerate the inputs at the outputs by using reversible logic. Since reversible logic has one to-one mapping between the input and output vectors, it is possible to determine the inputs from the outputs. Thus, the regenerated inputs can be checked with the original inputs for the detection of multiple faults For example, in the Fredkin gate, if its outputs P, Q, and R are fed to another Fredkin gate cascaded in series, the original inputs will be regenerated at the outputs of the second Fredkin gate. This property of Fredkin gate can be beneficial for the detection of multiple faults, as multiple faults or single fault in either first or second Fredkin gate, or in both will result in incorrect values of regenerated inputs. Therefore, the original input vector can be compared with the regenerated input vector for detection of faults. One of the areas for my research work will be to extend future efforts toward multiple fault detection in QCA computing. Taking the above mentioned approach of using latches for the implementation of concurrently testable sequential circuits for molecular QCA based on conservative reversible logic we will carry our further work, in the field of molecular QCA nanocomputing. A probabilistic-based design methodology for designing nano-scale computer architectures based on Markov Random Fields (MRF) has been given by [84]. The MRF can express arbitrary logic circuits and logic operation is achieved by maximizing the probability of state configurations in the logic network. Maximizing state probability is equivalent to minimizing a form of energy that depends on neighboring nodes in the network. [84]

7. CONCLUSION AND FUTURE SCOPE The phenomenon of spin transportation is of immense technological relevant for the development of nanoscale devices and electronics of nano devices. Gradually these models are becoming matured enough to demand an attention in engineering and technological context. The development is however embedded in the ongoing research on the study of the

Physics of the electron behavior under spin and developing new models for the quantum information processing. There is still need for adequate theoretical reasoning/modeling of the phenomena. Also in order to make the Spin applications more viable further improvement in spin relaxation (how spins are created and disappears) and spin transport (How spin moves in metal/Semiconductor interface?) is needed. The present work is guided to reconcile the effect of spin on quantum or nanoelectronics devices. Few probabilistic models with Quantum Dot Cellular automata (QCA) are given by [83, 84]. The outcome of the present work is expected to be significant in the sense that it will explore several aspects of the Quantum computing getting affected by the inherent physical properties shown by the electron under spin. The results on the proposed philosophies will be useful for the scientists and researchers to fully develop a quantum level algorithm for the nanoscale devices and will serve as a guide to theoretical and experimental workers for further advancements. The emphasis is based on the fact that the nanoscale devices under the spin effect have different characteristic parameters and in this proposal an investigation will be carried out on its effect on nanoscale spin-electronics. .

REFERENCES [1]. The Royal Society, Nano science and Nanotechnologies:

Opportunities and Uncertainties, London, U.K., 2004. [2]. Spintronics: A new paradigm for electronics for the new

millennium, Stuart A. Wolf & Daryl Treger, IEEE transactions on Magnetics,Vol.36,No.5,Sept.2000

[3]. Seth Copen Goldstein, Challenges and Opportunities of Nanoelectronics, School of Computer Science, Carnegie Mellon University, Pittsburg.

[4]. Bennett, C. H., and D. P. DiVincenzo. 2000. Quantum information and computation. Nature (404):267.

[5]. Das Sharma, Spintronics, American Scientist,Volume89,November-december 2001

[6]. Das Sarma, S., et al. 2000. Theoretical perspectives on spintronics and spin-polarized transport. IEEE Transactions on Magnetics 36:2821.

[7]. Das Sarma, S., et al. 2001. Spin electronics and spin computation. Solid State Communications 119:207.

[8]. Datta, S., and B. Das. 1990. Electronic analog of the electro optic modulator. Applied Physics Letters 56:665.

[9]. Fabian, J., and S. Das Sarma. 1999. Spin relaxation of conduction electrons. Journal of Vacuum Science and Technology B17:1780.

[10]. Sematech. International Technology Roadmap for Semiconductors, 2001.

