American Institute of Physics
SEARCH AIP
home contact us sitemap
Physics News Update
Number 595, June 26, 2002 by Phil Schewe, James Riordon, and Ben Stein

Ballistic Magnetoresistance

Ballistic magnetoresistance (BMR) is yet another way in which spin orientation, encoding information on a storage medium such as a hard drive, can modify electrical resistance in a nearby circuit, thereby accomplishing the sensing of that orientation. The sensitive part of the circuit might consist of sandwiches of alternating magnetic and nonmagnetic layers (giant magnetoresistance, GMR; and tunnel junctions, TMR) or might have no magnetic materials at all (extraordinary magnetoresistance, or EMR; see Update 589).

In ballistic magnetoresistance, the sensor size is reduced to just a cluster of ferromagnetic atoms, joined together by, say, two lead wires. "Ballistic" means that the sensor is smaller than the typical scattering path length for the electron, which therefore moves in a straight trajectory. This means that the scattering the electron suffers will be owing to magnetic effects and not to general scattering from atoms in the sensor itself, making the readout process very sensitive.

If the electrons flowing in the circuit have been spin-polarized then when they flow through the sensor they will scatter more or less (meaning larger or lesser resistance) depending on the magnetization state within the sliver of atoms constituting the contact, and on the faint force exerted by the tiny magnetic storage domain being read out by the sensor (see figure).

In a new BMR experiment conducted at SUNY-Buffalo (Harsh Deep Chopra, hchopra@eng.buffalo.edu, 716-645-2593, x2310, and Susan Hua), the size of the sensor is so small (only nm in width and length) that the electron, on its way through the contact, will have less of a chance to accommodate itself to the spin regime of the second electrode (if it is different from that of the first electrode) and will consequently scatter more prominently, translating into a large magnetoresistance effect.

In the Buffalo experiment a remarkably large magnetoresistance effect (change in resistance) of 3150% is observed at room temperature (compared to 100% for GMR, and 1300% for EMR, or 1300% for room temperature "colossal magnetoresistance," or CMR). This represents the highest room temperature spin dependent MR effect ever observed for a spintronic device.

And this was accomplished in a very weak magnetic fields (less than 160 gauss), which means that as the size of the domain being read out shrinks (as more and more data is crammed onto smaller spaces on the recording medium) the signal will continue to be strongly felt in the sensor. Since the size of the sensor is only a sliver of atoms, bits can be reduced to comparable size, which could lead to storage capacities approaching terabits/sq in. (Chopra and Hua, Physical Review B, rapid communications, 1 July).

Nanospintronics: A Single-Spin Transistor

Spintronics is a relatively new field in which the electron's spin, not just its charge, can be exploited in devices and circuits. The ultimate spintronics degree of control would come from controlling a circuit at the level of a single spin.

Physicists at the Institute for Microstructural Sciences (Ottawa) are the first to create a prototype of a single-spin transistor, which consists of a quantum dot connected to spin-polarized leads.

A quantum dot is an artificial atom in which electrons are confined spatially by an electrostatic potential much in the way that a nucleus localizes electrons in an atom. The dot can be emptied and then electrons added one at a time to create a "hydrogen" dot (with one electron onboard), "helium" dot (two electrons), "iron" dot, and so forth.

The spins of the electrons in the transistor are not random but depend on the number of electrons in the electron puddle, and on the applied magnetic field. Most importantly, by connecting the dot to spin-polarized reservoirs, one can insist that the electrons flowing in or out have their spins aligned up or down, and this criterion (is the electron's spin up or down?) can be used as a gate to allow a high or low current to flow through the dot. Hence the spin state of the dot is encoded in the difficulty of adding the extra electron.

In this way the group was able to "read" the spin properties of the dot. They could also in a sense "write" (i.e., change the spin state of the dot controllably) by either adding an electron or by tuning the magnetic field.

Such a unique combination of control at the single charge and single spin may play a role in the future solid state form of quantum computing where the unit of quantum manipulation, the qubit might consist of specially prepared spin states. (Ciorga et al., Physical Review Letters, 24 June 2002; contact Pawel Hawrylak, pawel.hawrylak@nrc.ca or Andrew Sachrajda, andy.sachrajda@nrc.ca.)

An Optically Pumped Nanocrystal Quantum Dot Laser

An optically pumped nanocrystal quantum dot laser has been demonstrated by a group at MIT. Lasers come in many sizes and can be made from a variety of resonant cavities and active laser materials.

Generally, increasing confinement enforces an increasing quantization in the energy of electrons. Therefore quantum dots, essentially zero-dimensional bits of material, will (once excited) re-emit light at nearly a single wavelength.

Quantum dots are therefore a good starting point for producing laser light. Some existing quantum dot lasers employ dots made epitaxially: the atoms in the dots are laid down meticulously using beams of atoms or molecules.

In the MIT laser the gain medium consists of nm-sized particles of CdSe coated with a layer of organic molecules and then immersed in a glassy film. The medium sits in a waveguide atop a grating. The fabrication advantage in this case derives from the fact that one uses simple solution processing rather than the more exacting technique of epitaxy usually needed for semiconductors.

Furthermore, the color of the output laser light can be varied by changing the size of the CdSe particles, the grating spacing, or the refractive index of the waveguide, giving great flexibility to the design and application of the laser. (Eisler et al., Applied Physics Letters, 17 June 2002; contact Moungi Bawendi, MIT, 617-253-9796, mgb@mit.edu.)