AVS2010 Session MI-WeA: Spintronic Devices and Proximity Effects

The use of a spin polarized current to rotate the magnetization direction of a ferromagnetic has a number of important implications for novel magnetic devices. Spin torque can switch the resistivity of a magnetic tunnel junction, create spin waves over a very wide range of frequencies, and move magnetic domain walls. I will describe some of the recent advances in Spin Torque Magnetic Random Access Memory (STT-RAM), spintronic nano-oscillators and flux shuttle memory. In addition I will describe a very novel logic array based on electrical control of magnetism called a Reconfigurable Array of magnetic Automata which uses the ferroelastic strain on a magnetic nanopillar to rotate the direction of magnetization. I will show how this can be used to perform logic at very low power approaching zeptojoules per switch..

The magnetic doping of the group IV semiconductors Si and Ge with Mn is coveted as valuable component in novel spintronics devices. Their magnetic doping is hampered by the low solubility of Mn in both elements, and the formation of silicide and germanide phases which modify or suppress the magnetic response of the material. We therefore investigate Mn-nanostructures which are synthesized on the Si(100)(2x1) surface and then capped with a layer of Si or Ge to protect the integrity of the nanomaterial. The Mn-nanostructure is therefore embedded as a delta-doped layer within a group IV semiconductor matrix.

The first section of our presentation focuses on nanostructure synthesis, in particular the growth of monoatomic Mn-wires on the Si(100)(2x1) surface, where the dimer row structure guides the wire formation. A scanning tunneling microscopic (STM) study establishes the relation between the nanostructure type and the quality of the substrate, which is expressed in the concentration of defects prior to Mn deposition. An increasing defect concentration (1.0 % to 18.2%) leads to a suppression of wire formation and the growth of ultrasmall Mn-clusters is favored, which allows for an excellent control of nanostructure type. Surface structuring through the presence of dimer vacancy lines, or narrow terraces offer an additional route to control Mn-wire formation. A Monte-Carlo based model is introduced to describe the surface processes which lead to wire-formation.

The second section of our presentation assesses the stability of the nanostructures during the growth of the Si or Ge caps. The capping of nanostructures is indispensible for preservation of the nanostructures’ unique capabilities in devices, and is critical for magnetic measurements such as VSM (vibrating sample magnetometry) and XMCD (X-ray magnetic circular dichroism, Advanced Light Source, Lawrence Berkeley National Laboratory). An STM observation of the growth process of the Si and Ge caps shows that the Mn-nanostructures are indeed preserved, and we are therefore able to assess the magnetic properties of Mn-wires and ultrasmall clusters.

The third section of our presentation is devoted to the discussion of the magnetic properties of the Mn-wire and Mn-cluster structures. The role of magnetic anisotropy, the ionic and metallic contributions from the Mn-nanostructure, and the nature of the magnetic coupling within the respective nanostructure, and with the cap-material will be discussed.

The support from NSF-Chemistry and DOE are gratefully acknowledged.

Exchange bias is the interaction at the interface between an antiferromagnetic material and a ferromagnetic thin film or nanoparticle which causes the center of the ferromagnetic hysteresis loop to shift away from zero field, effectively resulting in a unidirectional anisotropy. Despite the fact that this effect was discovered approximately fifty years ago, and that it is used in magnetic sensors found in hard drives, the fundamental mechanism responsible for this interaction was poorly understood until recently. Important advances using ideal antiferromagnets (antiferromagnets with well understood and relatively simple properties) have been made during the past few years to assess or validate theories that explain exchange bias. My group has used transition metal difluoride epitaxial thin films, such as Fe_xZn_1-xF₂ and Fe_xNi_1-xF₂, which allow us to vary the magnetic disorder and anisotropy of the antiferromagnet in a controlled manner. Traditional magnetometry techniques, as well as more sophisticated experiments sensitive to the depth profile of the magnetization, such as x-ray magnetic circular dichroism (XMCD) and polarized neutron reflectivity (PNR), have allowed us to understand the interface processes responsible for the effect. I will discuss the important results from these experiments, including 1) the effects of short range order at the surface of the antiferromagnet above its Néel temperature; 2) the observation of pinned and unpinned magnetic moments at the ferromagnet/antiferromagnet interface; and 3) the effects of the magnetic anisotropy of the antiferromagnet on the temperature dependence of the exchange bias and the possibility of reversing the effect at low temperatures.

This work was supported by the National Science Foundation.

In this talk I will introduce the general concepts of spintronics and highlight some of the key experimental results that have catalyzed the emerging research area known as “organic spintronics”. The confluence of spin-electronics, which combines transition metal ferromagnets and inorganic semiconductors, and molecular-electronics, which combines organic semiconductors and non-magnetic metals, has lead to the evolution of organic spintronics. Furthermore, the development of organic light emitting diodes and organic photovoltaic cells has proven that molecular thin films can reliably function as the active layer in commercial products. Consequently it often envisioned that spin dependent conduction in organic-based semiconductors, rather than traditional charge transport, can be manipulated and detected to fabricate low-power, non-volatile, multifunctional devices because electron spin coupling to orbital angular momentum and nuclear spin is weak. Recent experimental evidence is beginning to demonstrate this view is too simplistic and hyperfine coupling is very important in the hopping transport mechanism characteristic of disordered organic semiconductors.

The focus of this talk will be on the phenomena of magnetoresistance; specifically in vertical device structures with conducting electrodes used to separate an organic semiconductor layer(s). In many laboratories around the world such devices are observed to change resistance when placed in an external magnetic field. If both electrodes are magnetic and the change in the resistance corresponds to the coercive fields of the electrodes, then the magnetoresistance can be interpreted as spin-polarized injection, transport, and ejection of carriers from one ferromagnetic layer, through the non-magnetic spacer layer, and into the second ferromagnetic layer, respectively. However, when neither of the electrodes is magnetic the magnetoresistance must arise some other phenomena and there are several competing theories to describe the effects. The bulk of this talk will be devoted to reviewing case studies from the former class of devices, since there are well-accepted criteria to support the interpretation of device magnetoresistance when using magnetic electrodes. In addition I will cite examples where the interfaces of the devices have been examined by surface sensitive spectroscopy/microscopy techniques and correlated to device performance. Finally I will conclude by recommending some new experiments that could reveal more knowledge of the fundamental spintronics effects in molecular materials and suggest some possible applications.