[11]. M. Butts, A. DeHon, and S. Goldstein. ICCAD-2002, Nov. 2002.

[12]. Robert W. Keyes. Science, 230(4722):138144, October 1985.

[13]. S. Das Sarma, J. Fabian, X. Hu, and I. Zutid, LANL Preprint cond-mat/0002256; cond-mat/9912040; G. A prinz, science282



[14]. S. Das Sarma, J. Fabian, X. Hu, and I. Zutid,issues ,concepts and challenges in Spintronics, Dept. of Physics, University of Maryland

[15]. ITRS, Semiconductor Industry Association, 2005. http://www.itrs.net/Common/2005ITRS/Home2005.htm.

[16]. Ohno, H., Making nonmagnetic semiconductors ferromagnetic. Science, 1998. 281(5379): p. 951-6.

[17]. Chiba, D., et al., Control of magnetization reversal in ferromagnetic semiconductors by electrical means. Journal of Physics-Condensed Matter, 2004. 16(48): p. S5693-6.

[18]. Kato, Y.K., et al., Observation of the spin Hall Effect in semiconductors. Science, 2004. 306(5703): p. 1910-13.

[19]. Bandyopadhyay, S., B. Das, and A.E. Miller, Supercomputing with spin-polarized single electrons in a quantum coupled architecture. Nanotechnology, 1994. 5(2): p. 113-33.

[20]. G. Feher and A. Kip, Phys. Rev. 98, p. 337 (1955). [21]. R. J. Elliott, Phys. Rev. 96, p. 266 (1954); Y. Yafet, in

Solzd State Physacs, edited by F. Seitz and D. Turnbull [22]. J. Fabian and S. Das Sarma, Phys. Rev. Lett. 81, p. 5624

(1998); 83, p. 1211 (1999); J. Appl. Phys. 85, p. 5057 [23]. M. Johnson and R. H. Silsbee, Phys. Rev. Lett. 55, p.

1790 (1985). [23-A] P. R. Hammar, B. R. Bennett, Ivl. Y. Yang, and Ivl. Johnson, Phys. Rev. Lett. 83, p. 203 (1999).

[24]. R. Fiederling et al., Nature 402, p. 787 (1999); Y. Ohno et al., Nature 402, p. 790 (1999).

[25]. D. D. Awschalom and J. M. Kikkawa, Physics Today 52, p. 33 (1999). Prinz, Science 282, p. 1660 (1998); J. Gregg. et al., J. Magn. Magn. Mater. 175, p. 1 (1997). (Academic, Kew York, 1963), Vol. 14. (1999);

[26]. M. Kikkawa, I. P. Smorchkova, K. Samarth, and D. D. Awschalom, Science 277, p. 1284 (1997); J. M. Kikkawa

[27]. G. A. Prinz, Physics Today 45, 58 (1995); Science 282, 1660 (1998)

[28]. J. M. Kikkawa, Physics Today 52, 33 (1999). [29]. S. Datta and B. Das, Appl. Phys. Lett. 56, 665, (1990). [30]. R. Fiederling, M. Keim, G. Reischer, W. Ossau, G.

Schmidt, A. Waag, and L.W. Molenkamp, Nature (London) 402, 787 (1999).

[31]. Y. Ohno, D. K. Young, B. Beschoten, F. Matsukura, H. Ohno and D. D. Awschalom, Nature 402, 790 (1999).

[32]. G. Schmidt, D. Ferrand, L. W. Molenkamp, A. T. Filip, and B. J. vanWees, Phys. Rev. B 62, R4790 (2000).

[33]. M. Johnson and R. H. Silsbee, Phys. Rev. B 35, 4959 (1987).

[34]. P. C. van Son, H. van Kempen, and P. Wyder, Phys. Rev. Lett. 58, 2271 (1987).

[35]. T. Valet and A. Fert, Phys. Rev. B 48, 7099 (1993). [36]. W. Y. Lee, S. Gardelis, B.-C. Choi, Y. B. Xu, C. G.

Smith, C. H. W. Barnes, D. A. Ritchie, E. H. Lin eld, and J. A. C. Bland, J. Appl. Phys. 85, 6682 (1999).

[37]. P. R. Hammar, B. R. Bennett, M. J. Yang, M. Johnson, Phys. Rev. Lett. 83, 203 (1999).

[38]. F. G. Monzon, H. X. Tang, M. L. Roukes, Phys. Rev. Lett. 84, 5022 (2000).

[39]. B. J. van Wees, Phys. Rev. Lett. 84, 5023 (2000). [40]. P. R. Hammar, B. R. Bennett, M. J. Yang, M. Johnson,

Phys. Rev. Lett. 84, 5024(2000). [41]. A. T. Filip, B. H. Hoving, F. J. Jedema, B. J. van Wees,

B. Dutta, and S. Borghs, Phys. Rev. B 62, 9996 (2000). [42]. R. H. Silsbee, Phys. rev. B 63, 155305, (2001). [43]. E. I. Rashba, Phys. Rev. B 62, 16267 (2000). [44]. G. Kirczenow, Phys. Rev. B 63, 54422 (2001). [45]. S.F. Alvarado, P. Renaud, Phys. Rev. Lett. 68, 1387

(1992) [46]. V. P. LaBella, D. W. Bullock, Z. Ding, C. Emergy, A.

Venkatesen, W. F. Oliver, G. J.Salamo, P. M. Thibado, and M. Mortazavi, Science 292, 1518 (2001).

[47]. H.J. Zhu, M. Ramsteiner, H. Kostial, M.Wassermeier, H.P. Schonherr, K.H. Ploog, Phys. Rev. Lett. 87, 16601 (2001)

[48]. C.-M. Hu, J. Nitta, A. Jansen, J. B. Hansen, and H. Takayanagi, Phys. Rev. B 63, 125333 (2001).

[49]. S. Gardelis et al. Phys. Rev. B 60, 7764 (1999). [50]. P. R. Hammar and M. Johnson, Phys. Rev. Lett. 88,

066806 (2002). [51]. M. Johnson, Phys. Rev. B 58, 9635 (1998). M. Johnson

and R. H. Silsbee, Phys. Rev. B 37, 5326 (1988). [52]. D. Grundler, Phys. Rev. B 63, 161307, (2001). [53]. We note that the magnetic susceptibility of all-metal

ferromagnetic - non-magnetic - ferromagnetic junctions has been studied theoretically in the quantum coherent regime by S. T. Chui and J. R. Cullen, Phys. Rev. Lett. 74, 2118 1995.

[54]. R. Landauer, IBM J. Res. Dev. 1, 223 (1957), R. Landauer, Phys. Lett. 85A, 91 (1981).

[55]. Y. A. Bychkov E. I. Rashba, J. Phys, C 17, 6039 (1984). [56]. J. Luo et al. Phys. Rev. B 41, 7685 (1990). [57]. A. C. H. Rowe et al. Phys. Rev. B 63, 201307 (2001). [58]. The largest experimental value reported to date of the

Rashba spin-orbit coupling strength R in InAs-based heterojunctions is R = 31012eV m, which corresponds to a Rashba wave vector of kR = 1:5 105cm1 = 1:5ko; see Y. Sato, T. Kita, S. Gozu, and S. Yamada, J. Appl. Phys. 89, 8017 2001 and J. Nitta , T. Akasaki, H. Takayanagi, and T. Enoki, Phys. Rev. Lett. 78, 1335 1997.

[59]. S. Datta and B. Das, Appl. Phys. Lett. 56, p. 665 (1990). [60]. I. Zutid and S. Das Sarma, Phys. Rev. B 60, p. R16322

(1999). [61]. I. ZutiC and 0. T. Valls, Phys. Rev. B 60, p. 6320 (1999);

61 1555 (2000). [62]. I. Zutid, J. Fabian, and S. Das Sarma, unpublished. [63]. Khitun, A. and K. Wang, Nano scale computational

architectures with Spin Wave Bus. Superlattices & Microstructures, 2005. 38: p. 184 200.

[64]. Kostylev, M.P., et al., Spin-wave logical gates. Applied Physics Letters, 2005. 87(15): p. 153501-1-3.



[65]. A. Steane, Rep. Prog. Phys. 61, p. 117 (1996). [66]. X. Hu and S. Das Sarma, Phys. Rev. A 61, p. 2301

(2000). [67]. D. Loss and D. P. DiVincenzo, Phys. Rev. A 57, p. 120

(1998). [68]. R. Vrijen et al., LANL Preprint quant-ph/9905096. [69]. B. E. Kane, Nature 393, p. 133 (1998). [70]. S. L. Braunstein et al., Phys. Rev. Lett. 83, p. 1054

(1999). [71]. X. Hu, R. De Sousa, and S. Das Sarma, LANL Preprint

cond-mat/0004459. [72]. G. Burkard, D. Loss, D. P. DiVincenzo, and J. A. Smolin,

Phys. Rev. B 60, p. 11 404 (1999). [73]. J. Kempe, D. Bacon, D. A. Lidar, and K. B. Whaley,

LANL Preprint quant-ph/0004064. and D. D. Awschalom, Nature 397, p. 139 (1999); Science 287, p. 473 (2000).

[74]. P. Tougaw and C. Lent, J. Appl. Phys., vol. 75, no. 3, pp. 18181825, 1994.

[75]. A. O. Orlov, I. Amlani, G. H. Bernstein, C. S. Lent, and G. L. Snider, Science, vol. 277, pp. 928930, 1997.

[76]. X. Ma, J. Huang, C. Metra, and F. Lombardi, Springer J. Electron. Testing, vol. 24, no. 13, pp. 297311, Jun. 2008.

[77]. M. B. Tahoori, J. Huang, M. Momenzadeh, and F. Lombardi, IEEE Trans. Nanotechnology., vol. 3, no. 4, pp. 432442, Dec. 2004.

[78]. R. Landauer, IBM J. Res. Dev., vol. 5, pp. 183191, 1961. [79]. C. H. Bennett, IBM J. Res. Dev., vol. 17, pp. 525532,

Nov. 1973. [80]. P. Gupta, N. K. Jha, and L. Lingappan, IEEE Trans. Very

Large Scale Integr. (VLSI) Syst., vol. 15, no. 1, pp. 2436, Jan. 2007.

[81]. M. Momenzadeh, M. Ottavi, and F. Lombardi, Proc. DFT VLSI Syst., Monterey, CA, Oct. 2005, pp. 208216.

[82]. R. K. Kummamuru, A. O. Orlov, R. Ramasubramaniam, C. S. Lent, G. H. Bernstein, and G. L. Snider, IEEE Trans. Electron. Devices, vol. 50, no. 9, pp. 19061913, Sep. 2003.

[83]. Himanshu Thapliyal, Student Member, IEEE, and Nagarajan Ranganathan, Fellow, IEEE IEEE TRANSACTIONS ON NANOTECHNOLOGY, VOL. 9, NO. 1, JANUARY 2010

[84]. J chen, J. Mundy, Y. Bai, S.-M.C.Chan, P.Petrica and R.I. Bahar, Division of Engineering, Brown University, RI02912, USA

[85]. P. Gupta, N. K. Jha, and L. Lingappan, IEEE Trans. Very Large Scale Integr. (VLSI) Syst., vol. 15, no. 1, pp. 2436, Jan. 2007.

[86]. M. B. Tahoori, J. Huang, M. Momenzadeh, and F. Lombardi, IEEE Trans. Nanotechnology., vol. 3, no. 4, pp. 432442, Dec. 2004

226-libre

Documents